UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Incidence, survival and mortality among women with epithelial ovarian cancer by histotype : a population… Arora, Nimisha 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2018_november_arora_nimisha.pdf [ 1.56MB ]
Metadata
JSON: 24-1.0372092.json
JSON-LD: 24-1.0372092-ld.json
RDF/XML (Pretty): 24-1.0372092-rdf.xml
RDF/JSON: 24-1.0372092-rdf.json
Turtle: 24-1.0372092-turtle.txt
N-Triples: 24-1.0372092-rdf-ntriples.txt
Original Record: 24-1.0372092-source.json
Full Text
24-1.0372092-fulltext.txt
Citation
24-1.0372092.ris

Full Text

 INCIDENCE, SURVIVAL AND MORTALITY AMONG WOMEN WITH EPITHELIAL OVARIAN CANCER BY HISTOTYPE: A POPULATION STUDY IN BRITISH COLUMBIA, CANADA by  Nimisha Arora  B.Tech., Vellore Institute of Technology, 2015  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Reproductive and Developmental Sciences)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2018                                                               © Nimisha Arora, 2018 ii   The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled:  Incidence, survival and mortality among women with epithelial ovarian cancer by histotype: A population study in British Columbia, Canada.     Examining Committee: Dr. Gillian E. Hanley Supervisor  Dr. Jessica McAlpine Supervisory Committee Member  Dr. Aline Talhouk Supervisory Committee Member Dr. Gina Ogilvie, Dr. Paul Yong Additional Examiner   Additional Supervisory Committee Members: Dr. Michael Law Supervisory Committee Member submitted by Nimisha Arora in partial fulfillment of the requirements for the degree of Master of Science in Reproductive and Developmental Sciences iii  Abstract Objectives.  Epithelial Ovarian cancer (EOC) is composed of five distinct histologic subtypes. However, histotype- specific survival and mortality estimates among women with EOC are limited. Also, we lack information on long- term health conditions faced by ovarian cancer survivors. This thesis examined: 1) histotype- specific incidence and survival rates among women with EOC in British Columbia (BC); 2) causes of death among women with EOC by histotype; and, 3) compared these causes of death to age- standardized causes of death in the general population.   Methods. Using population- based administrative datasets, I built two population- based cohorts of all women with EOC diagnosed in BC: 1. women diagnosed between 1980 and 2015(cohort 1) and women diagnosed between 1990 and 2014 (cohort 2). Cohort 1 was used to answer question 1, whereas cohort 2 was used to answer question 2 and 3. For question 2, I compared the causes of death within histotypes, by age at diagnosis, BRCA status, and time since diagnosis. For question 3, I calculated all- cause and cause- specific standardized mortality ratios.    Results. Decreasing incidence rates and increasing survival rates were observed among women with EOC in BC. As expected, ovarian cancer was the most common cause of death among these women, which was first surpassed by other causes of deaths at 11 years after diagnosis. When stratified by serous and non- serous EOCs, ovarian cancer was the leading cause of death for 12 years and for 8 years respectively after diagnosis. Of particular interest, the number of deaths from other cancers (breast, colorectal and lung cancer) and external causes (falls) among long-term ovarian cancer survivors (5-9 and 10+ years post diagnosis) were higher than the expected.  iv   Conclusions. Although there was an improvement in the survival over time, ovarian cancer remains the leading cause of death for 11 years following diagnosis of EOC. My findings suggest that long- term survivors (those living 5-9 and 10+ years following diagnosis) are particularly vulnerable to deaths from other cancers and from falls in elderly survivors. Hence, these women may benefit from closer surveillance of other cancers and bone health.     v  Lay Summary The most recent survival data in BC dates to 2005 and is in desperate need of updating to understand whether we are making progress in treating ovarian cancer. We now understand that ovarian cancer is not a single disease, but rather 5 distinct diseases, separated according to histotype. Hence, I updated the survival statistics taking histotypes into consideration. I also studied the causes of death among women with ovarian cancer, and among women surviving ovarian cancer for 5-9 years and 10+ years after their diagnosis (long-term survivors). Through this analysis, we were able to illustrate that long-term survivors of ovarian cancer are at higher risk of lung, breast and colorectal cancer and of deaths to falls. This information is important as the end of cancer treatment is not the end of cancer experience, and we must focus on improving the life of ovarian cancer survivors. vi  Preface This statement is to certify that the work in this thesis is original. This thesis is manuscript- based and it was proposed, conducted and written by Nimisha Arora. All the research work discussed in this thesis was approved by the University of British Columbia’s Behavioural Research Ethics Board: UBC BREB Number: H16‐01538. Chapter 2 is based on work conducted by Nimisha Arora (NA), Aline Talhouk (AT), Jessica N. McAlpine (JM), Michael R. Law (ML), Gillian E. Hanley (GH). NA and GH came up with the study proposal and wrote the manuscript. AT and NA conducted the data analysis. AT and GH assisted NA in the final interpretation of the data. JM, ML contributed to manuscript revision. A version of Chapter 2 will be submitted for journal publication.  Chapter 3 is based on work conducted by Nimisha Arora (NA), Aline Talhouk (AT), Jessica N. McAlpine (JM), Michael R. Law (ML), Gillian E. Hanley (GH). NA and GH came up with the study proposal and wrote the manuscript. NA performed the data analysis, thereby prepared required figures and tables. GH assisted NA in the final analysis and interpretation of the data. AT, JM, ML contributed to manuscript revision. A version of Chapter 3 has received revisions from BMC Cancer.  Chapter 4 is based on work conducted by Nimisha Arora (NA), Aline Talhouk (AT), Jessica N. McAlpine (JM), Michael R. Law (ML), Gillian E. Hanley (GH). NA and GH came up with the study proposal and wrote the manuscript. NA conducted the data analysis. GH assisted NA in the final analysis and interpretation of the data. AT, JM, ML contributed to manuscript revision. A version of Chapter 4 is in press at the International Journal of Gynecological Cancer.   vii  Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ........................................................................................................................ vii List of Tables ............................................................................................................................... xii List of Figures ............................................................................................................................. xiii List of abbreviations ....................................................................................................................xv Acknowledgements .................................................................................................................... xvi Dedication .................................................................................................................................. xvii Chapter 1: Introduction ................................................................................................................... 1 1.1 Ovarian Cancer Statistics in Canada ............................................................................... 1 1.2 Ovarian cancer is not a single disease ............................................................................. 1 1.2.1 Histologic subtypes of ovarian cancer ................................................................ 3 1.2.1.1 High- grade serous ovarian cancers (HGSC) ...................................................... 3 1.2.1.2 Low- grade serous ovarian cancer (LGSC) ........................................................ 3 1.2.1.3 Clear cell ovarian cancer (CCC) ......................................................................... 3 1.2.1.4 Endometrioid ovarian cancer (EC) ..................................................................... 4 1.2.1.5 Mucinous ovarian cancer (MC) .......................................................................... 4 1.3 Risk factors ..................................................................................................................... 4 1.3.1 The ‘Incessant ovulation hypothesis’ and associated risk factors ...................... 4 1.3.2 Hereditary risk factors......................................................................................... 6 viii  1.3.3 Hormone replacement therapy and associated hormonal therapies risk factors . 7 1.3.4 Surgical protection risk factors ........................................................................... 8 1.3.4.1 Current surgical approaches in practice .............................................................. 9 1.3.5 Lifestyle risk factors ......................................................................................... 10 1.3.6 Other factors...................................................................................................... 11 1.4 Presentation ................................................................................................................... 12 1.5 Screening: Secondary Prevention of Ovarian Cancer ................................................... 13 1.5.1 Large prospective trials of Ovarian Cancer Screening ..................................... 13 1.5.1.1 The Prostate, Lung, Colorectal and Ovarian Cancer Screening (PLCO) ......... 13 1.5.2 The Japanese Multicentral Screening ............................................................... 14 1.5.3 The United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) ......................................................................................................................... 14 1.6 Current Diagnostic Evaluation for ovarian cancer ........................................................ 16 1.7 Ovarian cancer treatment .............................................................................................. 17 1.7.1 Surgery for ovarian cancer ................................................................................ 17 1.7.1.1 Staging Epithelial ovarian cancer ..................................................................... 17 1.7.2 Chemotherapy for ovarian cancer ..................................................................... 18 1.7.2.1 Intraperitoneal chemotherapy ........................................................................... 19 1.7.3 Radiation therapy .............................................................................................. 20 1.7.4 Molecular targeted therapy ............................................................................... 20 1.7.4.1 Anti- angiogenic therapy................................................................................... 20 1.7.4.2 PARP inhibitors ................................................................................................ 21 1.8 Survival statistics for ovarian cancer ............................................................................ 22 ix  1.8.1 Increased survival among women with ovarian cancer .................................... 22 1.9 Survivorship .................................................................................................................. 24 1.9.1 Using underlying causes of death to better understand long-term health consequences among women with ovarian cancer ............................................................... 25 1.10 Purpose .......................................................................................................................... 25 1.11 Study objectives and research questions ....................................................................... 26 1.12 Thesis roadmap ............................................................................................................. 27 Chapter 2: Recent trends in ovarian Cancer incidence and survival in British Columbia by histotype. ....................................................................................................................................... 28 2.1 Introduction ................................................................................................................... 28 2.2 Methods ......................................................................................................................... 30 2.2.1 Data sources ...................................................................................................... 30 2.2.2 Study cohort ...................................................................................................... 30 2.3 Statistical Analysis ........................................................................................................ 33 2.3.1 Incidence ........................................................................................................... 33 2.3.2 Survival ............................................................................................................. 33 2.4 Results ........................................................................................................................... 34 2.4.1 Clinical characteristics of women with epithelial ovarian cancer .................... 34 2.4.2 Age- standardized incidence rates of epithelial ovarian cancer by year of diagnosis 35 2.5 Discussion ..................................................................................................................... 44 Chapter 3: Long-term mortality among women with epithelial ovarian cancer: A population-based study in British Columbia, Canada ..................................................................................... 50 x  3.1 Introduction ................................................................................................................... 50 3.2 Methods ......................................................................................................................... 51 3.2.1 Data sources ...................................................................................................... 51 3.2.2 Study cohort ...................................................................................................... 52 3.2.3 Assessment of causes of death .......................................................................... 53 3.2.4 Statistical Analysis ............................................................................................ 53 3.3 Results ........................................................................................................................... 54 3.4 Discussion ..................................................................................................................... 65 Chapter 4: Causes of death among women with epithelial ovarian cancer by length of survival post-diagnosis: A population-based study in British Columbia, Canada ..................................... 69 4.1 Introduction ................................................................................................................... 69 4.2 Methods ......................................................................................................................... 70 4.2.1 Study context .................................................................................................... 70 4.2.2 Data sources ...................................................................................................... 70 4.2.3 Study cohort ...................................................................................................... 71 4.2.4 Assessment of causes of death .......................................................................... 72 4.2.5 Statistical Analysis ............................................................................................ 72 4.3 Results ........................................................................................................................... 73 4.4 Discussion ..................................................................................................................... 77 Chapter 5: Conclusion................................................................................................................... 82 5.1 Summary of key findings and implications .................................................................. 83 5.1.1 Trends in incidence and survival rates among women with epithelial ovarian cancer in British Columbia, 1980- 2015. .............................................................................. 83 xi  5.1.2 Long-term mortality among women with epithelial ovarian cancer in British Columbia, 1990- 2014 .......................................................................................................... 85 5.1.3 Causes of death among women with epithelial ovarian cancer by length of survival post-diagnosis, 1990 – 2014. .................................................................................. 86 5.2 Research contribution to literature ................................................................................ 87 5.3 Strengths and limitations ............................................................................................... 90 5.3.1 Limitations ........................................................................................................ 90 5.3.2 Strengths ........................................................................................................... 91 5.4 Recommendations for future research .......................................................................... 91 5.5 Conclusions ................................................................................................................... 93 Bibliography .................................................................................................................................94 Appendices ..................................................................................................................................108  Stepwise diagnostic evaluation of ovarian cancer ........................................................... 108  Underlying causes of death and their ICD-10 codes ........................................................ 109  1- year, 3- year and 5- year age standardised overall survival rates among women with ‘others and unspecified’ EOCs, 1980 – 2015 ....................................................................................... 111  1- year, 3- year and 5- year age standardised disease specific survival rates among women with ‘others and unspecified’ EOCs, 1980 – 2015 ............................................................................... 112  xii  List of Tables  Table 1: ICD- 10 codes for classifying EOC histotypes ............................................................... 31 Table 2: Clinical characteristics of the study cohort. .................................................................... 34 Table 3: Percentage distribution of EOC histotypes from 1980 to 2015. ..................................... 37 Table 4: Clinical characteristics of the study cohort. .................................................................... 54 Table 5: Cause of death stratified by histotype ............................................................................. 58 Table 6: Cause of death stratified by histotype and age at diagnosis ........................................... 59 Table 7: Cause of death stratified by histotype and BRCA status ................................................. 60 Table 8: Clinical characteristics of the study cohort. .................................................................... 74 Table 9: Standardized Mortality Ratios (SMRs) in women with ovarian cancer according to years of survival post-diagnosis by age. ................................................................................................. 75 Table 10: Causes of death among women with ovarian cancer and in the general population .... 76 Table 11: Standardized Mortality Ratios (SMRs) in women with ovarian cancer by years of survival post diagnosis according to cause of death. .................................................................... 77    xiii  List of Figures  Figure 1: Age standardized incidence rate of EOCs (all histotypes combined), 1980 – 2015. .... 36 Figure 2: Age standardized incidence rate, stratified by histotype, 1980 – 2015. ........................ 36 Figure 3:  1- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015. ...................................... 39 Figure 4: 3- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015.   .................................... 40 Figure 5: 5- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015. ...................................... 41 Figure 6: 1- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015. ........................ 42 Figure 7: 3- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015. ........................ 43 Figure 8: 5- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015. ........................ 44 Figure 9: Frequency distribution of deaths among patients diagnosed with all the histotypes. ... 62 Figure 10: Frequency distribution of deaths among patients diagnosed with serous epithelial ovarian cancers.............................................................................................................................. 62 Figure 11: Frequency distribution of deaths among patients diagnosed with non- serous epithelial ovarian cancers.............................................................................................................................. 63 Figure 12: Frequency distribution of deaths among older patients (>= 60 years) diagnosed ....... 63 xiv  Figure 13: Frequency distribution of deaths among younger patients (aged under 60 years) diagnosed with serous epithelial ovarian cancers. ........................................................................ 64 Figure 14: Frequency distribution of deaths among older patients (aged 60 years or more) diagnosed with non- serous epithelial ovarian cancers. ................................................................ 64 Figure 15: Frequency distribution of deaths among younger patients (aged under 60 years more) diagnosed with non- serous epithelial ovarian cancers. ................................................................ 65  xv  List of abbreviations          Body mass index (BMI) British Columbia (BC)           Epithelial ovarian cancer (EOC) Standard deviation (SD)          Standardized mortality ratios (SMR)  xvi  Acknowledgements  In this beautiful journey, I have found a teacher, an inspiration and a pillar of support in my supervisor, Dr. Gillian Hanley. Without Gillian’s faith in my abilities, this thesis would not be possible. She has always been patient with me, giving me enough space and time to translate my thoughts into an actual research. Her wisdom, guidance, dedication and positive attitude are no less than core elements of this research. Thank you for your precious time and effort in creating an amazing learning opportunity for me. I would also like to thank my thesis committee members, Dr. Jessica N. McAlpine, Dr. Michael R. Law and Dr Aline Talhouk for providing appreciable insights into the project. This research would not have reached a certain level without your support. Your invaluable guidance, motivation and to the greatest your time has meant a lot to me.    I am sincerely grateful to Dr. Anthony Karnezis for providing help with histotype- classification. Dr. Anna Tinker was willing to discuss my research findings and she was a major help in better understanding the results of Chapter 2. Also, I would like to thank Cathy Qiaoyue Tang for helping me with survival analysis. I must express my gratitude towards my ex program assistant Roshni Nair and her family. Rosh, thank you for blessing me with home away from home and showering me with your love and support.   My sincere thanks go to all the funding sources supporting my master’s study- Canadian Cancer Society Research Institute, the Canadian Institutes for Health Research, as well as by (donor) funds from the Vancouver General Hospital and University of British Columbia Hospital Foundation. Also, I am grateful to have received International Tuition Award and Faculty of Medicine Graduate Award from the University of British Columbia.   