SUNLIGHT AND LUNG CANCER SURVIVAL IN PATIENTS DIAGNOSED AND REFERRED TO THE BC CANCER AGENCY BETWEEN 1980 AND 1989 by ASEF GHOLAMABBAS JAVAN M.D. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Interdisciplinary Oncology) THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER) April 2011 © Asef Gholamabbas Javan, 2011 ii Abstract Background Lung cancer is the deadliest form of cancer in the world. Recent reports indicate that patients who are diagnosed and treated, especially with surgery, during sunny periods have better survival. Research questions (objectives) 1) Is there an association between a lung cancer patient’s survival and the amount of sunshine around the time of their diagnosis? 2) Is there an association between a lung cancer patient’s survival and the amount of sunshine around the time of their treatment? Design A population based retrospective study Materials and methods We examined BC patients diagnosed with lung cancer during 1980-1989 and referred to the BCCA for treatment. Data from BCCA computerized medical charts were linked with observations from the BC weather service. Survival analyses were adjusted for patient factors (i.e., age and gender), disease characteristics (i.e., stage and cell histology) and treatment information (i.e., surgery, radiation therapy and chemotherapy use).Each analysis consisted of a univariate and multivariate test of the sunshine variable’s association with the patients’ survival. The univariate analysis was a Kaplan-Meier plot with log-rank test. The multivariate analysis iii was the hazard ratio from a Cox proportional-hazards model adjusting for patient age, disease stage, treatment, tumor laterality and tumor presenting site. Results This study considered 9302 patient records. Our results indicate that the monthly mean vitamin D sunshine (UV) index might be a prognostic indicator for this disease, especially among patients with non-small cell lung cancer. It showed that sun exposure and subsequent vitamin D production increase the survival of some of the patient who diagnosed with lung cancer between the period of 1980 and 1989 in BC. Conclusions 90% of active vitamin D in our body is triggered by UV exposure. Different studies have shown that vitamin D has inhibitory effects in the development of cancers such as colon, breast and pancreas. Although the monthly mean vitamin D sunshine (UV) index is an indirect measure of vitamin D exposure, it might affect the prognosis for some lung cancer patients. iv Table of Contents Abstract ..................................................................................................................... ii Table of Contents ..................................................................................................... iv List of Tables .......................................................................................................... vii List of Figures ........................................................................................................ xiii List of graphs ......................................................................................................... xiv List of Abbreviations .............................................................................................xxv Acknowledgments ............................................................................................... xxvii Dedication .......................................................................................................... xxviii 1. Introduction ............................................................................................................ 1 1.1 Thesis overview ................................................................................................... 1 1.2 Lung cancer .......................................................................................................... 2 1.2.1 Descriptive epidemiology ............................................................................................................ 2 1.2.2 Risk factors .................................................................................................................................. 3 1.2.3 Pathology ..................................................................................................................................... 4 1.2.4 Diagnosis (signs and symptoms) .................................................................................................. 5 1.2.5 Staging ......................................................................................................................................... 6 1.2.6 Treatment ..................................................................................................................................... 9 1.2.7 Prognosis .................................................................................................................................... 11 1.3 Sunlight and vitamin D ......................................................................................12 1.4 Sunlight, vitamin D and cancer prognosis .........................................................13 1.5 Hypothesis ..........................................................................................................14 2. Materials and methods .........................................................................................15 2.1 Proposed variables and methods ........................................................................15 2.1.1 Lung cancer patient information ................................................................................................ 15 2.1.2 Sunshine measures ..................................................................................................................... 15 2.1.3 Methods ...................................................................................................................................... 15 2.2 Pilot study ..........................................................................................................16 v 2.2.1 Study .......................................................................................................................................... 16 2.2.2 Lessons learned from the pilot study ......................................................................................... 19 2.3 Final variables and methods ...............................................................................19 2.3.1 Lung cancer population .............................................................................................................. 19 2.3.2 Sunshine measures ..................................................................................................................... 20 2.3.3 Methods ...................................................................................................................................... 20 3. Results ..................................................................................................................22 3.1 Patient characteristics, disease factors and treatment ........................................22 3.2 Small-cell lung cancer (SCLC) ..........................................................................24 3.2.1 Patients who received surgery with or without other treatments ............................................... 24 3.2.2 Patients who received radiotherapy with or without other treatments ....................................... 26 3.2.3 Patients who received chemotherapy with or without other treatments ..................................... 28 3.3 Non-small-cell lung cancer (NSCLC) ...............................................................30 3.3.1 Patients who received surgery with or without other treatments ............................................... 30 3.3.2 Patients who received radiotherapy with or without other treatments ....................................... 32 3.3.3 Patients who received chemotherapy with or without other treatments ..................................... 35 3.4 Summary ............................................................................................................37 3.5 Additional analyses ............................................................................................42 3.5.1 Small-cell lung cancer ................................................................................................................ 42 3.5.2 Non-small-cell lung cancer ........................................................................................................ 44 3.5.3 Multiple testing analysis ............................................................................................................ 46 3.6 Main results’ graphs and tables..........................................................................48 4. Discussion ..........................................................................................................228 4.1 Sunlight and cancer survival association .........................................................228 4.2 Study strengths .................................................................................................229 4.3 Study limitations ..............................................................................................230 4.4 Comparing the results with previous studies ...................................................231 4.5 Future work ......................................................................................................231 References ..............................................................................................................232 Appendices .............................................................................................................240 vi Appendix A: MMVDSI data set generation process .............................................240 Appendix B: Lung cancer population details ........................................................243 Appendix C: Summary of additional analyses’ significant results ........................244 vii List of Tables Table 1.1- Histopathological and epidemiological characteristics of lung cancer ......................... 4 Table 1.2- Clinical signs and symptoms in lung cancer ................................................................. 5 Table 1.3- Paraneoplastic syndromes in lung cancer ...................................................................... 6 Table 1.4- TNM staging in lung cancer .......................................................................................... 7 Table 1.5- Usual staging of lung cancer for NSCLC ...................................................................... 8 Table 1.6- Usual staging of lung cancer for SCLC ......................................................................... 8 Table 1.7- Lung cancer treatment ................................................................................................. 10 Table 1.8- Lung cancer prognosis ................................................................................................. 11 Table 3.1- Treatment summary for study patients ........................................................................ 23 Table 3.2- Female patients with SCLC who died because of lung cancer: HR estimates and multivariate P-values result for each treatment group .................................................................. 39 Table 3.3- Female patients with NSCLC who died because of lung cancer: HR estimates and multivariate P-values result for each treatment group .................................................................. 40 Table 3.4- Male patients with NSCLC who died because of lung cancer: HR estimates and multivariate P-values result for each treatment group .................................................................. 41 Table 3.5- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received surgery ................................................................. 48 Table 3.6- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received surgery ................................................................. 50 Table 3.7- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received surgery ................................................................. 52 Table 3.8- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received surgery ................................................................. 54 Table 3.9- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received surgery ....................................... 56 Table 3.10- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received surgery .............................................................. 58 Table 3.11- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with SCLC who received surgery ............................................................ 60 Table 3.12- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received surgery .............................................................. 62 viii Table 3.13- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received surgery .............................................................. 64 Table 3.14- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received surgery .................................... 66 Table 3.15- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received surgery ............................... 68 Table 3.16- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with SCLC who received surgery .................................. 70 Table 3.17- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received surgery .................................. 72 Table 3.18- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received surgery .................................. 74 Table 3.19- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received surgery ........ 76 Table 3.20- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received radiotherapy ......................................................... 78 Table 3.21- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received radiotherapy ......................................................... 80 Table 3.22- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received radiotherapy ......................................................... 82 Table 3.23- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received radiotherapy ......................................................... 84 Table 3.24- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received radiotherapy ............................... 86 Table 3.25- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received radiotherapy...................................................... 88 Table 3.26- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with SCLC who received radiotherapy...................................................... 90 Table 3.27- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received radiotherapy...................................................... 92 Table 3.28- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received radiotherapy...................................................... 94 Table 3.29- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received radiotherapy ............................ 96 ix Table 3.30- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received radiotherapy .......................... 98 Table 3.31- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with SCLC who received radiotherapy ........................ 100 Table 3.32- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received radiotherapy ...................... 102 Table 3.33- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received radiotherapy ........................ 104 Table 3.34- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received radiotherapy .............................................................................................................................. 106 Table 3.35- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received chemotherapy .................................................... 108 Table 3.36- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received chemotherapy .................................................... 110 Table 3.37- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received chemotherapy .................................................... 112 Table 3.38- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received chemotherapy .................................................... 114 Table 3.39- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received chemotherapy .......................... 116 Table 3.40- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received chemotherapy .............................................. 118 Table 3.41- Hazard ratio (HR) and 95% confidence interval (CI) associated Table 3.41- Hazard ratio (HR) and 95% confidence interval (CI) associated ............................................................ 120 Table 3.42- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received chemotherapy ................................................. 122 Table 3.43- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received chemotherapy ................................................. 124 Table 3.44- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received chemotherapy ....................... 126 Table 3.45- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received chemotherapy ...................... 128 Table 3.46- Hazard ratio (HR) and 95% confidence interval (CI) associated Table 3.46- Hazard ratio (HR) and 95% confidence interval (CI) associated ............................................................ 130 x Table 3.47- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received chemotherapy ...................... 132 Table 3.48- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received chemotherapy ...................... 134 Table 3.49- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received chemotherapy .............................................................................................................................. 136 Table 3.50- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received surgery ............................................................ 138 Table 3.51- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received surgery ............................................................ 140 Table 3.52- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received surgery ............................................................ 142 Table 3.53- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with NSCLC who received surgery ............................................................ 144 Table 3.54- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received surgery .................................. 146 Table 3.55- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received surgery ......................................................... 148 Table 3.56- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received surgery ......................................................... 150 Table 3.57- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received surgery ......................................................... 152 Table 3.58- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received surgery ......................................................... 154 Table 3.59- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received surgery ............................... 156 Table 3.60- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with NSCLC who received surgery ............................. 158 Table 3.61- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received surgery ............................. 160 Table 3.62- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received surgery ............................. 162 Table 3.63- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received surgery ............................. 164 xi Table 3.64- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with NSCLC who received surgery......................................................................................................................................... 166 Table 3.65- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received radiotherapy .................................................... 168 Table 3.66- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received radiotherapy .................................................... 170 Table 3.67- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received radiotherapy .................................................... 172 Table 3.68- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with NSCLC who received radiotherapy .................................................... 174 Table 3.69- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received radiotherapy .......................... 176 Table 3.70- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received radiotherapy ................................................. 178 Table 3.71- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received radiotherapy ................................................. 180 Table 3.72- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received radiotherapy ............................................... 182 Table 3.73- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received radiotherapy .............................................. 184 Table 3.74- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received radiotherapy ....................... 186 Table 3.75- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with NSCLC who received radiotherapy ..................... 188 Table 3.76- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received radiotherapy ..................... 190 Table 3.77- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received radiotherapy ..................... 192 Table 3.78- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received radiotherapy ..................... 194 Table 3.79- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index male and female patients with NSCLC who received radiotherapy ................................................................................................................................ 196 xii Table 3.80- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received chemotherapy ................................................. 198 Table 3.81- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received chemotherapy ................................................. 200 Table 3.82- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received chemotherapy ................................................. 202 Table 3.83- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with NSCLC who received chemotherapy ................................................. 204 Table 3.84- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received chemotherapy ....................... 206 Table 3.85- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received chemotherapy .............................................. 208 Table 3.86- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received chemotherapy .............................................. 210 Table 3.87- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received chemotherapy .............................................. 212 Table 3.88- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received chemotherapy ............................................ 214 Table 3.89- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received chemotherapy .................... 216 Table 3.90- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with NSCLC who received chemotherapy ................... 218 Table 3.91- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received chemotherapy ................... 220 Table 3.92- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received chemotherapy ................... 222 Table 3.93- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received chemotherapy ................... 224 Table 3.94- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with NSCLC who received chemotherapy ....................................................................................................... 226 xiii List of Figures Figure 1.1- Lung anatomy............................................................................................................... 1 Figure 1.2- Lung cancer .................................................................................................................. 2 Figure 3.1- Patient age at diagnosis .............................................................................................. 22 Figure 3.2- Cancer laterality (Tumor location) ............................................................................. 22 Figure 3.3- Patient sex .................................................................................................................. 22 Figure 3.4- Patient cause of death ................................................................................................. 22 Figure 3.5- Disease stage (at time of diagnosis) ........................................................................... 