I would like to thank my family and friends for their love, care and support. To my parents who always trusted me with my decisions without a second thought. Thank you both for having my back and keeping me on track during challenging times. Your strength, love and blessings have brought the best out of me. To my sister Nikita who was amazingly understanding when it came to graduate school commitments. Thank you for always seeing the spark in me and helping me stay hopeful. Your laugh is the best medicine. To Jennet, thank you for sharing the same amount of craziness and being a genuine friend all this time. Vancouver would not have been the same without you. To Tam, Minnie, Sarah, Akshita, Jordan, Miyue, Jammay, Daphne and Brian for being generous with timely support and encouragement. To my friends in India, thank you all for always motivating me to step out of my comfort zone. Finally, I would like to thank Karan for being the most supportive and understanding partner I could ever imagine. Thank you for always believing in me, even when I did not believe in myself. On top of all, thank you for being my rock.  xvii  Dedication  This study is wholeheartedly dedicated to my beloved parents, Bharat Kumar Arora and Anita Arora, who have been my ultimate source of unconditional love, support and encouragement. This thesis would not be possible without their valuable presence and constant faith in me.   Mummy and Papa, I am truly proud to be your daughter and I would always appreciate the sacrifices you have made to bless me with the life that I have. Thank you for everything.   1  Chapter 1: Introduction  1.1 Ovarian Cancer Statistics in Canada Ovarian cancer is the most lethal cancer of the female reproductive system [1]. While the general population lifetime risk of ovarian cancer is relatively low at 1.4% to 1.7%, it is the seventh most commonly diagnosed cancer [2] and the eighth most common cause of cancer mortality for women worldwide [3]. Within Canada, there were 2,800 new cases of ovarian cancer in Canada in 2017, and 1700 women died of the disease [4], making ovarian cancer the ninth most commonly diagnosed cancer and the fifth most common cause of cancer deaths among Canadian women [5]. There is a crude incidence rate of 17.3 per 10,000 in British Columbia [6].  Although 5-year survival rates have marginally improved over the past decade, survival remains at less than 50% [7].   1.2 Ovarian cancer is not a single disease Ovaries are composed of three types of cells: germ cells, stromal cells and epithelial cells. Consequently, ovarian cancers are classified into cancers of non-epithelial origin (cancers arising in the germ cells or stromal cells) and epithelial origin (cancers arising in the epithelial cells). Generally, 90% of ovarian cancers are epithelial ovarian cancers (EOCs). In this thesis, I focus only on EOC. In the past, EOC was considered a single disease. However, recent studies have indicated that EOC represents a group of five distinct histologic subtypes [8,9]- high-grade serous cancer, endometrioid ovarian cancer, clear cell ovarian cancer, mucinous ovarian cancer, and low-grade serous cancer. Research has suggested that each of these histotypes have different cellular origins, morphology, and molecular features that dictate clinical behavior, including 2  treatment response and overall survival. In addition, there are different risk factors associated with the different histotypes of ovarian cancer, and different prevention approaches for each histologic subtype as well [10-14].  The ovary is the most frequent site of the dominant ovarian cancer tumor mass and this led to the focus on cortical inclusion cysts (tubal CICs) within the ovary as the probable source of the disease.  Tubal CICs were postulated to arise from transformation of the ovarian surface epithelium trapped within the ovary after ovulation. Our understanding of the fallopian tube’s involvement in ovarian cancer changed when examination of the fallopian tubes removed at risk reducing salpingectomy oophorectomy from BRCA mutation carriers (women at high risk for ovarian cancer, detailed further below) revealed the presence in the distal fallopian tube (the fimbriae) of tubal malignancies in 2-10% of these high-risk women [15-20]and preinvasive lesions in the fimbriae (serous tubal intraepithelial cancers; STICs) in 1-8% of these women [21-23].  The Sectioning and Extensive Examining of the Fimbria (SEE-FIM) protocol was developed to maximize the detection of ovarian cancer precursors or early fallopian tube cancers by sectioning and examining the fallopian tube fimbriae (FTFE) for pathology [24,25]. This protocol has revealed tubal involvement in up to 70% of unselected women diagnosed with ovarian or primary peritoneal High Grade Serous Cancer (HGSC) (with and without BRCA mutation), [26-29] including the presence of fimbrial STICs in 40-60% of these women [28]. Importantly, STICs occur extremely rarely in women with benign gynecologic conditions [30].  3  1.2.1 Histologic subtypes of ovarian cancer  1.2.1.1 High- grade serous ovarian cancers (HGSC) High-grade serous ovarian cancers are most common form of EOCs, comprising around 70%[31]. Although the pathogenesis is not completely understood, the majority of HSGCs are believed to originate in the fallopian tube [32]. These tumors are commonly diagnosed in older woman [11]. HGSCs usually present in advanced stages with a 5- year overall survival rate of 32% [33]. Women with HGSC initially respond well to cisplatin (first- line chemotherapy), but they eventually develop resistance to chemotherapy, relapse and succumb to their ovarian cancer [34,35].  1.2.1.2 Low- grade serous ovarian cancer (LGSC) Low- grade serous ovarian cancers are rare, representing only 3-5% of all EOCs [36]. These tumors are commonly diagnosed at younger age compared to HGSC [11], with a 5- year overall survival rate of 84% when detected early [33]. Women diagnosed with LGSC have markedly better survival compared to those diagnosed with HGSC [34].  1.2.1.3 Clear cell ovarian cancer (CCC) Clear cell carcinomas are the second most common subtype of ovarian cancer, representing 10-26% of all epithelial ovarian cancers[36]. Although the majority of CCCs are detected at early stages with 5- year overall survival rate of 82% [33], they do not respond well to chemotherapy. These cancers have higher rates of relapse than the other histotypes given similar treatment [37]. Some evidence has suggested an association between CCC and endometriosis [35].  4  1.2.1.4 Endometrioid ovarian cancer (EC) Representing 7-24% of all EOCs, endometrioid carcinomas are the third most common subtype of ovarian cancer [36]. The majority of EC cases are detected at early stages, thereby presenting favorable clinical outcomes with 5- year overall survival rate of 87% [33]. These tumors are also associated with endometriosis [36].   1.2.1.5 Mucinous ovarian cancer (MC)  Mucinous carcinomas are the least common subtype of ovarian cancer, accounting for 2-6% of all EOCs [32].  Mucinous ovarian cancers are often diagnosed at an early stage with a 5- year overall survival rate of 83% [35].  Like CCC, MC also respond poorly to chemotherapy [37].  1.3 Risk factors Risk factors for ovarian cancer have been relatively well studied [38-44]. Herein I will describe the incessant ovulation hypothesis and its many associated risk factors for ovarian cancer. I will also outline hereditary risk factors, associations between ovarian cancer and hormone therapy, lifestyle and surgical history.  1.3.1 The ‘Incessant ovulation hypothesis’ and associated risk factors  One of the earliest hypotheses surrounding the development of ovarian cancer was the ‘incessant ovulation hypothesis’ [45]. This theory posits that the inflammation in the fimbrial end of the fallopian tube associated with ovulation is believed to promote ovarian carcinogenesis. Epidemiologic studies have consistently observed that decreasing the lifetime number of ovulatory cycles decreases ovarian cancer risk, with multiple studies reporting a reduced risk associated with oral contraceptive use , greater risk reduction with longer duration of oral 5  contraceptive use, [46,47] and increased protection with increased numbers of pregnancies. The association between increased parity (or pregnancy) with lower ovarian cancer risk is well established [43,44,48]. With each birth, a risk reduction of 20% was observed, particularly for endometrioid and clear- cell EOCs [44].  The protective effect of pregnancy has also been associated with age at pregnancy as well.  For instance, risk of ovarian cancer was observed to be higher among women who were pregnant at an older age (~ 35 years) in comparison to women who were pregnant at a younger age (~25 years) [42]; however, age at first pregnancy requires further study as it is unclear whether the relationship is confounded by other factors associated with choosing a later age at first birth. Also, multiparity was also linked with 50% reduced risk of ovarian cancer [43]. Breastfeeding is also linked with reduced risk of ovarian cancer. A reduced risk of 20-25% in developing ovarian cancer among parous women who have breastfed compared to women with the same parity who did not breastfeed has been reported [48], and reduced risk is associated with longer durations of lactation time. However, most studies report that one pregnancy is considerably more protective than a single year of oral contraceptive use[49], making the direct link between pregnancy and number of ovulatory cycles seem overly simplistic and calling into question the validity of the incessant ovulation hypothesis.  Other risk factors posited to result from inflammatory processes include pelvic inflammatory disease and talcum powder use. Pelvic inflammatory disease results in inflammation of ovaries, fallopian tubes and endometrium [50]. Currently, the association between pelvic inflammatory disease and ovarian cancer risk is unclear. However, based on the limited evidence, women with pelvic inflammatory disease are at increased risk of developing ovarian cancer in comparison to those with no history of the disease [42]. Epidemiological studies have suggested an associated between perineal use of talc and ovarian cancer [51,52]. 6  This association came into the picture in 1960’s, when talcum powder may have contained asbestos (a potential carcinogen) [39]. It was believed that the powder particles resulted in inflammation in the ovaries by travelling up to the reproductive system. However, there are conflicting results in the literature regarding the link between talcum powder and ovarian cancer [53,54]. 1.3.2 Hereditary risk factors The most common hereditary cancer syndromes that cause ovarian cancer are hereditary breast-ovarian cancer syndrome and lynch syndrome. Hereditary breast- ovarian cancer syndrome, accounting for 65% - 85% of hereditary ovarian cancers, is caused by inherited mutations in BRCA1 and BRCA2 genes. Whereas, lynch syndrome is linked with inherited mutation in DNA mismatch repair genes and it is responsible for 10%-15% of the hereditary cases[55].  Women at high- risk of developing ovarian cancer due to their inheritance of germline BRCA1 and/or BRCA2 mutations have an average lifetime risk for ovarian cancer of between 35-46% and 13-23%, respectively [56-59]. Inherited germline mutations of BRCA1 and BRCA2 account for approximately 11.7% to 20% of all the invasive ovarian carcinomas [60-63]. Most hereditary ovarian cancers are high- grade serous cancers [48]. Historically women with ovarian cancer were tested for inherited BRCA mutations if they had a family history of breast and/or ovarian cancers and/or are of Ash¬kenazi Jewish ethnicity [1,39,42]. We are currently moving into an era of histotype-based referral for hereditary testing as up to 1/5th of women with HGSC will test positive for a BRCA1 or BRCA2 mutation [64], and many of these women would not have qualified for hereditary cancer testing based on family history criteria [65].  7  Lynch syndrome is associated with increased risk for non- serous ovarian cancers, particularly endometrioid and clear cell EOCs[55,66,67]. Although Lynch Syndrome is primarily known for its dramatic increase in lifetime risk of colorectal cancer (or Hereditary NonPolyposis Colorectal Cancer, HNPCC), women with Lynch syndrome have an approximately 12% lifetime risk of developing ovarian cancer (compared with 1.4% in the general population) [42]. This syndrome accounts for only 1% of all EOCs [41].  Even after accounting for genetic mutations known to increase risk for ovarian cancer (BRCA and Lynch), increased risk remains among women who have first-degree relatives (daughter, sister, mother) with ovarian cancer on either father’s or mother’s side of their family and are found not to have BRCA mutations or Lynch syndrome [39].  The risk gets higher with the increasing number of relatives with ovarian cancer. Family history of other cancers, including breast cancer, uterine cancer, pancreatic and colorectal cancer can also increase risk for ovarian cancer [39,40]. 1.3.3 Hormone replacement therapy and associated hormonal therapies risk factors The use of hormone replacement therapy has been linked with increased risk for ovarian cancer, particularly for serous and endometrioid cancers [43]. Research has suggested that the risk of ovarian cancer is 40% higher among women taking hormone replacement therapy compared with those not taking hormone replacement therapy [38]. Though, this increased risk seems to gradually disappear over time. For example, women who took hormone replacement therapy for 5 years or more are at a significant risk of developing ovarian cancer for at least 5 years since stopping treatment [43,48].  Secondly, the association between infertility (or fertility drugs) and ovarian cancer remains controversial [38,42,43]. Early studies have reported an 8  increased risk of ovarian cancer with the use of fertility drugs [68,69]. However, majority of recent studies have observed no association [70,71].  1.3.4 Surgical protection risk factors  Previous studies have reported a significant inverse relationship between gynecologic surgeries (Bilateral salpingooophorectomy, hysterectomy, salphingectomy and tubal ligation) and risk of ovarian cancer.  Among high-risk women (BRCA mutation carriers), a risk reduction of at least 90% is achieved by bilateral prophylactic salpingooophorectomy (a surgery involving removal of both fallopian tubes and ovaries) [50]. Surgical removal of fallopian tubes hinders the cancer initiation, specifically for high- grade serous EOCs which originates from the distal end of fallopian tube and block the transport of potentially cancerous cells to ovaries or peritoneal cavity[72].  Tubal ligation (a form of permanent sterilization that generally cauterizes fallopian tubes) has been shown to decrease ovarian cancer risk by 29% overall, but there are differences in this effect across subtype [73,74]. In the detailed analysis of primary individual data from studies in the Ovarian Cancer Association Consortium (OCAC), the greatest risk reduction was found in endometrioid EOCs (52%), followed by clear cell EOCs (48%), and there was a 20% reduction in risk of high grade serous carcinomas [74]. Surgical ligation is hypothesized to have a greater protective effect against endometrioid and clear EOCs (endometriosis- associated ovarian cancers) due to the blockage of retrograde menstruation flow (transport of endometrial cells to peritoneal cavity and on to ovary through the fallopian tubes) [74]. Another recent meta-analysis of published results reported similar findings [73], and a case-control study by the Hereditary Ovarian Cancer Clinical Study Group reported that tubal 9  ligation decreased the risk of ovarian cancer by 60% in BRCA1 or BRCA2 mutation carriers. Results by histologic subtype were not reported [75].  There are also encouraging data on a small number of excisional tubal surgery cases (defined as complete salpingectomy, distal fimbriectomy or partial salpingectomy). Researchers from the Rochester epidemiology project reported a 64% reduction in the risk of ovarian cancer after excisional tubal sterilization compared with those without sterilization or with non-excisional tubal sterilization (Odds ratio, OR =0.36; 95% confidence interval, 95%CI= 0.13, 1.02) [76]. Danish researchers using a national database, reported that bilateral salpingectomy (removal of both the fallopian tubes while sparing the ovaries) reduced risk for ovarian cancer by 42% (OR=0.58; 95% CI= 0.36, 0.95) [77]. Most recently, a retrospective population-based study using Swedish health registers reported that bilateral salpingectomy was associated with a 65% reduction in risk (Hazard ratio, HR=0.65, 95% CI= 0.52, 0.81) [78]. This Swedish study reported that other surgeries such as hysterectomy (removal of uterus), and hysterectomy with bilateral salpingectomy, were also associated with statistically significant reductions in ovarian cancer risk (HR = 0.79, 95% CI= 0.70, 0.80; HR= 0.06, 95% CI= 0.03, 0.12 respectively). A recent meta-analysis, including the three studies described above, concluded that there was a significant decrease in the risk of ovarian cancer occurrence in patients who underwent bilateral salpingectomy relative to controls (OR=0.51; 95% CI 0.35, 0.75) [79].  1.3.4.1 Current surgical approaches in practice The standard of care is bilateral salpingo- oophorectomy among women with BRCA1 or BRCA2 mutations. This surgical approach, involving removal of both the ovaries and fallopian tubes, is associated with a 80% to 90% reduction in ovarian cancer risk [18,19,25].  However, it is not recommended in the general population due the fact that it puts women into surgical 10  menopause, increasing their risk for osteoporosis, coronary heart disease and stroke [80]. Given the health risks associated with bilateral salpingo- oophorectomy and given our understanding of the fallopian tube as the sources of origin for most HGSCs, bilateral salpingectomy has been proposed a prevention strategy in the general population.  Recommendations for opportunistic salpingectomy were made by the Ovarian Cancer Research team (OVCARE) in British Columbia in 2010, similar recommendation from the Society of Gynecologic Oncology of Canada [81], and later by the Society of Gynecologic Oncology [82]. More recently the American College of Obstetricians and Gynecologists and the Society of Obstetricians and Gynecologists of Canada have supported the recommendation [83,84].  It has been advocated to consider: 1) performing bilateral salpingectomy at the time of hysterectomy; 2) performing bilateral salpingectomy instead of tubal ligation for permanent contraception- commonly referred to as opportunistic salpingectomy. OVCARE  reported encouraging results on the safety of opportunistic salpingectomy; no additional risks associated with this treatment procedure [85]. However, long- term studies are needed to evaluate the effectiveness of opportunistic salpingectomy among women in the general population.   1.3.5 Lifestyle risk factors The association between diet and ovarian cancer risk is highly inconsistent, ranging from positive to no effect of dietary factors on the risk of ovarian cancer. Based on a relatively limited literature, dietary factors such as vegetables, whole- grain food and low-fat milk are believed to reduce the risk of the disease [86], whereas, high intake of saturated fats increases the risk of ovarian cancer [87]. Studies report inconclusive effects of meat, tea, coffee, and Vitamin D consumption on ovarian cancer [50]. Consistent with diet- specific findings, the association 11  between exercise and ovarian cancer risk are also not firmly established. However, the most active women are observed to have 20% lower risk of ovarian cancer in comparison to the least active women [50]. Another study reported reduced risk among women performing recreational biking and walking compared with women who did not exercise [43].  Smoking was not considered a risk factor for ovarian cancer until the association was studied separately for histotypes [88,89]. A metanalysis involving 51 epidemiological studies reported 50% increase in ovarian cancer risk among women who smoke with mucinous EOCs compared to never smokers [89]. This finding supported a hypothesis that mucinous EOCs resemble mucinous gastrointestinal cancers, which have also been associated with smoking [50].  Some studies have reported (modest) increases in risk of ovarian cancer among obese women [42,48]. For instance, premenopausal women with BMI greater than 30 are at higher risk of developing ovarian cancer (adjusted OR = 2.19; 95% CI 1.19-4.04) in comparison with premenopausal women with BMI less than 25 [90]; however, BMI is not currently considered an important risk factor for ovarian cancer.  1.3.6 Other factors  There is an increased risk of developing ovarian cancer, particularly clear- cell EOCs and endometrioid EOCs, among women with endometriosis (a condition where a uterine tissue from grows outside the uterus). In a meta-analysis of 15 cohort and 20 case- control studies, the relative risks of developing clear cell and endometrioid EOCs among self reported endometrioid cases (reported history of endometrioid) were 2.606 and 1.759 respectively [91].    12  Table 1: Risk factors associated with increased or decreased risk for EOCs Decreased EOC risk Oral contraceptive use Gynecologic surgeries Breast feeding Diet Exercise Increased parity Increased EOC risk Mutated BRCA genes Lynch Syndrome Family history of ovarian cancer Endometriosis Personal history of cancer Hormonal replacement therapy  Infertility and fertility treatment Pelvic inflammatory disease  Perineal Use of talcum powder (or talc) Smoking Obesity  1.4 Presentation Ovarian cancer is often diagnosed at advanced stages (Stage III and IV), primarily because symptoms are not common in the early stages (Stage I and II), and when they do occur, the symptoms may be general and non-specific such as bloating, urinary frequency, loss of appetite, back pain [92]. Hence, ovarian cancer is often referred to as the silent disease. Approximately 60% of cases are detected at advanced stages, when cancer has spread beyond the ovary to distant lymph nodes or other organs [36]. Since the survival rate is correlated with the stage at diagnosis, women with ovarian cancer diagnosed in late stages have a significantly lower 5- year survival rate (17% to 28%) [1] than women diagnosed in early stages (92%) [1,93]. However, detecting ovarian cancer early remains very difficult, as despite the tremendous international effort [94-97], no effective screening methods are currently available [98].  13  1.5 Screening: Secondary Prevention of Ovarian Cancer Cancer screening is defined as the detection of cancer at an earlier and more curable stage, using a test or a combination of tests, on an asymptomatic population [99]. The objectives of a cancer-screening program are reducing the number of deaths from the disease in a population, followed by increasing the life expectancy of cancer patients [100]. Hence screening programs are widely applied only when they result in a significant reduction in mortality from the disease in question.  Below I am reviewing 3 major ovarian cancer screening trials.  These studies evaluated the efficacy of screening by determining whether they result in meaningful reductions in mortality and result in a significantly earlier detection in the cancer (measure by stage). A brief review of this research is relevant to understand why survival rates have been slow to improve in this disease.  1.5.1 Large prospective trials of Ovarian Cancer Screening 1.5.1.1 The Prostate, Lung, Colorectal and Ovarian Cancer Screening (PLCO) The PLCO trial was a randomized controlled trial conducted in the USA [94]. It included 78,216 women aged 55 to 74, of which 39,105 received annual screening (intervention group) and 39,111 received usual care (control group). The annual screening was performed, using CA-125 level measurements (CA-125) and transvaginal sonography (TVS). The intervention group received annual screening with CA-125 and TVS for 4 years, followed by CA-125 only for the next 2 years while the control group received usual medical care, which consisted of no screening for ovarian cancer.  Participants were followed for a median of 12.4 years. Ovarian cancer was diagnosed in 176 women (0.45%) in the control group and 212 women (0.54%) in the 14  intervention group. Among screen-detected cases, 77% of cases were detected at the advanced stages. Thus no mortality reduction was observed with the application of annual screening, and this method of screening for ovarian cancer was concluded to be ineffective. 1.5.2 The Japanese Multicentral Screening The Schizuoka Cohort Study of Ovarian Cancer Screening (SCOCS) was performed in Japan [96]. The study was a randomised controlled trial, including 82, 487 low-risk post menopausal women. Of 82, 487 participants, 41,688 received annual screening (intervention group) and 40,799 did not (control group). Corresponding to PLCO trial, this study performed annual screening using CA-125 and TVS.  In the intervention group, 27 women (0.06%) were diagnosed of ovarian cancer, with relatively good sensitivity (77.1%) and specificity (99.9%).  Among screen-detected cases, 17 (63%) were Stage I, 1 (4%) were Stage II, 7 (26%) were Stage III, and 2 (7%) were Stage IV. Thus over 2/3rds of cases were detected at the early stages, but this rise in the detection of early-stage ovarian cancer was not significant, and ovarian cancer mortality in the intervention and control groups was not reported, and thus we cannot conclude that screening reduced mortality. 1.5.3 The United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) The UKCTOCS is the largest screening trial to date [95]. It was a RCT of 202,638 women, including control group (n= 101,359); multimodal screening group (n= 50,640) and ultrasound screening group (n = 50,639). The multimodal screening (or MMS) group underwent annual screening, using CA-125 level with TVS as a second line test. A major step that differentiates this trial from other studies is “Risk of Ovarian Cancer (ROCA)” algorithm. The patented algorithm was used to interpret CA-125 levels in the MMS group, instead of using a fixed cut off 15  value. The ROCA algorithm calculates a woman’s level of risk based on her CA-125 levels and changes in these values over time combined with known risk factors [101]. The ultrasound screening (USS) group underwent screening by TVS alone, and the control group received usual care (consisting of no screening for ovarian cancer).  Participants were followed for 11.1 years. Of 202,638 women, ovarian cancer was diagnosed in 1282 women: 630 (0.6%) in the control group, 338 (0.7%) in the MMS group and 314 (0.6%) in the USS group [102]. Over the first 14 years of the study, an average mortality reduction of 15% (8% for years 0-7; 23% for years 7-14) was reported in the MMS group versus control group and 11% (2% for years 0-7; 21% for years 7-14) in the USS group versus control group. However, a significant mortality reduction of 28% was observed with the MMS group in years 7 to 14 years, when prevalent cases were excluded. False-positive surgeries were performed in 14 and 50 per 10,000 screen detected cases in the MMS and TVS groups, respectively.  Among women in the MMS group, a greater number of cancers were detected at an early stage in comparison to the control group. However, 641 patients need to be screened annually using the MMS method for 14 years to prevent one cancer death, raising questions about the cost-effectiveness of this method.  Recent cost-effectiveness studies suggest that the cost of the ROCA algorithm will be critical to determining cost-effectiveness [103,104], along with ensuring that mortality reductions do reach significance following further follow up in the MMS group.  In conclusion, despite rigorous research on ovarian cancer screening, none of these trials has observed a significant reduction in either ovarian cancer- specific or all cause mortality. While the results from the UKCTOCS are promising and we await further follow-up to 16  determine whether the mortality reduction will become statistically significant, until such evidence is available, ovarian cancer screening cannot be recommended in the general population. Even if the UKCTOCs trial shows a modest reduction in mortality among the MMS group, rigorous cost-effectiveness analyses would need to be done accounting for the price of the ROCA algorithm before recommendations for screening among the general population would be made. However, it is possible that women who are at higher risk for ovarian cancer, including BRCA and Lynch mutation carriers, could benefit from screening. Some experts have already recommended that BRCA mutation carriers who have not yet undergone bilateral salpingo-oophorectomy be screened with TVS, CA-125, and/or pelvic exam every six months beginning at one of the following scenarios: between 30 and 35 years of age; 35 years of age; 5 to 10 years before the onset of first ovarian case in their family [105].  1.6 Current Diagnostic Evaluation for ovarian cancer The first step in the diagnostic evaluation for ovarian cancer is the assessment of symptoms (abdominal pain, bloating, frequent urination etc) and risk factors. If the symptoms indicate ovarian cancer, a complete physical examination is conducted to investigate abdominal and pelvic masses. Women with abnormal masses (or suspected ovarian cancer) are evaluated by TVS and CA-125 biomarker test. If these women are suspected to have ovarian cancer based on these examinations, they should be referred to a gynaecologic oncologist who could conduct biopsy to confirm the presence of cancer (Appendix A) [1].  17  1.7 Ovarian cancer treatment Approximately 75% of affected women present at an advanced stage (Stage III and Stage IV), requiring platinum-based chemotherapy and optimal cytoreductive surgery. Owing to major advances in these standard treatment strategies (the addition of paclitaxel to platinum in the late 1990s, BRCA1 and BRCA2 testing with prophylactic surgery, use of intraperitoneal chemotherapy, anti-angiogenic therapies, and polyADP-ribose polymerase (PARP) inhibitors)[106], a few studies have suggested an improvement in 5- year survival [7]. However, cancer in at least 90% of women with diagnosed with (high-grade serous) ovarian cancer reoccurs, alongside resistance to chemotherapy [107] and most women succumb to the disease [108].   Presently, the standard of care for EOC involves surgical debulking and platinum-based chemotherapy. The order in which these two treatments are administered depends on several factors, including distribution of disease, health/nutritional status of the patient including frailty/age, tumor grade, histotype and molecular features (reflecting chemosensitivity). 1.7.1 Surgery for ovarian cancer If removal of all tumor burden is deemed achievable, surgery is often the preferred first step. Ovarian cancer is a surgically staged disease with the best outcomes observed in women who undergo surgery by a gynecologic oncologist [109]. Surgery is performed to assess the spread of cancer (staging) and to remove as much tumor as possible (optimal cytoreduction). Surgery for ovarian cancer should be performed by a gynaecologist oncologist [110,111].  1.7.1.1 Staging Epithelial ovarian cancer Staging determines where the cancer has spread from its origin (stage) and provides tumor for pathology interpretation (grade, histotype) to direct treatment. Staging consists of thorough 18  examination of the pelvis and abdomen. Peritoneal washings or removal of ascites is performed, followed by hysterectomy (removal of uterus), salpingectomy (removal of the fallopian tubes), oophorectomy (removal of the ovary), bilateral oophorectomy (removal of both the ovaries), omentectomy (removal of the omentum, the layer of fatty tissue covering abdominal intestines) and/or removal of lymph nodes. In the vast majority of cases, a hysterectomy with bilateral salpingo-oophorectomy is performed. 1.7.2 Chemotherapy for ovarian cancer Chemotherapy is a systemic therapy, where drugs are administered into the bloodstream to kill cancer cells through out the body. In simpler terms, systemic chemotherapy treats the entire body and not just the organ(s) affected by the cancer. Chemotherapy can be given after the surgery to remove remaining cancer cells (adjuvant chemotherapy) or before the surgery to reduce the tumor size (neoadjuvant chemotherapy). Adjuvant chemotherapy is specifically administered to the cases, where cancer has spread beyond the ovaries [1]. Neoadjuvant chemotherapy is meant for the advanced staged cases, when a patient is unable to tolerate surgery.  Chemotherapy for ovarian cancer uses a combination of a platinum drug (carboplatin or cisplatin) and a taxane drug (paclitaxel/taxol or docetaxel). The most commonly used combination is carboplatin with taxol. Typically, the drugs are administered intravenously every 3 to 4 weeks [112].  For adjuvant chemotherapy, chemotherapy can be primarily performed in two ways: g Intravenous (IV) chemotherapy (injecting drugs into a vein), and intraperitoneal (IP) chemotherapy (injecting drugs into abdominal cavity using a thin tube). Of note, IP chemotherapy is generally administered as a combination of IV and IP chemotherapy [113].   19  1.7.2.1 Intraperitoneal chemotherapy  Intraperitoneal chemotherapy involves direct delivery of chemotherapy drugs to the peritoneal cavity. This rationale is supported by preclinical and pharmacokinetic data. For instance, there was a 20- fold and 1000-fold greater concertation in the abdominal cavity with IP administration in comparison to IV administration [114]. IP chemotherapy is most suitable for optimally cytoreduced cases as direct penetration of this therapy is evident for tumors less than 3mm [113]. Three large chemotherapy trials for ovarian cancer have suggested improved survival for IP administration over IV administration among optimally debulked Stage III patients [115-117]. The most recent of these trials i.e. GOG-172 study compared IV cisplatin-paclitaxel (IV arm) and a course combining IV paclitaxel and IP cisplatin-paclitaxel (IV/IP arm)[115]. This trial observed a median overall survival of 65 months in the IV/IP arm in comparison to 49.7 months in the IV arm. Overall, a significant improvement in survival was noted (i.e median overall survival difference of 16 months) after IP administration compared with that after conventional IV chemotherapy.  Although IP chemotherapy showcased survival advantage over IV chemotherapy, it is associated with toxic effects such as metabolic toxicity, myelotoxicity, gastro- intestinal toxicity, neurologic toxicity and worse quality of life [118]. Hence, a significant proportion of patients were not able to complete the treatment in the IV/IP arm. For instance, in GOG-162 study, only 42% patients completed all cycles of the IV/IP arm compared to 83% in the IV arm. Likely owing to this increased toxicity, the use of IP chemotherapy has been slow in gaining acceptance in clinical practice [119]. 20  1.7.3 Radiation therapy Radiation is rarely used in treating ovarian cancer. However, it can be considered for certain histotypes of ovarian cancer, primarily clear cell, endometrioid, and mucinous EOCs [120].  1.7.4 Molecular targeted therapy Molecular targeted therapy acts on specific molecules within the cancer cells disrupting signals necessary for the cancer cells to grow or divide. Herein I have focused on: 1. Anti- angiogenic therapy and 2. PARP (Polyadenosine diphosphate ribose polymerase) inhibitors. Other therapies involve targeting the epidermal growth factor receptor, the folate receptor alpha, the WEE1 tyrosine kinase, the PI3 K/AKT and RAS/RAF/MEK pathways [7,106]. I have focused on the anti-angiogenic therapy and the PARP inhibitors as there is the most evidence for increased survival with these targeted therapies [7,107,108]. 1.7.4.1 Anti- angiogenic therapy Tumor angiogenesis is a key component in sustaining cancer. In the process, normal cells are stimulated by tumor cells to produce key mediators [121], also termed as angiogenic molecules such as the vascular endothelial growth factor (VEGF).  As a result, new blood vessels are produced, which facilitate growth, metastasis, and survival of tumor cells.  Moreover, VEGF is observed to be over-expressed in ovarian cancers and is associated with cancer progression and poor prognosis [7]. To disrupt tumor angiogenesis, anti- angiogenic agents are required which basically prevent the interaction between VEGF ligand and its receptors. Presently, bevacizumab (a monoclonal antibody) is the most studied anti-angiogenic agent in EOC.  However, it is associated with several toxicities, including hypertension, and bowel perforation. 21  There are also patients who are resistant to therapy, and there is a high cost for bevacizumab treatment ($79,086 CAD for 1.1 quality-adjusted life-year) [7,108,122]. 1.7.4.2 PARP inhibitors A single cell can experience at least 10,000 injuries in its DNA on a regular basis, resulting from environmental factors (UV light) or normal DNA replication [123]. The resulting DNA damage is mediated by key DNA- repair pathways such as homologous recombination (HR), and non-homologous end joining (NHEJ) [110]. The HR pathway is responsible for repairing double stranded DNA breaks, whereas NHEJ (or PARP) pathway is responsible for repairing single stranded DNA breaks. These pathways are controlled by two sets of genes, BRCA1 and BRCA2. Mutations in BRCA genes are a major risk factor in developing ovarian cancer. The HR pathway of DNA repair is disrupted in these mutated deficient cells. Consequently, these cells must depend solely on the PARP pathway for cell survival [106].  This biological mechanism was used to propose the idea of “synthetic lethality”. The PARP pathway, the only DNA repair pathway in BRCA1 and/or BRCA2 deficient cells, is targeted by PARP inhibitors. Inhibition of the PARP pathway leads to the accumulation of DNA damage (single- strand and double strand DNA breaks) and thereby cell death. Once both the repair pathways have non-functioning proteins, the cancerous cells cannot survive [123]. Olaparib is the most studied PARP inhibitor in ovarian cancer. Since normal cells can repair DNA damage using both the DNA repairing mechanisms, the cells are not susceptible to the synthetic lethality. Under the condition of PARP inhibition, the normal cells can still be repaired by the HR pathway [107].Hence, the introduction of PARP inhibitors was a milestone strategy in treating ovarian cancer among women with BRCA mutations, while sparing normal cells.  22  Previous literature has reported improved survival among ovarian cancer patients treated with olaparib. A randomised trial of 265 women with BRCA mutation and recurrent high- grade serous cancer received either olaparib (treatment group) or placebo (control group) as maintenance therapy following their platinum-based chemotherapy. Among women treated with olaparib, the progression free survival was 8.4 months compared to 4.8 months in the control group [124]. Another analysis studying the effects of Olaparib on progression free survival among women with relapsed serous ovarian cancer by BRCA mutation status reported similar results. In this study, 51% of women were BRCA mutation carriers. Of these women with BRCA mutation, the median progression free survival was 11.2 months for women treated with olaparib, whereas the median progression free survival was 4.3 months for women treated with placebo. While overall survival did not differ significantly between the groups (with and without BRCA mutation), women with relapsed serous ovarian cancer and a BRCA mutation were most likely to benefit from olaparib treatment [125].  1.8 Survival statistics for ovarian cancer  1.8.1 Increased survival among women with ovarian cancer Partly owing to some of these innovative therapies, improving treatment and post-operative care, some limited evidence has suggested that 2- and 5- year survival rates of women diagnosed with ovarian cancer have been increasing temporally [5,126,127]. In 2011, a population-based Canadian study compared relative survival ratios over the period of 1992-2005 among women with EOC by age, histology of tumor and geographic regions of residence. This study was based on the hypothesis that the survival among women with ovarian cancer has increased over time. 7771 women were included in the study after being diagnosed with epithelial ovarian cancer in 23  Canada over the period of 1992 and 2005. Of which, 4104 died by the end of 2006. The relative survival curves were plotted by age at diagnosis (<40, 40-49, 50-59, 60-69, 70-79, >= 80 years), histology of tumor (serous, mucinous, endometroid, clear cell, transitional cell), and geographical location (Ontario, British Columbia, Eastern Canada, Central- West and Northern Canada). Relative survival ratio by age group decreased with increasing age. Moreover, the survival curves of the oldest group were constant 5 years post diagnosis, indicating no excess mortality rate compared to undiagnosed women in the same age group.  Their histology-specific results indicated that women with serous cancers had the lowest 2- and 5-year survival rates followed by women with mucinous cancers. Relative survival by geographical location showed the highest survival for Ontario, followed by British Columbia. Also, trends in 2- and 5- year relative survival rates were evaluated for all age groups, based on year of diagnosis. The resulting upward trends supported the hypothesis that there have been modest improvements in survival among women with EOC between 1992 and 2005. This improvement was attributed to improvement in diagnostic methods (down staging), treatment therapies (IP chemotherapy), and post- operative care. A similar study was conducted by Statistics Canada, evaluating relative survival estimates for women with ovarian cancer between the ages 15 and 99. A steady decrease in five- year survival was observed for all age groups between 2006 and 2008. However, age standardized survival estimates at 1, 3 and 5 years post diagnosis depicted significant improvement in survival for women diagnosed between 2006 and 2008 in comparison to women diagnosed between 1992 and1994 [5].  As these studies reflect a limited amount of evidence indicating a potential improvement in survival among women with EOC, but all use data that are at least 10 years old, it is important to 24  update our knowledge on 1- year, 3-year, and 5-year survival curves among women diagnosed with EOC in BC. Given that a fewer studies have examined histology-specific survival rates, more evidence regarding histologic survival is also needed to better inform women with EOC and clinicians.  1.9 Survivorship The Institute of Medicine (IOM) defines cancer survivorship as a different dimension of cancer research and notes that survivorship has received less attention in education, research, advocacy, and clinical practice. Improving our understanding of the post- treatment medical needs (psychosocial, or health outcomes) will assist affected individuals and their health care providers in implementing suitable interventions to maximize survivors long-term health [36].  Gynaecological cancer survivors are expected to experience adverse health issues post treatment. These issues include general medical problems and potential side-effects of treatment, which can negatively impact their quality of life [128]. With respect to ovarian cancer, most of the published research has focused on psychological concerns [129-132], including emotional distress (anxiety, depression) and fear of recurrence, as well as pain, fatigue, sexual dysfunction, and cognitive dysfunction. Some of the most commonly reported medical conditions include bowel obstruction, bladder dysfunction, gastro-intestinal side effects, and peripheral neuropathy [128,129]. However, there is a lack of research into the long-term health consequences of ovarian cancer survivors. The survivorship findings of other, better studied, cancers (e.g. breast and prostate) are unlikely expected to apply to ovarian cancer survivors [130]. Hence, long-term health consequences among ovarian cancer survivors must be investigated to facilitate timely intervention. 25  1.9.1 Using underlying causes of death to better understand long-term health consequences among women with ovarian cancer Some preliminary research has suggested that women diagnosed with EOC typically die of cancer-related causes within five years of diagnosis [133].The risk, however, decreases with time and women become vulnerable to death from other causes. However, other leading causes of death among women surviving their ovarian cancer have received very little attention to date, and no extensive research has been done on causes of death among Canadian women surviving their ovarian cancer for greater than 5 years. Understanding whether ovarian cancer survivors are at increased risk of dying from certain causes compared to the age- standardized general population (women who have never had ovarian cancer) will shed light on key areas of ovarian cancer survivorship research that need to be explored in the future. In particular, it could improve our understanding of the health consequences faced by ovarian cancer survivors.  1.10 Purpose  We expect that women with ovarian cancer in British Columbia (BC) are also surviving longer following a diagnosis and treatment for ovarian cancer; however, BC survival rates need to be updated as the most recent reported survival curves are from 1992 to 2005.  These numbers suggest improvements in 2- year and 5- year relative survival rates over the period of 1992 to 2005, across all the age groups[126] . Existing epidemiological studies fail to incorporate our modern understanding of various ovarian cancer histologic subtypes. In other words, these survival rates also need to reflect our most current understanding of the disease, and thus survival rates should be reported considering histologic subtype. Given the evidence that survival is increasing following ovarian cancer, there is an urgent need to better understand the health 26  outcomes of ovarian cancer survivors, an area that has been largely ignored in research to date likely due to the paucity of survivors. There is a reason to be concerned about long-term health consequences in ovarian cancer survivors, particularly in relation to their causes of death. Therefore, this proposal seeks to examine temporal and histologic subtype trends and survival rates following an ovarian cancer diagnosis in British Columbia and to study major causes of death among ovarian cancer survivors.  1.11 Study objectives and research questions The overarching aim of my research is to improve our knowledge of ovarian cancer survivorship. To achieve this aim, I will ask the following research questions: 1) Have incidence rates and survival rates (1- year, 3- year and 5- year overall and disease specific survival) of ovarian cancer changed significantly in British Columbia between 1980 and 2015?  2) Among women with ovarian cancer what are the most common causes of death, and when do other causes of death surpass ovarian cancer as the leading cause of death? Does this differ by histologic subtype?  3) Among women with ovarian cancer, and among women surviving their ovarian cancer for 5-9 or 10+ years, what are the most common causes of death and how do they compare to causes of death among and age-standardized population of women who have never been diagnosed with ovarian cancer?   27  1.12 Thesis roadmap This thesis consists of five chapters.  Chapter 2 outlines the trends in incidence rates and survival rates (1- year, 3- year and 5- years overall and disease specific survival rates) for women with ovarian cancer diagnosed between 1980 and 2015. Chapter 3 reports the causes of death among women with ovarian cancer in British Columbia diagnosed between 1990 and 2014 by histotype, age at diagnosis and BRCA status, and examines how many years women need to survive before other causes of death surpass ovarian cancer as the leading cause of death. Chapter 4 evaluates age- specific and cause- specific standardized mortality ratios (SMR), by comparing the causes of deaths in women with ovarian cancer (including women surviving ovarian cancer for 5-9 and 10+ years post diagnosis) with the causes of death in the general population.  Finally, Chapter 5 summarizes the findings, discusses the implications, the strengths and limitations of this dissertation, and suggests potential areas of future research in ovarian cancer.        28  Chapter 2: Recent trends in ovarian Cancer incidence and survival in British Columbia by histotype.  2.1 Introduction  Ovarian cancer is the deadliest gynecologic cancer. It is commonly diagnosed in women aged between 55 and 64 years [1]. Despite a low lifetime risk of developing ovarian cancer in the general population (1.4%)  [134], it is the fifth most common cause of cancer death among women. Early detection of ovarian cancer is hindered by its non- specific symptoms.  Hence, the majority of women are diagnosed at the advanced stages (Stage III and IV) with a 5- year survival rate of less than 50%. Furthermore, no screening methods for ovarian cancer have yet demonstrated a significant reduction in disease specific or all- cause mortality [94-97]. In recent years, advances in treatment and management of women with ovarian cancer have suggested potentially improved survival rates [7]. Prior studies have shown that women with advanced ovarian cancer, receiving appropriate treatment (optimal surgical debulking surgery and/or standard of care chemotherapy), have better survival than those not receiving this treatment [135,136]. Research from the Netherlands has revealed increases in survival rates between 1989 and 2009 that appear directly connected to higher rates of optimal treatment (as defined above) [137]. While little recent Canadian research is available, a population-based Canada-wide study, showed a significant improvement in the trends of 2- and 5- year relative survival for all the age groups, over the period of 1992-2005 [126]. Statistics Canada also reported a significant increase in the age standardized survival at 1,3, 5 and 10 years post diagnosis between the early 1990’s and 2006 to 2008 [5].  