23 Figure 3.6- Disease histology classification ................................................................................. 23 Figure 3.7- Disease site (Location of tumor in lower respiratory tract) ....................................... 23 xiv List of graphs Graph 3.1- Survival according to season of diagnosis for males with SCLC who received surgery: lung cancer deaths ........................................................................................................... 49 Graph 3.2- Survival according to season of diagnosis for males with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 49 Graph 3.3- Survival according to season of first treatment for males SCLC who received surgery: lung cancer deaths ........................................................................................................... 51 Graph 3.4- Survival according to season of first treatment for males SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 51 Graph 3.5- Survival according to month of diagnosis for males with who received surgery: lung cancer deaths ................................................................................................................................. 53 Graph 3.6- Survival according to month of diagnosis for males with SCLC who received surgery: non-lung-cancer death..................................................................................................... 53 Graph 3.7- Survival according to month of first treatment for males with SCLC who received surgery: lung cancer deaths ........................................................................................................... 55 Graph 3.8- Survival according to month of first treatment for males with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 55 Graph 3.9- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received surgery: lung cancer deaths .................................................................................... 57 Graph 3.10- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received surgery: non-lung-cancer deaths.................................................................. 57 Graph 3.11- Survival according to season of diagnosis for females with SCLC who received surgery: lung-cancer deaths .......................................................................................................... 59 Graph 3.12- Survival according to season of diagnosis for females with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 59 Graph 3.13- Survival according to season of first treatment for females with SCLC who received surgery: lung cancer deaths ........................................................................................................... 61 Graph 3.14- Survival according to season of first treatment for females with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 61 Graph 3.15- Survival according to month of diagnosis for females with SCLC who received surgery: lung cancer deaths ........................................................................................................... 63 Graph 3.16- Survival according to month of diagnosis for females with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 63 xv Graph 3.17- Survival according to month of first treatment for females with SCLC who received surgery: lung cancer deaths ........................................................................................................... 65 Graph 3.18- Survival according to month of first treatment for females with SCLC who received surgery: non-lung-cancer deaths ................................................................................................... 65 Graph 3.19- Survival according to monthly mean Vitamin D sunshine index for females with SCLC who received surgery: lung-cancer deaths ......................................................................... 67 Graph 3.20- Survival according to monthly mean Vitamin D sunshine index for females with SCLC who received surgery: non-lung-cancer deaths.................................................................. 67 Graph 3.21- Survival according to season of diagnosis for male and female patients with SCLC who received surgery: lung cancer deaths .................................................................................... 69 Graph 3.22- Survival according to season of diagnosis for male and female patients with SCLC who received surgery: non-lung-cancer deaths............................................................................. 69 Graph 3.23- Survival according to season of first treatment for male and female patients with SCLC who received surgery: lung cancer deaths ......................................................................... 71 Graph 3.24- Survival according to season of first treatment for male and female patients with SCLC who received surgery: non-lung-cancer deaths.................................................................. 71 Graph 3.25- Survival according to month of diagnosis for male and female patients with SCLC who received surgery: lung cancer deaths .................................................................................... 73 Graph 3.26- Survival according to month of diagnosis for male and female patients with SCLC who received surgery: non-lung-cancer deaths............................................................................. 73 Graph 3.27- Survival according to month of first treatment for male and female patients with SCLC who received surgery: lung cancer deaths ......................................................................... 75 Graph 3.28- Survival according to month of first treatment for male and female patients with SCLC who received surgery: non-lung-cancer deaths.................................................................. 75 Graph 3.29- Survival according to monthly mean vitamin D sunshine index for male and female patients with SCLC who received surgery: lung cancer deaths .................................................... 77 Graph 3.30- Survival according to monthly mean Vitamin D sunshine index for male and female patients with SCLC who received surgery: non-lung-cancer deaths ............................................ 77 Graph 3.31- Survival according to season of diagnosis for males with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 79 Graph 3.32- Survival according to season of diagnosis for males with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 79 Graph 3.33- Survival according to season of first treatment for males with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 81 xvi Graph 3.34- Survival according to season of first treatment for males with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 81 Graph 3.35- Survival according to month of diagnosis for males with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 83 Graph 3.36- Survival according to month of diagnosis for males with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 83 Graph 3.37- Survival according to month of first treatment for males with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 85 Graph 3.38- Survival according to month of first treatment for males with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 85 Graph 3.39- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received radiotherapy: lung cancer deaths ................................................................. 87 Graph 3.40- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received radiotherapy: non-lung-cancer deaths ......................................................... 87 Graph 3.41- Survival according to season of diagnosis for females with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 89 Graph 3.42- Survival according to season of diagnosis for females with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 89 Graph 3.43- Survival according to season of first treatment for females with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 91 Graph 3.44- Survival according to season of first treatment for females with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 91 Graph 3.45- Survival according to month of diagnosis for females with SCLC who received radiotherapy: lung cancer deaths… .............................................................................................. 93 Graph 3.46- Survival according to month of diagnosis for females with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 93 Graph 3.47- Survival according to month of first treatment for females with SCLC who received radiotherapy: lung cancer deaths .................................................................................................. 95 Graph 3.48- Survival according to month of first treatment for females with SCLC who received radiotherapy: non-lung-cancer deaths ........................................................................................... 95 Graph 3.49- Survival according to monthly mean vitamin D sunshine index for females with SCLC who received radiotherapy: lung cancer deaths ................................................................. 97 Graph 3.50- Survival according to monthly mean Vitamin D sunshine index for females with SCLC who received radiotherapy: non-lung-cancer deaths ......................................................... 97 xvii Graph 3.51- Survival according to season of diagnosis for male and female SCLC patients who received radiotherapy: lung cancer deaths .................................................................................... 99 Graph 3.52- Survival according to season of diagnosis for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths ............................................................................ 99 Graph 3.53- Survival according to season of first treatment for male and female SCLC patients who received radiotherapy: lung cancer deaths .......................................................................... 101 Graph 3.54- Survival according to season of first treatment for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths .................................................................. 101 Graph 3.55- Survival according to month of diagnosis for male and female SCLC patients who received radiotherapy: lung cancer deaths .................................................................................. 103 Graph 3.56- Survival according to month of diagnosis for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths .......................................................................... 103 Graph 3.57- Survival according to month of first treatment for male and female SCLC patients who received radiotherapy: lung cancer deaths .......................................................................... 105 Graph 3.58- Survival according to month of first treatment for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths .................................................................. 105 Graph 3.59- Survival according to monthly mean vitamin D sunshine index for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths .......................................... 107 Graph 3.60- Survival according to monthly mean vitamin D sunshine index for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths .......................................... 107 Graph 3.61- Survival according to season of diagnosis for males with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 109 Graph 3.62- Survival according to season of diagnosis for males with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 109 Graph 3.63- Survival according to season of first treatment for males with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 111 Graph 3.64- Survival according to season of first treatment for males with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 111 Graph 3.65- Survival according to month of diagnosis for males with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 113 Graph 3.66- Survival according to month of diagnosis for males with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 113 Graph 3.67- Survival according to month of first treatment for males with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 115 xviii Graph 3.68- Survival according to month of first treatment for males with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 115 Graph 3.69- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received chemotherapy: lung cancer deaths ............................................................. 117 Graph 3.70- Survival according to monthly mean vitamin D sunshine index for males with SCLC who received chemotherapy: non-lung-cancer deaths ..................................................... 117 Graph 3.71- Survival according to season of diagnosis for females with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 119 Graph 3.72- Survival according to season of diagnosis for females with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 119 Graph 3.73- Survival according to season of first treatment for females with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 121 Graph 3.74- Survival according to season of first treatment for females with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 121 Graph 3.75- Survival according to month of diagnosis for females with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 123 Graph 3.76- Survival according to month of diagnosis for females with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 123 Graph 3.77- Survival according to month of first treatment for females with SCLC who received chemotherapy: lung cancer deaths .............................................................................................. 125 Graph 3.78- Survival according to month of first treatment for females with SCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 125 Graph 3.79- Survival according to monthly mean vitamin D sunshine index for females with SCLC who received chemotherapy: lung cancer deaths ............................................................. 127 Graph 3.80- Survival according to monthly mean vitamin D sunshine index for females with SCLC who received chemotherapy: non-lung-cancer deaths ..................................................... 127 Graph 3.81- Survival according to season of diagnosis for male and female patients with SCLC who received chemotherapy: lung cancer deaths ........................................................................ 129 Graph 3.82- Survival according to season of diagnosis for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths ................................................................ 129 Graph 3.83- Survival according to season of first treatment for male and female patients with SCLC who received chemotherapy: lung cancer deaths ............................................................. 131 Graph 3.84- Survival according to season of first treatment for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths ..................................................... 131 xix Graph 3.85- Survival according to month of diagnosis for male and female patients with SCLC who received chemotherapy: lung cancer deaths ........................................................................ 133 Graph 3.86- Survival according to month of diagnosis for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths ................................................................ 133 Graph 3.87- Survival according to month of first treatment for male and female patients with SCLC who received chemotherapy: lung cancer deaths ............................................................. 135 Graph 3.88- Survival according to month of first treatment for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths ..................................................... 135 Graph 3.89- Survival according to monthly mean vitamin D sunshine index for male and female patients with SCLC who received chemotherapy: lung cancer deaths ....................................... 137 Graph 3.90- Survival according to monthly mean vitamin D sunshine index for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths ............................... 137 Graph 3.91- Survival according to season of diagnosis for males with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 139 Graph 3.92- Survival according to season of diagnosis for males with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 139 Graph 3.93- Survival according to season of first treatment for males with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 141 Graph 3.94- Survival according to season of first treatment for males with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 141 Graph 3.95- Survival according to month of diagnosis for males with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 143 Graph 3.96- Survival according to month of diagnosis for males with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 143 Graph 3.97- Survival according to month of first treatment for males with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 145 Graph 3.98- Survival according to month of first treatment for males with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 145 Graph 3.99- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received surgery: lung-cancer deaths .................................................................... 147 Graph 3.100- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 147 Graph 3.101- Survival according to season of diagnosis for females with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 149 xx Graph 3.102- Survival according to season of diagnosis for females with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 149 Graph 3.103- Survival according to season of first treatment for females with NSCLC who received surgery: lung cancer deaths .......................................................................................... 151 Graph 3.104- Survival according to season of first treatment for females with NSCLC who received surgery: non-lung-cancer deaths .................................................................................. 151 Graph 3.105- Survival according to month of diagnosis for females with NSCLC who received surgery: lung cancer deaths ......................................................................................................... 153 Graph 3.106- Survival according to month of diagnosis for females with NSCLC who received surgery: non-lung-cancer deaths ................................................................................................. 153 Graph 3.107- Survival according to month of first treatment for females with NSCLC who received surgery: lung cancer deaths .......................................................................................... 155 Graph 3.108- Survival according to month of first treatment for females with NSCLC who received surgery: non-lung-cancer deaths .................................................................................. 155 Graph 3.109- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received surgery: lung cancer deaths .................................................................... 157 Graph 3.110- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 157 Graph 3.111- Survival according to season of diagnosis for male and female patients with NSCLC who received surgery: lung cancer deaths .................................................................... 159 Graph 3.112- Survival according to season of diagnosis for male and female patients with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 159 Graph 3.113- Survival according to season of diagnosis for male and female patients with NSCLC who received surgery: lung cancer deaths .................................................................... 161 Graph 3.114- Survival according to season of diagnosis for male and female patients with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 161 Graph 3.115- Survival according to month of diagnosis for male and female patients with NSCLC who received surgery: lung cancer deaths .................................................................... 163 Graph 3.116- Survival according to month of diagnosis for male and female patients with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 163 Graph 3.117- Survival according to month of first treatment for male and female patients with NSCLC who received surgery: lung cancer deaths .................................................................... 165 Graph 3.118- Survival according to month of first treatment for male and female patients with NSCLC who received surgery: non-lung-cancer deaths ............................................................. 165 xxi Graph 3.119- Survival according to monthly mean vitamin D sunshine index for male and female patients with NSCLC who received surgery: lung cancer deaths ............................................... 167 Graph 3.120- Survival according to monthly mean vitamin D sunshine index for male and female patients with NSCLC who received surgery: non-lung-cancer deaths ....................................... 167 Graph 3.121- Survival according to season of diagnosis for males with NSCLC who received radiotherapy: lung cancer deaths ................................................................................................ 169 Graph 3.122- Survival according to season of diagnosis for males with NSCLC who received radiotherapy: non-lung-cancer deaths ......................................................................................... 169 Graph 3.123- Survival according to season of first treatment for males with NSCLC who received radiotherapy: lung cancer deaths .................................................................................. 171 Graph 3.124- Survival according to season of first treatment for males with NSCLC who received radiotherapy: non-lung-cancer deaths .......................................................................... 171 Graph 3.125- Survival according to month of diagnosis for males with NSCLC who received radiotherapy: lung cancer deaths ................................................................................................ 173 Graph 3.126- Survival according to month of diagnosis for males with NSCLC who received radiotherapy: non-lung-cancer deaths ......................................................................................... 173 Graph 3.127- Survival according to month of first treatment for males with NSCLC who received radiotherapy: lung cancer deaths .................................................................................. 175 Graph 3.128- Survival according to month of first treatment for males with NSCLC who received radiotherapy: non-lung-cancer deaths .......................................................................... 175 Graph 3.129- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received radiotherapy: lung cancer deaths ............................................................ 177 Graph 3.130- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received radiotherapy: non-lung-cancer deaths .................................................... 177 Graph 3.131- Survival according to season of diagnosis for females with NSCLC who received radiotherapy: lung cancer deaths ................................................................................................ 179 Graph 3.132- Survival according to season of diagnosis for females with NSCLC who received radiotherapy: non-lung-cancer deaths ....................................................................................... 179 Graph 3.133- Survival according to season of first treatment for females with NSCLC who received radiotherapy: lung cancer deaths .................................................................................. 181 Graph 3.134- Survival according to season of first treatment for females with NSCLC who received radiotherapy: non-lung-cancer deaths .......................................................................... 181 Graph 3.135- Survival according to month of diagnosis for females with NSCLC who received radiotherapy: lung cancer deaths ................................................................................................ 183 xxii Graph 3.136- Survival according to month of diagnosis for females with NSCLC who received radiotherapy: non-lung-cancer deaths ......................................................................................... 183 Graph 3.137- Survival according to month of first treatment for females with NSCLC who received radiotherapy: lung cancer deaths .................................................................................. 185 Graph 3.138- Survival according to month of first treatment for females with NSCLC who received radiotherapy: non-lung-cancer deaths .......................................................................... 185 Graph 3.139- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received radiotherapy: lung cancer deaths ............................................................ 