29  Our understanding of the histologic diversity in epithelial ovarian cancer, accounting for 90% of all ovarian malignancies, has also changed [138]. We now understand that epithelial ovarian cancer comprises five distinct histologic subtypes: high grade serous cancer (HGSC), low grade serous cancer(LGSC), endometrioid cancer, clear cell cancer and mucinous cancer [37]. These subtypes differ in their cells of origin, molecular features, clinical features (risk factors, prognosis, treatment response) and survival [11-13]. A retrospective study evaluating differences in survival outcomes between high- grade serous cancers and low- grade serous cancers reported that women with high- grade serous cancers had less than half the median survival of women with low- grade serous cancers [34]. Recently, the CONCORD programme, a large international study, has compared survival among women with ovarian cancer by histotype. The study observed that the survival rates among women with high-grade serous cancers were lowest [13].   Survival statistics are commonly used to assess the progress against cancer, and thus should be regularly updated. For ovarian cancer, histotype- specific survival estimates would provide information for patients and clinicians to inform prognosis and direct treatment/management. Beginning to consider histotype-specific survival rates will be part of our progress toward a histotype-specific approach to ovarian cancer treatment and management wherein referral for hereditary cancer testing, surgical approaches, adjuvant treatment and survivorship plans will all be informed by the histologic subtype of ovarian cancer. In the present study, I examined age-standardized trends in the incidence and in the overall survival and disease- specific survival rates (1- year, 3- year and 5- year survival rates) of different EOC histotypes in British Columbia, between 1980 and 2015.   30  2.2 Methods  2.2.1 Data sources  I used data from the population-based BC Cancer Registry [139]. The BC Cancer Registry receives notifications of cancer from many sources including pathology, cytology, other labs, death certificates, and admissions to cancer centers operated by the BC Cancer Agency. The Registry contains personal and demographic information, information about the specific cancer diagnosis, as well as mortality information received from the BC Vital Statistics Agency [140].   I accessed information on all women diagnosed with ovarian cancer between 1980 and 2015 in British Columbia. My dataset included detailed information such as date of diagnosis, age at diagnosis, death status, last appointment date, date of death, and cancer histotype. 2.2.2 Study cohort  My study population consisted of all patients diagnosed with ovarian cancer in British Columbia between 1980 and 2015. The International Statistical Classification of Diseases and Related Health problems: tenth revision (ICD-10) was used to identify women with ovarian cancer with the code C48.1(peritoneum, specified parts), C48.2 (peritoneum, non- specified parts), C56.0 (ovary), and C57. 0 (fallopian tube) in the Canadian cancer registry. I then used the ICD-O morphology codes to identify the histologic subtype of cancer. Informed by the WHO criteria, I worked with an anatomical pathologist who is an expert in ovarian cancer, Dr. Anthony Karnezis to assign the ICD-O morphology codes to their relevant histotype.   The initial stage of this review ensured that only epithelial tumors were included, and thus codes were reviewed with an eye to removing all women with non-epithelial tumors from our study cohort. Once we had developed a list of codes that pertained only to EOCs, the codes were next classified according their malignant versus benign versus borderline status. All women with 31  non-malignant EOCs were excluded at this stage (i.e. patients with benign and borderline epithelial tumors were excluded). Using his knowledge of the pathology, the remaining codes (those for malignant EOCs) were stratified into serous, mucinous, endometrioid, and clear cell and ‘others and unspecified’ epithelial ovarian cancers based on the descriptions provided for each code (see Table 1 for further details).  Our final breakdown of the ICD-O morphology into relevant histotypes was then compared with other published lists and differences were considered and discussed [33], and Dr. Karnezis had the final say as to classification. The codes do not distinguish by grade, thus high-grade and low-grade serous cancers are grouped together. Mixed carcinoma, malignant Brenner tumors, and other unspecified epithelial tumors (codes that were too general to enable histotype-specific classification) labelled as ‘others and unspecified’ histologic subtype were presented separately so as not to contaminate the histotype-specific analysis.   Table 1: ICD- 10 codes for classifying EOC histotypes  Histotype ICD-O codes Serous  84603 Papillary serous cystadenocarcinoma (C56.9) 84413 Serous cystadenocarcinoma, NOS (C56.9) 84613 Serous surface papillary carcinoma 82603 Papillary adenocarcinoma, NOS 80203 Carcinoma, undifferentiated, NOS 84403 Cystadenocarcinoma, NOS 80503 Papillary carcinoma, NOS 84503 Papillary cystadenocarcinoma, NOS (C56.9) 81203 Transitional cell carcinoma, NOS 80213 Carcinoma, anaplastic, NOS 80223 Pleomorphic carcinoma 81303 Papillary transitional cell carcinoma (C67.0)  Mucinous  84803 Mucinous adenocarcinoma 84703 Mucinous cystadenocarcinoma, NOS (C56.9) 84713 Papillary mucinous cystadenocarcinoma (C56.9) 32  84813 Mucin-producing adenocarcinoma 84711 Papillary mucinous cystadenocarcinoma, borderline (C56.9) 84823 Mucinous adenocarcinoma, endocervical type  Endometrioid   83803 Endometrioid adenocarcinoma, NOS 85603 Adenosquamous carcinoma 85703 Adenocarcinoma with squamous metaplasia 83813 Endometrioid adenofibroma, malignant  83823 Endometrioid adenocarcinoma, secretory variant  Clear cell  83103 Clear cell adenocarcinoma, NOS 91103 Mesonephroma, malignant 83133 Clear cell adenocarcinofibroma (C56.9) 82903 Oxyphilic adenocarcinoma 83101 Clear cell adenoca, borderline  Others and unspecified  Mixed carcinomas     Malignant brenner tumors and mixed mullerian tumour     Not otherwise specified (NOS)                  83233 Mixed cell adenocarcinoma  82553 Adenocarcinoma with mixed subtypes 80453 Combined small cell carcinoma    90003 Brenner tumor, malignant (C56.9) 89503 Mullerian mixed tumor (C54) 89513 Mesodermal mixed tumor 89803 Carcinosarcoma, NOS        81403 Adenocarcinoma, NOS    80103 Carcinoma, NOS    80413 Small cell carcinoma, NOS   80463 Non-small cell carcinoma (C34)   81402 Adenocarcinoma in situ, NOS   82463 Neuroendocrine carcinoma, NOS   85623 Epithelial-myoepithelial carcinoma   82303 Solid carcinoma, NOS   80123 Large cell carcinoma, NOS  85042 Non-infiltrating intracystic carcinoma        80013    Tumor cells, malignant      80102 Carcinoma, NOS    33    2.3 Statistical Analysis  2.3.1 Incidence  I report age standardized incidence rates. I do not report crude incidence rates (number of cancer cases per 100, 000 individuals in the population) because they are influenced by the underlying age distribution of the population in the province. An aging population will result in higher crude rates because incidence for most cancers, including ovarian cancer, increases with age [42]. Age-adjusting the rates ensures that differences in incidence from one year to another are not due to differences in the age distribution of the populations being compared.  To calculate age- standardized rates, the crude incidence rates, stratified by histotype, were standardized to the year 2011 BC standard million population[141]. These rates were categorized into 4- year calendar periods from 1980-1984 to 2011-2015. 2.3.2 Survival To evaluate trends in survival over time, the overall survival and disease- specific survival probabilities of ovarian cancer patients were calculated using the cohort method. The cohort method is based on evaluating survival experience of specific cohorts, diagnosed with ovarian cancer and followed for a specific length of time (1, 3, and 5 years post diagnosis).  Consequently, patients diagnosed between 2011 and 2015 were excluded from the survival analysis since they would not have a minimum 5- year follow up until December 31st, 2015 (the most recent data available).   In addition, I implemented right censoring at five years by marking patients who experienced an event beyond 5 years post diagnosis as censored. The timeline for each patient 34  began from the date of diagnosis and ended at the censored date (December 31st in the fifth-year post diagnosis), for patients who did not experience an event in that time period and the death date for those who did. A death was considered an event only if occurred on or prior to the censored date.  2.4 Results 2.4.1 Clinical characteristics of women with epithelial ovarian cancer A total of 12,754 women were identified as having been diagnosed with ovarian cancer between 1980 and 2015 in British Columbia.  After excluding women who did not have epithelial ovarian cancer (n= 3209), 9545 were included in our study. The clinical characteristics of BC women diagnosed with EOC between 1980 and 2015 are outlined in Table 1. The entire study cohort included 5245 (55.0 %) serous, 662 (6.9 %) mucinous, 963 (10.1 %) endometrioid, 541 (5.7 %) clear cell and 2134 (22.4 %) ‘others and unspecified’ EOCs. The majority (48.5 %) of affected women were diagnosed with EOC between 60 and 79 years. All patients were 15 years of age or older, and only 38 (0.4%) patients were under the age of 25.  Table 2: Clinical characteristics of the study cohort. Year of diagnosis  N % 1980 – 1983 769 8.1 1984 -1987 876 9.2 1988 – 1991 901 9.4 1992 – 1995 942 9.9 1996 – 1999 1094 11.5 2000 – 2003 1078 11.3 2004 – 2007 1183 12.4 2008 – 2011 1248 13.1 2012 – 2015 1454 15.2 35   2.4.2 Age- standardized incidence rates of epithelial ovarian cancer by year of diagnosis Figure 1 displays the decreasing pattern of age- adjusted incidence over the period of 1980 to 2015. The age- adjusted incidence rate of EOC decreased gradually from 14.4 cases per 100,000 population in 1980-1983 to 12.5 cases per 100,000 population in 2012-2015. The age- adjusted incidence, stratified by histotype, reported a decreasing trend in serous and mucinous EOCs (Figure 2).  A slight increasing trend was observed for clear-cell (from 0.5 per 100,000 in 1980-1983 to 1 in 100,000 in 2012-2015) EOCs, and for ‘others and unspecified’ EOCs (from 0.3 per 100,000 in 1980 – 1983 to 0.6 per 100,000 in 2012 – 2-15) EOCs across the entire time period. For endometrioid cancers, the age- adjusted incidence rates increased from 1.1 cases per 100,000 in 1980-1983 to 2 cases per 100,000 population in 1992-1995 and then slightly decreased to 1.5 cases per 100,000 by 2015. Between 1980 to 2015, the age- adjusted incidences of serous EOCs decreased from 10.7 per 100,000 in 1980-1984 to 9.3 per 100,000 in 2012-2015; however, the lowest point was reached in 2008-2011 where age-adjusted incidence was 7.1 per    Histology N % Serous 5245 55.0 Mucinous 667 6.9 Endometrioid 963 10.1 Clear cell 541 5.7 Others and unspecified epithelial             Mixed carcinoma              Malignant Brenner              Not otherwise specified (NOS) 2134             273             186             1656 22.4             2.9             1.9 17.5    Age at diagnosis N % <40 440 4.6 40 – 59 3275 34.3 60 – 79 4630 48.5 80+ 1200 12.6 36  100,000 population. Age-adjusted incidence also decreased for mucinous EOCs, from 2.2 to 0.7 per 100,000 population. Figure 1: Age standardized incidence rate of EOCs (all histotypes combined), 1980 – 2015.                        Figure 2: Age standardized incidence rate, stratified by histotype, 1980 – 2015.  37  The distribution of each histotype by 4- year time periods relative to the entire EOC population are presented in Table 2. The percentages increased for each histotype except for mucinous and for ‘others and unspecified’ EOCs throughout the study period. In serous EOCs, the percentage was fairly stable around 50 except in the period 2012- 2015, where the percentage value was 61.9. The percentage of the clear- cell and endometrioid EOCs steadily increased with a value of 2.6 (1980 – 1983) to 6.4 (2012 – 2015) and 6.7 (1980 -1983) to 9.7 (2012 – 2015) respectively. However, the percentage drastically decreased from 11.7 (1980 – 1983) to 3.9 (2012 – 2015) in mucinous EOCs and from 22.1 (1980 – 1983) to 18.1 (2012 – 2015) in ‘others and unspecified’ EOCs.  Table 3: Percentage distribution of EOC histotypes from 1980 to 2015.  I plotted the 1- year (Figure 3), 3- year (Figure 4) and 5- year (Figure 5) age- standardized overall survival rates (OS) separately for all the ovarian cancers (stratified by serous and non serous (clear cell, mucinous and endometrioid) histotype; and excluding ‘others and unspecified’ EOCs), based on the cohort of diagnosis. In these figures, I added the 95% confidence band to quantify uncertainty in estimates at each point. 1- year, 3- year and 5- year age- standardized Year of diagnosis  N (%) Serous Mucinous  Endometrioid Clear cell Others and unspecified epithelial  1980 – 1983 441 (57.3) 90 (11.7) 48 (6.2) 20 (2.6) 170 (22.1) 1984 -1987 513 (58.6) 92 (10.5) 74 (8.4) 33 (3.8) 164 (18.7) 1988 – 1991 499 (55.4) 93 (10.3) 101 (11.2) 47 (5.2) 161 (17.9) 1992 – 1995 510 (54.1) 79 (8.4) 123 (13.1) 52 (5.5) 178 (18.9) 1996 – 1999 583 (53.3) 73 (6.7) 127 (11.6) 67 (6.1) 244 (22.3) 2000 – 2003 571 (53.0) 56 (5.2) 112 (10.4) 67 (6.2) 272 (25.2) 2004 – 2007 607 (51.3) 59 (5.0) 133 (11.2) 81 (6.8) 303 (25.6) 2008 – 2011 621 (49.8) 63 (5.0) 104 (8.3) 81 (6.5) 379 (30.4) 2012 – 2015 900 (61.9) 57 (3.9) 141 (9.7) 93 (6.4) 263 (18.1) 38  survival rates increased during the study period when examining all ovarian cancers (Figure 3, 4 and 5). I was unable to further stratify into serous, endometrioid, clear cell and mucinous as this resulted in unstable lines due to small numbers of cases within each group. When stratified into serous and nonserous, the survival curves trended downwards over the study period among women with serous EOCs. In general, the 1- year survival rate (80.8%) was nearly twice the 5- year survival rate (43.2%) for all EOC cases by 2010. Whereas, among women with serous EOCs, the 1- year survival rate (77.2%) was approximately four times the 5- year survival rate (25.2%) (Figure 3, Figure 5). Moreover, women with serous EOCs had the lowest 3- year survival rate (43%) and 5- year survival rate (43%), and no major change was observed in their 3- year and 5- year survival rates between 1980 and 2015 (Figure 4, Figure 5).  Among women with non- serous EOCs, approximately 69.3% survived 3 years post diagnosis (Figure 4) and 52.3% of women survived 5 years post diagnosis by the end of the study (Figure 5). These non- serous cases showed a slight improvement in their 3- year survival rates, however, the 5- year survival rates decreased throughout the study period. The 1- year and 5- year survival curves for ‘others and unspecified’ EOCs increased between 1980 and 2015 (Appendix C).  39   Figure 3:  1- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015.   40  Figure 4: 3- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015.                         41  Figure 5: 5- year age standardized overall survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 - 2015.      I plotted the 1- year (Figure 6), 3- year (Figure 7) and 5- year (Figure 8) age- standardized disease specific survival rates (DSS) separately for all ovarian cancers (stratified by serous and non serous histotype; and excluding ‘others and unspecified’ EOCs), based on the cohort of diagnosis.  Consistent with the overall survival rates for all EOCs, 1- year, 3- year and 5- year disease specific survival rates increased during the study period. For all the women with EOC, the 1- year survival rates (82.3%) were nearly twice the 5- year survival rates (45.5%) by 2010. The disease- specific survival rates were either stable or decreasing for serous and non- serous histotypes over the study period. Whereas, there was a dramatic increase in 1- year and 5- year disease specific survival for women with ‘others and unspecified’ EOCs (Appendix D).   42  Figure 6: 1- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015.             43  Figure 7: 3- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015.              44   Figure 8: 5- year age standardized disease specific survival rates for all specified epithelial ovarian cancers and stratified by serous and non-serous subtypes, 1980 – 2015.    2.5 Discussion Herein I report that the incidence rate of epithelial ovarian cancer in British Columbia decreased from 1980 to 2015. However, this differed significantly based on histotype. For instance, there was a slight increase in the incidence of endometrioid and clear cell EOCs. Whereas, the incidence of serous and mucinous EOCs decreased in our study period. As a result, endometrioid and clear cells made up a larger proportion/share of epithelial ovarian cancer over time.  Decreasing incidence of epithelial ovarian cancer might be explained by increased use of oral contraceptive pills (OCP) in Canada in the 1960’s. Unfortunately, there is a minimal 45  Canadian literature on the proportion of oral contraceptive users over time. However, a study conducted in France reports that the percentage of women (aged between 20 and 44 years) regularly taking these pills increased from 28.3% in 1978 to 45.4% in 2000. There were approximately 79% of women (aged between 15 and 29 years) who were taking oral contraceptive pills by 2010 [142]. However, the oral contraceptive pill use for many of the women in our study would have occurred many years earlier. A woman diagnosed with ovarian cancer in 1990, at the age of 61, would likely have used oral contraceptive pills during the 1950s and 1960s. A similarly aged women diagnosed in 2010 would likely have used oral contraceptive pills during the 1970s and 1980s. Epidemiologic studies have consistently reported the protective effect of reproductive factors, including oral contraceptives, against ovarian cancer [46,47]. By reducing the number of lifetime ovulatory cycles, oral contraceptives can reduce the risk of developing ovarian cancer [143]. However, the formulation of oral contraceptive pills has changed in important ways since the women in our cohort would have been using them. It is imperative that future research seeks to understand whether the new formulations of oral contraceptive pills, being used by women currently in their 20s and 30s have the same protective effects as the older generation of these medicines.  Previous studies have reported a significant inverse relationship between gynecologic surgeries and risk of ovarian cancer. This likely results from the role of fallopian tube, both as the site of origin for HGSC, and as a conduit for retrograde menstruation from the uterus for endometrioid and clear cell cancers. Thus tubal ligation (a form of permanent sterilization that generally cauterizes fallopian tubes) is associated with a 29% reduction in ovarian cancer with differences in this effect across subtype [73,74]. The rates of these gynaecologic surgeries are changing significantly in British Columbia. Most importantly, there was an educational 46  campaign in September of 2010 in British Columbia, recommending to all gynecologic surgeons in the province of British Columbia (BC) Canada that, when operating on women at general population risk for ovarian cancer, they consider: 1) performing bilateral salpingectomy at the time of hysterectomy (even when the ovaries are being preserved); and 2) performing bilateral salpingectomy in place of tubal ligation for permanent sterilization. These procedures have come to be referred to as opportunistic salpingectomy (OS). Moreover, hysterectomy (removal of uterus) is also associated with statistically significant reductions in risk for ovarian cancer (HR = 0.79, 95% CI= 0.70, 0.80) [78]. While hysterectomy rates have decreased since the 1980s, [144] rates of hysterectomy with bilateral salpingectomy have dramatically increased in British Columbia. Salpingectomy is performed in lieu of tubal ligation in over half of all tubal sterilization procedures (as of 2014), meaning that many women in British Columbia no longer have fallopian tubes [31]. I expect this will dramatically decrease incidence of ovarian cancer in the future, the effects of which should be noticeable by 2020.  The most common histotype of ovarian cancer in this study was serous EOCs. Though the incidence of this histotype declined between 1980 and 2005, the percentage was fairly stable, whereas the percentages of mucinous EOCs decreased dramatically. This might reflect an improvement in the classification of mucinous tumors, involving differentiation of these tumors from extraovarian metastases (intestinal mucinous carcinomas), benign and borderline tumors [145-147]. This improvement, however, is more applicable to recent cases (cases diagnosed after 2008), hence many tumors might have been misclassified as mucinous in the early years.  Our data also showed an increasing trend of 1- year, 3- year and 5- year overall and disease specific survival rates for epithelial ovarian cancer over the past 35 years. Our findings are consistent with other literature [5,126,127]. A population study comparing the relative survival 47  ratios over the period of 1992-2005 by age at diagnosis, histotype and geographical location, reported improvement in 2- and 5- year survival based on year of diagnosis across different age groups [126]. Another study has also shown a significant improvement in 1- year, 3- year, and 5- year age- standardized survival estimates for women diagnosed between 2006 and 2008 in comparison to women diagnosed between 1992 and1994 [5].  When examining increasing overall survival and disease specific survival for all EOCs, I suspect this is partly driven by the increasing proportion of endometrioid and clear cell cancers over time, which both have better survival outcomes than serous cancers and advanced stage mucinous cancers [148]. Hence, the observed increase in overall and disease specific survival rates for all EOCs may not be attributable to recent advances in treatment of ovarian cancer, including platinum-based chemotherapy, (intraperitoneal) IP chemotherapy, and surgeries, which are commonly reported to be improving survival among women with this disease[126].  The population-based nature of the study and its inclusion of all women diagnosed with EOC in British Columbia between 1980 and 2015 is an important strength of our research; however, some limitations are noted. Our reliance on the ICD morphologic codes to classify tumours into histologic subtypes is our most important limitation. This resulted in a large number of ovarian cancers that I was unable to classify by histotype as a significant number of women (22.4%) had ICD morphologic codes that map to “ovarian cancer, not otherwise specified”. These included unclassified cases (17.9%), mixed carcinoma (2.9%), malignant Brenner tumors (1.9%). These cases showed a dramatic increase in their 1- year, and 5- year survival rates.  It is unclear why women with unclassified ovarian cancers would have improved overall and disease specific survival rates while survival among women with serous and nonserous EOCs slightly decreased. Research is currently underway to properly classify these unspecified cases.  48  While there have been considerable advances in the categorization of epithelial ovarian cancer subtypes with high interobserver agreement in histotype assignment [12,149] of this disease, our cases include many cancers subtyped prior to these publications, and most cases did not have the benefit of additional immunohistochemical tests to help characterize challenging cases, meaning there is likely histotype misclassification. Recent research illustrated the considerable shift in understanding of histotype classification over our study period. An expert gynecopathologist re-reviewed 286 pathology specimens that he had previously examined in 2002. When re-reviewing these specimens in 2014, he only agreed with his initial 2002 assessment in 54% of cases. In contrast, 98% of these cases were assigned the same histotype independently by two gynecopathologists using 2014 WHO diagnostic criterion[12]. This reflects how better understanding of ovarian cancer pathology has impacted the diagnosis of histotypes over the years and suggests that our histotype classification was likely more accurate in more recent years.  The rarity of ovarian cancer combined with the relatively small BC population has resulted in small numbers of women with histotypes other than serous ovarian cancer, and thus I had to group these histotypes in our analyses. I was also unable to examine survival trends by stage, which is a well-established prognostic factor for survival. Lastly, estimating the effect of treatment therapies, such as cytoreductive surgery and chemotherapy on survival over time were beyond the scope of this study [1].  The results of this study indicate that incidence of ovarian cancer in British Columbia decreased between 1980- 2015, and that both the age- standardized overall and disease- specific survival rate of ovarian cancer increased after 1985. Increasing survival coupled with a 49  decreasing number of cases over time presents a positive picture of ovarian cancer status among affected women in British Columbia.   