187 Graph 3.140- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received radiotherapy: non-lung-cancer deaths .................................................... 187 Graph 3.141- Survival according to season of diagnosis for male and female NSCLC patients who received radiotherapy: lung cancer deaths .......................................................................... 189 Graph 3.142- Survival according to season of diagnosis for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths .................................................................. 189 Graph 3.143- Survival according to season of first treatment for male and female NSCLC patients who received radiotherapy: lung cancer deaths ............................................................. 191 Graph 3.144- Survival according to season of first treatment for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths ..................................................... 191 Graph 3.145- Survival according to month of diagnosis for male and female NSCLC patients who received radiotherapy: lung cancer deaths .......................................................................... 193 Graph 3.146- Survival according to month of diagnosis for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths .................................................................. 193 Graph 3.147- Survival according to month of first treatment for male and female NSCLC patients who received radiotherapy: lung cancer deaths ............................................................. 195 Graph 3.148- Survival according to month of first treatment for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths ..................................................... 195 Graph 3.148- Survival according to monthly mean vitamin D sunshine index for male and female NSCLC patients who received radiotherapy: lung cancer deaths ............................................... 197 Graph 3.150- Survival according to monthly mean vitamin D sunshine index for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths ....................................... 197 Graph 3.151- Survival according to season of diagnosis for males with NSCLC who received chemotherapy: lung cancer deaths .............................................................................................. 199 Graph 3.152- Survival according to season of diagnosis for males with NSCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 199 xxiii Graph 3.153- Survival according to season of first treatment for males with NSCLC who received chemotherapy: lung cancer deaths ............................................................................... 201 Graph 3.154- Survival according to season of first treatment for males with NSCLC who received chemotherapy: non-lung-cancer deaths ........................................................................ 201 Graph 3.155- Survival according to month of diagnosis for males with NSCLC who received chemotherapy: lung cancer deaths .............................................................................................. 203 Graph 3.156- Survival according to month of diagnosis for males with NSCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 203 Graph 3.157- Survival according to month of first treatment for males with NSCLC who received chemotherapy: lung cancer deaths ............................................................................... 205 Graph 3.158- Survival according to month of first treatment for males with NSCLC who received chemotherapy: non-lung-cancer deaths ........................................................................ 205 Graph 3.159- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 207 Graph 3.160- Survival according to monthly mean vitamin D sunshine index for males with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 207 Graph 3.161- Survival according to season of diagnosis for females with NSCLC who received chemotherapy: lung cancer deaths .............................................................................................. 209 Graph 3.162- Survival according to season of diagnosis for females with NSCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 209 Graph 3.163- Survival according to season of first treatment for females with NSCLC who received chemotherapy: lung cancer deaths ............................................................................... 211 Graph 3.164- Survival according to season of first treatment for females with NSCLC who received chemotherapy: non-lung-cancer deaths ........................................................................ 211 Graph 3.165- Survival according to month of diagnosis for females with NSCLC who received chemotherapy: lung cancer deaths .............................................................................................. 213 Graph 3.166- Survival according to month of diagnosis for females with NSCLC who received chemotherapy: non-lung-cancer deaths ...................................................................................... 213 Graph 3.167- Survival according to month of first treatment for females with NSCLC who received chemotherapy: lung cancer deaths ............................................................................... 215 Graph 3.168- Survival according to month of first treatment for females with NSCLC who received chemotherapy: non-lung-cancer deaths ........................................................................ 215 Graph 3.169- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 217 xxiv Graph 3.170- Survival according to monthly mean vitamin D sunshine index for females with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 217 Graph 3.171- Survival according to season of diagnosis for male and female patients with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 219 Graph 3.172- Survival according to season of diagnosis for male and female patients with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 219 Graph 3.173- Survival according to season of first treatment for male and female patients with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 221 Graph 3.174- Survival according to season of first treatment for male and female patients with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 221 Graph 3.175- Survival according to month of diagnosis for male and female patients with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 223 Graph 3.176- Survival according to month of diagnosis for male and female patients with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 223 Graph 3.177- Survival according to month of first treatment for male and female patients with NSCLC who received chemotherapy: lung cancer deaths .......................................................... 225 Graph 3.178- Survival according to month of first treatment for male and female patients with NSCLC who received chemotherapy: non-lung-cancer deaths .................................................. 225 Graph 3.179- Survival according to monthly mean vitamin D sunshine index for male and female patients with NSCLC who received chemotherapy: lung cancer deaths .................................... 227 Graph 3.180- Survival according to monthly mean vitamin D sunshine index for male and female patients with NSCLC who received chemotherapy: non-lung-cancer deaths ............................ 227 xxv List of Abbreviations Apr April Aug August BC British Columbia BCCA British Columbia Cancer Agency BCCR British Columbia Cancer Registry CCS Canadian Cancer Statistics CI Confidence interval CMC Computerized medical charts CT Computerized tomography (scan) Dec December Feb February HR Hazard ratio Jan January Jul July Jun June lat Latitude lon Longitude Mar March MMVDSI Monthly mean vitamin D sunshine index MnDph Mean hourly dose of vitamin D action spectrum xxvi n Total number of cases Nov November NSCLC Non-small cell lung cancer NS Not significant Oct October PET Positron Emission Tomography RSR Relative survival ratio SCC Squamous cell carcinoma SCLC Small cell lung cancer Sep September Tis Tumor in situ TNM Tumor size, Node (lymph node) involvement, Metastasis UV Ultraviolet WHO World Health Organization xxvii Acknowledgments First of all, I would like to dedicate my thesis to Dr. Chris Bajdik, my supervisor. He was not only my supervisor, but also my mentor. He has taught me a lot and I am sure I will use all of them throughout my life. I also want to thank my thesis committee members, Dr. Tim Lee and Dr. Sam Wiseman. Their advices were very constructive and helpful during doing my project. xxviii Dedication I also like to thank my lovely wife, Yalda Mahmoudi. Her motivations and positive energy has had a great impact on me. Additionally, I am thankful for my family, specially my mom and dad who have always supported and encouraged me in my life. Last but not least my gratitude goes to my friend, Ehsan Bayaki. Since I have moved to Vancouver, he has helped me in every way possible without any hesitation. I have valued his friendship for the last fifteen years and I am looking forwards to many more years. 1 1. Introduction 1.1 Thesis overview This thesis is about lung cancer survival and sunlight exposure. My study is explained in four chapters. Chapter one is about lung cancer, sunlight and how sunlight exposure might affect a lung cancer patient’s survival. The study methods are discussed in chapter two and the results are presented in chapter three. Finally, chapter four is a discussion in which I explain the implications of this study in cancer research. Figure 1.1 is an illustration from a standard medical text of human lung anatomy. Figure 1.1- Lung anatomy (Gray's Anatomy of the Human Body, 20th ed. 1918) 2 1.2 Lung cancer Lung cancer, like other types of cancer, is the result of an imbalance in cell growth. Uncontrolled division and proliferation of lung tissue cells eventually forms a mass known as a lung tumor. Figure 1.2 illustrates human lung cancer. Figure 1.2-Lung cancer (Copyright of Cleveland clinic organization.) 1.2.1 Descriptive epidemiology Lung cancer is the most common cause of death due to cancer worldwide 1 . According to the World Health Organization (WHO), it is the most common newly-diagnosed cancer in men worldwide, and ranks second after prostate cancer in the western world 1 . Further, it is the second most common cancer in women worldwide. Canadian Cancer Statistics (CCS) in 2010 reports that lung cancer is the most common cause of death due to cancer in the last decade nationally 3 and the second-most-incident cancer 2 . Based on this report, mortality and incidence rates are the highest in Quebec and the lowest in British Columbia among men 3 . And Nova Scotia is predicted to have the highest rate of lung cancer among women 3 . 1.2.2 Risk factors One of the most established risk factors for lung cancer is tobacco smoking 4 and almost 18 of every 20 lung cancer deaths in developed countries are caused by it 5 . In the United States, smoking is the cause of about 87% of lung cancer cases 6 (90% of male cases and 85% of female cases). In Canada, about 85% of new lung cancer cases are related to smoking 7 . There is less evidence about other factors associated with lung cancer. Radon exposure is the second major risk factor for the disease 8 . Asbestos has a synergistic effect with tobacco smoking in lung cancer formation 9 . Heavy metals such as chromium, nickel and arsenic are also believed to be associated with this cancer 10 . There is evidence that a 1% airborne particulate matter concentration increases lung cancer risk by as much as 14% 11-12 . Lung cancer is initiated by oncogene activation or tumor suppressor gene inactivation 13 . For example, K-ras proto-oncogene mutations are responsible for 10-13% of lung adenocarcinoma 14-15 . The oncogene epidermal growth factor receptor (EGFR) regulates cell proliferation, apoptosis, angiogenesis and tumor invasion – all of which are amplified in non- small cell lung cancer 14 . Chromosomal damage also can cause inactivation of tumor suppressor genes on 3p, 5q, 13q, and 17p – which is common in small cell lung cancer14. Other genes such as c-MET, NKX2-1, LKB1, PIK3CA, and BRAF also may be amplified or mutated in lung cancer patients 14 . Mutations also might occur when someone develops lung cancer. Examples are variations in the genes coding interleukin-1 16 , cytochrome P450 17 , apoptosis 18 and DNA repair molecules 19 . 4 1.2.3 Pathology Most lung cancers arise from epithelial cells and are carcinomas 20 . These are mainly categorized as small cell lung cancer (SCLC) or non-small cell lung carcinoma (NSCLC). This is based on a cooperative effort by the WHO and International Association for Study of Lung Cancer. The categorization is summarized in Table 1.1. Table 1.1: Histopathological and epidemiological characteristics of lung cancer (Incidence rates are summarized from Travis et all: Cancer 75:191,1995) Lung cancer type Incidence proportion Smoking correlation Location in the lung Cell histology Metastasis Small Cell Lung Cancer (SCLC) strong central Oat cell, neuroendocrine origin in endobronchial cells Disseminated at presentation Non-Small Cell Lung Cancer (NSCLC) A d en o ca rc in o m a 39-43% weak peripheral Glandular, mucin producing Early, distant S q u am o u s ce ll ca rc in o m a (S C C ) 30% strong Central Keratin, intracellular bridges Slow, local invasion, may cavitate L ar g e ce ll ca rc in o m a 9% yes peripheral Anaplastic, undifferentiated Early, distant 5 Lung cancers are remarkably heterogeneous tumors that may contain more than one histological subtype 21 . There are other types of lung cancer than SCLC and NSCLC. For example, carcinoid tumors, carcinoma of salivary glands and adenosquamous carcinoma are those who are least frequent cancers with epithelial origin. These types of lung cancer were not present in the data for this thesis. 1.2.4 Diagnosis (signs and symptoms) A summary of lung cancer patients’ signs and symptoms is given in Tables 1.2 and 1.3. Many of these can be expressed based on the tumor location 22 , but paraneoplastic phenomena are better classified by clinical characteristics. These phenomena are mediated by humoral factors (by hormones or cytokines) or an immune response against the tumor 23 . Other features are the result of metastasis 24 . Table 1.2: Clinical signs and symptoms in lung cancer based on the extent of disease spread Disease spread Signs and symptoms (% of patients that are affected) Locoregional Cough (75%), shortness of breath (60%), Chest pain (45%), spitting of blood (35%), Wheezing, Difficulty swallowing, Hoarseness, Sputum (salty suggests bronchoalveolar) Metastatic Bone pain, Jaundice, Seizures, Headaches, Adrenal lesions, Skin lesions 6 Table 1.3: Paraneoplastic syndromes in lung cancer System Clinical presentation Type of lung cancer Skeletal Clubbing Not SCLC Dermatologic Acanthosis nigricance, dermatomyositis Carcinoma Endocrine Hypercalcemia, hypophosphatemia SCC (Squamous cell carcinoma) Cushing syndrome, syndrome of inappropriate antidiuretic hormone hypersecretion (SIADH) SCLC Neuromyopathic Eaton-Lambert Syndrome, polymyositis, subacute cerebellar degeneration, spinocerebellar degeneration, peripheral neuropathy SCLC Vascular/Hematologic Nonbacterial endocarditis, Trousseau’s syndrome, disseminated intravascular coagulation (DIC) Carcinoma Renal Nephrotic syndrome 1.2.5 Staging Staging a tumor is a process to assess the extent of its spread. Based on the type of tumor the staging systems differ. A well-known staging method for lung cancer is the TNM system which is mostly used for NSCLC. In this system, a tumor is staged based on its size (T), the lymph node involvement (N) and the presence of metastasis to distant organs (M). There are different clinical investigations for staging lung cancer using imaging methods (e.g., ultrasound) and surgical methods (e.g., diagnostic thoracotomy). Using the TNM system, lung cancer is categorized in 4 stages, as described in Table 1.4-1.6. 7 Table 1.4: TNM staging in lung cancer (The contents is obtained from different sources) Description P ri m a ry t u m o r si ze ( T ) Tis Carcinoma in situ T1 Tumor <= 3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus* (i.e., not in the main bronchus) T2 Tumor with any of the following features of size or extent: > 3 cm in greatest dimension, involves main bronchus, >= 2 cm distal to the carina, invades the visceral pleura, associated with atelectasis or obstructive pneumontitis that extends to the hilar region but does not involve the entire lung T3 Tumor of any size that directly invades any of the following: chest wall (including superior sulcus tumors), diaphragm, mediastinal pleura, parietal pericardium, or tumor in the main bronchus < 2 cm distal to the carina, but without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea,esophagus, vertebral body, carina; or tumor with a malignant pleural or pericardial effusion, or with satellite tumor nodule(s) within the ipsilateral primary-tumor lobe of the lung R eg io n a l ly m p h n o d es in v o lv em en t (N ) N0 No regional lymph node metastasis N1 Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes, and intra-pulmonary nodes involved by direct extension of the primary tumor N2 Metastasis to ipsilateral mediastinal and/or subcarinal lymph node(s) N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s) D is ta n t m et a st a si s (M ) M0 No distant metastasis M1 Distant metastasis present 8 Table 1.5: Usual staging of lung cancer for NSCLC Stage TNM Subset Stage TNM Subset 0 Carcinoma in situ IIIA T3N1M0, T1N2M0, T2N2M0, T3N2M0 IA T1N0M0 IB T2N0M0 IIIB T4N0M0, T4N1M0, T4N2M0 T1N3M0, T2N3M0, T3N3M0 T4N3M0 IIA T1N1M0 IIB T2N1M0, T3N0M0 IV Any T, Any N, M1 Table 1.6: Usual staging of lung cancer for SCLC Limited disease Primary tumor confined to hemithorax, ipsilateral hilar nodes, ipsilateral and contralateral, supraclavicular nodes, ipsilateral and contralateral, mediastinal nodes, pleural effusion Extensive disease More advanced than limited disease, metastases to contralateral lung, or distant metastases 9 1.2.6 Treatment Treatment for lung cancer patients is based on the tumor’s cell type, the tumor’s degree of invasion and the patient’s comorbidity status. There are 3 major types of treatments: surgery, chemotherapy and radiotherapy 25-26 . Table 1.7 shows how lung cancer patients are treated based on their tumor stage. Surgery Physicians do imaging investigations such as CT and PET scans to determine whether the cancer is surgically resectable 27. The appropriateness of surgery is also determined by a patient’s comorbidity status, using blood tests and spirometry. Surgery only is usually an option for stage IA or earlier unilateral NCSLC only. Surgical procedures in lung cancer include wedge resection, lobectomy and pneumonectomy. Chemotherapy The most common chemotherapeutic medications for SCLC patients are cisplatin and etoposide 28 . NSCLC patients are often treated with cisplatin or carboplatin, in combination with gemcitabine, paclitaxel, docetaxel, etoposide, or vinorelbine 29. Depending on a patient’s age and other characteristics, various molecules also can be used for treating advanced lung cancer. For example, Gefitinib targets the EGFR gene and is useful for some patient types 30-31 . Radiotherapy If surgery is not used, physicians usually administer radical radiotherapy with chemotherapy for NSCLC patients 32 . For SCLC, chest radiotherapy would accompany chemotherapy 33. Based on a tumor’s location and type, other types of radiation therapy might be used such as brachytherapy or prophylactic cranial irradiation 34-35 . 10 Table 1.7: Lung cancer treatment Stage Treatment Non-Small Cell Lung Cancer (NSCLC) IA Lobectomy/Pneumonectomy IB Lobectomy/Pneumonectomy adjuvant chemotherapy if high risk features (e.g. large tumor, high grade II Lobectomy/Pneumonectomy adjuvant chemotherapy IIIA (T1N2M0) Concurrent chemoradiation followed either by lobectomy/pneumonectomy or radiation boost IIIA (Unresectable) and IIIB Concurrent chemoradiation with potential chance of cure, and possible sequential chemoradiation IIIB (with malignant pleural effusion) and IV Radiation, palliative chemotherapy, resection if solitary central nervous system (CNS) metastasis, and possible targeted therapy Small Cell Lung Cancer (SCLC) Limited stage Radiation, concurrent chemotherapy, and prophylactic cranial (head) irradiation (PCI) Extensive stage Palliative chemotherapy, and possible targeted therapy 11 1.2.7 Prognosis In general, SCLC has the poorest prognosis of all lung cancer types, and NSCLC patients have a better prognosis if they are diagnosed at an early disease stage. Table 1.8 summarizes 5-year survival in lung cancer patients. The prognosis of NSCLC patients depends on their pulmonary symptoms, tumor size, tumor histology, disease stage, disease metastasis to lymph nodes and vascular invasion of the disease 36 . The prognosis for SCLC patients depends on patient’s status, patient gender, disease stage, and CNS or liver involvement at the time of diagnosis 37 . Table 1.8: Lung cancer prognosis Cell type Stage Prognosis Non-Small Cell Lung Cancer (NSCLC) IA 70% 5-year survival rate IB 60% 5-year survival rate IIA 50% 5-year survival rate IIB 40% 5-year survival rate IIIA 30% 5-year survival rate IIIB 15% 5-year survival rate IV 5-10% 5-year survival rate Small Cell Lung Cancer (SCLC) Limited stage 15-20% 5-year survival rate Extensive stage <5% 5-year survival rate 12 In Canada, the 5-year lung cancer relative survival ratio (RSR) in 2002-2004 was 13% for men and 17% for women. The relative survival rate (RSR) is the chance that a patient will survive a set time period after diagnosis divided by the percentage of the corresponding age-and- sex-specific general population. For example, the five-year relative survival rate for lung cancer describes the percentage of patients with that disease that are alive five years after diagnosis, divided by the percentage of the corresponding Age-and-sex-specific general population that are alive after five years. The RSR is calculated to adjust for the chance of death due to conditions other than the cancer. Canadian Cancer Statistics (CCS) in 2009 reports that lung cancer value is among the RSR values for the least favorable cancer types. (RSR=6% for pancreatic cancer and RSR=14% for esophageal cancer) 38 . Based on this report, lung cancer relative survival was highest among young patients (20-39 y) 39 . Further, the age-standardized RSR during 2002-2004 was highest in Manitoba (19%) 40 and lowest in Prince Edward Island (11%) 40 . The age- standardized 5-year RSR in British Columbia was 12-14% 40 . 1.3 Sunlight and vitamin D Vitamin D belongs to the group of fat-soluble vitamins (i.e., vitamins K, A, D and E). It has two physiologically relevant forms: 25(OH) D (calcidiol) and 1,25(OH)D (calcitriol). The most important physiologic effect of vitamin D is regulating the concentration of calcium and phosphate in the bloodstream, promoting the healthy mineralization, growth and remodeling of bone. Vitamin D is carried in the blood stream to the liver where it is converted to 25(OH) D. The immune system or kidneys convert the circulating 25(OH) D to 1, 25(OH) D, which is the biologically active form of vitamin D. In humans, having an optimal vitamin D level is related to sun exposure or artificial UV-B (280-320 nm) sources 41. It’s also related to consumption of 13 vitamin D enriched foods such as eggs, margarine, oily fish and vitamin supplements 42 . Sun- induced vitamin D is usually the main source of this vitamin, but vitamin D status changes seasonally for people in Canada 43 . 1.4 Sunlight, vitamin D and cancer prognosis Recent studies have shown the effects of vitamin D in survival of patients with prostate, breast, ovarian, pancreatic, and colorectal cancer as well as Hodgkin disease 44-50 . Those studies were motivated by the idea that active forms of vitamin D (D3) affect proliferation, differentiation and apoptosis in different kinds of cells 51-53 . (This effect is likely to be an immunomodulatory one.) The effects are probably the results of interactions between active vitamin D derivatives and/or membrane targets 51, 54 . In vitro studies performed with lung cancer cell-lines showed that vitamin D derivatives have an inhibitory effect on cell-growth and proliferation 55 . Animal investigations have demonstrated the ability of vitamin D derivatives to suppress angiogenesis, metastasis and invasion 56-58 Recently it has been hypothesized that seasonal variation of vitamin D might affect the prognosis of patients with colon cancer, breast cancer, prostate cancer as well as Hodgkin’s disease 59-60 . In a separate study, the same researchers found that sun-induced vitamin D improves lung cancer prognosis specifically in males under age 50 years 61 . In a separate US study, scientists found a positive effect for season of surgery and recent vitamin D intake on lung cancer survival 62 . A comprehensive UK study has produced similar results 63 . 