50  Chapter 3:   Long-term mortality among women with epithelial ovarian cancer: A population-based study in British Columbia, Canada 3.1 Introduction Ovarian cancer is the leading cause of death among gynecologic cancers. Although lifetime risk of ovarian cancer in the general population is relatively low (1.4%) [134], it is the fifth leading cause of cancer deaths among women in Canada, with a 5-year survival rate of 44% [150]compared with nearly 90% [4] for breast cancer, more than 80% [150] for endometrial cancer, and nearly 73% [150] for cervical cancer. While survival is much improved when ovarian cancer is detected in the early stages, there are presently no effective screening methods demonstrated to reduce mortality [98]. Ovarian cancer is also largely asymptomatic in early stages, thus approximately 70% of women are diagnosed when the disease is already at advanced stages (Stage III and IV) [133].  Causes of death among cancer patients has been relatively well studied across many different forms of cancer. It is often reported that within 5 years of a cancer diagnosis, cancer is the most common cause of death. However, the risk of dying from cancer decreases with time from diagnosis and cancer patients become more likely to die from other causes [133]. This has been reported for breast cancer [151], prostate cancer [152], head and neck cancer [153], and lung cancer [154]. With respect to ovarian cancer, a previous American study using Surveillance Epidemiology, and End Results (SEER) data has reported that the probability of dying from ovarian cancer decreases with time, but that ovarian cancer remains the leading cause of death for 15 years among women diagnosed in advanced stages [133].  51  Here, I have focused on epithelial ovarian cancer (EOC), which represents 90% of all ovarian cancers [155]. Over the past decade, it has become apparent that EOC is a heterogenous disease, comprising of distinct histotypes that differ in presentation, response to therapy, molecular features, hereditary predisposition, site of origin and clinical outcomes. Although EOC histotypes share an anatomical location (the ovary), they are now considered distinct diseases [10-14]. This histotype-specific approach has completely changed the approach to clinical care and research. Histotype and stage remain the strongest prognostic factors in EOC. Some international work examining survival among ovarian cancer patients by histologic-subtypes has reported that survival rates were lowest among women with high-grade serous cancers [13]. However, little is known about causes of death among women with epithelial ovarian cancer, and there is no current evidence on whether causes of death differ among these women by histotype.  Histotype-specific mortality estimates are of clinical importance as information may impact advice given or interventions undertaken for patients and physicians involved in their care. For women with OC histotypes less likely to recur, information is currently lacking on their potential health concerns.  I therefore examined survival and causes of death among women with EOC in BC between 1990 and 2014 by histotype and years since EOC diagnosis.    3.2 Methods  3.2.1 Data sources  I built a population-based cohort using data from the BC cancer registry [139], Vital statistics [140], BC hereditary cancer program (HCP) [139], and BC’s insurance registry file (the Consolidation file) [156]. The BC Cancer Registry is a population-based registry of all cancers diagnosed in British Columbia residents. It receives notifications of cancer from many sources 52  including pathology, cytology and other labs, hospital charts, death certificates, and admissions to cancer centers operated by the BC Cancer Agency. The Registry contains personal and demographic information and information about the specific cancer diagnosis. The vital statistics death file is an extract of the deaths registration file provided by British Columbia vital statistics agency. It contains information on all deaths in BC, including underlying cause of death (UCOD) and exact date of death. I accessed data from the HCP, the source of all BRCA1 and BRCA2 testing in the province of British Columbia. I classified patients as having a BRCA mutation if they had either a BRCA1 or BRCA2 mutation or both. The consolidation file is a comprehensive data set, containing information on individuals receiving health services and/or individuals eligible to receive health services in British Columbia, Canada (˜4.6 million people). It contains demographic and regional information on every individual residing in British Columbia. With the permission of all relevant data stewards, and ethics approval form the University of British Columbia’s behavioural research ethics board, all these data were retrieved from PopData BC [157]. All inferences, opinions, and conclusions drawn are those of the authors and do not reflect the opinions or policies of the Data Stewards. 3.2.2 Study cohort  Our study population consists of all patients diagnosed with ovarian cancer in British Columbia between 1990 and 2014. The International Statistical Classification of Diseases codes: tenth revision (ICD-10) was used to identify these women with ovarian cancer with the codes of C56.0 (ovary), and C57. 0 (fallopian tube) in the British Columbia cancer registry. We used the same histotype classification table as reported in section 2.2.2 (Table 1). To ensure I had complete follow-up on women included in our cohort and to prevent misclassifying women as 53  alive, among women who were not recorded in the death registry, I required that women be registered for health care in BC for at least 5 years after their diagnosis. 3.2.3 Assessment of causes of death  I classified the UCODs using ICD-10 categories. The UCODs were classified into specific categories such as ovarian cancer, breast cancer, colorectal cancer, ‘other’ cancers (lung cancer, gastrointestinal tract cancer, blood lymph cancer, other malignancy, non malignant and unspecified), cardiovascular diseases (rheumatic, hypertension, ischemic, heart failure, congenital, pulmonary, cardiomyopathy etc), other chronic conditions (diabetes, COPD, AIDS/HIV, pneumonia, other infectious and parasitic disease, asthma, cerebro and other vascular disease, liver disease, pulmonary fibrosis etc),  external causes (Motor vehicle accidents, poisoning, falls, suicide, other unintentional injuries etc), and unclassified causes (causes of death that did not meet the criterion of the above categories) (Appendix B). Patients were considered to have died of ovarian cancer if the cause of death was reported as ovarian cancer or cancer-related likely due to ovarian cancer (which included deaths from neoplasm of uterus, cervix, placenta, ovary and adnexa, vagina and external genitalia following an initial diagnosis of ovarian cancer). 3.2.4 Statistical Analysis  In this study, I have described data descriptively. Women were monitored as of the date of their ovarian cancer diagnosis (as recorded in the BC Cancer Registry database) until their death or until December 31st, 2014 (the end of the follow-up period). Causes of death were stratified based on histotype categories (Serous, Mucinous, Endometrioid, Clear cell, Not classified). I further stratified based on age at diagnosis (<60 or >=60 years) and BRCA mutation status.  54  Causes of death were calculated as rates with 95% confidence intervals. All analyses were performed with R version 3.3.2[158].  3.3 Results  A total of 6,975 women were identified as having been diagnosed with ovarian cancer between 1990 and 2014 in British Columbia.  After excluding women who did not have epithelial ovarian cancer (n=407), and after excluding women who were not captured in the death registry and were not registered for health care in BC for at least 5 years post diagnosis (n= 141), 6427 were included in our study. The study cohort included 2996 (46.6 %) serous, 366 (5.7 %) mucinous, 719 (11.2 %) endometrioid, 431 (6.7 %) clear cell and 1915 (29.8 %) not classified EOCs.  The clinical characteristics of BC women diagnosed with EOC between 1990 and 2014 are outlined in Table 3.  Among all histotypes, serous carcinomas were commonly observed, accounting for approximately 46.6% of total EOCs (66.4% if we remove EOCs that were not classified by histotype). The majority (68.2 %) of affected women were diagnosed with EOC between 50 and 79 years of age and were not BRCA mutation carriers (96.9%).  Table 4: Clinical characteristics of the study cohort. Year of diagnosis  N % 1990 – 1994 1106 17.2 1995 – 1999 1265 19.7 2000 – 2004 1252 19.5 2005 – 2009 1395 21.7 2010 - 2014  1409 21.9    Histology N % Serous 2996 46.6 Mucinous 366 5.7 Endometrioid 719 11.2 55   By the end of this study, 55.9% of the study cohort (all histologies) died from their ovarian cancer, 33.9% were alive, 10.2% of women died from causes other than ovarian cancer; 0.8% from breast cancer, 0.5% from colorectal cancer, 3.5% from ‘other’ cancers, 1.8% from cardiovascular disease, 1.8% from other chronic conditions, 0.3% from external causes and 1.5% from unclassified causes (Table 4).  Serous: Among women with serous EOCs, 29.3% were alive at the end of follow-up while 62.2% had died from their disease. Of those women who died from other causes, 0.9% died from breast cancer, 0.5% died from colorectal cancer, 2.7% died from other cancers, 1.5% died from cardiovascular disease, 1.2% died from other chronic conditions, 0.4% died from external causes, and 1.3% died from unclassified causes (Table 4).  Mucinous: 54.4% women with mucinous EOCs were alive while 32% died from ovarian cancer. The majority of women with mucinous EOCs died from other cancers (5.2%), followed by deaths from cardiovascular diseases (3.0%) and deaths from other chronic conditions (3.0%).  Although minimal in percentage, I also observed deaths from external causes (0.3%), deaths Clear cell 431 6.7 Not classified 1915 29.8    Age at diagnosis N % <40 410 6.4 40 – 49 815 12.7 50 – 59 1477 23.0 60 – 69 1457 22.7 70 – 79 1448 22.5 80+ 820 12.8    BRCA status N % BRCA wild type 199 3.1 BRCA null 6228 96.9 56  from colorectal cancer (0.3%), deaths from breast cancer (0.5%), and 1.4% died from unclassified causes (Table 4). Endometrioid: In all histotypes, the greatest number of women alive at the end of follow-up was observed among women with endometrioid EOC (62.0%).  Among the women who had died, 26.7% died from ovarian cancer, 0.7% died from breast cancer, 0.8% died from colorectal cancer, 2.6% died from other cancers, 2.5% died from cardiovascular disease, 2.4% died from other chronic conditions, 0.3% died from external causes, and 2% died from unclassified causes (Table 4).  Clear cell: There was a considerable number of women with clear cell EOCs who were alive at the end of follow-up (55.9%).  Among women who died, 35.3% died from ovarian cancer, 0.7% died from breast cancer, 0.5% died from colorectal cancer, 2.1% died from other cancers, 0.9% died from cardiovascular disease, 1.9% died from other chronic conditions, 0.9% died from external causes, and 1.9% died from unclassified causes (Table 4). Not classified: Among women with EOCs that could not be classified into histotypes, 21.8% of cases were alive at the end of follow-up while 66.2% died from ovarian cancer. Of those women who died from other causes, 0.7% died from breast cancer, 0.6% died from colorectal cancer, 5.1% died from ‘other’ cancers, 1.9% died from cardiovascular disease, 2.1% died from other chronic conditions, 0.1% died from external causes, and 1.5% died from unclassified causes (Table 4). Differences between age groups in the causes of death, stratified by histotype, are reported in Table 5. In each histotype group, a greater number of women who were diagnosed under 60 years of age survived than older women. Women diagnosed with serous cancer at 60 or older were most likely to die from ovarian cancer (66.8%), whereas women diagnosed with 57  endometrioid cancer under the age of 60 were least likely to die from ovarian cancer (19.5%) followed closely by women diagnosed with mucinous cancer under the age of 60 (21.3%). For each age cohort, and for each histotype ovarian cancer was the leading cause of death. In all, death from ovarian cancer was most common, followed by ‘other’ cancers.  The outcomes of women with ovarian cancer and a BRCA mutation, stratified by serous or non- serous histotype is reported in Table 6. I was unable to stratify by other histotypes due to small sample sizes. Among women diagnosed with serous cancer, those with a BRCA mutation (BRCA null) face greater risk of death from breast cancer (4%) in comparison to women without a mutation (0.7%). In all, regardless of a woman’s BRCA mutation status, and for all histotypes ovarian cancer was the leading cause of death.   58             -- suppressed due to small cell sizes    Table 5: Cause of death stratified by histotype    Cause of death, N (%; 95% CI) All EOC patients  (n= 6427) Serous   (n=2996) Endometrioid  (n = 719) Clear cell   (n= 431) Mucinous   (n= 366) Not classified (n= 1915) Alive 2181  (33.9; 32.8, 35.1) 877  (29.3; 27.7, 30.9) 446  (62.0; 58.4, 65.5) 241  (55.9; 51.2, 60.5) 199  (54.4; 49.3, 59.4) 418  (21.8; 20.0, 23.7) Ovarian cancer 3592 (55.9; 54.7, 57.1) 1864 (62.2; 60.5, 63.9) 192 (26.7; 23.6, 30.1) 152 (35.3; 30.9, 39.9) 117 (32; 27.4, 36.9) 1267 (66.2; 64.0, 68.2) Breast cancer 49  (0.8; 0.6, 1) 26 (0.9; 0.6, 1.3) 5 (0.7; 0.3, 1.6) -- -- 13 (0.7; 0.4, 1.2) Colorectal cancer 34 (0.5; 0.6, 0.7) 14 (0.5; 0.3, 0.8) 6 (0.8; 0.4, 1.8) -- -- 11 (0.6; 0.3, 1) Other cancer 227 (3.5; 3.1, 4.0) 82 (2.7; 2.2, 3.4) 19 (2.6; 1.7, 4.1) 9 (2.1; 1.1, 3.9) 19  (5.2; 3.4, 8) 98 (5.1; 4.2, 6.2) Cardiovascular disease 116 (1.8; 1.5, 2.2) 46 (1.5; 1.2, 2.1) 18 (2.5; 1.6, 3.9) -- 11 (3.0; 1.7, 5.3) 37 (1.9; 1.4, 2.7) Other chronic 114 (1.8; 1.5, 2.1) 37 (1.2; 0.9, 1.7) 17 (2.4; 1.5, 3.8) 8 (1.9; 1, 3.6) 11 (3.0; 1.7, 5.3) 41 (2.1; 1.6, 2.9) External causes 20  (0.3; 0.2, 0.5) 12 (0.4; 0.2, 0.7) -- -- -- -- Unclassified causes 94  (1.5; 1.2, 1.8) 38 (1.3; 0.9, 1.7) 14 (2; 1.2, 3.3) 8 (1.9; 1, 3.6) 5 (1.4; 0.6, 3.2) 29 (1.5; 1.1, 2.2) 59  Table 6: Cause of death stratified by histotype and age at diagnosis - suppressed due to small cell sizes Cause of death, N (%; 95% CI) Serous   (n=2996) Endometrioid  (n = 719) Clear cell   (n= 431) Mucinous   (n= 366) Not classified (n= 1915)  <60 (n= 1153) >=60 (n= 1843) <60 (n= 446) >=60 (n= 273) <60 (n= 293) >=60 (n=138) <60 (n= 216) >=60 (n=150) <60 (n= 594) >=60 (n= 1321) Alive 462 (40.0;  32.3, 42.9) 415 (22.5;  20.7, 24.5) 340 (76.2;  72.1, 79.9) 106 (38.3;  33.2, 44.7) 191 (65.2;  59.6, 70.4) 50 (36.2; 28.7, 44.5) 158 (73.1;  66.9, 78.6) 41 (27.3; 20.8,34.9) 299 (50.3;  46.3, 54.4) 119 (9;  7.6, 10.7) Ovarian cancer 633 (54.9;  52, 57.8) 1231 (66.8;  64.6, 68.9) 87 (19.5;  16.1, 23.4) 105 (38.5;  32.9, 44.4) 92 (31.4;  26.4, 36.9) 60 (43.5;  35.5, 51.8) 46 (21.3;  16.4, 27.2) 71 (47.3;  39.5, 55.3) 253 (42.6; 38.7, 46.6) 1014 (76.8;  74.4, 79) Breast cancer 13 (1.1; 0.7, 1.9) 13 (0.7;  0.4, 1.2) -- -- -- -- -- -- -- 9 (0.7;  0.4, 1.3) Colorectal cancer -- 12 (0.7;  0.4, 1.1) -- -- -- -- -- -- -- 8 (0.6;  0.3, 1.2) Other cancer 15 (1.3;  0.8, 2.1) 67 (3.6;  2.9, 4.6) 6 (1.4;  0.6, 2.9) 13 (4.8;  2.8, 8) -- 8 (5.8;  3, 11) 7 (3.2;  1.6, 6.5) 12 (8;  4.6, 13.5) 22  (3.7; 2.5, 5.5) 76 (5.8;  4.6, 7.1) Cardiovascular disease 8 (0.7;  0.4, 1.4) 38 (2.1;  1.5, 2.8) -- 17 (6.2;  3.9, 9.8) -- -- -- 8 (5.3;  2.7, 10.2) -- 35 (2.7;  1.9, 3.7) Other chronic 11 (1.0;  0.5, 1.7) 26 (1.4; 1,  2.1) -- 13 (4.8;  2.8, 8) -- 5 (0.3;  0.1, 0.8) -- 11 (7.3;  4.1, 12.7) 5 (0.8;  0.4, 2) 36 (2.7;  2, 3.8)  External causes -- 9 (0.5;  0.3, 0.9) -- -- -- -- -- -- -- -- Unclassified causes 6 (0.5;  0.2, 1.1) 32 (1.7;  1.2, 2.5) -- 12 (4.4;  2.5, 7.5) -- 8 (5.8;  3, 11.0) -- 5 (3.3;  1.4, 7.6) 6 (1.0;  0.5, 2.2) 23 (1.7;  1.2, 2.6) 60                -- suppressed due to small cell sizes Table 7: Cause of death stratified by histotype and BRCA status   Cause of death, N (%; 95% CI) Serous   (n=2996) Non- serous  (n = 1516)  BRCA wild type (n= 2846) BRCA null (n= 150) BRCA wild type (n= 1502) BRCA null (n= 14) Alive 798  (28.0; 26.4, 29.7)  79 (52.7; 44.7, 60.5) 879 (58.5; 56.0, 60.1) 7  (50; 26.8, 73.2) Ovarian cancer 1802 (63.3; 61.5, 65.1) 62 (41.3; 33.8, 49.3) 455 (30.3; 28.0, 32.7) -- Breast cancer 20 (0.7; 0.5, 1.1) 6 (4; 1.8, 8.4) 10 (0.7; 0.4, 1.2) -- Other causes 214 (7.5; 6.6, 8.6) -- 151 (10.1; 8.6, 11.7) 6  (42.9; 21.4, 67.4) 61   Figure 7 displays the frequency distribution of deaths for all histotypes. It reveals that ovarian cancer is the leading cause of death among women diagnosed with ovarian cancer for 10 years post diagnosis. It is first surpassed by other causes of deaths 11 years post diagnosis. Figure 8 displays the frequency distribution of deaths for serous EOCs. Ovarian cancer remains the leading cause of death among women diagnosed with serous EOCs for 12 years following diagnosis. Figure 9 depicts the frequency distribution of deaths for non- serous (endometrioid, clear cell, mucinous) EOCs. Other causes of death surpass ovarian cancer as the leading cause among women diagnosed with non-serous EOCs at 8 years post diagnosis. Figure 10 and figure 11 display the frequency distribution of deaths for serous EOCs by age group. Figure 12 and Figure 13 displays the frequency distribution of deaths for non-serous EOCs by age group. Ovarian cancer remains the leading cause of death for longer among younger women (<60 years) than among older women (60 years or more).  Other causes of death first surpass ovarian cancer as the leading cause of death among older women with serous EOCs at 8 years post diagnosis. Whereas ovarian cancer is the leading cause of cause among younger women with serous EOCs for 12 years post diagnosis. Among women with non- serous EOCs, ovarian cancer is the leading cause of death among older women for 5 years after diagnosis in comparison to younger women where it is 8 years after diagnosis. There were too few deaths from causes other than ovarian cancer in the women with a BRCA mutation to reliably examine causes of death over time in this group.       62  Figure 9: Frequency distribution of deaths among patients diagnosed with all the histotypes.  Figure 10: Frequency distribution of deaths among patients diagnosed with serous epithelial ovarian cancers.                                                63  Figure 12: Frequency distribution of deaths among older patients (>= 60 years) diagnosed  with serous epithelial ovarian cancers.  Figure 11: Frequency distribution of deaths among patients diagnosed with non- serous epithelial ovarian cancers.                               64  Figure 13: Frequency distribution of deaths among younger patients (aged under 60 years) diagnosed with serous epithelial ovarian cancers.                              Figure 14: Frequency distribution of deaths among older patients (aged 60 years or more) diagnosed with non- serous epithelial ovarian cancers.            65  Figure 15: Frequency distribution of deaths among younger patients (aged under 60 years more) diagnosed with non- serous epithelial ovarian cancers.   3.4 Discussion Herein I report that among EOC patients, ovarian cancer is the leading cause of death for 10 years post diagnosis before it is surpassed by other causes of death. However, this differed significantly based on histotype and age at diagnosis. For example, among women with serous EOC, ovarian cancer was the leading cause of death for 12 years compared to 8 years among women with non- serous EOCs. When stratified by age, ovarian cancer was the leading cause of death for 8 years post diagnosis among women with serous EOCs diagnosed after age 60 compared to 12 years among younger women (<60 years of age) with serous EOCs. Our results suggest that women with non- serous EOCs were more likely to die from causes other than ovarian cancer in comparison to serous patients, as the majority of women with serous EOCs died from ovarian cancer. For instance, women with endometrioid and mucinous EOCs were 66  more vulnerable to die from cardiovascular diseases, chronic conditions and unclassified causes than women with serous EOC.  Although the rates of death from breast cancer were not significantly different across the histotypes, I found that these rates varied when stratified by BRCA status. Women with a BRCA mutation and with serous carcinomas were less likely to die from other causes, and more likely to die from ovarian cancer, than women without a BRCA mutation. Whereas women with a BRCA mutation and a non- serous cancer were at relatively comparable risk of death from other causes as women without a BRCA mutation and a non- serous cancer.  Our results reporting a significant risk of death from ovarian cancer for many years post diagnosis is consistent with those previously reported using SEER data [133]. The SEER study reported that the probability of death from ovarian cancer decreases with increased survival years post diagnosis and the probability of death from all other causes increases. I observed that ovarian cancer is the leading cause of death among BC women diagnosed with EOC for 10 years post diagnosis. The SEER study reported that ovarian cancer was surpassed by other causes of death 7 years after diagnosis.  Our results are also consistent with a small body of literature reporting minimal risk of breast cancer among women with EOC and a BRCA mutation [159]. Our findings indicate a low incidence of death from breast cancer among women with ovarian cancer and a BRCA mutation. When stratified by histotype, BRCA mutation carriers diagnosed with serous EOCs were at the greater risk of death from breast cancer, breast cancer was only responsible for 4% of deaths among these women. This supports the assertion that there is no need to rush to perform mastectomy to prevent breast cancer among women with a BRCA mutation and EOC.  67  The population-based nature of the study and its inclusion of all women diagnosed with EOC in British Columbia between 1990 and 2014 is an important strength of our research; however, some limitations are noted. Our reliance on the ICD morphologic codes to classify tumours into histologic subtypes likely introduced some misclassification. As outlined in section 2.5, there have been considerable advances in categorization of epithelial ovarian cancer subtypes with high interobserver agreement in histotype assignment for this disease over our study period [12,149], and we have reason to expect that our histotype classification improved considerably toward the end of our study period, resulting in higher rates of misclassification in the earlier years of our study. In addition, the rarity of ovarian cancer combined with the relatively small BC population has resulted in small numbers of women with histotypes other than serous ovarian cancer, and thus I had to group these histotypes in some analyses. Also, I cannot comment on the lynch syndrome status of our study population as I lack these data. Based on the past literature, women with endometroid and clear cell EOCs were observed to be at risk of Lynch syndrome [66]. Associated with a high risk of colorectal cancer, lynch syndrome status may have factored into the risk of death from colorectal cancer.  Importantly, I was unable to examine long-term mortality by stage because these data were missing in our cancer registry. Previous research has shown that women diagnosed with early stage EOC are less likely to die from ovarian cancer [133], but this has never been examined by histotype. Future research should examine whether stage influences long-term mortality differently by histotype. The high number of deaths from unclassified causes is unexpected and may reflect inaccuracies or missing information in the death certificates for some of those women. To limit this possible bias, our main analysis compared ovarian cancer specific mortality to all other causes of death.  68  The results of this study have implications for clinicians and ovarian cancer patients. The findings can help clinicians better understand the differences in outcomes among women with ovarian cancer based on histotype, age at diagnosis and BRCA status. 69  Chapter 4: Causes of death among women with epithelial ovarian cancer by length of survival post-diagnosis: A population-based study in British Columbia, Canada 4.1 Introduction  Ovarian Cancer is the most lethal gynecologic cancer with a 5- year survival rate ranging between 40 and 45% [150]. It is the fifth most common cause of cancer-related deaths among Canadian women [5]. While 5-year survival rates are better when ovarian cancer is detected in Stages I and II, the majority of women are diagnosed in advanced stages (Stage III and IV), due to its asymptotic presentation in early stages [133]. There is also no effective screening method for ovarian cancer. Despite considerable international effort to develop one [94-97], no screening trial to date has reported a significant reduction in mortality [160].  