14 1.5 Hypothesis Lung cancer is one of the most important and deadliest cancers in Canada and BC. There are lots of researches that have been done in this field but a few ones have addressed the significance of vitamin D in lung cancer prognosis. In addition, there is only some published research in which sun exposure used as an indicator of vitamin D effects in lung cancer prognosis. Based on this background, my research questions are: 1) Is there an association between a lung cancer patient’s survival and the amount of sunshine during their diagnosis? 2) Is there an association between a lung cancer patient’s survival and the amount of sunshine during their treatment? Based on these questions and the possible adjusting factors, my formal hypothesis is: Primary Hypothesis: Sunlight at the time of diagnosis and first treatment affects the length of lung cancer patients’ survival in BC. The effect of sunlight depends on (1) a patient’s age, (2) a patient’s gender, (3) the lung cancer’s stage, (4) the lung cancer’s histology, (5) types of treatment, (6) the primary tumor’s location in the lung and (7) the tumor’s laterality. This study attempts to provide evidence about whether sunlight improves the effectiveness of lung cancer treatment in BC. It might lead to interventions that involve vitamin D supplementation for lung cancer patients during their treatment. 15 2. Materials and methods 2.1 Proposed variables and methods 2.1.1 Lung cancer patient information Based on our hypothesis, our study population was those who were diagnosed because of lung cancer in British Columbia. In addition, we needed some information for each patient regarding on date of birth and death, sex, age, place of residence, primary tumour site and stage of disease at diagnosis, pathology of tumour, cause of death and method of treatment. Since we assumed that sunshine variables would be available, we decided to do a pilot study for the period of 2000 to 2005. The study period of the full analysis is from 1980 to 1989. 2.1.2 Sunshine measures The production of active vitamin D in our body depends on sun exposure. Vitamin D has shown to have inhibitory effects in development and progression of different cancers. It has been proven that vitamin D varies monthly because of the sun exposure in those months. In fact, we can use the weather data as an indirect measure of population’s active vitamin D level. 2.1.3 Methods We proposed to separately analyze data based on the histological types of tumor (small- cell lung cancer and non-small cell lung cancer), kinds of patient death (lung-cancer reasons or non-lung-cancer reasons), and type of treatment received by the patient (surgery, chemotherapy, radiation). Each analysis was to consist of a univariate and multivariate test of the variables. The univariate analysis was a Kaplan-Meier plot with log-rank test. The multivariate analysis used 16 the Hazard Ratio estimate from a Cox proportional-hazards regression model adjusting for patient age, disease stage, other treatment modalities received by the patient (i.e., chemo=yes/no; radiation=yes/no), tumor laterality (being in the right or left side of the lung) and tumor anatomic site in lower respiratory tract. 2.2 Pilot study 2.2.1 Study At the beginning of a project, the first thing that should be considered is evaluating the available data. To get this, I decided to do a pilot study to find how I can gather the data for my project and what would be the available ones. Lung cancer population When a patient in BC is diagnosed with a cancer, he or she is often referred to a BCCA centre. The centre records the patient’s information and stores it in the BC Cancer Registry (BCCR) and computerized medical charts (CMC). From the BCCR, I requested a sample of information for lung cancer patients who were referred to BCCA between 2000 and 2005. Each patient had a unique ID which had information regarding of date of birth, date of death, sex, age at the diagnosis, place of residence, BCCA referred centre, type of tumor and its location in the lung, treatment received and its date, stage of the tumor. In addition, it had the name of physicians who were involved in diagnosing, managing and treating of a patient. 17 Sunshine measures After this, I discussed my project with different meteorologists to find a reliable sunshine measure. I also reviewed previous publications in this field again to find what they used in their studies. Based on my findings, I considered five potential sunshine measures. A- Season of diagnosis Some previous studies used this variable because of seasonal variation of vitamin D production in their area. They found that the level of production was higher in studies conducted in sunny seasons. I discussed this with an expert meteorologist, and found that there is seasonal variation in vitamin D production in British Columbia. B- Season of first treatment According to the sample data, I found that it usually takes a month for a patient to receive any treatment after diagnosis in BC. In many cases, the season changed between diagnosis and treatment. This change could substantially affect the level of vitamin D in patients because the new season could be more or less sunnier. C- Bright sunshine This is the total hours of sunshine for each day. It was measured in different centres across BC but recording stopped after 2000. Another problem is that there is lots of missing information in some centers. 18 D- Total cloud amount This the total amount of cloud recorded hourly, which is an indirect measure of sunshine. If we calculate the period between sunrise and sunset, then it is necessary to subtract the total cloudy hours for each day. This might not be accurate because people can still be exposed to UV during cloudy periods. E: MnDph (Monthly mean hourly dose of vitamin D action spectrum) This is one of the most accurate measures of UV that was available. There are different subdivisions of UV, but wavelength 315-280 nm is related to vitamin D production. This information was recorded during 1980-1989 for various values of latitude (lat) and longitude (lon) across Canada. The 10-year average of this measure was calculated, and there were no significant year-to-year variability for a significant month of the year in the values for each lon and lat. Accordingly, the measure has not been recorded in any subsequent year because month- to-month changes can be determined from the existing years of data. (Please see the MMVDSI generation diagram in A1 in appendix section) F- Month of diagnosis and month of treatment As discussed earlier, vitamin D production in our body has a direct association with sun exposure. Further, there was at least one month period between the diagnosis and treatment for lung cancer patients in BC. Therefore, Month of diagnosis and month of treatment is an indirect measure of sunshine exposure. 19 2.2.2 Lessons learned from the pilot study In terms of the lung cancer population, I found that the BCCR and CMC have comprehensive information regarding the patients, their tumors and their treatments. In the case of sunshine measures, I decided that some measures couldn’t be used because of substantial missing information. I decided to perform analyses using month of diagnosis, month of first treatment, season of diagnosis, and season of first treatment. The most useful sunshine measure was MnDph but it was recorded based on lat and lon values. Patients’ addresses were recorded using postal codes. If we wanted to use this data, we had to find a way to convert postal codes to lat and lon. 2.3 Final variables and methods 2.3.1 Lung cancer population MnDph hasn’t been used in previous studies, but offers the best available measure of sunshine in BC. The variable is only available for the period of 1980-1989. Therefore, I decided to base the study on lung cancer patients referred to BCCA between 1980 and 1989. I excluded lung cancer diagnoses where: a patient has been diagnosed previously with another type of cancer, a patient lived less than 30 days after diagnosis (Because usually it took a 1-month period for patients to receive treatment. Therefore, those who died in less than a month could not be used in the analyses since we wanted to adjust our analyses for treatment options.), and any non- bronchogenic carcinomas (i.e., diagnoses that were mesothelioma or carcinoid). For MnDph, we excluded 1432 additional cases because there was another 781 cases whose postal code was 20 missing. The final study population for MnDph analysis was 8521 cases and 9302 cases for the other sunshine measures’ analyses. (Please see the A2 in appendix) 2.3.2 Sunshine measures My final choices of sunshine measurement were month of diagnosis, month of first treatment, season of diagnosis, season of first treatment, and MnDph. Season was categorized as Spring (March-April-May), Summer (June-July-August), Fall (September-October-November) and Winter (December-January-February). As mentioned earlier, MnDph was calculated as the 10-year average for each hour for each lat and lon. I calculated the mean value of each month based on the mean of each day for each lat and lon, and then I converted postal codes (I used the online system from www.postalcodedownload.com) to lat and lon, and linked the monthly values to the patient data. The result was that each patient has a sunshine measure (based on new MnDph data set). I named the value “the monthly mean vitamin D sunshine index (MMVDSI)”. 2.3.3 Methods I separately fitted a model to the data for histological types of tumor (small-cell lung cancer and non-small cell lung cancer), patient cause of death (lung-cancer reasons or non-lung- cancer reasons), and type of treatment received by the patient (surgery, chemotherapy, radiation). Each analysis consisted of a univariate and multivariate test of the variables. The univariate analysis was a Kaplan-Meier plot with log-rank test. The multivariate analysis was a Hazard Ratio estimate from a Cox proportional-hazards regression model adjusting for patient age (less than 50, 50-60, 60-70 and more than 70), disease stage (below or equal to stage ΙΙ or above stage ΙΙ for NSCLC; extended and limited for SCLC), other treatment modalities received by the patient (surgery, chemo, or radiation), tumor laterality (being in the 21 right or left side of the lung) and tumor anatomic site in lower respiratory tract. The last date of follow-up was May 30 th 2007 for all cases. We used SPSS software for all the analyses. All p-values <0.05 were considered statistically significant, and all p-values <0.1 are reported as “not significant” or NS. 22 3. Results This chapter begins with a descriptive summary of the study’s main variables. Detailed results of the analysis are given for small-cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC) in section 3.2 and 3.3. There is a summary and overview of the results in section 3.4. Some simple variations of the main analyses are presented in section 3.5. All the graphs and tables are at the end of this chapter in section 3.6. 3.1 Patient characteristics, disease factors and treatment Figures 3.1-3.7 show demographic characteristics of patients in the study and characteristics of their disease. Table 1 shows the main form of treatment that study patients received. Figure 3.1- Patient age at diagnosis Figure 3.2- Cancer laterality (Tumor location) Figure 3.3- Patient sex Figure 3.4- Patient cause of death 8% 23% 39% 30% ≤50 y 51-60 y 61-70 y >70 y 55% 43% 2% Right Left N/A 66.4% 33.6% Male Female 84% 14% 2% Lung Cancer Other causes Alive 23 Figure 3.5- Disease stage (at time of diagnosis) Figure 3.6- Disease histology classification (N/A: not available based on the BBCA registry data set) Figure 3.7- Disease site (Location of tumor in lower respiratory tract) Table 3.1- Treatment summary for study patients Yes No N/A Only Total Chemotherapy 999 3538 4765 167 9302 Radiotherapy 5048 846 3408 2059 9302 Surgery 1373 3165 4764 319 9302 N/A: not available based on the BCCA registry data set (They might have been treated somewhere else for example) 26% 45% 29% ≤Stage 2B >Stage 2B N/A 19% 81% SCLC (n=1811) NSCLC (n=7491) 2% 8% 38% 23% 23% 6% Overlap lesion of bronchus and lung Main Bronchus Middle lobe, bronchus or lung Lower lobe, bronchus or lung Upper lobe, bronchus or lung Bronchus or lung, unsepecified 24 3.2 Small-cell lung cancer (SCLC) In this part, there are references to 90 graphs (Graph 3.1 – Graph 3.90) and 45 tables (Table 3.5 – Table 3.49). Results are presented for 3 different subgroups of patients according to the type of therapy that they received as part of their treatment: those who received surgery, those who received radiotherapy, and those who received chemotherapy. In each subgroup, results are presented for the variables:  season of diagnosis and season of first treatment  month of diagnosis and month of first treatment  monthly mean vitamin D sunshine index (MMVDSI) Results are presented for men and women separately, and for men and women together. 3.2.1 Patients who received surgery with or without other treatments There are 30 graphs (Graph 3.1 – Graph 3.30) and 15 tables (Table 3.5 – Table 3.16) for patients who received surgery as part of their treatment. 3.2.1.1Men The survival curves associated with season of diagnosis are shown in Graphs 3.1 and 3.2; the survival curves associated with season of first treatment are shown in Graphs 3.3 and 3.4. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.5 and 3.6. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.5 and 3.6; the survival curves associated with month of first treatment are shown in Graphs 3.7 and 3.8. The log-rank tests only were significant for those patients who died because of non-lung-cancer 25 reasons. The corresponding multivariate tests and HRs are shown in Tables 3.7 and 3.8. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.9 and 3.10. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.9. None of the associations were significant. 3.2.1.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.11 and 3.12; the survival curves associated with season of first treatment are shown in Graphs 3.13 and 3.14. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.10 and 3.11. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.15 and 3.16; the survival curves associated with month of first treatment are shown in Graphs 3.17 and 3.18. The log-rank test only was significant for month of first treatment in those patients who died because of lung cancer. The corresponding multivariate tests and HRs are shown in Tables 3.12 and 3.13. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.19 and 3.20. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.14. The association only was significant in those patients who died because of lung cancer. 3.2.1.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.21 and 3.22; the survival curves associated with season of first treatment are shown in Graphs 3.23 and 26 3.24. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.15 and 3.16. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.25 and 3.26; the survival curves associated with month of first treatment are shown in Graphs 3.27 and 3.28. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.17 and 3.18. The association was only significant for month of diagnosis in those patients who died because of lung cancer. The survival curves associated with MMVDSI are shown in Graphs 3.29 and 3.30. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.19. None of the associations were significant. 3.2.2 Patients who received radiotherapy with or without other treatments There are 30 graphs (Graph 3.31 – Graph 3.60) and 15 tables (Table 3.20 – Table 3.44) for patients who received radiotherapy as part of their treatment. 3.2.2.1Men The survival curves associated with season of diagnosis are shown in Graphs 3.31 and 3.32; the survival curves associated with season of first treatment are shown in Graphs 3.33 and 3.34. The association was only significant for month of first treatment in those patients who died because of lung cancer. The corresponding multivariate tests and HRs are shown in Tables 3.20 and 3.21. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.35 and 3.36; the survival curves associated with month of first treatment are shown in Graphs 3.37 and 27 3.38. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.22 and 3.223. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.39 and 3.40. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.24. None of the associations were significant. 3.2.2.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.41 and 3.42; the survival curves associated with season of first treatment are shown in Graphs 3.43 and 3.44. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.25 and 3.26. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.45 and 3.46; the survival curves associated with month of first treatment are shown in Graphs 3.47 and 3.48. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.27 and 3.28. The test only was significant for month of first treatment in those patients who died because of lung cancer. The survival curves associated with MMVDSI are shown in Graphs 3.49 and 3.50. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.29. None of the associations were significant. 3.2.2.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.51 and 3.52; the survival curves associated with season of first treatment are shown in Graphs 3.53 and 3.54. None of the associations were significant. The corresponding multivariate tests and HRs 28 are shown in Tables 3.30 and 3.31. The test only was significant in those patients who died because of lung cancer. The survival curves associated with month of diagnosis are shown in Graphs 3.55 and 3.56; the survival curves associated with month of first treatment are shown in Graphs 3.57 and 3.58. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.32 and 3.33. The association only was significant in those patients who died because of lung cancer. The survival curves associated with MMVDSI are shown in Graphs 3.59 and 3.60. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.34. None of the associations were significant. 3.2.3 Patients who received chemotherapy with or without other treatments There are 30 graphs (Graph 3.61 – Graph 3.90) and 15 tables (Table 3.35 – Table 3.49) for patients who received chemotherapy as part of their treatment. 3.2.3.1 Men The survival curves associated with season of diagnosis are shown in Graphs 3.61 and 3.62; the survival curves associated with season of first treatment are shown in Graphs 3.63 and 3.64. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.35 and 3.36. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.65 and 3.66; the survival curves associated with month of first treatment are shown in Graphs 3.67 and 3.68. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.37 and 3.38. None of the associations were significant. 29 The survival curves associated with MMVDSI are shown in Graphs 3.69 and 3.70. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.39. None of the associations were significant. 3.2.3.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.71 and 3.72; the survival curves associated with season of first treatment are shown in Graphs 3.73 and 3.74. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.40 and 3.41. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.75 and 3.76; the survival curves associated with month of first treatment are shown in Graphs 3.77 and 3.78. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.42 and 3.43. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.79 and 3.80. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.44. None of the associations were significant. 3.2.3.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.81 and 3.82; the survival curves associated with season of first treatment are shown in Graphs 3.83 and 3.84. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.45 and 3.46. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.85 and 3.86; the survival curves associated with month of first treatment are shown in Graphs 3.87 and 30 3.88. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.47 and 3.48. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.89 and 3.90. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.49. None of the associations were significant. 3.3 Non-small-cell lung cancer (NSCLC) In this part, there are references to 90 graphs (Graph 3.91 – Graph 3.180) and 45 tables (Table 3.50 – Table 3.94). Results are presented for 3 different subgroups of patients according to the type of therapy that they received as part of their treatment: those who received surgery, those who received radiotherapy, and those who received chemotherapy. In each subgroup, results are presented for the variables:  season of diagnosis and season of first treatment  month of diagnosis and month of first treatment  monthly mean vitamin D sunshine index (MMVDSI) Results are presented for men and women separately, and for men and women together. 3.3.1 Patients who received surgery with or without other treatments There are 30 graphs (Graph 3.91 – Graph 3.120) and 15 tables (Table 3.50 – Table 3.64) for patients who received surgery as part of their treatment. 31 3.3.1.1 Men The survival curves associated with season of diagnosis are shown in Graphs 3.91 and 3.92; the survival curves associated with season of first treatment are shown in Graphs 3.93 and 3.94. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.50 and 3.51. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.95 and 3.96; the survival curves associated with month of first treatment are shown in Graphs 3.97 and 3.98. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.52 and 3.53. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.99 and 3.100. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.54. None of the associations were significant. 3.3.1.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.101 and 3.102; the survival curves associated with season of first treatment are shown in Graphs 3.103 and 3.104. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.55 and 3.56. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.105 and 3.106; the survival curves associated with month of first treatment are shown in Graphs 3.107 and 3.108. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.57 and 3.58. The association was only significant for month of first treatment in those patients who died because of non-lung-cancer reasons. 32 The survival curves associated with MMVDSI are shown in Graphs 3.109 and 3.110. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.59. The association only was significant in those patients who died because of lung cancer. 3.3.1.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.111 and 3.112; the survival curves associated with season of first treatment are shown in Graphs 3.113 and 3.114. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.60 and 3.61. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.115 and 3.116; the survival curves associated with month of first treatment are shown in Graphs 3.117 and 3.118. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.62 and 3.63. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.119 and 3.120. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.64. None of the associations were significant. 3.3.2 Patients who received radiotherapy with or without other treatments There are 30 graphs (Graph 3.121 – Graph 3.150) and 15 tables (Table 3.65 – Table 3.78) for patients who received radiotherapy as part of their treatment. 33 3.3.2.1Men The survival curves associated with season of diagnosis are shown in Graphs 3.121 and 3.122; the survival curves associated with season of first treatment are shown in Graphs 3.123 and 3.124. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.65 and 3.66. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.125 and 3.126; the survival curves associated with month of first treatment are shown in Graphs 3.127 and 3.128. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.67 and 3.68. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.129 and 3.130. The log-rank test only was significant in those patients who died because of lung cancer. The corresponding multivariate tests are shown in Table 3.69. The association only was significant in those patients who died because of lung cancer. 3.3.2.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.131 and 3.132; the survival curves associated with season of first treatment are shown in Graphs 3.133 and 3.134. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.70 and 3.71. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.135 and 3.136; the survival curves associated with month of first treatment are shown in Graphs 3.137 and 3.138. The association only was significant for month of first treatment in those patients who died because of non-lung-cancer reasons. The corresponding multivariate tests and HRs are 34 shown in Tables 3.72 and 3.73. The test only was significant for month of first treatment in those patients who died because of lung cancer. The survival curves associated with MMVDSI are shown in Graphs 3.139 and 3.140. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.74. None of the associations were significant. 