Among cancer patients, the most common cause of death within 5 years of diagnosis is cancer itself. However, as the length of time from diagnosis increases, the risk of death from cancer significantly decreases, and patients begin to die from other causes [133]. Largely because 5-year ovarian cancer survival rates have remained relatively low, little attention has been paid to long-term ovarian cancer survivors. Most studies examining ovarian cancer survivors have focused on improving their quality of life and reported primarily on psychosocial and emotional needs [130-132,161]. The most commonly reported health issues among long-term ovarian cancer survivors are bowel obstruction, bladder dysfunction, gastro-intestinal side effects, and peripheral neuropathy [130]. However, the research on the health of ovarian cancer survivors remains sparse.  70  Understanding whether ovarian cancer survivors are at increased risk of dying from causes other than ovarian cancer compared to the general population can improve our understanding of the health consequences they face, which can in turn improve follow-up and care provided to ovarian cancer patients. Therefore, I examined the causes of death among women diagnosed with ovarian cancer between 1990 and 2014 in BC. In particular, I focused on long-term ovarian cancer survivors (women surviving for 5 to 9 and 10 or more years after their ovarian cancer diagnosis) and compared their causes of death with those of women in the general population following age-standardization.  4.2 Methods  4.2.1 Study context  British Columbia has a provincially administered, universal medical and hospital health insurance program that is implemented and governed by the province and performs according to the specified conditions and criteria established in the Canada Health Act. It covers everyone residing in the province (4.6 million Canadians) and the provincial ministry of health collects detailed health data, which is regularly made available for research purposes. 4.2.2 Data sources  I built a population-based cohort of all women diagnosed with ovarian cancer between 1990 and 2014 using data from the BC cancer registry [139], vital statistics data [140], and the BC Ministry of Health (the consolidation file)[156]. The BC Cancer Registry is a population-based registry of all cancers diagnosed in British Columbia resident that contains personal and demographic information, and information about the specific cancer diagnosis. The vital statistics death file is an extract of the deaths registration file provided by British Columbia vital 71  statistics agency. It contains information on all deaths in BC, including underlying cause of death (the primary condition that lead to death) and the exact date of death. The consolidation file contains information on individuals receiving health services and/or individuals eligible to receive health services in British Columbia, Canada. It contains demographic and regional information on every individual residing in British Columbia. With the permission of all relevant data stewards, and ethics approval from the University of British Columbia’s behavioural research ethics board, all these data were retrieved from PopData BC [157]. All inferences, opinions, and conclusions drawn are those of the authors and do not reflect the opinions or policies of the Data Stewards. 4.2.3 Study cohort I examined a population based- cohort of women diagnosed with epithelial ovarian cancer (EOC) between 1990 and 2014 in British Columbia (BC). The International Statistical Classification of Diseases and Related Health problems: tenth revision (ICD-10) was used to identify women with ovarian cancer with the code of C56.0 (ovary), and C57. 0 (fallopian tube) in the British Columbia cancer registry. To ensure I had complete follow-up on women included in our cohort and to prevent misclassifying women as alive if they were not recorded in the death registry but had actually moved out of the province and then perished, I required that women be registered for health care in BC for at least 5 years after their diagnosis. I stratified women into three groups. The first group contained all the patients diagnosed with EOC between 1990 and 2014. The second group, (drawn from the first) includes only women who survived 5-9 years following their ovarian cancer diagnosis, while the third group contains women who have survived ovarian cancer for 10 years or more post diagnosis.  72  Deaths in the general population were obtained from publicly available tables, publishing sex-specific death rates for the BC population by detailed causes of deaths (https://www2.gov.bc.ca/gov/content/life-events/statistics-reports/annual-reports). I averaged across three time periods (2005, 2010 and 2015) in order not to rely too heavily on one specific year of data. 4.2.4 Assessment of causes of death Among ovarian cancer patients, I classified the underlying cause of death (UCOD) using ICD-10 categories. The UCODs were classified into specific categories such as ovarian cancer, ‘other’ cancers (lung cancer, breast cancer, colorectal cancer, etc.), cardiovascular diseases (rheumatic, hypertension, heart failure etc), other chronic conditions (diabetes, COPD, other infectious and parasitic disease, cerebro and other vascular disease etc),  external causes (motor vehicle accidents, poisoning, falls, suicide, other unintentional injuries etc), and unclassified causes (causes of death that did not fall into any of the above categories) (Appendix B). Patients were classified as having died from ovarian cancer if the cause of death was reported as ovarian cancer or a cancer likely related to ovarian cancer (which included deaths from neoplasm of uterus, cervix, placenta, ovary and adnexa, vagina, external genitalia and cancer without specified site following an initial diagnosis of ovarian cancer). The publicly available death tables had enough detail to use these same groupings for causes of death among the general population. 4.2.5 Statistical Analysis  Women diagnosed with EOC were monitored as of the date of their diagnosis until their death or until December 31st, 2014 (the end of follow up period). I calculated the length of survival from the date of ovarian cancer diagnosis to the date of death or the end of follow-up. 73  Causes of death (COD) analyses were stratified based on age categories (<45, 45-64, 65-79, 80+). Age-standardized mortality ratios (SMRs) with two- sided 95% confidence interval (95% CI), were calculated to assess significant differences in causes of death between the women with ovarian cancer and the general population.  All analyses were performed with R version 3.3.2 [158].   4.3 Results  A total of 6,975 women were identified as having been diagnosed with ovarian cancer between 1990 and 2014 in British Columbia.  After excluding women who did not have EOC (n=407), and after excluding women who were not captured in the death registry and were not registered for health care in BC for at least 5 years post diagnosis (n= 141), 6427 were included in our study.  The clinical characteristics of BC women diagnosed with EOC between 1990 and 2014, including long-term survivors of EOC are outlined in Table 1. Of the 6427 women with ovarian cancer, 810 patients survived for 5-9 years and 962 women survived for 10 years or more (10+ years). The median age of diagnosis was around 60 among all EOC patients (63; SD, 15) and EOC patients surviving 5 to 9 years (59.5; SD, 14.1). However, the median age of diagnosis for EOC patients surviving 10+ years was 52.5 years (SD, 14.6). The rate of women diagnosed under 45 years (24.7%) was more than double in the group surviving 10+ years compared to all EOC patients (11.1%). While the number of deaths among all EOC patients (66.1 %) was high, only 18.3 % of EOC patients surviving 10+ years died by the end of follow-up (Table 1).    74  Table 8: Clinical characteristics of the study cohort.  All EOC patients (N= 6427) EOC patients surviving 5-9 years (N= 810) EOC patients surviving 10+ years (N= 962) Age by end of follow-up (2014), n (%)    <45 364 (5.7) 45 (5.6) 56 (5.8) 45 – 64 2204 (34.3) 321 (39.6) 278 (28.9) 65 – 79 2566 (39.9) 296 (36.5) 406 (42.2) 80+ 1295 (20.1) 146 (18.0) 222 (23.1)     Age at diagnosis,  n (%)    <45 714 (11.1) 110 (13.6) 238 (24.7) 45 – 64 2665 (41.5) 394 (48.6) 529 (55.0) 65 – 79 2228 (34.7) 260 (32.1) 169 (17.6) 80+ 820 (12.8) 46 (5.7) 26 (2.7)     Age at diagnosis median (SD) 63 (15) 59.5 (14.1) 52.5 (14.6)     Year of diagnosis mean (SD) 2002.5 (7.1) 2003.3 (5.8) 1997.3 (4.3)     Deaths, n (%) 4246 (66.1) 417 (51.5) 176 (18.3) Year at Death mean(SD) 2003.6 (6.7) 2005.9 (5.4) 2008.9 (3.7)  Table 8 displays calculated age-standardized SMRs. All- cause mortality among all the EOC patients, under the age of 45, was 886 times higher than the mortality in the age- standardized general population (SMR= 886; 95% CI, 748.1 - 1042). And mortality from all- causes among the EOC patients, under the age of 45, surviving 5 to 9 years was 421 times higher than the mortality in the age- standardized general population (SMR= 421; 95% CI, 195.5 – 799.6). The SMRs observed among EOC patients surviving 10+ years were considerably smaller.  For instance, among women aged 45 years or less surviving 10+ years, the SMR was 83 75  (95% CI, 14 – 275.3), and dropped to 5 (95% CI, 3.8 – 5.7) among those aged 80 years or older surviving 10+years.  Table 9: Standardized Mortality Ratios (SMRs) in women with ovarian cancer according to years of survival post-diagnosis by age.  Age group (years), SMR (95% CI) SMR all EOC patients SMR EOC patients surviving 5-9 years SMR EOC patients surviving 10+ years <45 886 (748.1 – 1042) 421 (195.5 – 799.6) 83 (14 – 275.3) 45 – 64 187 (176.8 – 197.4) 133.7 (112.3 –158.1) 27 (17.5 – 39.7) 65 – 79 47 (44.7 – 49.0) 38 (32.7 – 44.3) 11 (8.3 – 13.6) 80+ 9.8 (9.2 – 10.4) 9 (7.4 – 10.8) 5 (3.8 – 5.7) Abbreviations: CI, confidence interval. The number of deaths in all the study cohorts and in the general population, stratified by causes is reported in Table 9. By the end of this study, 55.9% of all the EOC patients had died from ovarian cancer, and 33.9% were still alive. Of those women who had died of causes other than ovarian cancer (10.2%), 4.8% died from other cancers, 1.8% died from cardiovascular diseases, 1.8% died from other chronic conditions, 0.3% died from external causes and 1.5% died from unclassified causes. Among EOC patients surviving 5 to 9 years post diagnosis, 37.9% died from ovarian cancer by the end of follow-up and 48.5% were alive. The second most common cause of death was other cancers, representing 5.6% of women who died. In all the groups of women with EOC, the greatest number women alive at the end of follow-up were among EOC patients surviving 10+ years post diagnosis (81.7%). Among these women, 6.4% died from ovarian cancer by the end of follow-up. The next most common causes of death were other cancers (3.3%) and cardiovascular diseases (2.8%). Ovarian cancer was the leading cause of death among all women with EOC, regardless of years of survival since diagnosis. 76  Table 10: Causes of death among women with ovarian cancer and in the general population Cause of death, N (%) All EOC patients   (N=6427) EOC patients surviving 5-9 years  (N = 810) EOC patients surviving 10+ years  (N = 962) General population  (N = 2242627)  Alive 2181 (33.9) 393 (48.5) 786 (81.7) 2228849 (99.4) Ovarian cancer 3592 (55.9) 307 (37.9) 62 (6.4) 0 (0) Other cancers 310 (4.8) 45 (5.6) 32 (3.3) 4344 (0.2) CVD 116 (1.8) 18 (2.2) 27 (2.8) 4776 (0.2) Other chronic 114 (1.8) 24 (3.0) 18 (1.9) 909 (0) External cause 20 (0.3) 4 (0.5) 8 (0.8) 562 (0) Unclassified 94 (1.5) 19 (2.3) 29 (3.0) 3187 (0.1) Abbreviations: N, sample size of the cohort; CVD, cardiovascular diseases.  Cause specific SMRs in women with ovarian cancer are outlined in Table 10. When compared with the general population, women with ovarian cancer were more likely to experience death from ovarian cancer, and other causes (other cancers, cardiovascular diseases, other chronic conditions, external causes and unclassified causes) regardless of length of survival since diagnosis. Women with EOC were 28 times more likely to die from all causes than women of the same age in the general population (SMR= 28; 95% CI, 26.9 – 28.5). Among EOC patients surviving 5 to 9 years post diagnosis, all cause mortality rate was 24 times higher than in the age- standardized general population (SMR= 24; 95% CI, 21.5 – 26). Women surviving 10+ years since their ovarian cancer diagnosis were still at higher risk of death than would be expected based on the age-standardized general population (SMR= 7; 95% CI, 5.9 – 8).     77  Table 11: Standardized Mortality Ratios (SMRs) in women with ovarian cancer by years of survival post diagnosis according to cause of death. Cause of death, SMR (95% CI) All EOC patients   EOC patients surviving 5-9 years   EOC patients surviving 10+ years   All causes  28 (26.9 – 28.5) 24 (21.5 – 26) 7 (5.9 – 8) Other cancers 8 (7.4 – 9.2) 10 (7.6 – 13.6) 5 (3.7 – 7.3) CVD 2 (1.8 – 2.7) 3 (1.9 – 4.7) 3 (2.1 – 4.4) Other chronic 4 (3.5 – 5.1) 8 (5.1 – 11.4) 4 (2.4 – 6.2) External cause 5 (3.2 – 7.8) 9 (2.8 – 21.2) 12 (5.8 – 23.7) Unclassified 3 (2.4 – 3.6) 1 (0.7 – 1.7) 1 (0.8 – 1.6) Abbreviations: CI, confidence interval; CVD, cardiovascular diseases.  The SMRs for specific causes of death indicate that women diagnosed with EOC are significantly more likely to die from other cancers and external causes than those in the age-standardized general population, with SMRs of 5 or higher across all ovarian cancer groups. The SMRs for other cancers indicates that risk of death from other cancers was significantly higher compared to what would be expected in the general population among those surviving 5-9 years after their diagnosis (SMR=10; 95% CI, 7.6 – 13.6), and the same was true for external causes (SMR=9; 95% CI, 2.8 – 21.2). Women surviving 10+ years since their EOC diagnosis had the highest SMR for deaths from external causes (SMR=12; 95% CI, 5.8 – 23.7), followed by SMR for other cancers (SMR= 5; 95% CI, 3.7 – 7.3).   4.4 Discussion Unsurprisingly, our study found significantly higher all-cause mortality among women with ovarian cancer compared with what would be expected in an age-standardized general 78  population. However, this differed significantly based on age and the number of years a woman had survived since her ovarian cancer diagnosis. The SMRs for all EOC patients stratified by age paint a clear picture of how devastating a diagnosis of ovarian cancer is—EOC patients under age 45 years had 886- fold increase in mortality from all causes compared with the general population, followed by 187- fold increase among those between 45 and 64 years, 47- fold increase among those between 65 and 79 years and 9.8- fold increase among those aged 80 years or older.   I also found that ovarian cancer remains the leading cause of death for all women diagnosed with EOC regardless of time since diagnosis. When compared with the general population, the next highest number of deaths were from other cancers, followed by deaths from external causes and deaths from other chronic conditions. Deaths from other cancers can largely be explained by deaths from breast cancer (15.8%), lung cancer (12.3%), and colorectal cancer (11%). Of note, the mortality associated with external causes were predominately deaths from falls (44.4% of deaths associated with external causes were deaths from falls). Deaths from other chronic conditions are primarily due to cerebro and vascular diseases, chronic obstructive pulmonary disease (COPD), and diabetes. Among long term survivors of EOCs, deaths from other cancers and external causes were particularly problematic with SMRs of 5 or higher in all groups; however, the mean age of the women dying from falls was 91.9 (SD: 7.9) years for EOC patients surviving 10+ years and this is based on 8 EOC women, so the result should be interpreted with caution.  Our results are consistent with those previously reported using Surveillance Epidemiology and End Results (SEER) data. As per the Dinkelspiel et al study using SEER data [133], the risk of ovarian cancer- related death decreases as the time since diagnosis of ovarian cancer 79  increases. I observed that the proportion of deaths from ovarian cancer decreased dramatically with increased years since diagnosis. The SEER study also reported that other causes of death (e.g. cardiovascular diseases, other cancers, and other chronic conditions) surpassed ovarian cancer as the leading cause of death 15 years after diagnosis among women with advanced EOCs. To this existing literature, I have shown that women diagnosed with EOC remain at increased risk for virtually all causes of death (except unclassified causes) compared with the general population. While the SMRs become smaller as years since diagnosis increase, even long-term survivors are at increased risk of death from other cancers, other chronic conditions, cardiovascular disease and external causes.   Prior to this investigation, little was known about the all- cause specific mortality and cause specific mortality among patients with EOC compared with the general population. However, previous studies have described major health conditions experienced by women after an ovarian cancer diagnosis. Women with ovarian cancer, particularly women of 50 years or under, were observed to have an increased risk of ischemic stroke post diagnosis [162]. Another study reported higher incidence of venous thromboembolism among women with ovarian cancer [163]. Moreover, previous research has illustrated that women with ovarian borderline tumors are at increased risk of colorectal cancer and breast cancer (although this result was not statistically significant) [164,165].  Our study also revealed a considerable number of women with ovarian cancer who died from colorectal cancer (11%). I also observed a large number of deaths from breast cancer (15.8%) and lung cancer (12.3%). In fact, other cancers were a leading cause of death among long-term ovarian cancer survivors after ovarian cancer. Moreover, the mortality risk from other chronic conditions was primarily attributable to deaths from 80  cerebrovascular disease, disease of arteries and veins, hypotension and other circulatory system disease (37.7%). Our study is strengthened by using several large population-based datasets dating back to 1990 in British Columbia. However, the study is not without limitations. First, the death certificates in British Columbia include codes for contributing causes of death.  Herein I have considered only underlying cause of death, and there is likely some amount of misclassification as it can be arbitrary to assign a single cause to some deaths. Second, while I chose to report major groups of causes of deaths, I ended up with nearly 100 deaths that were from unclassified causes. Larger studies in the future may want to further categorize to detect any meaningful trends in ‘unclassified’ causes of death that our study has missed. Third, the new era of ovarian cancer research has illustrated the importance of histotype-specific research. Ovarian cancer is not one single disease, but five distinct diseases, with important clinical differences (i.e. high-grade serous, mucinous, endometrioid, clear cell, and low-grade serous ovarian carcinoma) [10,12,13]. While I was unable to classify our study cohort based on histotype due to small sample size, future research should examine age and cause specific SMRs stratified by histotype.  Our study has implications for women with ovarian cancer and their clinicians. The findings can help clinicians better understand the differences in mortality risk among women with ovarian cancer based on age and years since diagnosis. For example, I show that risk of death from other cancers was high in all groups when compared with the general population, and looking in further detail, our data suggest that breast, lung and colorectal cancer are especially common among women with a previous EOC diagnosis.  If corroborated by future research, more careful screening for these cancers could be considered among women surviving 5 or more years since their ovarian cancer diagnosis. Women surviving their ovarian cancer were also at 81  increased risk of death from external causes, primarily explained by deaths from falls. While numbers were small and thus should be interpreted with caution, if corroborated in future research, factors leading to deaths from falls among long term survivors would be worth exploring in future studies.  Our results also illustrate the continued devastation of, ovarian cancer, as it remains the leading cause of death among women with EOC, regardless of how many years they have survived since their diagnosis. However, women surviving 10 or more years since their EOC diagnosis were nearly twice as likely to die from another cause than to die from their ovarian cancer.    82  Chapter 5: Conclusion Ovarian cancer is the deadliest cancer of the female reproductive system. However, some sparse evidence has hinted at a possible improvement in survival rates over the past few decades [5,126]. It is important to understand whether survival rates are increasing among ovarian cancer patients, as historically very little attention has been paid to understanding the long-term health consequences and challenges faced by ovarian cancer survivors, likely a result of the poor 5-year survival rates for women with this disease. If women are, in fact, surviving their ovarian cancer at higher rates and for longer times, that suggests that research on survivorship should be given a higher priority. It is also important to update survival rates in order to provide accurate counsel to patients. At present, epidemiological evidence investigating possible outcomes among women with ovarian cancer and among women surviving ovarian cancer is limited.   Previously it was believed that ovarian cancer originated from the ovarian surface epithelium. However, this is not true for all histotypes of EOCs and different histotypes can have different cellular origins [166]. For instance, most high- grade serous cancers appear to originate in the fallopian tube, and clear cell and endometrioid ovarian cancers appear to originate from endometrial cells [167]. Owing to such drastic differences in molecular features, ovarian cancer is not considered a single disease. There are five main distinct histotypes of EOC- high- grade serous, low- grade serous, endometrioid, clear cell and mucinous cancers with varying clinical behavior (genetics, treatment response and survival) [13,37].  At present, limited data is available that could reflect this new understanding of the diversity of ovarian cancer. This histotype- specific information is needed to better understand the differences in epidemiological features across histotypes, which may be effective in improving clinical outcomes from EOC.   83  The primary goal of this thesis was to address these current gaps in ovarian cancer epidemiology, with a focus on understanding how survival varied by histotype. This concluding chapter is structured to answer four questions. First, what were the key findings and implications of this thesis? I briefly summarize the results and implications of the three studies presented in this thesis (chapter 2 to 4). Second, how have these findings have contributed to the existing literature? Third, what were the main strengths and limitations of the work presented in this thesis? Fourth, what are the recommendations for future research that stem from this thesis?  5.1 Summary of key findings and implications The objectives of this thesis were to: 1) evaluate the incidence rates and survival rates (1- year, 3- year and 5- year overall and disease specific survival rates) of women with epithelial ovarian cancer (EOC) in British Columbia (BC) between 1980 and 2015 and to conduct histotype-specific analysis of survival rates; 2) determine the causes of death among women with epithelial ovarian cancer by histotype, age at diagnosis and BRCA status and examine when other causes of death surpass ovarian cancer as the leading cause of death following diagnosis; 3) compare causes of death among women with epithelial ovarian cancer according to length of survival post-diagnosis with the causes of death among ovarian cancer- free women in the general population of the same age group.   5.1.1 Trends in incidence and survival rates among women with epithelial ovarian cancer in British Columbia, 1980- 2015.  In chapter 2, ‘Trends in incidence and survival rates among women with epithelial ovarian cancer in British Columbia, 1980- 2015’, I presented the most recent statistics on incidence and survival (overall survival and disease specific survival) of ovarian cancer in British Columbia. 84  These estimates were also studied to evaluate the incidence and survival of different histotypes over time. I observed declining incidence rates of ovarian cancer between 1980 and 2015 for all women with EOCs. This declining trend might be explained by the increased use of oral contraceptive pills (OCP) in Canada since the 1960’s, which is a well-established protective factor for ovarian cancer. Another factor that may have contributed to this decline are changing rates of gynecological surgeries, for e.g. hysterectomy and tubal ligation. These surgeries reduce the risk of ovarian cancer and increased during key parts of our study period that would affect incidence rates. I expect that more recent decreasing rates of hysterectomy and tubal ligation will be more than offset by the risk reduction of opportunistic salpingectomy, a surgical practice change that occurred far too late to be reflected in our incidence statistics but will likely reduce future incidence of EOC in British Columbia.  With respect to histotype- specific incidence, I found a slight increase in the incidence of endometrioid and clear cell EOCs, whereas a declining trend was reported for serous and mucinous EOCs.   This study also showed an increasing trend of 1- year, 3- year and 5- year overall and disease specific age- standardized rates for epithelial ovarian cancer over the past 35 years. When examining the increasing survival rates for all EOCs, I suspect that it was driven by the increasing proportion of endometrioid and clear cell cancers over time, which both have better survival outcomes than serous cancers and advanced stage mucinous cancers[148].   In all, I reported a positive picture of ovarian cancer in British Columbia. Previously, not many population-based studies have provided incidence and survival statistics by histology in Canada. This study is able to give insights into the progress against ovarian cancer from a histotype- specific perspective.  85  5.1.2 Long-term mortality among women with epithelial ovarian cancer in British Columbia, 1990- 2014 In chapter 3, ‘Long-term mortality among women with epithelial ovarian cancer in British Columbia, 1990- 2014’, I studied the most common causes of death among women with epithelial ovarian cancer by histotype, age at diagnosis and BRCA status. Specifically, I focused on evaluating the duration of time in which ovarian cancer is the leading cause of mortality following diagnosis. Among BC women with EOC, ovarian cancer is the leading cause of death for 10 years post diagnosis before deaths from other causes surpass it. I also report that: 1) women with serous EOCs are more likely to die from ovarian cancer than any other cause for 12 years in comparison to 8 years among women with non- serous EOCs; 2) women with non- serous EOCs are more likely to die from causes other than ovarian cancer compared to women with serous EOCs; 3) Among older women (60 years or older) with serous EOCs, death from other causes surpasses ovarian cancer as the leading cause of death at 8 years post diagnosis. Whereas, younger women (under 60 years) with serous EOCs had ovarian cancer as the leading cause of death for 12 years post diagnosis. These histotype-specific findings have not been reported elsewhere.  The results of my study display a timeline when risk of mortality from ovarian cancer decreases enough to be surpassed by other causes during the clinical course of the disease. This timeline can be used, when combined with further research such as that done in Chapter 4, to inform follow-up and management of women who have survived their ovarian cancer for more than 5 years.  86  5.1.3 Causes of death among women with epithelial ovarian cancer by length of survival post-diagnosis, 1990 – 2014.  Chapter 4, ‘Causes of death among women with epithelial ovarian cancer by length of survival post-diagnosis, 1990 – 2014’, investigate the causes of death among women with EOC, stratified by years since diagnosis and compares these with causes of death in an age-standardized population of women who were never diagnosed with ovarian cancer, using standardized mortality ratios (SMR). When compared with an age-standardized general population, all- cause and cause- specific SMRs among women with EOC, including women surviving 5 to 9 years post diagnosis and women surviving 10 years or more post diagnosis, were significantly higher compared to what would be expected in the general population. This was not surprising given the high mortality rate of EOC.   More interesting were the results among women who had survived their ovarian cancer for 5-9 and 10+ years. I report that among women who have survived their ovarian cancer at least 5 years, women with a previous diagnosis of ovarian cancer were significantly more likely to die from other cancers, deaths from external causes and from other chronic conditions compared to age-standardized women who had never been diagnosed with ovarian cancer.  Women who had survived their EOC for 5-9 or 10+ years were particularly at risk for from breast cancer (15.8%), lung cancer (12.3%), and colorectal cancer (11%). These women were also at increased risk for death from external causes, the majority of which were from falls (44.4%). Deaths from chronic conditions can largely be explained by deaths from cerebral and vascular diseases (45.9%), chronic obstructive pulmonary disease (COPD) (16.2%), and diabetes (8.1%).  I focused mainly on the causes of death with an SMR value of 5 or more (deaths from other cancers and external causes). These causes of death represent areas of important future 87  inquiry to inform surveillance among women surviving their ovarian cancer for 5 or more years. The findings can assist affected women and their health care providers in facilitating timely intervention to augment survivors long- term health. In particular, our research suggests that long- term survivors may benefit from closer surveillance for breast, lung and colorectal cancer. Also, careful attention to bone health may be helpful in this population.   5.2 Research contribution to literature   In the context of epidemiological research in Canada, the survival statistics of ovarian cancer are limited, particularly data for trends over time by histotype. Also, there is little information on how survival from ovarian cancer would impact the long- term health among women with EOC and/or among women at risk of developing EOC. Health care practitioners, specifically gynaecologic oncologists have affirmed that there is a need for updated statistics that could reflect differences in survival outcomes across ovarian cancer histotypes.  Furthermore, a better understanding of long- term health conditions among ovarian cancer survivors could be used to assist evidence- based clinical practice and to improve quality of life in long- term survivors of ovarian cancer.  Chapter 2 reported the most- current trends in incidence and survival in ovarian cancer in a Canadian setting. Additionally, it is the only BC- specific study that reports these trends by histotype. This study contributed to the existing literature by evaluating incidence and survival rates over the span of 35 years. As a result, my study sample covered a long time period than previous work, including women diagnosed between 1980 and 2015[5,126].  The results presented in chapter 3 represent a Canadian study on long- term morality among women with EOC. This work was possible because I had access to high quality 88  administrative datasets, containing information on histotype, age at diagnosis, BRCA status, and causes of death for all BC women diagnosed with EOC. A similar study was done by H.E. Dinkelspiel et al in the US, evaluating the causes of death among long- term ovarian cancer survivors by stage and years since diagnosis. This research included women with EOC diagnosed between 1980 and 2012. In my research, I focused primarily on recently diagnosed cohorts of women in BC (1990 – 2014), providing more recent assessment on causes of death among long- term ovarian cancer survivors, with an additional benefit of evaluating mortality in the context of ovarian cancer histotypes.  Among women with EOC and a BRCA mutation, few deaths from causes other than ovarian cancer were reported. In particular, I observed a low incidence of death from breast cancer among these BRCA mutation carriers. Deaths attributed to breast cancer among women with serous EOCs and a BRCA mutation were only 4% among these women.  This informs clinicians that rushing to mastectomy among BRCA null  women with EOC is unlikely to offer a significant survival advantage. While my study is not the first one to report low breast cancer incidence among women with EOC and a BRCA mutation, it is the first one to implement histotype- specific approach to estimating breast cancer risk among women with EOC and a BRCA mutation.  Chapter 4 offers a greater understanding on causes of death among women with EOC, particularly among women surviving their ovarian cancer by comparing them to the causes of death in an age-standardized population of women without ovarian cancer. This type of investigation is a first step to painting a more complete picture on the possible adverse outcomes following ovarian cancer diagnosis, and to informing a more appropriate and specific way of following up with women who survive their ovarian cancer for at least 5 years. 89  As a whole my thesis indicates that 1- 3- and 5-year survival rates have improved in BC over the past 35 years (although this is a largely a result of an increasing distribution of non-serous cancers). Regardless, this increases the importance of understanding the challenges faced by ovarian cancer patients who survive their cancers. Chapter 3 examining when other causes of death overtake ovarian cancer as the leading cause of death suggests that after surviving 12 years for women with serous cancer and 8 years for women with non- serous cancers, women are more likely to die from causes other than ovarian cancer. This can have important implications for follow-up in this group of women. While vigilance for recurrences of their ovarian cancer is still required, they also need to pay attention to other leading causes of death. Chapter 4, presenting SMRs for women who survived 5-9 and 10+ years following their diagnosis points to a few specific areas that warrant increased vigilance in women surviving their ovarian cancer for extended periods of time. Most notably, I report significantly higher risk of death from breast cancer, lung cancer, and colorectal cancer, all of which are cancers that have effective screening methods. Thus, if these findings are replicated in other populations, future work could examine whether women surviving their ovarian cancer for more than 5 years would benefit from enhanced screening for these cancers. Other noteworthy findings include data on minimal risk of breast cancer after ovarian cancer diagnosis among BRCA mutation carriers. This information is aimed at clinicians, suggesting that preventive mastectomy (surgical removal of breasts to prevent breast cancer) need not be rushed among women with ovarian cancer and a BRCA mutation.   90  5.3 Strengths and limitations  I have included a discussion on the strengths and limitations associated with the methods and the data used in each of the research chapters (Chapter 2 to 4). This section underlines the strengths and limitations of the thesis as a whole.  5.3.1 Limitations First, there were considerable number of women with EOC who were not classified in any histotype category in my study. For instance, 22.4% cases were unclassified among women diagnosed between 1980 and 2015 (chapter 2) and 29.8% cases were unclassified among women diagnosed between 1990 and 2014 (chapter 3 and chapter 4). This reduces power to understand the differences between histotypes, and creates this large unclassified category of EOCs for which we can draw few conclusions. Second, there have been considerable changes in the classification of EOC histotypes over the past decade. For instance, a study evaluating the diagnostic shift in histotype classification re-reviewed 286 pathology specimens in 2014, which were previously examined in 2002. Only 54% of cases were rendered the same ovarian cancer histotype when re-reviewed in 2014. In contrast, 98% of these cases were assigned the same histotype independently by two gynecopathologists using 2014 WHO diagnostic criterion [12]. Further changes include the fact that the most recent 2014 WHO diagnostic classification guidelines relabeled high- grade endometrioid, transitional and undifferentiated tumors as high grade serous EOCs [167].  Recently, this EOC histotype classification was further refined using immunohistochemical tests [37]. Cases that were defined prior to the new WHO guidelines and/or immunochemical tests may have been misclassified or not classified at all, and there is good reason to believe that misclassification of histotypes would be considerably higher in the earlier years of our study period.  91  Third, despite my best efforts to present histology-specific estimates in this thesis, I was frequently forced to combine endometrioid, clear cell and mucinous cancers into a non-serous category. Given that these are three distinct diseases, I was missing important nuance and distinction by combining these groups. Future work should combine data from BC with data from other provinces in order to properly study each of these histologic subtypes separately. Fourth, this work was conducted using BC- specific datasets, hence there are some questions about the generalizability to other provinces of Canada. However, I expect that most results would apply beyond British Columbia.   5.3.2 Strengths Despite these limitations, there are some notable strengths of the studies presented. One of such important strengths is the inclusion of data on the entire population of British Columbia diagnosed with EOC, thus eliminating any source of selection bias. By using the population- based administrative data, I was able to eliminate biases related to self- reported data such as recall bias and non- response biasMuch of this work was possible by merging the datasets: the BC Cancer Registry, Vital statistics data, BC Cancer hereditary cancer program, and BC Ministry of Health file (the consolidation file). All the studies reported in this thesis are built on de-identified data; ensuring confidentiality of all the women involved in my research.  5.4 Recommendations for future research Recommendations specific to each of the study findings have been highlighted in each of the research chapter (chapter 2 to 4). However, there are several recommendations that should be taken into consideration in future research. First and foremost, studies should be conducted to update statistics of ovarian cancer at the national level. This is important given the changing trends of ovarian cancer and introduction of advanced treatment strategies. Future research 92  should validate the findings in this thesis after properly classifying the histotypes of the cancers that I considered ‘unclassified or other’ in this thesis. By properly classifying these cases, I could identify differences in survival by histotype with increased power [167].   In Chapter 1, my results report the latest trends in ovarian cancer incidence and survival among women with EOC in BC. However, further research on how changes in the patterns of OCP use, which have been shown to be protective against ovarian cancer [168],  have impacted ovarian cancer incidence over time would be insightful. There has been considerable changes in OCP formulations (decreased doses of estrogen and inclusion of progastrin), usage among different age- groups, and intake patterns (monthly v/s continuously cycles), and there is limited evidence regarding whether the protective effects of OCP against ovarian cancer reported in earlier studies will remain after accounting for all of these important changes in modern day OCP use [167].  The findings in Chapter 3 suggest variations in outcomes of ovarian cancer based on histotype, and age at diagnosis. For instance, among older women (60 years or older) with serous EOCs, death from other causes surpasses ovarian cancer as the leading cause of death at 8 years post diagnosis. Further research on whether this timeline differs based on stage, and grade information would be helpful. Better understanding regarding the combined effects of stage, grade and histotype data on outcomes following ovarian cancer diagnosis would be useful for counselling patients at the time of diagnosis.   As observed in Chapter 4, all- cause and ovarian cancer- specific mortality risks among women with EOC and among women surviving EOC are significantly higher than what would be expected in the general population. While this study has offered insights into possible outcomes, histotype- specific data is needed to reflect new understanding of the heterogenous nature of 93  ovarian cancer. For instance, a study focused on evaluating SMRs among women with EOC and women surviving EOC based on histotypes would be helpful.  5.5 Conclusions My thesis suggests that ovarian cancer survival rates are improving over time and that histotype plays an important role in long-term survival. My thesis also reports that other causes of death surpass ovarian cancer as the leading cause of death 10 years after diagnosis among women with EOC. When stratified by histotype, the timeline of mortality from ovarian cancer and other causes differs. For instance, among women with serous EOCs, ovarian cancer was the leading cause of death for 12 years post diagnosis, whereas ovarian cancer was the leading cause of death for 8 years among women with non- serous EOCs.  While continued vigilance for recurrence in women surviving ovarian cancer is important, my research does suggest potential timeframes when women and their providers can feel less concerned about ovarian cancer recurrence. My research also suggests that after survival for 5-9 and 10 plus years there may be reason to increase vigilance for other causes of death, including breast, lung and colorectal cancer. If supported with future research, some enhanced screening for these cancers could potentially result in an even longer survival time for women who survived their ovarian cancer.  This may be worth consideration as the quality of life of all women at risk and/or diagnosed with ovarian cancer could be improved by introducing necessary interventions to prevent and/or manage these possible long- term health outcomes of EOC.  94  Bibliography  1. Doubeni, Chyke A.,MD, MPH, Doubeni, Anna R.B.,MD, MPH, Myers, Allison E.,MD, MPH. Diagnosis and management of ovarian cancer. Am Fam Physician. 2016;93(11):937-944. 2. Ovarian cancer statistics. http://www.wcrf.org/int/cancer-facts-figures/data-specific-cancers/ovarian-cancer-statistics. Accessed 15 January 2017.  3. Lowe KA, Chia VM, Taylor A, et al. An international assessment of ovarian cancer incidence and mortality. Gynecol Oncol. 2013;130(1):107-114. 4. Nuttall R. Canadian cancer statistics 2017. Canadian Cancer Society; 2017. 5. Navaneelan T, Ellison L, Canadian Publications From 2, Statistics Canada (English), Canada. Ovarian cancer: Survival statistics. Statistics Canada; 2015. 6. Ovary - epithelial carcinoma. http://www.bccancer.bc.ca/health-professionals/professional-resources/cancer-management-guidelines/gynecology/ovary-epithelial-carcinoma. Accessed 17 March 2017. 7. Lopez J, Banerjee S, Kaye SB. New developments in the treatment of ovarian cancer--future perspectives. Ann Oncol. 2013;24 Suppl 10(suppl 10):x76.  8. Gilks CB. Molecular abnormalities in ovarian cancer subtypes other than high-grade serous carcinoma. Journal of Oncology. 2010; 2010:1-7.  9. Kalloger SE, Köbel M, Leung S, et al. Calculator for ovarian carcinoma subtype prediction. Modern Pathology. 2011;24(4):512-521.  10. Köbel M, Kalloger SE, Boyd N, et al. Ovarian carcinoma subtypes are different diseases: Implications for biomarker studies. PLoS medicine. 2008;5(12):e232.  11. Matz M, Coleman MP, Sant M, et al. The histology of ovarian cancer: Worldwide distribution and implications for international survival comparisons (CONCORD-2). Gynecol Oncol. 2017;144(2):405-413.  12. Kommoss S, Gilks CB, du Bois A, Kommoss F. Ovarian carcinoma diagnosis: The clinical impact of 15 years of change. Br J Cancer. 2016;115(8):993-999.  13. Matz M, Coleman MP, Carreira H, Salmerón D, Chirlaque MD, Allemani C. Worldwide comparison of ovarian cancer survival: Histological group and stage at diagnosis (CONCORD-2). Gynecol Oncol. 2017;144(2):396-404.  95  14. Gilks CB, Ionescu DN, Kalloger SE, et al. Tumor cell type can be reproducibly diagnosed and is of independent prognostic significance in patients with maximally debulked ovarian carcinoma. Hum Pathol. 2008;39(8):1239-1251.  15. Colgan TJ, Murphy J, Cole DE, Narod S, Rosen B. Occult carcinoma in prophylactic oophorectomy specimens: Prevalence and association with BRCA germline mutation status. Am J Surg Pathol. 2001;25(10):1283-1289.  16. Lamb JD, Garcia RL, Goff BA, Paley PJ, Swisher EM. Predictors of occult neoplasia in women undergoing risk-reducing salpingo-oophorectomy. Obstet Gynecol. 2006;194(6):1702-1709.  17. Leeper K, Garcia R, Swisher E, Goff B, Greer B, Paley P. Pathologic findings in prophylactic oophorectomy specimens in high-risk women. Gynecol Oncol. 2002;87(1):52-56.  18. Powell CB, Kenley E, Chen L, et al. Risk-reducing salpingo-oophorectomy in BRCA mutation carriers: Role of serial sectioning in the detection of occult malignancy. Journal of Clinical Oncology. 2005;23(1):127-132.  19. Powell CB, Chen L, McLennan J, et al. Risk-reducing salpingo-oophorectomy (RRSO) in BRCA mutation carriers: Experience with a consecutive series of 111 patients using a standardized surgical-pathological protocol. International Journal of Gynecological Cancer. 2011;21(5):846-851.  20. Yates MS, Meyer LA, Deavers MT, et al. Microscopic and early-stage ovarian cancers in BRCA1/2 mutation carriers: Building a model for early BRCA-associated tumorigenesis. Cancer prevention research (Philadelphia, Pa.). 2011;4(3):463-470.  21. Wethington SL, Park KJ, Soslow RA, et al. Clinical outcome of isolated serous tubal intraepithelial carcinomas (STIC). International Journal of Gynecological Cancer. 2013;23(9):1603-1611.  22. Reitsma W, de Bock GH, Oosterwijk JC, Bart J, Hollema H, Mourits MJE. Support of the ‘fallopian tube hypothesis’ in a prospective series of risk-reducing salpingo-oophorectomy specimens. Eur J Cancer. 2013;49(1):132-141.  23. Candidate serous cancer precursors in fallopian tube epithelium of mutation carriers. Modern Pathology. 2009;22(9):1133-1138.  24. Medeiros F, Muto MG, Lee Y, et al. The tubal fimbria is a preferred site for early adenocarcinoma in women with familial ovarian cancer syndrome. Am J Surg Pathol. 2006;30(2):230-236.  25. Kauff ND, Satagopan JM, Robson ME, et al. Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med. 2002;346(21):1609-1615.  96  26. Carlson JW, Miron A, Jarboe EA, et al. Serous tubal intraepithelial carcinoma: Its potential role in primary peritoneal serous carcinoma and serous cancer prevention. Journal of Clinical Oncology. 2008;26(25):4160-4165.  27. Gao, Faye F.,MD, PhD, Bhargava R, MD, Yang, Huaitao,MD, PhD, Li, Zaibo,MD, PhD, Zhao C, MD. Clinicopathologic study of serous tubal intraepithelial carcinoma with invasive carcinoma: Is serous tubal intraepithelial carcinoma a reliable feature for determining the organ of origin? Hum Pathol. 2013;44(8):1534-1543.  28. Tang S, Onuma K, Deb P, et al. Frequency of serous tubal intraepithelial carcinoma in various gynecologic malignancies: A study of 300 consecutive cases. International Journal of Gynecological Pathology. 2012;31(2):103-110. 29. Seidman JD, Zhao P, Yemelyanova A. "Primary peritoneal" high-grade serous carcinoma is very likely metastatic from serous tubal intraepithelial carcinoma: Assessing the new paradigm of ovarian and pelvic serous carcinogenesis and its implications for screening for ovarian cancer. Gynecol Oncol. 2011;120(3):470.  30. Rabban JT, Garg K, Crawford B, Chen L, Zaloudek CJ. Early detection of high-grade tubal serous carcinoma in women at low risk for hereditary breast and ovarian cancer syndrome by systematic examination of fallopian tubes incidentally removed during benign surgery. Am J Surg Pathol. 2014;38(6):729-742.  31. Hanley GE, McAlpine JN, Kwon JS, Mitchell G. Opportunistic salpingectomy for ovarian cancer prevention. Gynecologic oncology research and practice. 2015;2:5.  32. Types of ovarian cancer. http://www.ovcare.ca/about_ovarian_cancer/types_of_ovarian_cancer/. Accessed 15 January 2017.   33. Peres LC, Cushing-Haugen KL, Köbel M, et al. Invasive epithelial ovarian cancer survival by histotype and disease stage. JNCI: Journal of the National Cancer Institute. 2018.  34. Gockley A, Melamed A, Bregar AJ, et al. Outcomes of women with high-grade and low-grade advanced-stage serous epithelial ovarian cancer. Obstetrics & Gynecology. 2017;129(3):439-447.  35. Jayson GC, Prof, Kohn EC, MD, Kitchener HC, Prof, Ledermann JA, Prof. Ovarian cancer. Lancet, The. 2014;384(9951):1376-1388.  36. Services, Board on Health Care, NCBI Bookshelf. Ovarian cancers: Evolving paradigms in research and care. National Academies Press (US); 2016.1. Doubeni, Chyke A.,MD, MPH, Doubeni, Anna R.B.,MD, MPH, Myers, Allison E.,MD, MPH. Diagnosis and management of ovarian cancer. Am Fam Physician. 2016;93(11):937-944. 97    37. Doherty JA, Peres LC, Wang C, Way GP, Greene CS, Schildkraut JM. Challenges and opportunities in studying the epidemiology of ovarian cancer subtypes. Current Epidemiology Reports. 