3.3.2.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.141 and 3.142; the survival curves associated with season of first treatment are shown in Graphs 3.143 and 3.144. The association only was significant for season of first treatment in those patients who died because of lung cancer. The corresponding multivariate tests and HRs are shown in Tables 3.75 and 3.76. The association only was significant for season of first treatment in those patients who died because of lung cancer. The survival curves associated with month of diagnosis are shown in Graphs 3.145 and 3.146; the survival curves associated with month of first treatment are shown in Graphs 3.147 and 3.148. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.77 and 3.78. The association only was significant for month of diagnosis in those patients who died because of lung cancer. The survival curves associated with MMVDSI are shown in Graphs 3.149 and 3.150. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.79. The association only was significant in those patients who died because of lung cancer. 35 3.3.3 Patients who received chemotherapy with or without other treatments There are 30 graphs (Graph 3.151 – Graph 3.180) and 15 tables (Table 3.80 – Table 3.94) for patients who received chemotherapy as part of their treatment. 3.3.3.1 Men The survival curves associated with season of diagnosis are shown in Graphs 3.151 and 3.152; the survival curves associated with season of first treatment are shown in Graphs 3.153 and 3.154. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.80 and 3.81. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.155 and 3.156; the survival curves associated with month of first treatment are shown in Graphs 3.157 and 3.158. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.82 and 3.83. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.159 and 3.160. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.84. None of the associations were significant. 3.3.3.2 Women The survival curves associated with season of diagnosis are shown in Graphs 3.161 and 3.162; the survival curves associated with season of first treatment are shown in Graphs 3.163 and 3.164. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.85 and 3.86. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.165 and 3.166; the survival curves associated with month of first treatment are shown in Graphs 3.167 36 and 3.168. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.87 and 3.88. The association only was significant for month of first treatment in those patients who died because of lung cancer. The survival curves associated with MMVDSI index are shown in Graphs 3.169 and 3.170. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.89. None of the associations were significant. 3.3.3.3 Men and women together The survival curves associated with season of diagnosis are shown in Graphs 3.171 and 3.172; the survival curves associated with season of first treatment are shown in Graphs 3.173 and 3.174. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.90 and 3.91. None of the associations were significant. The survival curves associated with month of diagnosis are shown in Graphs 3.175 and 3.176; the survival curves associated with month of first treatment are shown in Graphs 3.177 and 3.178. None of the associations were significant. The corresponding multivariate tests and HRs are shown in Tables 3.92 and 3.93. None of the associations were significant. The survival curves associated with MMVDSI are shown in Graphs 3.179 and 3.180. None of the associations were significant. The corresponding multivariate tests are shown in Table 3.94. None of the associations were significant. 37 3.4 Summary We reviewed the results, focusing on relationships for which there was a significant multivariate association between survival and monthly mean vitamin D sunshine index for patients who died because of lung cancer. We chose this focus because the vitamin D index is the best sunshine measure in our study, and the multivariate tests adjusted for other prognostic variables. Based on this focus, we found three groups for which there was a significant relationship between survival and sunshine: 1- Female SCLC patients 2- Female NSCLC patients 3- Male NSCLC patients Tables 3.2 – 3.4 summarize the results for each group. In Table 3.2, I present the multivariate results for different treatment groups in female SCLC patients. Patients who received surgery are not very informative because there are few of them (i.e., n<30). For patients who received radiotherapy, multivariate estimates for month of first treatment, season of diagnosis and season of first treatment were significant. For patients who received chemotherapy, none of the sunshine measures were significantly associated with survival in multivariate tests. In Table 3.3, I present the multivariate results for different treatment groups in female NSCLC patients. For patients who received surgery, MMVDSI was significantly associated with survival in multivariate tests. The HR values for this variable increased for each level of exposure except for the highest category. For patients who received radiotherapy, none of the sunshine measures were significantly associated with survival in multivariate tests. For patients 38 who received chemotherapy, only the sunshine measure month of diagnosis was significantly associated with survival in multivariate tests. In Table 3.4, I present the multivariate results for different treatment groups in male NSCLC patients. For patients who received surgery, none of the sunshine measures was significantly associated with survival in multivariate tests. For patients who received radiotherapy, MMVDSI was significantly associated with survival in multivariate tests. The HR values for this variable increased for the first three categories of exposure, but not thereafter. For patients who received chemotherapy, none of the sunshine variables were significantly associated with survival in multivariate tests. 39 Table 3.2- Female patients with SCLC who died because of lung cancer: HR estimates and multivariate P-values for each treatment group. T re a tm en t HR (95% CI) for MMVDSI Multivariate P A B C D E M M V D S I M o n th o f D ia g n o si s M o n th o f fi rs t T re at m en t S ea so n o f D ia g n o si s S ea so n o f fi rs t T re at m en t S u rg er y (n = 2 5 ) 1 N/A 0.07 <0.01-1.1 N/A N/A 0.05 NS NS NS NS R a d io th er a p y (n = 1 5 7 ) 1 1.14 0.6-2.0 2.02 1.0 -3.9 1.44 0.8-2.5 1.33 0.7-2.3 NS 0.062 0.045 0.043 0.038 C h em o th er a p y (n = 3 1 0 ) 1 1.13 0.7-1.6 0.99 0.6 -1.4 0.78 0.5-1.1 1.04 0.6-1.5 NS NS 0.096 NS NS A: ≤0.406 B: 0.406-0.812 C: 0.812-1.218 D: 1.218-1.624 E: ≥1.624 HR is hazard ratio; MMVDSI is monthly mean vitamin D sunshine index; CI is confidence interval; N/A is not available; NS is not significant with p>0.1 40 Table 3.3- Female patients with NSCLC who died because of lung cancer: HR estimates and multivariate P-values for each treatment group. T re a tm en t HR (95% CI) for MMVDSI Multivariate P A B C D E M M V D S I M o n th o f D ia g n o si s M o n th o f fi rs t T re at m en t S ea so n o f D ia g n o si s S ea so n o f fi rs t T re at m en t S u rg er y (n = 4 3 1 ) 1 1.03 0.7-1.4 1.14 0.8-1.5 1.43 1.0-1.9 0.79 0.5-1.1 0.050 NS NS NS NS R a d io th er a p y (n = 1 3 7 7 ) 1 1.09 0.9-1.3 1.21 1.0-1.4 1.12 0.9-1.3 1.01 0.8-1.2 NS NS NS 0.072 NS C h em o th er a p y (n = 8 1 ) 1 0.70 0.3-1.4 1.82 0.8-4.0 2.77 1.0-7.6 0.66 0.2-2.0 NS 0.025 NS NS NS A: ≤0.406 B: 0.406-0.812 C: 0.812-1.218 D: 1.218-1.624 E: ≥1.624 HR is hazard ratio; MMVDSI is monthly mean vitamin D sunshine index; CI is confidence interval; N/A is not available; NS is not significant with p>0.1 41 Table 3.4- Male patients with NSCLC who died because of lung cancer: HR estimates and multivariate P-values for each treatment group. T re a tm en t HR (95% CI) for MMVDSI Multivariate P A B C D E M M V D S I M o n th o f D ia g n o si s M o n th o f fi rs t T re at m en t S ea so n o f D ia g n o si s S ea so n o f fi rs t T re at m en t S u rg er y (n = 8 4 7 ) 1 1.08 0.8-1.3 1.26 1.0-1.5 1.08 0.8-1.3 1.00 0.7-1.3 NS NS NS NS NS R a d io th er a p y (n = 2 6 8 7 ) 1 1.07 0.9-1.2 1.22 1.0-1.3 0.92 0.8-1.0 1.03 0.9-1.1 <0.01 0.060 NS NS NS C h em o th er a p y (n = 1 5 7 ) 1 1.14 0.6-2.0 2.02 1.0-3.9 1.44 0.8-2.5 1.33 0.7-2.3 NS 0.057 NS NS NS A: ≤0.406 B: 0.406-0.812 C: 0.812-1.218 D: 1.218-1.624 E: ≥1.624 HR is hazard ratio; MMVDSI is monthly mean vitamin D sunshine index; CI is confidence interval; N/A is not available; NS is not significant with p>0.1 42 3.5 Additional analyses Having completed the main analyses, we fitted some extra models to investigate alternative patient groups, treatment groups and variable categories. In particular, I was interested in separately considering patients age <50 years and age > 50 years. I was also interested is considering patients who received monotherapy (i.e., only chemotherapy, only radiotherapy or only surgery). Those results are summarized in sections 3.5.1 and 3.5.2. A final analysis re-evaluated the test outcomes using a correction for multiple testing. That analysis is presented in section 3.5.3. 3.5.1 Small-cell lung cancer 3.5.1.1 Patients age <50 years who received monotherapy Patients who received only surgery There were no significant results in any of the univariate or multivariate analyses. Patients who received only radiotherapy The univariate and multivariate analyses showed a significant association for season of diagnosis among females who died because of lung cancer. MMVDSI was also significantly associated with survival in univariate analysis for females who died because of lung cancer. We also performed the analyses for females and males together. The univariate and multivariate analyses for month of diagnosis and season of first treatment were significant in patients who died because of lung cancer. The multivariate analyses for month of diagnosis and season of diagnosis were significant in patients who died because of lung cancer. 43 Patients who received only chemotherapy The multivariate analyses for season of diagnosis and MMVDSI were significant among males who died because of lung cancer. The univariate and multivariate analyses for season of first treatment were significant in males who died because of lung cancer. We also performed the analyses for both females and males together. The univariate analyses for month of first treatment were significant in patients who died because of lung cancer. The multivariate analysis was significant for season of diagnosis in patients who died because of lung cancer. The univariate and multivariate analyses for season of first treatment were significant in patients who died because of lung cancer. 3.5.1.2- Patients age ≥50 years who received monotherapy Patients who received only surgery The multivariate analysis for month of diagnosis was significant in men and women combined for patients who died because of lung cancer. Patients who received only radiotherapy The multivariate analysis for season of first treatment was significant separately in males and females who died because of lung cancer. In analyses that considered males and females together, only the multivariate analysis for season of first treatment was significant. Patients who received only chemotherapy There were no significant results in any of the univariate or multivariate analyses. 3.5.1.3- Monotherapy without age categorization Patients who received only surgery There were no significant results in any of the univariate or multivariate analyses. 44 Patients who received only radiotherapy The univariate analysis for month of diagnosis and the multivariate analysis for month of first treatment were significant in females who died because of lung cancer. The univariate and multivariate analyses were significant for month of first treatment for males who died because of lung cancer. And. The univariate and multivariate analyses were significant for season of diagnosis for males who died because of lung cancer. The univariate and multivariate analyses were significant for season of first treatment for males and females separately who died because of lung cancer. We also performed the analyses for males and females together. The univariate and multivariate analyses were significant for all of the sunshine variables in patients who died because of lung cancer. Patients who received only chemotherapy The univariate analysis for month of first treatment was significant for males who died because of lung cancer. 3.5.2 Non-small-cell lung cancer 3.5.2.1 Patients age <50 years who received monotherapy Patients who received only surgery The univariate analysis showed a significant association with survival for season of diagnosis in females who died because of lung cancer. We also performed the analyses for males and females together. The univariate analyses for season of diagnosis and season of first treatment were significant in those patients who died because of lung cancer. The multivariate analysis was also significant for month of first treatment in patients who died because of lung 45 cancer. The univariate and multivariate analyses for month of diagnosis was significant in patients who died because of lung cancer. Patients who received only radiotherapy The univariate and multivariate analyses for season of first treatment were significant for patients who died because of lung cancer. Patients who received only chemotherapy There were no significant results in either univariate or multivariate analyses. 3.5.2.2 Patients age ≥50 years who received monotherapy Patients who received only surgery The only significant result was in multivariate analyses of the association of survival with MMVDSI for males and females together who died because of lung cancer. Patients who received only radiotherapy The multivariate analyses for season of first treatment were significant separately in males and females who died because of lung cancer. The multivariate analysis for MMVDSI was significant in males who died because of lung cancer. Also, the univariate analysis for month of diagnosis was significant in males who died because of lung cancer. We also performed the analyses for both females and males. The univariate and multivariate analyses for season of first treatment were significant in patients who died because of lung cancer. The multivariate analysis for MMVDSI was significant in patients who died because of lung cancer. Patients who received only chemotherapy The multiivariate analyses for month of first treatment and MMVDSI were significant in males who died because of lung cancer. 46 3.5.2.3- Monotherapy without age categorization Patients who received only surgery There were no significant results in either univariate or multivariate analyses. Patients who received only radiotherapy The univariate and multivariate analyses for season of first treatment were significant separately in females and males who died because of lung cancer. The univariate and multivariate analyses for MMVDSI were significant in males who died because of lung cancer. We also performed the analyses for females and males together. The univariate and multivariate analyses for season of first treatment were significant in patients who died because of lung cancer. The multivariate analysis for MMVDSI was significant in patients who died because of lung cancer. Patients who received only chemotherapy The multivariate analysis for season of diagnosis was significant in males who died because of lung cancer. We also performed the analyses for females and males together. The univariate and multivariate analyses for month of first treatment were significant in patients who died because of lung cancer. The multivariate analysis for month of diagnosis was significant in those patients who died because of lung cancer. (Please see A3 in appendix) 3.5.3 Multiple testing analysis I re-evaluated the statistical results adjusting for the multiple tests. I used an informal Bonferroni method for this re-evaluation, in which I multiplied the original test P-values by the number of tests performed in the main analysis. (The main analysis is the 180 tests involving 47 MMVDSI.) The only test in which the P-value remained significant after the Bonferroni re- analysis was that for males with NSCLC who received radiotherapy. 48 3.6 Main results’ graphs and tables Table 3.5- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 45 8 NS NS 3.07 0.5-18.1 3.28 0.5-18.6 1 2.44 0.4-13.5 Other causes 45 37 NS NS N<5 N<5 1 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 49 Graph 3.1- Survival according to season of diagnosis for males with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.2- Survival according to season of diagnosis for males with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 50 Table 3.6- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 45 8 NS NS 1 N<5 N<5 N<5 Other causes 45 37 NS NS 1 6.56 0.1->100 2.97 0.1-74.4 0.38 <0.01-25.2 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases. Number of caes is the cumulative number of SCLC and NSCLC before censoring for cause of death. 51 Graph 3.3- Survival according to season of first treatment for males with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.4- Survival according to season of first treatment for males with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 52 Table 3.7- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS 0.030 Multivariate P value NS NS Number of cases 45 45 Number of censored cases 8 37 H R ( 9 5 % C I) Jan 1 1 Feb n<5 N<5 Mar n<5 N<5 Apr n<5 N<5 May n<5 N<5 Jun n<5 N<5 Jul n<5 N<5 Aug n<5 N<5 Sep n<5 N<5 Oct n<5 N<5 Nov n<5 N<5 Dec n<5 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of diagnosis. NS is not significant N is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 53 Graph 3.5- Survival according to month of diagnosis for males with SCLC who received surgery: lung cancer deaths Graph 3.6- Survival according to month of diagnosis for males with SCLC who received surgery: non-lung-cancer death 54 Table 3.8- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS <0.01 Multivariate P value NS NS Number of cases 45 45 Number of censored cases 8 37 H R ( 9 5 % C I) Jan 1 1 Feb N<5 N<5 Mar N<5 N<5 Apr N<5 N<5 May N<5 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct N<5 N<5 Nov N<5 N<5 Dec N<5 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of caes is the cumulative number of SCLC and NSCLC before censoring for cause of death. 55 Graph 3.7- Survival according to month of first treatment for males with SCLC who received surgery: lung cancer deaths Graph 3.8- Survival according to month of first treatment for males with SCLC who received surgery: non-lung-cancer deaths 56 Table 3.9- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 45 8 NS NS 1 0.48 0.1-2.1 4.79 0.7-29.2 2.27 0.6- 8.0 n<5 Other causes 45 37 NS NS 1 N<5 N <5 N <5 N <5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant; N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 57 Graph 3.9- Survival according to MMVDSI for males with SCLC who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.10- Survival according to MMVDSI for males with SCLC who received surgery: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 58 Table 3.10- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 25 9 NS NS N<5 N<5 1 6.0 0.2-<100 Other causes 25 16 NS NS N<5 N<5 1 1 0.0-30.0 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 59 Graph 3.11- Survival according to season of diagnosis for females with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.12- Survival according to season of diagnosis for females with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 60 Table 3.11- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 25 9 NS NS 1 N<5 N<5 1.71 0.1-20.4) Other causes 25 16 NS NS 1 N<5 N<5 1 0.0-18.1 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 61 Graph 3.13- Survival according to season of first treatment for females with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.14- Survival according to season of first treatment for females with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 62 Table 3.12- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value <0.01 NS Multivariate P value NS NS Number of cases 25 25 Number of censored cases 9 16 H R ( 9 5 % C I) Jan 1 1 Feb N<5 N<5 Mar N<5 N<5 Apr N<5 N<5 May N<5 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct N<5 N<5 Nov N<5 N<5 Dec N<5 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 63 Graph 3.15- Survival according to month of diagnosis for females with SCLC who received surgery: lung cancer deaths Graph 3.16- Survival according to month of diagnosis for females with SCLC who received surgery: non-lung-cancer deaths 64 Table 3.13- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value <0.01 NS Multivariate P value NS NS Number of cases 25 25 Number of censored cases 9 16 H R ( 9 5 % C I) Jan 1 1 Feb N<5 N<5 Mar N<5 N<5 Apr N<5 N<5 May N<5 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct N<5 N<5 Nov N<5 N<5 Dec N<5 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 65 Graph 3.17- Survival according to month of first treatment for females with SCLC who received surgery: lung cancer deaths Graph 3.18- Survival according to month of first treatment for females with SCLC who received surgery: non-lung-cancer deaths 66 Table 3.14- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 25 9 NS 0.05 1 N<5 N<5 N<5 N<5 Other causes 25 16 NS NS 1 N<5 N<5 N<5 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 67 Graph 3.19- Survival according to MMVDSI for females with SCLC who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.20- Survival according to MMVDSI for females with SCLC who received surgery: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 68 Table 3.15- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 70 17 NS NS 1.39 (0.4-4.1) 1.72 (0.5-5.1) 1 1.16 (0.4-3.0) Other causes 70 53 NS NS N<5 N<5 1 10.33 (0.0-1.8) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 69 Graph 3.21- Survival according to season of diagnosis for male and female patients with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.22- Survival according to season of diagnosis for male and female patients with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 70 Table 3.16- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 70 17 NS NS 1 1.95 0.6-5.7 0.81 0.2-2.2 1.22 0.5-2.7 Other causes 70 53 NS NS 1 N<5 N<5 4.68 0.4-45.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 71 Graph 3.23- Survival according to season of first treatment for male and female patients with SCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.24- Survival according to season of first treatment for male and female patients with SCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 72 Table 3.17- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received surgery Treatment Surgery Cause of death Lung cancer Other causes Univariate P value NS NS Multivariate P value 0.010 NS Number of cases 70 70 Number of censored cases 17 53 H R ( 9 5 % C I) Jan 1 1 Feb 0.37 0.0-1.6 N<5 Mar 0.07 0.0-0.4 N<5 Apr 0.38 0.0-2.4 N<5 May N<5 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct 0.30 0.0-1.7 N<5 Nov N<5 N<5 Dec 0.02 0.0-0.1 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 73 Graph 3.25- Survival according to month of diagnosis for male and female patients with SCLC who received surgery: lung cancer deaths Graph 3.26- Survival according to month of diagnosis for male and female patients with SCLC who received surgery: non-lung-cancer deaths 74 Table 3.18- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received surgery Treatment Surgery Cause of death Lung cancer Other causes Univariate P value NS NS Multivariate P value NS NS Number of cases 70 70 Number of censored cases 17 53 H R ( 9 5 % C I) Jan 1 1 Feb 1.26 0.1-8.1 N<5 Mar 0.44 0.1-1.7 N<5 Apr 0.32 0.0-1.4 N<5 May N<5 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct 0.48 0.1-2.2 N<5 Nov N<5 N<5 Dec 0.10 0.0-0.6 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and month of first NS is not significant N is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 75 Graph 3.27- Survival according to month of first treatment for male and female patients with SCLC who received surgery: lung cancer deaths Graph 3.28- Survival according to month of first treatment for male and female patients with SCLC who received surgery: non-lung-cancer deaths 76 Table 3.19- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 70 17 NS NS 1 0.81 0.2-2.2 2.33 0.8-6.7 2.43 0.8-6.7 N<5 Other causes 70 53 NS NS 1 N<5 N<5 N<5 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 77 Graph 3.29- Survival according to MMVDSIfor male and female patients with SCLC who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.30- Survival according to MMVDSIfor male and female patients with SCLC who received surgery: non-lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 78 Table 3.20- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 532 68 NS 0.043 1.04 0.8-1.3 0.88 0.7-1.1 1 1.13 0.8-1.4 Other causes 532 469 NS NS 0.52 0.2-1.1 0.582 0.2-1.2 1 0.86 0.4-1.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 79 Graph 3.31- Survival according to season of diagnosis for males with SCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.32- Survival according to season of diagnosis for males with SCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 80 Table 3.21- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 433 55 NS 0.038 1 0.78 0.5-1.0 0.76 0.5-1.0 0.96 0.7-1.2 Other causes 433 381 NS NS 1 1.28 0.5-2.9 0.75 0.3-1.7 1.08 0.4-2.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death. 81 Graph 3.33- Survival according to season of first treatment for males with SCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.34- Survival according to season of first treatment for males with SCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 82 Table 3.22- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 532 532 Number of censored cases 68 469 H R ( 9 5 % C I) Jan 1 1 Feb 1.35 0.8-2.1 N<5 Mar 1.01 0.6-1.5 N<5 Apr 1.02 0.6-1.6 N<5 May 0.88 0.5-1.3 1.11 0.3-3.8 Jun 0.74 0.4-1.1 N<5 Jul 0.76 0.4-1.2 1.06 0.2-3.9 Aug 0.93 0.6-1.4 N<5 Sep 0.95 0.6-1.4 1.52 0.4-5.5 Oct 0.91 0.5-1.4 1.10 0.2-4.4 Nov 0.92 0.5-1.4 1.44 0.4-5.1 Dec 0.89 0.5-1.4 1.46 0.3-5.5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of diagnosis. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 83 Graph 3.35- Survival according to month of diagnosis for males with SCLC who received radiotherapy: lung cancer deaths Graph 3.36- Survival according to month of diagnosis for males with SCLC who received radiotherapy: non-lung-cancer deaths 84 Table 3.23- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 433 433 Number of censored cases 55 381 H R ( 9 5 % C I) Jan 1 1 Feb 1.05 0.6-1.7 0.80 0.1-3.9 Mar 1.28 0.7-2.0 N<5 Apr 0.89 0.5-1.46 0.77 0.1-3.6 May 1.15 0.6-1.9 N<5 Jun 0.71 0.4-1.2 N<5 Jul 0.95 0.5-1.6 1.73 0.4-7.2 Aug 1.05 0.6-1.7 1.92 0.4-8.4 Sep 0.67 0.4-1.0 0.51 0.1-2.3 Oct 0.79 0.4-1.3 N<5 Nov 0.92 0.5-1.5 N<5 Dec 1.12 0.6-1.9 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 85 Graph 3.37- Survival according to month of first treatment for males with SCLC who received radiotherapy: lung cancer deaths Graph 3.38- Survival according to month of first treatment for males with SCLC who received radiotherapy: non-lung-cancer deaths 86 Table 3.24- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 531 68 NS NS 1 0.93 0.7-1.2 1.01 0.7-1.3 0.85 0.6-1.1 0.75 0.5-1.0 Other causes 531 468 NS NS 1 0.70 0.3-1.5 0.57 0.2-1.4 0.69 0.3-1.4 0.81 0.3-1.8 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 87 Graph 3.39- Survival according to MMVDSI for males with SCLC who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.40- Survival according to monthly MMVDSI for males with SCLC who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 88 Table 3.25- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 416 50 0.043 NS 1.19 0.8-1.6 0.86 0.6-1.1 1 1.32 0.9-1.7 Other causes 416 374 NS NS 1.02 0.3-2.7 1.41 0.5-3.3 1 1.05 0.3-2.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 89 Graph 3.41- Survival according to season of diagnosis for females with SCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.42- Survival according to season of diagnosis for females with SCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 90 Table 3.26- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 348 40 NS NS 1 0.67 0.48-9.4 0.68 0.5-0.9 0.69 0.5-0.9 Other causes 348 314 NS NS 1 1.12 0.31-3.9 1.31 0.4-3.6 1.40 0.4-4.0 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 91 Graph 3.43- Survival according to season of first treatment for females with SCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.44- Survival according to season of first treatment for females with SCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 92 Table 3.27- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.062 NS Number of of cases 416 416 Number of censored cases 50 374 H R ( 9 5 % C I) Jan 1 1 Feb 1.85 1.0-3.2 N<5 Mar 1.63 0.9-2.9 N<5 Apr 1.23 0.6-2.2 N<5 May 1.14 0.6-2.0 N<5 Jun 0.84 0.4-1.5 0.71 0.2-2.4 Jul 1.35 0.7-2.5 n<5 Aug 0.89 0.5-1.6 n<5 Sep 1.27 0.7-2.3 n<5 Oct 1.00 0.5-1.7 n<5 Nov 1.09 0.6-1.9 n<5 Dec 1.44 0.7-2.7 n<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 93 Graph 3.45- Survival according to month of diagnosis for females with SCLC who received radiotherapy: lung cancer deaths Graph 3.46- Survival according to month of diagnosis for females with SCLC who received radiotherapy: non-lung-cancer deaths 94 Table 3.28- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.045 NS Number of cases 348 348 Number of censored cases 40 314 H R ( 9 5 % C I) Jan 1 1 Feb 1.05 0.5-2.0 N<5 Mar 1.61 0.8-3.0 N<5 Apr 1.51 0.8-2.7 N<5 May 0.97 0.6-1.8 N<5 Jun 0.91 0.4-1.7 1.31 0.0-17.9 Jul 0.83 0.4-1.6 n<5 Aug 1.06 0.5-2.0 N<5 Sep 0.63 0.3-1.2 N<5 Oct 1.02 0.5-1.9 N<5 Nov 1.29 0.6-2.4 N<5 Dec 0.74 0.3-1.4 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 95 Graph 3.47- Survival according to month of first treatment for females with SCLC who received radiotherapy: lung cancer deaths Graph 3.48- Survival according to month of first treatment for females with SCLC who received radiotherapy: non-lung-cancer deaths 96 Table 3.29- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received radiotherapy C au se o f D ea th T o ta l n u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 414 49 NS NS 1 0.98 0.7-1.3 0.95 0.6-1.3 0.75 0.5-1.0 0.80 0.5-1.1 Other causes 414 373 NS NS 1 1.14 0.4-3.1 1.42 0.5-3.6 0.96 0.3-2.6 1.40 0.4-4.3 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 97 Graph 3.49- Survival according to MMVDSI for females with SCLC who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.50- Survival according to MMVDSI for females with SCLC who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Table 98 3.30- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 948 118 NS 0.041 1.07 0.8-1.2 0.89 0.7-1.0 1 1.19 0.9-1.4 Other causes 948 843 NS NS 0.84 0.4-1.4 0.85 0.5-1.4 1 1.03 0.5-1.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 99 Graph 3.51- Survival according to season of diagnosis for male and female SCLC patients who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.52- Survival according to season of diagnosis for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 100 Table 3.31- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 782 96 NS <0.01 1 0.78 0.6-0.9 0.73 0.6-0.9 0.85 0.6-1.0 Other causes 782 696 NS NS 1 1.28 0.6-2.4 1.06 0.5-1.9 1.12 0.5-2.1 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 101 Graph 3.53- Survival according to season of first treatment for male and female SCLC patients who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.54- Survival according to season of first treatment for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 102 Table 3.32- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.033 NS Number of cases 948 948 Number of censored cases 118 843 H R ( 9 5 % C I) Jan 1 1 Feb 1.53 0.1-2.1 N<5 Mar 1.20 0.8-1.6 0.64 0.2-1.6 Apr 1.08 0.7-1.5 N<5 May 0.91 0.6-1.3 0.67 0.2-1.6 Jun 0.78 0.5-1.1 0.74 0.3-1.8 Jul 0.99 0.6-1.4 0.72 0.2-1.8 Aug 0.92 0.6-1.2 0.33 0.1-0.9 Sep 1.02 0.7-1.4 0.84 0.3-2.0 Oct 0.97 0.6-1.3 0.45 0.1-1.1 Nov 1.00 0.7-1.4 0.84 0.3-2.0 Dec 1.07 0.7-1.5 0.68 0.2-1.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 103 Graph 3.55- Survival according to month of diagnosis for male and female SCLC patients who received radiotherapy: lung cancer deaths Graph 3.56- Survival according to month of diagnosis for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths 104 Table 3.33- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.017 NS Number of cases 781 781 Number of censored cases 95 695 H R ( 9 5 % C I) Jan 1 1 Feb 1.12 0.7-1.6 1.87 0.5-6.5 Mar 1.41 0.9-2.0 2.16 0.5-7.8 Apr 1.21 0.8-1.7 1.75 0.4-6.2 May 1.05 0.7-1.5 N<5 Jun 0.80 0.5-1.2 1.02 0.2-3.9 Jul 0.96 0.6-1.4 2.92 0.8-10.1 Aug 1.13 0.7-1.6 2.63 0.6-10.0 Sep 0.69 0.4-1.0 1.22 0.3-4.1 Oct 0.92 0.6-1.3 1.62 0.4-6.0 Nov 1.15 0.7-1.6 2.05 0.5-7.7 Dec 0.96 0.6-1.4 1.80 0.5-6.5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 105 Graph 3.57- Survival according to month of first treatment for male and female SCLC patients who received radiotherapy: lung cancer deaths Graph 3.58- Survival according to month of first treatment for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths 106 Table 3.34- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 945 117 NS 0.097 1 0.97 0.7-1.1 0.95 0.7-1.1 0.80 0.6-0.9 0.77 0.6-0.9 Other causes 945 841 NS NS 1 0.87 0.4-1.5 0.86 0.4-1.6 0.86 0.4-1.5 0.91 0.4-1.7 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and chemotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 107 Graph 3.59- Survival according to MMVDSI for male and female SCLC patients who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.60- Survival according to MMVDSI for male and female SCLC patients who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 108 Table 3.35- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 450 58 NS NS 1.17 0.8-1.5 1.24 0.9-1.7 1 1.25 0.9-1.7 Other causes 450 398 NS NS 0.62 0.2-1.5 1.40 0.6-3.2 1 1.19 0.4-2.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 109 Graph 3.61- Survival according to season of diagnosis for males with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.62- Survival according to season of diagnosis for males with SCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 110 Table 3.36- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 448 57 NS NS 1 0.95 0.7-1.2 0.92 0.69-1.12 1.03 0.77-1.37 Other causes 448 396 NS NS 1 1.9 0.7-4.6 1.51 0.6-3.6 1.98 0.8-4.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 111 Graph 3.63- Survival according to season of first treatment for males with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.64- Survival according to season of first treatment for males with SCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 112 Table 3.37- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 450 450 Number of censored cases 58 398 H R ( 9 5 % C I) Jan 1 1 Feb 0.83 0.5-1.3 N<5 Mar 0.79 0.5-1.2 0.47 0.1-1.9 Apr 1.09 0.6-1.8 N<5 May 0.77 0.4-1.2 0.66 0.1-2.6 Jun 0.92 0.5-1.5 N<5 Jul 1.08 0.6->100 1.73 0.4-6.3 Aug 0.73 0.4-1.2 1.04 0.2-3.7 Sep 0.77 0.4-1.2 0.78 0.1-3.1 Oct 0.97 0.5-1.6 N<5 Nov 0.53 0.3-0.9 0.62 0.1-2.7 Dec 0.88 0.5-1.4 1.27 0.2-5.5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 113 Graph 3.65- Survival according to month of diagnosis for males with SCLC who received chemotherapy: lung cancer deaths Graph 3.66- Survival according to month of diagnosis for males with SCLC who received chemotherapy: non-lung-cancer deaths 114 Table 3.38- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 448 448 Number of censored cases 57 396 H R ( 9 5 % C I) Jan 1 1 Feb 0.79 0.4-1.3 N<5 Mar 1.00 0.6-1.6 0.93 0.2-4.0 Apr 0.58 0.3-1.0 N<5 May 1.05 0.6-1.7 0.26 0.0-2.6 Jun 0.67 0.3-1.1 0.52 0.1-2.5 Jul 0.92 0.5-1.5 1.21 0.2-5.3 Aug 0.86 0.5-1.4 1.16 0.2-5.2 Sep 0.69 0.4-1.1 0.82 0.2-3.4 Oct 1.04 0.6-1.8 N<5 Nov 0.65 0.3-1.1 0.51 0.0-52.7 Dec 0.75 0.4-1.3 0.98 0.2-4.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 115 Graph 3.67- Survival according to month of first treatment for males with SCLC who received chemotherapy: lung cancer deaths Graph 3.68- Survival according to month of first treatment for males with SCLC who received chemotherapy: non-lung-cancer deaths 116 Table 3.39- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 450 58 NS NS 1 0.70 0.7-1.4 0.38 0.8-1.5 0.69 0.7-1.4 0.65 0.7-1.4 Other causes 450 398 NS NS 1 0.67 0.2-1.6 0.59 0.2-1.6 0.99 0.4-2.2 1.21 0.5-2.8 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 117 Graph 3.69- Survival according to MMVDSI for males with SCLC who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.70- Survival according to MMVDSI for males with SCLC who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 118 Table 3.40- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 311 38 NS NS 1.01 0.7-1.4 0.91 0.6-1.2 1 1.01 0.7-1.4 Other causes 311 278 NS NS 0.92 0.2-2.8 1.42 0.4-4.4 1 0.88 02-2.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 119 Graph 3.71- Survival according to season of diagnosis for females with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.72- Survival according to season of diagnosis for females with SCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 120 Table 3.41- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring Summer Fall Winter Lung cancer 308 37 NS NS 1 0.84 0.5-1.2 0.86 0.6-1.1 0.73 0.5-1.0 Other causes 308 276 NS NS 1 1.37 0.4-4.2 0.85 0.2-2.4 1.17 0.4-3.1 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 121 Graph 3.73- Survival according to season of first treatment for females with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.74- Survival according to season of first treatment for females with SCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 122 Table 3.42- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 311 311 Number of censored cases 38 278 H R ( 9 5 % C I) Jan 1 1 Feb 1.89 1.0-3.5 N<5 Mar 1.71 0.9-3.1 N<5 Apr 1.35 0.7-2.4 N<5 May 1.01 0.5-1.8 N<5 Jun 1.02 0.5-1.9 N<5 Jul 1.51 0.7-2.9 N<5 Aug 1.27 0.6-2.3 N<5 Sep 1.39 0.7-2.6 N<5 Oct 1.34 0.7-2.4 N<5 Nov 1.34 0.7-2.4 N<5 Dec 1.40 0.7-2-8 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of diagnosis NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 123 Graph 3.75- Survival according to month of diagnosis for females with SCLC who received chemotherapy: lung cancer deaths Graph 3.76- Survival according to month of diagnosis for females with SCLC who received chemotherapy: non-lung-cancer deaths 124 Table 3.43- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.096 NS Number of cases 308 308 Number of censored cases 37 276 H R ( 9 5 % C I) Jan 1 1 Feb 0.70 0.3-1.3 N<5 Mar 1.46 0.7-2.7 N<5 Apr 1.11 0.6-1.9 N<5 May 0.74 0.4-1.3 N<5 Jun 0.74 0.3-1.4 N<5 Jul 1.10 0.5-2.2 N<5 Aug 0.97 0.5-1.8 N<5 Sep 0.64 0.3-1.1 N<5 Oct 1.37 0.7-2.5 N<5 Nov 1.11 0.6-2.0 N<5 Dec 0.65 0.3-1.2 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 125 Graph 3.77- Survival according to month of first treatment for females with SCLC who received chemotherapy: lung cancer deaths Graph 3.78- Survival according to month of first treatment for females with SCLC who received chemotherapy: non-lung-cancer deaths 126 Table 3.44- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 310 38 NS NS 1 1.13 0.7-1.6 0.99 0.6-1.4 0.78 0.5-1.1 1.04 0.6-1.5 Other causes 310 277 NS NS 1 1.32 0.3-4.6 1.10 0.3-3.3 1.30 0.4-3.9 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 127 Graph 3.79- Survival according to MMVDSI for females with SCLC who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.80- Survival according to MMVDSI for females with SCLC who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 128 Table 3.45- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 761 96 NS NS 1.07 0.8-1.3 1.08 0.8-1.3 1 1.15 0.9-1.4 Other causes 761 676 NS NS 0.83 0.4-1.5 1.36 0.7-2.5 1 1.10 0.5-2.0 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 129 Graph 3.81- Survival according to season of diagnosis for male and female patients with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.82- Survival according to season of diagnosis for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 130 Table 3.46- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 757 95 NS NS 1 0.94 0.7-1.1 0.94 0.7-1.1 0.92 0.7-1.1 Other causes 757 673 NS NS 1 1.89 0.9-3.7 1.29 0.6-2.3 1.56 0.8-2.9 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 131 Graph 3.83- Survival according to season of first treatment for male and female patients with SCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.84- Survival according to season of first treatment for male and female patients with SCLC who received chemotherapy: non-lung-cancer death (A is spring, B is summer, C is fall and D is winter) 132 Table 3.47- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 761 761 Number of censored cases 96 676 H R ( 9 5 % C I) Jan 1 1 Feb 1.15 0.7-1.6 0.54 0.1-1.7 Mar 1.03 0.7-1.4 0.55 0.1-1.5 Apr 1.16 0.7-1.7 N<5 May 0.87 0.6-1.2 0.78 0.2-2.0 Jun 0.97 0.6-1.4 0.74 0.2-2.0 Jul 1.25 0.8-1.8 1.62 0.5-4.6 Aug 0.90 0.6-1.3 0.86 0.3-2.2 Sep 0.95 0.6-1.4 0.68 0.2-1.8 Oct 1.12 0.7-1.6 0.84 0.2-2.5 Nov 0.80 0.5-1.1 0.70 0.2-1.9 Dec 1.16 0.7-1.7 0.88 0.2-2.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of diagnosis. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 133 Graph 3.85- Survival according to month of diagnosis for male and female patients with SCLC who received chemotherapy: lung cancer deaths Graph 3.86- Survival according to month of diagnosis for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths 134 Table 3.48- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with SCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 756 756 Number of censored cases 94 672 H R ( 9 5 % C I) Jan 1 1 Feb 0.82 0.5-1.2 1.50 0.4-5.0 Mar 1.12 0.7-1.6 1.95 0.5-6.8 Apr 0.78 0.5-1.1 0.63 0.1-2.4 May 0.94 0.6-1.4 N<5 Jun 0.73 0.4-1.1 1.22 0.3-4.4 Jul 1.00 0.6-1.5 1.94 0.5-7.3 Aug 0.89 0.6-1.3 2.28 0.6-8.3 Sep 0.70 0.4-1.0 1.12 0.3-3.7 Oct 1.17 0.7-1.7 1.75 0.4-6.5 Nov 0.89 0.6-1.3 N<5 Dec 0.75 0.4-1.1 1.59 0.4-5.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and month of first treatment. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 135 Graph 3.87- Survival according to month of first treatment for male and female patients with SCLC who received chemotherapy: lung cancer deaths Graph 3.88- Survival according to month of first treatment for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths 136 Table 3.49- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with SCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 760 96 NS NS 1 1.09 0.8-1.3 1.05 0.8-1.3 0.93 0.7-1.1 1.11 0.8-1.4 Other causes 760 675 NS NS 1 0.87 0.4-1.6 0.77 0.3-1.5 1.08 0.5-1.9 1.42 0.7-2.8 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (surgery and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 137 Graph 3.89- Survival according to MMVDSIfor male and female patients with SCLC who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.90- Survival according to MMVDSI for male and female patients with SCLC who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 138 Table 3.50- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 860 210 NS NS 1.11 0.9-1.3 0.99 0.7-1.2 1 0.92 0.7-1.1 Other causes 860 690 NS NS 0.90 0.5-1.4 1.05 0.6-1.6 1 1.05 0.6-1.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 139 Graph 3.91- Survival according to season of diagnosis for males with NSCLC who received surgery: lung cancer deaths only (A is spring, B is summer, C is fall and D is winter) Graph 3.92- Survival according to season of diagnosis for males with NSCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 140 Table 3.51- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 855 209 NS NS 1 0.90 0.7-1.1 0.94 0.7-1.1 0.94 0.7-1.1 Other causes 855 690 NS NS 1 1.2 0.7-1.9 1.32 0.8-2.1 1.57 1.0-2.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 141 Graph 3.93- Survival according to season of first treatment for males with NSCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.94- Survival according to season of first treatment for males with NSCLC who received surgery: non-lung-deaths (A is spring, B is summer, C is fall and D is winter) 142 Table 3.52- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 860 860 Number of censored cases 210 690 H R ( 9 5 % C I) Jan 1 1 Feb 1.11 0.7-1.7 0.1.08 0.5-2.3 Mar 1.26 0.8-1.8 0.72 0.3-1.4 Apr 1.36 0.9-2.0 0.81 0.3-1.7 May 1.17 0.7-1.7 0.92 0.4-1.8 Jun 1.04 0.6-1.5 1.07 0.5-2.1 Jul 1.02 0.6-1.5 0.81 0.3-1.7 Aug 1.27 0.8-1.8 0.88 4.1-1.8 Sep 1.32 0.8-1.9 0.62 0.2-1.4 Oct 0.95 0.6-1.4 1.49 0.7-3.0 Nov 1.10 0.7-1.6 0.69 0.3-1.4 Dec 1.03 0.6-1.5 0.81 0.4-1.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of diagnosis. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 143 Graph 3.95- Survival according to month of diagnosis for males with NSCLC who received surgery: lung cancer deaths Graph 3.96- Survival according to month of diagnosis for males with NSCLC who received surgery: non-lung-cancer deaths 144 Table 3.53- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 855 855 Number of censored cases 209 685 H R ( 9 5 % C I) Jan 1 1 Feb 1.09 0.7-1.6 1.28 0.6-2.5 Mar 1.02 0.7-1.5 0.59 0.2-1.2 Apr 1.42 0.9-2.0 0.96 0.4-2.2 May 1.08 0.7- 1.5 0.88 0.4-1.8 Jun 1.10 0.7-1.6 1.11 0.5-2.2 Jul 0.92 0.6-1.3 0.87 0.4-1.8 Aug 1.16 0.7-1.7 1.05 0.5-2.2 Sep 1.16 0.7-1.7 1.23 0.5-2.6 Oct 1.06 0.7-1.5 1.10 0.5-2.3 Nov 1.08 0.7-1.5 0.86 0.4-1.8 Dec 1.12 0.7-1.6 1.36 0.6-0.2 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of first treatment. NS is not significantNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 145 Graph 3.97- Survival according to month of first treatment for males with NSCLC who received surgery: lung cancer deaths Graph 3.98- Survival according to month of first treatment for males with NSCLC who received surgery: non-lung-cancer deaths 146 Table 3.54- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 847 205 NS NS 1 1.08 0.8-1.3 1.26 1.0-1.5 1.08 0.8-1.3 1.00 0.7-1.3 Other causes 847 678 NS NS 1 1.12 0.7-1.7 0.79 0.4-1.3 1.04 0.6-1.6 0.99 0.6-1.6 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 147 Graph 3.99- Survival according to MMVDSI for males with NSCLC who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.100- Survival according to MMVDSI for males with NSCLC who received surgery: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 148 Table 3.55- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 443 104 NS NS 1.2 0.9-1.1 1.09 0.8-1.4 1 1.00 0.7-1.3 Other causes 443 372 NS NS 0.91 0.4-1.8 1.10 0.5-2.0 1 0.63 0.3-1.2 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 149 Graph 3.101- Survival according to season of diagnosis for females with NSCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.102- Survival according to season of diagnosis for females with NSCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 150 Table 3.56- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 439 102 NS NS 1 0.88 0.6-1.2 0.88 0.6-1.2 0.85 0.6-1.1 Other causes 439 372 NS NS 1 0.96 0.4-2.0 1.26 0.6-2.6 0.66 0.2-1.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 151 Graph 3.103- Survival according to season of first treatment for females with NSCLC who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.104- Survival according to season of first treatment for females with NSCLC who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 152 Table 3.57- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 443 443 Number of censored cases 104 372 H R ( 9 5 % C I) Jan 1 1 Feb 0.71 0.4-1.2 0.30 0.0-1.0 Mar 1.22 0.7-2.1 N<5 Apr 1.07 0.6-1.7 0.75 0.2-2.2 May 0.96 0.5-1.6 N<5 Jun 0.82 0.4-1.4 1.07 0.3-3.0 Jul 0.66 0.3-1.1 0.96 0.3-2.6 Aug 1.52 0.9-2.4 0.42 0.1-1.5 Sep 0.96 0.5-1.5 1.21 0.4-3.2 Oct 0.73 0.4-1.2 0.69 0.2-2.0 Nov 0.91 0.5-1.5 0.42 0.1-1.5 Dec 0.85 0.5-1.4 0.19 0.0-0.8 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 153 Graph 3.105- Survival according to month of diagnosis for females with NSCLC who received surgery: lung cancer deaths Graph 3.106- Survival according to month of diagnosis for females with NSCLC who received surgery: non-lung-cancer deaths 154 Table 3.58- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS 0.046 Number of cases 439 439 Number of censored cases 102 369 H R ( 9 5 % C I) Jan 1 1 Feb 1.12 0.6-1.9 N<5 Mar 1.03 0.5-1.8 1.73 0.5-5.3 Apr 1.44 0.8-2.4 N<5 May 1.24 0.6-2.2 N<5 Jun 1.39 0.7-2.4 0.94 0.3-2.9 Jul 0.82 0.4-1.4 1.59 0.5-4.4 Aug 1.23 0.7-2.0 N<5 Sep 1.43 0.8-2.4 2.59 0.9-6.7 Oct 1.13 0.6-1.9 0.96 0.3-2.7 Nov 0.96 0.5-1.6 0.78 0.2-2.2 Dec 1.22 0.7-2.1 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 155 Graph 3.107- Survival according to month of first treatment for females with NSCLC who received surgery: lung cancer deaths Graph 3.108- Survival according to month of first treatment for females with NSCLC who received surgery: non-lung-cancer deaths 156 Table 3.59- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 431 102 NS 0.050 1 1.03 0.7-1.4 1.14 0.8-1.5 1.43 1.0-1.9 0.79 0.5-1.1 Other causes 431 361 NS NS 1 1.50 0.6-3.4 2.08 1.0-4.1 1.43 0.6-3.1 2.09 0.9-4.5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 157 Graph 3.109- Survival according to MMVDSI for females with NSCLC who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.110- Survival according MMVDSI for females with NSCLC who received surgery: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 158 Table 3.60- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring Summer Fall Winter Lung cancer 1303 314 NS NS 1.15 0.9-1.3 1.02 0.8-1.2 1 0.94 0.7-1.1 Other causes 1303 1062 NS NS 0.96 0.6-1.3 1.07 0.7-1.5 1 0.92 0.6-1.3 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 159 Graph 3.111- Survival according to season of diagnosis for male and female NSCLC patients who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.112- Survival according to season of diagnosis for male and female NSCLC patients who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 160 Table 3.61- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 1294 311 NS NS 1 0.90 0.7-1.0 0.95 0.7-1.1 0.90 0.7-1.0 Other causes 1294 1054 NS NS 1 1.18 0.8-1.7 1.34 0.9-1.9 1.20 0.8-1.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 161 Graph 3.113- Survival according to season of first treatment for male and female NSCLC patients who received surgery: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.114- Survival according to season of first treatment diagnosis for male and female NSCLC patients who received surgery: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 162 Table 3.62- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.067 NS Number of cases 1303 1303 Number of censored cases 314 1062 H R ( 9 5 % C I) Jan 1 1 Feb 0.89 0.6-1.2 0.69 0.3-1.3 Mar 1.20 0.8-1.6 0.79 0.4-1.4 Apr 1.24 0.9-1.6 0.83 0.4-1.5 May 1.06 0.7-1.4 0.81 0.4-1.4 Jun 0.95 0.6-1.3 1.06 0.6-1.8 Jul 0.85 0.6-1.1 0.84 0.4-1.5 Aug 1.30 0.9-1.7 0.82 0.4-1.5 Sep 1.16 0.8-1.5 0.80 0.4-1.4 Oct 0.85 0.6-1.1 1.18 0.6-2.0 Nov 1.02 0.7-1.3 0.60 0.3-1.1 Dec 0.95 0.6-1.2 0.65 0.3-1.1 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of diagnosis. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 163 Graph 3.115- Survival according to month of diagnosis for male and female NSCLC patients who received surgery: lung cancer deaths Graph 3.116- Survival according to month of diagnosis for male and female NSCLC patients who received surgery: non-lung-cancer deaths 164 Table 3.63- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received surgery Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 1294 1294 Number of censored cases 311 1054 H R ( 9 5 % C I) Jan 1 1 Feb 1.09 0.8-1.4 0.91 0.5-1.6 Mar 1.01 0.7-1.3 0.78 0.4-1.4 Apr 1.40 1.0-1.8 0.87 0.4-1.7 May 1.11 0.8-1.5 0.86 0.4-1.6 Jun 1.15 0.8-1.5 1.09 0.6-1.9 Jul 0.90 0.6-1.2 1.07 0.5-1.9 Aug 1.15 0.8-1.5 0.82 0.4-1.5 Sep 1.24 0.9-1.7 1.63 0.9-2.9 Oct 1.10 0.8-1.5 1.03 0.5-1.8 Nov 1.05 0.7-1.4 0.91 0.5-1.6 Dec 1.10 0.8-1.5 1.14 0.6-2.0 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and month of first treatment. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 165 Graph 3.117- Survival according to month of first treatment for male and female NSCLC patients who received surgery: lung cancer deaths Graph 3.118- Survival according to month of first treatment for male and female NSCLC patients who received surgery: non-lung-cancer deaths 166 Table 3.64- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with NSCLC who received surgery C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 1278 307 0.085 0.052 1 1.06 0.8-1.2 1.22 1.0-1.4 1.18 0.9-1.4 0.92 0.7-1.1 Other causes 1278 1039 NS NS 1 1.26 0.8-1.8 1.05 0.7-1.5 1.18 0.8-1.7 1.15 0.7-1.7 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and radiotherapy), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 167 Graph 3.119- Survival according to MMVDSI for male and female NSCLC patients who received surgery: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.120- Survival according to MMVDSI for male and female NSCLC patients who received surgery: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 168 Table 3.65- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 2706 311 NS NS 1.03 0.9-1.16 0.92 0.8-1.0 1 0.97 0.8-1.0 Other causes 2706 2419 NS NS 0.85 0.5-1.2 1.04 0.7-1.4 1 1.02 0.7-1.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 169 Graph 3.121- Survival according to season of diagnosis for males with NSCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.122- Survival according to season of diagnosis for males with NSCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 170 Table 3.66- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 2251 258 NS NS 1 0.92 0.8-1.0 0.88 0.7-1.0 0.94 0.8-1.0 Other causes 2251 2013 NS NS 1 1.09 0.7-1.6 1.30 0.8-1.8 1.22 0.8-1.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 171 Graph 3.123- Survival according to season of first treatment for males with NSCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.124- Survival according to season of first treatment for males with NSCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 172 Table 3.67- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.060 NS Number of cases 2706 2706 Number of censored cases 311 2419 H R ( 9 5 % C I) Jan 1 1 Feb 1.04 0.8-1.2 1.26 0.7-2.1 Mar 1.12 0.9-1.3 0.91 0.5-1.6 Apr 1.31 1.0-1.6 1.22 0.6-2.2 May 1.05 0.8-1.2 0.83 0.4-1.5 Jun 1.02 0.8-1.2 1.28 0.7-2.1 Jul 1.11 0.9-1.3 0.94 0.5-1.7 Aug 0.94 0.7-1.1 1.28 0.7-2.2 Sep 1.22 0.9-1.5 1.37 0.7-2.5 Oct 1.10 0.9-1.3 1.30 0.7-2.3 Nov 1.02 0.8-1.2 0.90 0.5-1.5 Dec 1.21 0.9-1.4 1.24 0.7-2.1 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of diagnosis. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 173 Graph 3.125- Survival according to month of diagnosis for males with NSCLC who received radiotherapy: lung cancer deaths Graph 3.126- Survival according to month of diagnosis for males with NSCLC who received radiotherapy: non-lung-cancer deaths 174 Table 3.68- Hazard ratio (HR) and 95% confidence interval (CI) according to month of first treatment for males with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 2251 2251 Number of censored cases 258 2013 H R ( 9 5 % C I) Jan 1 1 Feb 0.97 0.7-1.2 1.39 0.7-2.5 Mar 1.04 0.8-1.2 0.88 0.4-1.7 Apr 1.04 0.8-1.2 0.94 0.4-1.8 May 1.07 0.8-1.3 1.16 0.5-2.3 Jun 0.89 0.7-1.1 1.00 0.5-1.9 Jul 0.99 0.8-1.2 1.15 0.6-2.1 Aug 0.99 0.8-1.2 1.08 0.5-2.0 Sep 0.81 0.6-1.0 1.18 0.6-2.2 Oct 1.03 0.8-1.2 1.44 0.7-2.7 Nov 0.95 0.7-1.1 1.24 0.6-2.2 Dec 0.97 0.7-1.2 1.17 0.6-1.2 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of first treatment. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 175 Graph 3.127- Survival according to month of first treatment for males with NSCLC who received radiotherapy: lung cancer deaths Graph 3.128- Survival according to month of first treatment for males with NSCLC who received radiotherapy: non-lung-cancer deaths 176 Table 3.69- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 2687 306 0.024 <0.01 1 1.07 0.9-1.2 1.22 1.0-1.3 0.92 0.8-1.0 1.03 0.9-1.1 Other causes 2687 2402 NS NS 1 0.92 0.6-1.3 1.13 0.7-1.6 0.90 0.6-1.2 1.11 0.7-1.5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and to monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 177 Graph 3.129- Survival according to MMVDSI for males with NSCLC who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.130- Survival according to MMVDSI for males with NSCLC who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 178 Table 3.70- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 1394 201 0.089 NS 1.10 0.9-1.3 1.0 0.8-1.1 1 0.88 0.7-1.0 Other causes 1394 1221 NS NS 0.95 0.5-1.5 1.06 0.6-1.6 1 1.18 0.7-1.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 179 Graph 3.131- Survival according to season of diagnosis for females with NSCLC who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.132- Survival according to season of diagnosis for females with NSCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 180 Table 3.71- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 1175 161 NS NS 1 0.98 0.8-1.1 0.87 0.7-1.0 0.88 0.7-1.0 Other causes 1175 1037 NS NS 1 0.91 0.5-1.5 0.99 0.6-1.6 1.08 0.6-1.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 181 Graph 3.133- Survival according to season of first treatment for females with NSCLC who received radiotherapy: lung cancer death (A is spring, B is summer, C is fall and D is winter) Graph 3.134- Survival according to season of first treatment for females with NSCLC who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 182 Table 3.72- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 1394 1394 Number of censored cases 201 1221 H R ( 9 5 % C I) Jan 1 1 Feb 0.90 0.6-1.2 0.53 0.2-1.0 Mar 1.34 1.0-1.7 1.02 0.5-2.0 Apr 1.29 0.9-1.6 0.73 0.3-1.4 May 1.09 0.8-1.4 0.35 0.1-0.8 Jun 1.10 0.8-1.4 0.54 0.2-1.1 Jul 1.00 0.7-1.3 0.69 0.3-1.3 Aug 1.26 0.9-1.6 1.01 0.5-1.9 Sep 1.16 0.8-1.5 0.63 0.3-1.2 Oct 0.99 0.7-1.3 0.69 0.3-1.3 Nov 1.20 0.9-1.6 0.76 0.3-1.5 Dec 1.04 0.7-1.3 0.93 0.4-1.7 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of diagnosis. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 183 Graph 3.135- Survival according to month of diagnosis for females with NSCLC who received radiotherapy: lung cancer deaths Graph 3.136- Survival according to month of diagnosis for females with NSCLC who received radiotherapy: non-lung-cancer deaths 184 Table 3.73- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS <0.01 Multivariate P value NS NS Number of cases 1175 1175 Number of censored cases 161 1037 H R ( 9 5 % C I) Jan 1 1 Feb 0.92 0.6-1.2 0.49 0.2-1.1 Mar 1.00 0.7-1.3 0.72 0.3-1.6 Apr 1.23 0.9-1.6 0.50 0.2-1.1 May 1.11 0.8-1.5 0.73 0.3-1.5 Jun 1.30 0.9-1.7 0.43 0.1-1.1 Jul 1.01 0.7-1.3 0.87 0.4-1.7 Aug 1.05 0.7-1.4 0.46 0.2-1.0 Sep 1.12 0.8-1.5 1.19 0.6-2.3 Oct 0.89 0.6-1.2 0.23 0.0-0.6 Nov 0.94 0.6-1.2 0.70 0.3-1.4 Dec 1.06 0.7-1.4 0.64 0.2-1.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of first treatment. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 185 Graph 3.137- Survival according to month of first treatment for females with NSCLC who received radiotherapy: lung cancer deaths Graph 3.138- Survival according to month of first treatment for females with NSCLC who received radiotherapy: non-lung-cancer deaths 186 Table 3.74- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 1377 198 NS NS 1 1.09 0.9-1.3 1.21 1.0-1.4 1.12 0.9-1.3 1.01 0.8-1.2 Other causes 1377 1206 NS NS 1 1.05 0.6-1.6 0.84 0.5-1.3 0.85 0.5-1.3 0.72 0.4-1.2 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 187 Graph 3.139- Survival according to MMVDSI for females with NSCLC who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.140- Survival according to MMVDSI for females with NSCLC who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 188 Table 3.75- Hazard ratio (HR) and 95% confidence interval (CI) according to season of diagnosis for male and female patients with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 4100 512 NS 0.064 1.05 0.9-1.1 0.95 0.8-1.0 1 0.93 0.8-1.4 Other causes 4100 3640 NS NS 0.89 0.6-1.1 1.06 0.8-1.3 1 1.12 0.8-1.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 189 Graph 3.141- Survival according to season of diagnosis for male and female NSCLC patients who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.142- Survival according to season of diagnosis for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 190 Table 3.76- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 3426 419 <0.01 <0.01 1 0.94 0.8-1.0 0.88 0.8-0.9 0.92 0.8-1.0 Other causes 3426 3050 NS NS 1 1.03 0.7-1.4 1.17 0.8-1.5 1.13 0.8-1.5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 191 Graph 3.143- Survival according to season of first treatment for male and female NSCLC patients who received radiotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.144- Survival according to season of first treatment for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 192 Table 3.77- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.046 NS Number of cases 4100 4100 Number of censored cases 512 3640 H R ( 9 5 % C I) Jan 1 1 Feb 0.99 0.8-1.1 0.91 0.5-1.3 Mar 1.18 1.0-1.3 0.87 0.5-1.3 Apr 1.29 1.1-1.5 0.94 0.5-1.4 May 1.05 0.8-1.2 0.59 0.3-0.9 Jun 1.05 0.8-1.2 0.92 0.6-1.4 Jul 1.07 0.9-1.2 0.80 0.5-1.2 Aug 1.04 0.8-1.2 1.06 0.7-1.6 Sep 1.20 1.0-1.4 0.96 0.6-1.5 Oct 1.05 0.8-1.2 0.91 0.5-1.4 Nov 1.08 0.9-1.2 0.79 0.5-1.2 Dec 1.13 0.9-1.3 1.06 0.7-1.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of diagnosis. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 193 Graph 3.145- Survival according to month of diagnosis for male and female NSCLC patients who received radiotherapy: lung cancer deaths Graph 3.146- Survival according to month of diagnosis for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths 194 Table 3.78- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received radiotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 3426 3426 Number of censored cases 419 3050 H R ( 9 5 % C I) Jan 1 1 Feb 0.95 0.8-1.1 0.85 0.5-1.3 Mar 1.03 0.8-1.2 0.81 0.4-1.3 Apr 1.09 0.9-1.2 0.70 0.4-1.1 May 1.08 0.9-1.2 0.85 0.5-1.4 Jun 1.01 0.8-1.2 0.70 0.4-1.1 Jul 1.00 0.8-1.1 0.98 0.6-1.5 Aug 1.00 0.8-1.1 0.76 0.4-1.2 Sep 0.95 0.7-1.0 1.12 0.7-1.7 Oct 0.97 0.8-1.1 0.73 0.4-1.2 Nov 0.94 0.7-1.1 0.97 0.6-1.5 Dec 1.00 0.8-1.2 0.88 0.5-1.4 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and month of first treatment. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 195 Graph 3.147- Survival according to month of first treatment for male and female NSCLC patients who received radiotherapy: lung cancer deaths Graph 3.148- Survival according to month of first treatment for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths 196 Table 3.79- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index male and female patients with NSCLC who received radiotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 4064 504 0.06 <0.01 1 1.07 0.9-1.1 1.21 1.0-1.3 0.98 0.8-1.0 1.03 0.9-1.13 Other causes 4064 3608 NS NS 1 0.92 0.6-1.2 0.97 0.7-1.3 0.87 0.6-1.1 0.95 0.7-1.2 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (chemotherapy and surgery), disease site and monthly mean vitamin D sunshine index. NS is not significant Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 197 Graph 3.149- Survival according to MMVDSI for male and female NSCLC patients who received radiotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.150- Survival according to MMVDSI for male and female NSCLC patients who received radiotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 198 Table 3.80- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for males with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 157 28 NS NS 1.81 0.9-3.5 1.26 0.7-2.1 1 0.90 0.5-1.5 Other causes 157 132 NS NS N<5 0.60 0.1-2.4 1 0.55 0.1-2.2 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 199 Graph 3.151- Survival according to season of diagnosis for males with NSCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.152- Survival according to season of diagnosis for males with NSCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 200 Table 3.81- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for males with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 154 28 NS NS 1 1.20 0.6-2.1 1.19 0.6-2.0 1.16 0.6-2.0 Other causes 154 129 NS NS 1 N<5 N<5 6.13 1.2-29.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 201 Graph 3.153 Survival according to season of first treatment for males with NSCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.154- Survival according to season of first treatment for males with NSCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 202 Table 3.82- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for males with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.057 NS Number of cases 157 157 Number of censored cases 28 132 H R ( 9 5 % C I) Jan 1 1 Feb 0.19 0.0-0.5 N<5 Mar 0.62 0.2-1.7 N<5 Apr 0.90 0.3-2.6 N<5 May 0.68 0.1-2.5 N<5 Jun 0.41 0.1-1.0 N<5 Jul 0.43 0.1-1.1 N<5 Aug 0.88 0.3-2.4 N<5 Sep 0.61 0.1-1.9 N<5 Oct 0.28 0.0-0.7 N<5 Nov 0.37 0.1-1.0 N<5 Dec 0.45 0.1-1.4 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and month of diagnosis. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 203 Graph 3.155- Survival according to month of diagnosis for males with NSCLC who received chemotherapy: lung cancer deaths Graph 3.156- Survival according to month of diagnosis for males with NSCLC who received chemotherapy: non-lung-cancer deaths 204 Table 3.83- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for males with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 154 154 Number of censored cases 28 129 H R ( 9 5 % C I) Jan 1 1 Feb 1.18 0.4-3.2 N<5 Mar 0.64 0.2-1.7 N<5 Apr 1.22 0.4-3.4 N<5 May 1.83 0.5-6.07 N<5 Jun 0.63 0.2-1.8 N<5 Jul 1.18 0.4-3.3 N<5 Aug 2.18 0.8-5.9 N<5 Sep 1.75 0.6-0.4 N<5 Oct 0.78 0.2-2.2 N<5 Nov 0.85 0.3-2.3 N<5 Dec 1.01 0.4-2.4 18.91 1.2->100 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and month of first treatment. NS is not significantN is number of uncensored casesNumber of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 205 Graph 3.157- Survival according to month of first treatment for males with NSCLC who received chemotherapy: lung cancer deaths Graph 3.158- Survival according to month of first treatment for males with NSCLC who received chemotherapy: non-lung-cancer deaths 206 Table 3.84- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for males with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 157 28 NS NS 1 1.14 0.6-2.0 2.02 1.0-3.9 1.44 0.8-2.5 1.33 0.7-2.3 Other causes 157 132 NS NS 1 N<5 N<5 N<5 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 207 Graph 3.159- Survival according to MMVDSI for males with NSCLC who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.160- Survival according to MMVDSI for males with NSCLC who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 208 Table 3.85- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for females with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 81 14 NS NS 1.06 0.4-2.2 0.98 0.3-2.6 1 1.05 0.5-2.1 Other causes 81 72 NS NS N<5 N<5 1 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 209 Graph 3.161- Survival according to season of diagnosis for females with NSCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.162- Survival according to season of diagnosis for females with NSCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 210 Table 3.86- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for females with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 81 14 NS NS 1 0.75 0.3-1.6 1.02 0.4-2.4 0.95 0.5-1.8 Other causes 81 72 NS NS 1 N<5 N<5 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 211 Graph 3.163- Survival according to season of first treatment for females with NSCLC who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.164- Survival according to season of first treatment for females with NSCLC who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 212 Table 3.87- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for females with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.025 NS Number of cases 81 81 Number of censored cases 14 72 H R ( 9 5 % C I) Jan 1 1 Feb 0.87 0.2-3.2 N<5 Mar 0.55 0.1-1.6 N<5 Apr 1.43 0.4-4.9 N<5 May 2.93 0.7-12.2 N<5 Jun N<5 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep 2.35 0.6-9.1 N<5 Oct 0.68 0.2-2.2 N<5 Nov 0.81 0.2-2.5 N<5 Dec 1.15 0.3-3.7 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and month of diagnosis. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 213 Graph 3.165- Survival according to month of diagnosis for females with NSCLC who received chemotherapy: lung cancer deaths Graph 3.166- Survival according to month of diagnosis for females with NSCLC who received chemotherapy: non-lung-cancer deaths 214 Table 3.88- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for females with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.025 NS Number of cases 81 81 Number of censored cases 14 72 H R ( 9 5 % C I) Jan 1 1 Feb 0.16 0.0-0.5 N<5 Mar N<5 N<5 Apr 0.26 0.0-0.9 N<5 May 0.44 0.1-1.4 N<5 Jun 0.55 0.1-2.3 N<5 Jul N<5 N<5 Aug N<5 N<5 Sep N<5 N<5 Oct N<5 N<5 Nov 0.21 0.0-0.8 N<5 Dec 0.19 0.0-0.8 N<5 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of first treatment. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 215 Graph 3.167- Survival according to month of first treatment for females with NSCLC who received chemotherapy: lung cancer deaths Graph 3.168- Survival according to month of first treatment for females with NSCLC who received chemotherapy: non-lung-cancer deaths 216 Table 3.89- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for females with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 81 14 0.071 NS 1 0.70 0.3-1.4 1.82 0.8-4.0 2.77 1.0-7.6 0.66 0.2-2.0 Other causes 81 72 NS NS 1 N<5 N<5 N<5 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and Monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 217 Graph 3.169- Survival according to MMVDSI for females with NSCLC who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.170- Survival according to MMVDSI for females with NSCLC who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 218 Table 3.90- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of diagnosis for male and female patients with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 238 42 NS NS 1.07 0.6-1.6 1.22 0.7-1.8 1 0.86 0.5-1.3 Other causes 238 204 NS NS 1.81 0.5-6.2 1.12 0.3-3.9 1 1.41 0.4-4.3 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of diagnosis. NS is not significant; A is spring, B is summer, C is fall and D is winter. Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 219 Graph 3.171- Survival according to season of diagnosis for male and female NSCLC patients who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.172- Survival according to season of diagnosis for male and female NSCLC patients who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 220 Table 3.91- Hazard ratio (HR) and 95% confidence interval (CI) associated with season of first treatment for male and female patients with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt iv ar ia te P -v al u e HR (95% CI) Spring (A) Summer (B) Fall (C) Winter (D) Lung cancer 235 42 NS NS 1 1.15 0.7-1.7 1.31 0.8-2.0 1.07 0.7-1.6 Other causes 235 201 NS NS 1 0.57 0.1-1.9 N<5 1.90 0.7-4.9 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and season of first treatment. NS is not significant; A is spring, B is summer, C is fall and D is winter. N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 221 Graph 3.173- Survival according to season of first treatment for male and female NSCLC patients who received chemotherapy: lung cancer deaths (A is spring, B is summer, C is fall and D is winter) Graph 3.174- Survival according to season of first treatment for male and female NSCLC patients who received chemotherapy: non-lung-cancer deaths (A is spring, B is summer, C is fall and D is winter) 222 Table 3.92- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of diagnosis for male and female patients with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value 0.065 NS Number of cases 238 238 Number of censored cases 42 204 H R ( 9 5 % C I) Jan 1 1 Feb 0.40 0.1-0.8 0.59 0.0-4.5 Mar 0.60 0.3-1.1 n<5 Apr 0.91 0.4-1.9 6.79 0.8-57.4 May 1.02 0.4-2.4 N<5 Jun 0.85 0.4-1.6 N<5 Jul 0.70 0.3-1.4 N<5 Aug 1.22 0.5-2.5 N<5 Sep 1.37 0.6-2.9 N<5 Oct 0.57 0.2-1.1 N<5 Nov 0.59 0.3-1.1 2.07 0.2-16.3 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and month of diagnosis. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 223 Graph 3.175- Survival according to season of month of diagnosis for male and female NSCLC patients who received chemotherapy: lung cancer deaths Graph 3.176- Survival according to season of month of diagnosis for male and female NSCLC patients who received chemotherapy: non-lung-cancer deaths 224 Table 3.93- Hazard ratio (HR) and 95% confidence interval (CI) associated with month of first treatment for male and female patients with NSCLC who received chemotherapy Cause of death Lung cancer Other Causes Univariate P value NS NS Multivariate P value NS NS Number of cases 235 235 Number of censored cases 42 201 H R ( 9 5 % C I) Jan 1 1 Feb 0.77 0.3-1.5 2.78 0.4-18.7 Mar 0.52 0.2-1.0 N<5 Apr 0.90 0.4-1.9 7.39 0.9-56.7 May 0.90 0.4-1.9 N<5 Jun 0.70 0.3-1.5 N<5 Jul 0.94 0.4-2.0 N<5 Aug 0.98 0.4-2.1 N<5 Sep 1.34 0.6-2.8 N<5 Oct 0.90 0.4-1.9 N<5 Nov 0.71 0.3-1.5 1.35 0.1-10.6 Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and month of first treatment. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 225 Graph 3.177- Survival according to month of first treatment for male and female NSCLC patients who received chemotherapy: lung cancer deaths Graph 3.178- Survival according to month of first treatment for male and female NSCLC patients who received chemotherapy: non-lung-cancer deaths 226 Table 3.94- Hazard ratio (HR) and 95% confidence interval (CI) associated with monthly mean vitamin D sunshine index for male and female patients with NSCLC who received chemotherapy C au se o f D ea th N u m b er o f ca se s N u m b er o f C en so re d C as es U n iv ar ia te P -v al u e M u lt ii v ar ia te P -v al u e HR (95% CI) A B C D E Lung cancer 238 42 NS 0.058 1 0.95 0.6-1.4 1.61 1.0-2.5 1.70 1.0-2.7 1.41 0.9-2.2 Other causes 238 204 NS NS 1 N<5 1.58 0.4-5.0 0.84 0.2-2.9 N<5 (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Multivariate P-value is adjusted for age at diagnosis, cancer laterality, stage at time of diagnosis, other treatments (radiotherapy and surgery), disease site and monthly mean vitamin D sunshine index. NS is not significant N is number of uncensored cases Number of cases is the cumulative number of SCLC and NSCLC before censoring for cause of death 227 Graph 3.179- Survival according to MMVDSI for male and female NSCLC patients who received chemotherapy: lung cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) Graph 3.180- Survival according to MMVDSI for male and female NSCLC patients who received chemotherapy: non-lung-cancer deaths (A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218-1.624 IU E: ≥1.624 IU) 228 4. Discussion 4.1 Sunlight and cancer survival association These results suggest that sun exposure could improve prognosis in some lung cancer patients. This might imply sun-induced vitamin D affects some lung cancer patients. In particular, I found: A- According to table 3.2, survival was significantly associated with season of diagnosis, season of first treatment, month of diagnosis and month of first treatment in females with SCLC who received radiotherapy and died because of lung cancer. However, survival was not significantly associated with monthly mean MMVDSI in these same patients. B- According to table 3.3, survival was significantly associated with (MMVDSI) in females with NSCLC who received surgery and died because of lung cancer. However, survival was not significantly associated with other sunshine variables (i.e., month and season) in these same patients. C- According to table 3.4, survival was significantly associated with (MMVDSI) in males with NSCLC who received radiotherapy and died because of lung cancer. However, survival was not significantly associated with other sunshine variables (i.e., month and season) in these same patients. In additional analyses, I separately examined patients who received monotherapy, (i.e., only chemotherapy, surgery or radiotherapy), patients age <50 and ≥50 years (at time of diagnosis). The results were: 229  Survival was significantly associated with all of the sunshine variables in SCLC patients who received only radiotherapy and died because of lung cancer.  Survival was significantly associated with (MMVDSI) in NSCLC patients age ≥50y who received only surgery and died because of lung cancer.  Survival was significantly associated with (MMVDSI) in NSCLC patients who received only radiotherapy and died because of lung cancer. 4.2 Study strengths Compared to previous work that was done by other groups, my study has some strong points:  The study population was large. I considered more than 9000 patients with lung cancer and a remarkable amount of information.  I used five different sunshine variables, which allowed us to compare with previous studies’ results. This allowed me to compare this study with the previous ones and also analyze some novel sunshine variables.  In comparing our study to other studies we used a novel method to find the association of sun-induced vitamin D and cancer survival which was monthly mean vitamin D sunshine index.  Unlike previous studies, our study’s analyses were stratified by different types of treatment. Moreover, not only we considered the types of treatment, but also we analyzed each type of treatment individually.  The amount of missing information was small in my study. BC cancer registry, between 1980 and 1989, gathered treatment information of about 70% of patients who had lung cancer, which was an outstanding percentage at that period. 230 4.3 Study limitations Having discussed our study’s strength, it is important to mention that this study is not flawless. Most of the following limitations were difficult to avoid.  Our data didn’t include smoking information, which could be a potential effect modifier for the sunshine measure. There are many reasons why this might be true, but smoking is one of the important prognostic factors in lung cancer patients and smoking could interfere with sun-induced vitamin D production. Smoking information is not usually recorded in the BCCR, and it would have been very time consuming, or even impossible, to call each of the patients’ families for the information.  Our data didn’t include a patient’s race. Skin pigmentation is different in races, and vitamin D production depends on race. Like smoking information, The BCCR doesn’t usually record information about race. A partial solution might involve taking a sample of our data set and trying to find the patients’ races. Another solution might involve using each patient’s name to determine his or her race.  The sample size for SCLC patients who received surgery was not big. SCLC has the worst survival among different types of lung cancer and many patients already have metastases disease. Accordingly, few patients are candidates for surgery.  Since we have done a considerable number of analyses, the significant results may be because of chance.  We did not find out the amount of actual sunlight exposure in our study.  One of the sources of vitamin D for our body is diet and vitamin D supplements. It was not possible for us to consider this variable since we had no access to this information. 231 4.4 Comparing the results with previous studies Few studies have assessed the association between sun exposure and cancer survival. Most of the existing studies have been done in Norway by Porojnicu and colleagues. They have done several studies in which they found the prognostic advantage of sun-induced 25-hydroxy vitamin D in Hodgkins disease, colon, breast, prostate and lung cancers. In addition, there are some other, non-epidemiologic, studies in which the association of vitamin D and cancer survival was evaluated. In all of these studies, researchers have tried to show the protective effect of vitamin D, either directly or indirectly (sun exposure) in different types of cancer which I have already mentioned them in section 1.4. I also performed analyses using season of diagnosis, thereby allowing me to compare results with those of earlier studies. In Porojnicu’s results, male NSCLC patient’s age < 50 years had better survival during sunny seasons. In my study, of patients with the same age, sex and histology group, only those who received surgery had a better prognosis (univariate analysis only). 4.5 Future work There are several studies to further the things I discovered in this project. My study shows that there is a good chance that sunlight exposure, specifically the range which induces vitamin D production, could increase the survival in lung cancer patients. But, I would like to repeat my study for another time period to determine whether the result would be similar. Another project would involve a prospective patient cohort so that I could address some of our project’s weaknesses. If vitamin D improves a lung cancer patient’s prognosis, then I might propose vitamin D supplements as adjuvant or neoadjuvant treatment for lung cancer patients. 232 References 1- Commonly diagnosed cancers worldwide. Cancer Research UK. April 2005. http://info.cancerresearchuk.org/cancerstats/, Retrieved 2011-04-15 2- Incidence and mortality by cancer type, Canadian Cancer Society’s Steering Committee: Canadian Cancer statistics 2010.Toronto: Canadian Cancer Society, 2010, page 11, April 2010, ISSN 0835-2976 3- Incidence and mortality by province, Canadian Cancer Society’s Steering Committee: Canadian Cancer statistics 2010.Toronto: Canadian Cancer Society, 2010, page 16, April 2010, ISSN 0835-2976 4- Biesalski, HK; Bueno de Mesquita B, Chesson A et al. (1998). European Consensus Statement on Lung Cancer: risk factors and prevention. Lung Cancer Panel, CA Cancer J Clin 48 (3): 167– 176; discussion 164–166 5- Peto, R; Lopez AD, Boreham J et al. (2006), Mortality from smoking in developed countries 1950–2000: Indirect estimates from National Vital Statistics, American J of Epidemiology 143 (5):529-520 6- Samet, JM; Wiggins CL, Humble CG, Pathak DR (May 1988). Cigarette smoking and lung cancer in New Mexico, American Review of Respiratory Disease 137 (5): 1110–1113 7- Health Canada, Cancer Updates; Lung cancer in Canada 1998 8- Catelinois O, Rogel A, Laurier D, et al. (2006). Lung cancer attributable to indoor radon exposure in france: impact of the risk models and uncertainty analysis, Environ Health Perspect 114 (9): 1361–6 9- O'Reilly, KM; Mclaughlin AM, Beckett WS, Sime PJ (2007). Asbestos-related lung disease, American Family Physician 75 (5): 683–688 233 10- Tintinalli JE, Ruiz E, Krome RL (1996). Emergency Medicine: A Comprehensive Study Guide, 4th ed. McGraw-Hill; 1996:833-41. 11- Pope, CA 3rd; Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002). Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution, Journal of the American Medical Association 287 (9): 1132–1141 12- Valavanidis A, Fiotakis K, Vlachogianni T (2008). Airborne particulate matter and human health: toxicological assessment and importance of size and composition of particles for oxidative damage and carcinogenic mechanisms, J Environ Sci Health C Environ Carcinog Ecotoxicol Rev. 26 (4): 339-62 13- Fong, KM; Sekido Y, Gazdar AF, Minna JD (2003). Lung cancer 9: Molecular biology of lung cancer: clinical implications. Thorax (BMJ Publishing Group Ltd.) 58 (10): 892–900 14- Herbst, RS; Heymach JV, Lippman SM (2008). Molecular origins of cancer: lung cancer. N Engl J Med 359 (13): 1367–1380. 15- Aviel-Ronen, S; Blackhall FH, Shepherd FA, Tsao MS (2006), K-ras mutations in non- small-cell lung carcinoma: a review, Clinical Lung Cancer (Cancer Information Group) 8 (1): 30–38 16- Engels, EA; Wu X, Gu J et al. (2007). Systematic evaluation of genetic variants in the inflammation pathway and risk of lung cancer, Cancer Research (American Association for Cancer Research) 67 (13): 6520–6527 17- Wenzlaff, AS; Cote ML, Bock CH et al. (2005). CYP1A1 and CYP1B1 polymorphisms and risk of lung cancer among never smokers: a population-based study, Carcinogenesis (Oxford University Press) 26 (12): 2207–2212. 234 18- Son, JW; Kang HK, Chae MH et al. (2006). Polymorphisms in the caspase-8 gene and the risk of lung cancer, Cancer Genetics and Cytogenetics 169 (2): 121–127 19- Yin, J; Vogel U, Ma Y et al. (2007). The DNA repair gene XRCC1 and genetic susceptibility 20- Travis, WD; Travis LB, Devesa SS (1995), Lung cancer, Cancer 75 (Suppl. 1): 191–202 21- Roggli VL, Vollmer RT, Greenberg SD, McGavran MH, Spjut HJ, Yesner R(1985), Lung cancer heterogeneity: a blinded and randomized study of 100 consecutive cases, Hum Pathology 1985; 16: 569-79. 22- Hamilton, W; Peters TJ, Round A, Sharp D (2005). What are the clinical features of lung cancer before the diagnosis is made? A population based case-control study, Thorax (BMJ Publishing Group) 60 (12): 1059–1065 23- Honnorat, J; Antoine JC (2007). Paraneoplastic neurological syndromes, Orphanet Journal of Rare Diseases (BioMed Central) 2: 22 24- Greene, Frederick L. (2002). AJCC cancer staging manual, Berlin: Springer-Verlag 25- Minna, JD; Schiller JH (2008). Harrison's Principles of Internal Medicine, (17th ed.). McGraw-Hill, pp. 551–562 26- Schiller JH, Vidaver RM, Novello S, Brahmer J, Monroe L (2007), Living with a Diagnosis of Lung Cancer, http://www.nationallungcancerpartnership.org/index.cfm?page=treatment, Retrieved 2011-04-15 27- Colice, GL; Shafazand S, Griffin JP et al. (2007). Physiologic evaluation of the patient with lung cancer being considered for resectional surgery: ACCP evidenced-based clinical practice guidelines (2nd edition), Chest 132 (Suppl. 3): 161S–177S 28- N; Turrisi AT (2006). A review of first-line treatment for small-cell lung cancer, Journal of Thoracic Oncology 1 (3): 270–278 235 29- Clegg, A; Scott DA, Hewitson P et al. (2002). Clinical and cost effectiveness of paclitaxel, docetaxel, gemcitabine, and vinorelbine in non-small cell lung cancer: a systematic review Thorax (BMJ Publishing Group) 57 (1): 20–28 30- Raz, DJ; He B, Rosell R, Jablons DM (2006).Bronchioloalveolar carcinoma: a review, Clinical Lung Cancer 7 (5): 313–322. 31- Bencardino, K; Manzoni M, Delfanti S et al. (2007). Epidermal growth factor receptor tyrosine kinase inhibitors for the treatment of non-small-cell lung cancer: results and open issues, Internal and Emergency Medicine 2 (1): 3–12 32. Arriagada, R; Goldstraw P, Le Chevalier T (2002). Oxford Textbook of Oncology (2nd ed.). Oxford University Press. p. 2094. 33- Wagner, H (1998). Radiation therapy in the management of limited small cell lung cancer: when, where, and how much?, Chest (American College of Chest Physicians) 113 (Suppl. 1): 92S–100S 34- Celebioglu, B; Gurkan OU, Erdogan S et al. (2002). High dose rate endobronchial brachytherapy effectively palliates symptoms due to inoperable lung cancer, Japanese Journal of Clinical Oncology (Oxford University Press) 32 (11): 443–448 35- Ng, M; Chong J, Milner A et al. (2007).Tolerability of accelerated chest irradiation and impact on survival of prophylactic cranial irradiation in patients with limited-stage small cell lung cancer: review of a single institution's experience, Journal of Thoracic Oncology (International Association for the Study of Lung Cancer) 2 (6): 506–513. 36- Mountain, CF (1997). Revisions in the international system for staging lung cancer, Chest (American College of Chest Physicians) 111: 1710–1717. 236 37- Small Cell Lung Cancer Treatment (2008), PDQ for Health Professionals, National Cancer Institute. Retrieved 2008-11-22 38- Table 7.1, Estimated five-year relative survival ratio for selected cancers by sex, Canada, 2002-2004, Canadian Cancer Society’s Steering Committee: Canadian Cancer statistics 2009.Toronto: Canadian Cancer Society, 2009, page 59, April 2009, ISSN 0835-2976 39- Table 7.3, Estimated five-year relative survival ratio by age group for the most common cancers, Canada, 2002-2004, Canadian Cancer Society’s Steering Committee: Canadian Cancer statistics 2009.Toronto: Canadian Cancer Society, 2009, page 60, April 2009, ISSN 0835-2976 40- Figure 7.1, Estimated age-standardized five-year relative survival ratio for selected cancers, both sexes combined, Canada, 2002-2004, Canadian Cancer Society’s Steering Committee: Canadian Cancer statistics 2009.Toronto: Canadian Cancer Society, 2009, page 61, April 2009, ISSN 0835-2976 41- Holick MF. (1994) Vitamin D: Photobiology, metabolism and clinical application, In: Heersche NJM, Kanis JA, editors. Bone and mineral research, Amsterdam: Elsevir; 1994.p.543-62. 42- Calvo MS, Whiting SJ, Barton CN. (2005) Vitamin D intake: A global perspective of current status, J Nutr 2005; 135:310-6. 43- Fioletov, V. E., L. J. B. McArthur, J. B. Kerr, and D. I. Wardle (2001), Long-term variations of UV-B irradiance over Canada estimated from Brewer observations and derived from ozone and pyranometer measurements, J Geophys Res 106(D19), 23,009–23,027. 44- Tuohimaa P, Pukkala E, Scélo G, et al (2007). Does solar exposure, as indicated by the non- melanoma skin cancers, protect from solid cancers: vitamin D as a possible explanation, Eur. J. Cancer 43 (11): 1701–12 237 45- Gorham ED, Garland CF, Garland FC et al. (2007). Optimal vitamin D status for colorectal cancer prevention: a quantitative meta analysis, Am J Prev Med. 32:210-216. 46- Garland CF, Mohr SB, Gorham ED et al. (2006). Role of ultraviolet B irradiance and vitamin D in prevention of ovarian cancer, Am J Prev Med. 31:512-514. 47. Freedman DM, Looker AC, Chang SC, Graubard BI (2007). Prospective study of serum vitamin D and cancer mortality in the United States, J. Natl. Cancer Inst. 99 (21): 1594–602 48- Lappe JM, Travers-Gustafson D, Davies KM, Recker RR, Heaney RP. (2007). Vitamin D and calcium supplementation reduces cancer risk: results of a randomized trial, Am J Clin Nutr. 85 (6): 1586–91 49- Beer T, Myrthue A (2006), Calcitriol in the treatment of prostate cancer. Anticancer Res 26 (4A): 2647–51. 50- Buyru N, Tezol A;,Yosunkaya-Fenerci E, Dalay, N. (2003), Vitamin D receptor gene polymorphisms in breast cancer, Experimental and Molecular Medicine. 2003; 35(6):550-555 51- Metha RG, Meta RR, (2002), Vitamin D and cancer, J Nutr Biochem 2002; 13:252-64 52- Van den Bemd, G.T.G. Chang. (2002), Vitamin D and Vitamin D analogs in cancer treatment, Current Drug Targets 2002; 3:85-94 53- Napel S, Na S, Rathanachalam R. (2005), Non-calcemic Actions of Vitamin D receptor ligands, Endocr Rev 2005; 26:662-87 54- Lin R, White JH. (2204), The pleiotropic actions of Vitamin D, Bioassays 2004; 26:21-8 55- Guezy m, Sattler C, Deluca HF. (1998) Combinational effects of Vitamin D3 and retinoic acid (all trans and 9 cis) on proliferation, differentiation, and programmed cell death in two small cell lung carcinoma sell lines, Biochem Biophys res Commun 1998;249:735-44 238 56- Nakagawa K, Kawaura A, Kato S, Takeda E, OkanoT. (2204), Metastatic growth of lung cancer cells in extremely reduced in Vitamin D receptor knockout mice, J Steroid Biochem Mol Biol 2004; 89-90:545-7 57- Nakagawa K, Kawaura A, Kato S, Takeda E, Okano T. (2005) 1Alpha,25-Dihydroxy Vitamin D(3) is a preventive factor in the metastasis of Lung cancer, Carcinogenesis 2005;26:429-40 58- Nakagawa K, Sasaki Y, Kato S, Kubodera N, Okano T.(2005), 22-Oxa-1Alpha, 25-dihydroxy Vitamin D3 inhibits metastasis and angiogenesis in Lung cancer, Carcinogenesis 2005; 26:1044-54 59- Robsahm TE, Tretli S, Dahlback A, Moan J. (2004) Vitamin D3 from sun light may improve the prognosis of breast, colon and prostate cancer (Norway), Cancer Cause Control 2004; 15:149-58. 60- Moan J, Porojnicu AC, Robsahm TE, Dahlback A, Juzeniene A, Tretli S, et al. (2005) Solar Radiation, Vitamin D and survival rate of colon cancer in Norway. J Photochem Photobiol B 2005; 78:189-93 61- Porojnicu Alina Carmen ; Robsahm Trude Eid; Dahlback Arne; Berg Jens Petter; Christiani David; Buland Oyvind Sverre ; Moan Johan, (2007) Seasonal and geographical variations in lung cancer prognosis in Norway Does Vitamin D from the sun play a role, Lung Cancer 2007; 55, pp. 263-70 62- Zhou W, Suk R, Park S, Neuberg DSA, Wain JC, et al. (2005),Vitamin D is associated with improved survival in early-stage non-small cell lung cancer patients, Cancer Epidemiol Biomarkers Prev 2005; 14:2303-9. 63. Liam HS, Roychoudhuri R, Peto J, Schwartz G, Baade P, Moller H. (2006), 239 Cancer survival is dependent on season of diagnosis and sunlight exposure, Int J Cancer 2006;119:1530-6 240 Appendices Appendix A: MMVDSI data set generation process Table A-1 on next page, is a small version of MnDph data set. For example the number 0.3423 represents the10- average (1980-1980) of vitamin D UV-B spectrum in lat 54 and lon - 119 in 1 am in January 1 st . In the first step, I calculated the average of the first day of each month for each corresponding lat and lon. For example, in table A-2, the number 0.3451 represents the10- average (1980-1980) of vitamin D UV-B spectrum in lat 54 and lon -119 in first day of January. In the next step, I calculated the average of each month for each corresponding lat and lon. For example, in table A-3, the number 0.3482 represents the10- average (1980-1980) of vitamin D UV-B spectrum in lat 54 and lon -119 in January. I called the last number for each month, Monthly mean vitamin D sunshine index. The next step in generating the new data set was converting each patient postal code at the time of diagnosis to lat and lon. After this step, I had a lat and lon for each patient at the month of diagnoses. Then, I merged the data set of MMVDSI to this new data set. Finally, I had a new data set in which I had a number for each patient at the time of diagnosis representing of MMVDSI for that patient. (You can see the process in the diagram) In the analyses since I used the Cox-proportional model, I decided to divide the MMVDSI actual range to five subgroups; A: ≤0.406 IU B: 0.406-0.812 IU C: 0.812-1.218 IU D: 1.218- 1.624 IU E: ≥1.624 IU. I thought it would be fine to do it this way specially when the number of cases in each group were approximately the same 241 Lat 54 Lon -119 J a n u a ry F ir st (2 4 h rs M n D p h ) 1 0.3423 2 0.3425 3 ………… 4 ………… 5 ………… 6 ………… 7 ………… 8 ………… 9 ………… 10 ………… 11 ………… 12 ………… 13 ………… 14 ………… 15 ………… 16 ………… 17 ………… 18 ………… 19 ………… 20 ………… 21 ………… 22 ………… 23 ………… 24 ………… Lat 54 Lon -119 J a n u a ry (M ea n o f h o u rl y M n D p h f o r e a ch d a y ) 1 0.3451 2 0.3672 3 ----------- 4 ----------- 5 ----------- 6 ----------- 7 ----------- 8 ----------- 9 ----------- 10 ----------- 11 ----------- 12 ----------- 13 ----------- 14 ----------- 15 ----------- 16 ----------- 17 ----------- 18 ----------- 19 ----------- 20 ----------- 21 ----------- 22 ----------- 23 ----------- 24 ----------- 25 ----------- 26 ----------- 27 ----------- 28 ----------- 29 ----------- 30 ----------- 31 ----------- Lat 54 Lon -119 (M ea n o f h o u rl y M n D p h f o r e a ch m o n th ) Jan 0.3482 Feb 0.3582 Mar 0.3921 Apr ---------- ---- May ---------- ---- Jun ---------- ---- Jul ---------- ---- Aug ---------- ---- Sep ---------- ---- Oct ---------- ---- Nov ---------- ---- Dec ---------- ---- Table A-1: Hourly average of MnDph Table A-2: Daily average of MnDph Table A-3: Monthly average of MnDph 242 Latitude longitude 54 -119 52 -120 ------ ------- Patient ID Patient postal code Month of diagnosis 1234 V7T--- Jan 4321 V6M--- March 5321 ------ Jun January lat: 54 lan: -119 0.3482 Patient ID Patient postal code Month of diagnosis MMVDSI 1234 V7T--- Jan 0.3482 4321 V6M--- March 0.4523 5321 ------ Jun -------- Figure A-1: Diagram of MMVDSI calculation 243 Appendix B: Lung cancer population details Total number of cases in data set=9953, Number of exclusion due to 30-day period=651, Number of exclusion due to another type of cancer=6, Number of exclusion due to missing postal code=781) Table A-4: Total number of cases in all sunshine measures except MMVDSI Total number of cases Exclusion due to 30-day period exclusion due to another type of cancer Final study population for the rest of sunshine measures but MnDph (after considering exclusion criteria 9953 651 6 9302 Table A-5: Total number of cases in MMVDSI Total number of cases after considering exclusion criteria exclusion due to missing postal code Final study population for the MnDph 9302 781 8521 244 Appendix C: Summary of additional analyses’ significant results M=male, F=female, B= both sexes together, LC= lung cancer, NL=non-lung cancer, Tx=treatment, U=univariate, M=multivariate, n=number of uncensored cases in each analysis. How to read this table: + Sign means that the topic of that column is considered in that row for analysis. For example red row means: Univariate and multivariate analyses for month of first treatment were significant (0.032, <0.01) for SCLC patients who died because of lung cancer when the analyses considered both sexes. Histology Sex Cause of death Type of Tx M M V D S I S ea so n o f D ia g n o si s S ea so n o f 1 st T x M o n th o f d ia g n o si s M o n th o f 1 st T x P-value n S C L C N S C L C M F B LC NLC C S R U M + + + + + 0.020 0.012 37 + + + + + 0.042 37 + + + + + 0.032 <0.01 74 + + + + + 0.041 74 + + + + + 0.033 74 + + + + + 0.019 74 + + + + + 0.028 28 + + + + + 0.017 28 + + + + + 0.047 <0.01 + + + + + <0.01 56 + + + + + 0.036 56 + + + + + 0.019 <0.01 56 + + + + + 0.018 52 + + + + + 0.045 <0.01 161 + + + + + 0.41 118 + + + + + 0.035 118 + + + + + 0.043 117 + + + + + <0.01 <0.01 118 + + + + + <0.01 <0.01 321 Table A-6: Significant results for Patients age <50 years who received monotherapy 245 M=male, F=female, B= both sexes together, LC= lung cancer, NL=non-lung cancer, Tx=treatment, U=univariate, M=multivariate, n=number of uncensored cases in each analysis. How to read this table: + Sign means that the topic of that column is considered in that row for analysis. For example red row means: Multivariate analyses for season of first treatment was significant (<0.01) for females who had SCLC and died because of lung cancer. Histology Sex Cause of death Type of Tx M M V D S I S ea so n o f D ia g n o si s S ea so n o f 1 st T x M o n th o f d ia g n o si s M o n th o f 1 st T x P-value n S C L C N S C L C M F B LC NLC C S R U M + + + + + 0.017 66 + + + + + 0.011 495 + + + + + <0.01 379 + + + + + <0.01 874 + + + + + <0.01 3745 + + + + + <0.01 1233 + + + + + 0.022 3779 + + + + + 0.041 122 + + + + + <0.01 49 + + + + + <0.01 2546 + + + + + <0.01 2528 + + + + + 0.012 <0.01 3779 + + + + + 0.030 49 + + + + + 0.043 1162 + + + + + <0.01 49 Table A-7: Significant results for patients age ≥50 years who received monotherapy 246 M=male, F=female, B= both sexes together, LC= lung cancer, NL=non-lung cancer, Tx=treatment, U=univariate, M=multivariate, n=number of uncensored cases in each analysis. How to read this table: + Sign means that the topic of that column is considered in that row for analysis. For example red row means: Univariate and multivariate analyses for season of first treatment was significant (<0.01, <0.01) for males who had NSCLC and died because of lung cancer. Histology Sex Cause of death Type of Tx M M V D S I S ea so n o f D ia g n o si s S ea so n o f 1 st T x M o n th o f d ia g n o si s M o n th o f 1 st T x P-value n S C L C N S C L C M F B LC NLC C S R U M + + + + + <0.01 153 + + + + + 0.023 <0.01 153 + + + + + <0.01 <0.01 190 + + + + + <0.01 <0.01 153 + + + + + 0.021 87 + + + + + <0.01 0.01 92 + + + + + 0.028 118 + + + + + 0.22 0.25 341 + + + + + <0.01 <0.01 343 + + + + + <0.01 <0.01 343 + + + + + 0.013 0.017 343 + + + + + <0.01 0.028 179 + + + + + <0.01 <0.01 1019 + + + + + <0.01 <0.01 2001 + + + + + 0.045 0.040 1991 + + + + + <0.01 <0.01 3020 + + + + + 0.022 3000 + + + + + 0.043 28 + + + + + 0.027 0.012 40 + + + + + 0.039 41 Table A-8: Significant results for Monotherapy without age categorization