2017;4(3):211-220. 38. Risk and protection factors of ovarian cancer. https://www.targetovariancancer.org.uk/information-and-support/what-ovarian-cancer/risk-and-protection-factors-ovarian-cancer. Accessed 20 March 2017. 39. Risk factors for ovarian cancer. http://www.cancer.ca/en/cancer-information/cancer-type/ovarian/risks/?region=qc. Accessed 20 March 2017. 40. Risk factors. https://ocrfa.org/patients/about-ovarian-cancer/risk-factors/. https://ocrfa.org/patients/about-ovarian-cancer/risk-factors/. Accessed 20 March 2017. 41. What are the risk factors for ovarian cancer? https://www.cancer.org/cancer/ovarian-cancer/causes-risks-prevention/risk-factors.html. Accessed 20 March 2017.   42. HUNN J, RODRIGUEZ GC. Ovarian cancer: Etiology, risk factors, and epidemiology. Clin Obstet Gynecol. 2012;55(1):3-23. 43. La Vecchia C. Ovarian cancer: Epidemiology and risk factors. European Journal of Cancer Prevention. 2017;26(1):55-62. 44. Wentzensen N, Poole EM, Trabert B, et al. Ovarian cancer risk factors by histologic subtype: An analysis from the ovarian cancer cohort consortium. Journal of Clinical Oncology. 2016;34(24):2888-2898. 45. Tung K, Wilkens LR, Wu AH, et al. Effect of anovulation factors on pre- and postmenopausal ovarian cancer risk: Revisiting the incessant ovulation hypothesis. Am J Epidemiol. 2005;161(4):321-329. 46. Ness RB, Cottreau C. Possible role of ovarian epithelial inflammation in ovarian cancer. J Natl Cancer Inst. 1999;91(17):1459-1467. 47. McGuire V, Felberg A, Mills M, et al. Relation of contraceptive and reproductive history to ovarian cancer risk in carriers and noncarriers of BRCA1 gene mutations. Am J Epidemiol. 2004;160(7):613-618. 48. Permuth-Wey J, Sellers TA. Epidemiology of ovarian cancer. Methods Mol Biol. 2009;472:413. 49. Risch HA. Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst. 1998;90(23):1774-1786. 98  50. Brett M. Reid Jennifer B. Permuth Thomas A. Sellers. Epidemiology of ovarian cancer: a review. 癌症生物学与医学:英文版. 2017;14(1):9-32. 51. Penninkilampi R, Eslick GD. Perineal talc use and ovarian cancer: A systematic review and meta-analysis. Epidemiology. 2018;29(1):41-49. 52. Cramer DW, Vitonis AF, Terry KL, Welch WR, Titus LJ. The association between talc use and ovarian cancer: A retrospective Case–Control study in two US states. Epidemiology. 2015;27(3):334-346. 53. Merritt MA, Green AC, Nagle CM, et al. Talcum powder, chronic pelvic inflammation and NSAIDs in relation to risk of epithelial ovarian cancer. International Journal of Cancer. 2008;122(1):170-176. 54. Cook LS. Use of talcum powder in the genital area was not associated with an increased risk of epithelial ovarian cancer. Evidence-based Obstetrics & Gynecology. 2000;2(2):53-54. 55. Ketabi, Zohreh|Bartuma, Katarina|Bernstein, Inge|Malander, Susanne|Grönberg, Henrik|Björck, Erik|Holck,Susanne|Nilbert, Mef, Onkologi och Patologi, M V, Lund University, Oncology and Pathology, M V, Lunds universitet. Ovarian cancer linked to lynch syndrome typically presents as early-onset, non-serous epithelial tumors. Gynecol Oncol. 2011;121(3):462-465. 56. Boyd J. Specific keynote: Hereditary ovarian cancer: What we know. Gynecol Oncol. 2003;88(1):S10. 57. King M, Marks JH, Mandell JB, The New York Breast Cancer Study Group, New York Breast Cancer Study Group. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science. 2003;302(5645):643-646. 58. Prat J, Ribé A, Gallardo A. Hereditary ovarian cancer. Hum Pathol. 2005;36(8):861-870. 59. Mavaddat N, Peock S, Frost D, et al. Cancer risks for BRCA1 and BRCA2 mutation carriers: Results from prospective analysis of EMBRACE. J Natl Cancer Inst. 2013;105(11):812-822. 60. Alsop K, Fereday S, Meldrum C, et al. BRCA mutation frequency and patterns of treatment response in BRCA Mutation–Positive women with ovarian cancer: A report from the australian ovarian cancer study group. Journal of Clinical Oncology. 2012;30(21):2654-2663. 61. Risch HA, McLaughlin JR, Cole DEC, et al. Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. The American Journal of Human Genetics. 2001;68(3):700-710. 99  62. Pal T, Permuth‐Wey J, Betts JA, et al. BRCA1 and BRCA2 mutations account for a large proportion of ovarian carcinoma cases. Cancer. 2005;104(12):2807-2816. 63. Risch HA, McLaughlin JR, Cole DEC, et al. Population BRCA1 and BRCA2 mutation frequencies and cancer penetrances: A kin-cohort study in ontario, canada. J Natl Cancer Inst. 2006;98(23):1694-1706. 64. Hanley GE, McAlpine JN, Miller D, et al. A population-based analysis of germline BRCA1 and BRCA2 testing among ovarian cancer patients in an era of histotype-specific approaches to ovarian cancer prevention. BMC Cancer. 2018;18(1). 65. Schrader KA, Hurlburt J, Kalloger SE, et al. Germline BRCA1 and BRCA2 mutations in ovarian cancer: Utility of a histology-based referral strategy. Obstetrics & Gynecology. 2012;120(2, Part 1):235-240. 66. Vierkoetter KR, Ayabe AR, VanDrunen M, Ahn HJ, Shimizu DM, Terada KY. Lynch syndrome in patients with clear cell and endometrioid cancers of the ovary. Gynecol Oncol. 2014;135(1):81-84. 67. Blake Gilks, C.|Clarke, Blaise A.|Foulkes,William D. Ovarian carcinoma histotype in lynch syndrome. Gynecologic Oncology Reports. 2017;20:140-141. 68. Seabrook C. Fertility drug may increase risk of ovarian cancer. The Atlanta Constitution. 1994.  69. Rossing MA, Daling JR, Weiss NS, Moore DE, Self SG. Ovarian tumors in a cohort of infertile women. N Engl J Med. 1994;331(12):771-776. 70. Jensen A, Sharif H, Frederiksen K, Kjær SK. Use of fertility drugs and risk of ovarian cancer: Danish population based cohort study. BMJ. 2009;338(7694):580-583. 71. Kurta ML, Moysich KB, Weissfeld JL, et al. Use of fertility drugs and risk of ovarian cancer: Results from a U.S.-based case-control study. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2012;21(8):1282. 72. Tone AA, Salvador S, Finlayson SJ, et al. The role of the fallopian tube in ovarian cancer. Clinical advances in hematology & oncology : H&O. 2012;10(5):296. 73. Rice MS, Murphy MA, Tworoger SS. Tubal ligation, hysterectomy and ovarian cancer: A meta-analysis. Journal of ovarian research. 2012;5(1):13. 74. Sieh W, Salvador S, McGuire V, et al. Tubal ligation and risk of ovarian cancer subtypes: A pooled analysis of case-control studies. Int J Epidemiol. 2013;42(2):579-589. 100  75. Narod SA, Sun P, Ghadirian P, et al. Tubal ligation and risk of ovarian cancer in carriers of BRCA1 or BRCA2 mutations: A case-control study. The Lancet. 2001;357(9267):1467-1470. 76. Lessard-Anderson CR, Handlogten KS, Molitor RJ, et al. Effect of tubal sterilization technique on risk of serous epithelial ovarian and primary peritoneal carcinoma. Gynecol Oncol. 2014;135(3):423-427. 77. Madsen C, Baandrup L, Dehlendorff C, Kjær SK. Tubal ligation and salpingectomy and the risk of epithelial ovarian cancer and borderline ovarian tumors: A nationwide case–control study. Acta Obstet Gynecol Scand. 2015;94(1):86-94. 78. Falconer H, Yin L, Grönberg H, Altman D. Ovarian cancer risk after salpingectomy: A nationwide population-based study. J Natl Cancer Inst. 2015;107(2):dju410. 79. Yoon S, Kim S, Shim S, Kang S, Lee S. Bilateral salpingectomy can reduce the risk of ovarian cancer in the general population: A meta-analysis. Eur J Cancer. 2016;55:38-46. 80. Parker WH, Feskanich D, Broder MS, et al. Long-term mortality associated with oophorectomy compared with ovarian conservation in the nurses' health study. Obstet Gynecol. 2013;121(4):709-716. 81. GOC statement regarding salpingectomy and ovarian cancer prevention. https://g-o-c.org/wp-content/uploads/2015/09/7GOCStmt_2011Sep_SalpOvCa_EN.pdf. Accessed 18 July 2018.  82. SGO clinical practice statement: Salpingectomy for ovarian cancer. https://www.sgo.org/clinical-practice/guidelines/sgo-clinical-practice-statement-salpingectomy-for-ovarian-cancer-prevention/. Accessed 18 July 2018. 83. Salvador, Shannon, MD|Scott, Stephanie, MD|Francis, Julie Ann, MD|Agrawal, Anita, MD|Giede,Christopher, MD. No. 344-opportunistic salpingectomy and other methods of risk reduction for ovarian/fallopian tube/peritoneal cancer in the general population. Journal of Obstetrics and Gynaecology Canada (JOGC). 2017;39(6):480-493. 84. Committee on Gynecologic Practice. Committee opinion no. 620: Salpingectomy for ovarian cancer prevention. Obstet Gynecol. 2015;125(1):279-281. 85. Hanley, Gillian Elizabeth, PhD|McAlpine, Jessica Nell, MD|Pearce, Celeste Leigh, PhD|Miller,Dianne, MD. The performance and safety of bilateral salpingectomy for ovarian cancer prevention in the united states. Obstet Gynecol. 2017;216(3):270.e9. 86. Schulz M, Lahmann PH, Riboli E, Boeing H. Dietary determinants of epithelial ovarian cancer: A review of the epidemiologic literature. Nutr Cancer. 2004;50(2):120-140. 101  87. Qiu W, Lu H, Qi Y, Wang X. Dietary fat intake and ovarian cancer risk: A meta-analysis of epidemiological studies. Oncotarget. 2016. 88. Marchbanks PA, Wilson H, Bastos E, Cramer DW, Schildkraut JM, Peterson HB. Cigarette smoking and epithelial ovarian cancer by histologic type. Obstetrics & Gynecology. 2000;95(2):255-260. 89. ..., Beral V, Gaitskell K, et al. Ovarian cancer and smoking: Individual participant meta-analysis including 28 114 women with ovarian cancer from 51 epidemiological studies. Lancet Oncology, The. 2012;13(9):946-956. 90. Beehler GP, Sekhon M, Baker JA, et al. Risk of ovarian cancer associated with BMI varies by menopausal status. J Nutr. 2006;136(11):2881. 91. Torng P. Clinical implication for endometriosis associated with ovarian cancer. Gynecology and Minimally Invasive Therapy. 2017;6(4):152-156.  92. What are the signs & symptoms of ovarian cancer? http://ovarian.org/about-ovarian-cancer/what-are-the-signs-a-symptoms. Accessed 10 April 2018. 93. Ezzati M, Abdullah A, Shariftabrizi A, et al. Recent advancements in prognostic factors of epithelial ovarian carcinoma. International Scholarly Research Notices. 2014; 2014:1-10. 94. Andriole GL, Crawford ED, Grubb RL, et al. Prostate cancer screening in the randomized prostate, lung, colorectal, and ovarian cancer screening trial: Mortality results after 13 years of follow-up. J Natl Cancer Inst. 2012;104(2):125-132. 95. Jacobs IJ, Menon U, Ryan A, et al. Ovarian cancer screening and mortality in the UK collaborative trial of ovarian cancer screening (UKCTOCS): A randomised controlled trial. The Lancet. 2016;387(10022):945-956. 96. KOBAYASHI H, YAMADA Y, SADO T, et al. A randomized study of screening for ovarian cancer: A multicenter study in japan. International Journal of Gynecological Cancer. 2008;18(3):414-420. 97. van Nagell J,John R., DePriest PD, Ueland FR, et al. Ovarian cancer screening with annual transvaginal sonography: Findings of 25,000 women screened. Cancer. 2007;109(9):1887-1896. 98. Gérvas J. Ovarian cancer screening: Could you recommend it? no. Evidence-based medicine. 2016;21(4):134-135. 99. Koshiyama M, Matsumura N, Konishi I. Subtypes of ovarian cancer and ovarian cancer screening. Diagnostics. 2017;7(1):12. 102  100. Narod SA, Sopik V, Giannakeas V. Should we screen for ovarian cancer? A commentary on the UK collaborative trial of ovarian cancer screening (UKCTOCS) randomized trial. Gynecol Oncol. 2016;141(2):191-194. 101. Balas C, Barley D, Baugh E, et al. What women and their physicians need to know about the UKCTOCS study and ovarian cancer screening. Am Fam Physician. 2016;93(11):903-904. 102. van NAGELL J, JOHN R., PAVLIK EJ. Ovarian cancer screening. Clin Obstet Gynecol. 2012;55(1):43-51. 103. Menon U, Mcguire AJ, Raikou M, et al. The cost-effectiveness of screening for ovarian cancer: Results from the UK collaborative trial of ovarian cancer screening (UKCTOCS). The British Journal of Cancer. 2017;117(5):619. 104. Naumann RW, Brown J. Ovarian cancer screening with the risk of ovarian cancer algorithm (ROCA): Good, bad, or just expensive? Gynecol Oncol. 2018;149(1):117-120. 105. Hodny EM. Increasing awareness and knowledge about ovarian cancer to enhance health outcomes of women. 2017.  106. Jelovac D, Armstrong DK. Recent progress in the diagnosis and treatment of ovarian cancer. CA: A Cancer Journal for Clinicians. 2011;61(3):183-203. 107. Schmid BC, Oehler MK. New perspectives in ovarian cancer treatment. Maturitas. 2014;77(2):128-136. 108. Banerjee S, Kaye SB. New strategies in the treatment of ovarian cancer: Current clinical perspectives and future potential. Clinical cancer research : an official journal of the American Association for Cancer Research. 2013;19(5):961-968. 109. Minig L, Padilla-Iserte P, Zorrero C. The relevance of gynecologic oncologists to provide high-quality of care to women with gynecological cancer. Frontiers in oncology. 2015;5:308. 110. Engelen M, Kos HE, Willemse P, et al. Surgery by consultant gynecologic oncologists improves survival in patients with ovarian carcinoma. Cancer. 2006;106(3):589-598. 111. Earle CC, Schrag D, Neville BA, et al. Effect of surgeon specialty on processes of care and outcomes for ovarian cancer patients. J Natl Cancer Inst. 2006;98(3):172-180. 112. Chemotherapy for ovarian cancer.  https://www.cancer.org/cancer/ovarian-cancer/treating/chemotherapy.html. Accessed 10 July 2018.  103  113. Webber K, Michael. Chemotherapy for epithelial ovarian, fallopian tube and primary peritoneal cancer. Best Practice & Research: Clinical Obstetrics & Gynaecology. 2017;41:126-138. 114. Fung-Kee-Fung, M.|Provencher, D.|Rosen, B.|Hoskins, P.|Rambout, L.|Oliver, T.|Gotlieb,W.|Covens, A., IP Chemotherapy Working Group/Society of Gynecologic Oncologists of Canada. Intraperitoneal chemotherapy for patients with advanced ovarian cancer: A review of the evidence and standards for the delivery of care. Gynecol Oncol. 2007;105(3):747-756. 115. Armstrong DK, Bundy B, Wenzel L, et al. Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med. 2006;354(1):34-43. 116. Alberts DS, Liu PY, Hannigan EV, et al. Intraperitoneal cisplatin plus intravenous cyclophosphamide versus intravenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med. 1996;335(26):1950-1955. 117. Markman M, Bundy BN, Alberts DS, et al. Phase III trial of standard-dose intravenous cisplatin plus paclitaxel versus moderately high-dose carboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin in small-volume stage III ovarian carcinoma: An intergroup study of the gynecologic oncology group, southwestern oncology group, and eastern cooperative oncology group. Journal of Clinical Oncology. 2001;19(4):1001-1007. 118. Rao G, Crispens M, Rothenberg ML. Intraperitoneal chemotherapy for ovarian cancer: Overview and perspective. Journal of Clinical Oncology. 2007;25(20):2867-2872. 119. Despite survival benefits, role of IP chemotherapy remains unclear. https://www.healio.com/hematology-oncology/hematology/news/print/hemonc-today/%7Bc9de946e-80d4-478f-86f3-c548225b1afa%7D/despite-survival-benefits-role-of-ip-chemotherapy-remains-unclear. Accessed 10 July 2018. 120. Swenerton KD, Santos JL, Gilks CB, et al. Histotype predicts the curative potential of radiotherapy: The example of ovarian cancers. Annals of oncology : official journal of the European Society for Medical Oncology. 2011;22(2):341-347. 121. Angiogenesis inhibitors. https://www.cancer.gov/about-cancer/treatment/types/immunotherapy/angiogenesis-inhibitors-fact-sheet. Accessed 10 July 2018. 122. Ball G, Xie F, Tarride J. Economic evaluation of bevacizumab for treatment of platinum-resistant recurrent ovarian cancer in canada. PharmacoEconomics - Open. 2018;2(1):19-29. 123. Parkes EE, Kennedy RD. Clinical application of poly(ADP‐Ribose) polymerase inhibitors in High‐Grade serous ovarian cancer. Oncologist. 2016;21(5):586-593. 104  124. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance therapy in platinum-sensitive relapsed ovarian cancer. N Engl J Med. 2012;366(15):1382-1392. 125. Ledermann J, Prof, Harter P, MD, Gourley C, Prof, et al. Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: A preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial. Lancet Oncology, The. 2014;15(8):852-861. 126. Akhtar-Danesh N, Elit L, Lytwyn A. Temporal trends in the relative survival among patients diagnosed with ovarian cancer in canada 1992–2005: A population-based study. Gynecol Oncol. 2011;123(2):192-195. 127. Coleman M, Prof, Forman D, PhD, Bryant H, Prof, et al. Cancer survival in australia, canada, denmark, norway, sweden, and the UK, 1995–2007 (the international cancer benchmarking partnership): An analysis of population-based cancer registry data. Lancet, The. 2011;377(9760):127-138. 128. Westin SN, Sun CC, Tung CS, et al. Survivors of gynecologic malignancies: Impact of treatment on health and well-being. Journal of Cancer Survivorship. 2016;10(2):261-270. 129. Stavraka C, Ford A, Ghaem-Maghami S, et al. A study of symptoms described by ovarian cancer survivors. Gynecol Oncol. 2012;125(1):59-64. 130. Teng FF, Kalloger SE, Brotto L, McAlpine JN. Determinants of quality of life in ovarian cancer survivors: A pilot study. Journal of obstetrics and gynaecology Canada : JOGC = Journal d'obstétrique et gynécologie du Canada : JOGC. 2014;36(8):708-715. 131. Fitch MI, Steele R. Identifying supportive care needs of women with ovarian cancer. Can Oncol Nurs J. 2010;20(2):66-74. 132. Greimel E, Daghofer F, Petru E. Prospective assessment of quality of life in long‐term ovarian cancer survivors. International Journal of Cancer. 2011;128(12):3005-3011. 133. Dinkelspiel HE, Champer M, Hou J, et al. Long-term mortality among women with epithelial ovarian cancer. Gynecol Oncol. 2015;138(2):421-428. 134. Canadian cancer statistics 2014 - special topic: Skin cancers. 2014.  135. Fairfield KM, Lucas FL, Earle CC, Small L, Trimble EL, Warren JL. Regional variation in cancer-directed surgery and mortality among women with epithelial ovarian cancer in the medicare population. Cancer. 2010;116(20):4840-4848. 136. Goff BA, Matthews BJ, Wynn M, Muntz HG, Lishner DM, Baldwin L. Ovarian cancer: Patterns of surgical care across the united states. Gynecol Oncol. 2006;103(2):383-390. 105  137. van Altena AM, Karim-Kos HE, de Vries E, Kruitwagen, Roy F P M, Massuger, Leon F A G, Kiemeney LA. Trends in therapy and survival of advanced stage epithelial ovarian cancer patients in the netherlands. Gynecol Oncol. 2012;125(3):649-654. 138. Kim SI, Lim MC, Lim J, et al. Incidence of epithelial ovarian cancer according to histologic subtypes in korea, 1999 to 2012. Journal of Gynecologic Oncology. 2016;27(1):e5. 139. BC cancer agency registry data (2011). Population data BC. BC cancer agency (2011). http://www.popdata.bc.ca/data. Accessed 19 April 2017.  140. BC vital statistics agency (2011): vital statistics deaths. Population data BC. BC vital statistics agency (2011). Accessed 20 May 2017. 141. Population estimates. https://www2.gov.bc.ca/gov/content/data/statistics/people-population-community/population/population-estimates. Accessed 5 January 2018. 142. Trétarre, Brigitte|Molinié, Florence|Woronoff, Anne-Sophie|Bossard, Nadine|Bessaoud, Faiza|Marrer, Emilie|Grosclaude, Pascale|Guizard, Anne-Valérie|Delafosse, Patricia|Bara, Simona|Velten, Michel|Lapôtre-Ledoux, Bénédicte|Ligier, Karine|Léone, Nathalie|Arveux,Patrick|Uhry, Zoé. Ovarian cancer in france: Trends in incidence, mortality and survival, 1980–2012. Gynecol Oncol. 2015;139(2):324-329. 143. Collaborative Group on Epidemiological Studies of Ovarian Cancer, Collaborative Group on Epidemiological Studies of Ovarian Cancer, Cancerepidemiologi, et al. Ovarian cancer and oral contraceptives: Collaborative reanalysis of data from 45 epidemiological studies including 23 257 women with ovarian cancer and 87 303 controls. Lancet, The. 2008;371(9609):303-314. 144. Canadian Institute for Health Information, Canadian Electronic Library (Firm), desLibris - Documents, Statistics Canada. Health care in canada 2010. Canadian Institute for Health Information; 2011. 145. Chiang Y, Chen C, Chiang C, et al. Trends in incidence and survival outcome of epithelial ovarian cancer: 30-year national population-based registry in taiwan. Journal of Gynecologic Oncology. 2013;24(4):342-351. 146. Hart WR. Mucinous tumors of the ovary: A review. International Journal of Gynecological Pathology. 2005;24(1):4-25. 147. Gottschau M, Mellemkjaer L, Hannibal CG, Kjaer SK. Ovarian and tubal cancer in denmark: An update on incidence and survival. Acta Obstet Gynecol Scand. 2016;95(10):1181-1189. 106  148. Zaino RJ, Brady MF, Lele SM, Michael H, Greer B, Bookman MA. Advanced stage mucinous adenocarcinoma of the ovary is both rare and highly lethal: A gynecologic oncology group study. Cancer. 2011;117(3):554-562. 149. Köbel M, Bak J, Bertelsen BI, et al. Ovarian carcinoma histotype determination is highly reproducible, and is improved through the use of immunohistochemistry. Histopathology. 2014;64(7):1004-1013. 150. Media advisory - canadian cancer society releases canadian cancer statistics 2016 on wednesday, october 19, 2016. Canada NewsWire. 2016.  151. Colzani E, Liljegren A, Johansson ALV, et al. Prognosis of patients with breast cancer: Causes of death and effects of time since diagnosis, age, and tumor characteristics. Journal of Clinical Oncology. 2011;29(30):4014-4021. 152. Van Hemelrijck M, Folkvaljon Y, Adolfsson J, et al. Causes of death in men with localized prostate cancer: A nationwide, population-based study. BJU Int. 2016;117(3):507. 153. Massa ST, Osazuwa-Peters N, Christopher KM, et al. Competing causes of death in the head and neck cancer population. Oral Oncol. 2017;65:8-15. 154. Janssen-Heijnen MLG, van Erning FN, De Ruysscher DK, Coebergh JWW, Groen HJM. Variation in causes of death in patients with non-small cell lung cancer according to stage and time since diagnosis. Annals of Oncology. 2015;26(5):902-907. 155. Webb PM, Jordan SJ. Epidemiology of epithelial ovarian cancer. Best Practice & Research Clinical Obstetrics & Gynaecology. 2017;41:3-14. 156. British columbia ministry of heath (2011): Consolidation file (MSP registration and premium billing). Population data BC. MOH (2011). Accessed 20 May 2017.  157. About popdata. https://www.popdata.bc.ca/. Accessed 20 May 2017.  158. RStudio team (2016). RStudio: Integrated development for R. RStudio, inc., boston, MA URL http://www.rstudio.com/.  Accessed 20 January 2016.   159. Gangi A, Cass I, Paik D, et al. Breast cancer following ovarian cancer inBRCAMutation carriers. JAMA Surgery. 2014;149(12):1306. 160. Menon U, Griffin M, Gentry-Maharaj A. Ovarian cancer screening—Current status, future directions. Gynecol Oncol. 2014;132(2):490-495. 161. Lockwood-Rayermann S. Survivorship issues in ovarian cancer: A review. Oncol Nurs Forum. 2006;33(3):553-562. 107  162. Kuan A, Teng C, Wu H, et al. Risk of ischemic stroke in patients with ovarian cancer: A nationwide population-based study. BMC medicine. 2014;12(1):53. 163. Rodriguez AO, Wun T, Chew H, Zhou H, Harvey D, White RH. Venous thromboembolism in ovarian cancer. Gynecol Oncol. 2007;105(3):784-790. 164. Travis LB, Curtis RE, Boice J,J D., Platz CE, Hankey BF, Fraumeni J,J F. Second malignant neoplasms among long-term survivors of ovarian cancer. Cancer Res. 1996;56(7):1564. 165. Bouchardy C, Fernandez S, Merglen A, et al. Increased risk of second cancer among patients with ovarian borderline tumors. Gynecol Oncol. 2008;109(2):210-214. 166. Sood AK, Drapkin R, Berchuck A, et al. Rethinking ovarian cancer: Recommendations for improving outcomes. Nature Reviews Cancer. 2011;11(10):719-725. 167. Doherty JA, Jensen A, Kelemen LE, et al. Current gaps in ovarian cancer epidemiology: The need for new population-based research. J Natl Cancer Inst. 2017;109(10). 168. Jatoi A, Foster NR, Kalli KR, et al. Prior oral contraceptive use in ovarian cancer patients: Assessing associations with overall and progression-free survival. BMC Cancer. 2015;15(1):711.    108  Appendices   Stepwise diagnostic evaluation of ovarian cancer                                                                       Yes                                                                                                               No     Suspect ovarian cancer?   No                                                                                  Yes                                                                                           Abnormal or suspicious features **    **CA-125 >200U per mL in premenopausal women or elevated levels in after menopause; volume of ovaries >20mL in non- pregnant premenopausal women or >10mL at postmenopausal stage; abnormal or persistent pelvic mass, evidence of abdominal metastasis.  Evaluation of symptoms (or early warning signs); abdominal pain, bloating, frequent urination, and loss of appetite)  -Rectovaginal examination or pelvic examination -Risk assessment (Family history of cancers, particularly breast cancer and ovarian cancer).  - Assessment of other risk factors such as personal history of cancer, nulliparity etc.   -Genetic risk assessment (BRCA mutation carrier or not)  -Risk assessment  -Genetic testing and counselling (if BRCA mutation carrier) -TVS (assessment of ovarian outlook, detecting ascites and differentiating benign tumors (adnexal lesions) from malignant) - (Cancer antigen) CA-125 biomarker test  -Risk assessment  -Genetic testing and counselling (if BRCA mutation carrier) -Any strategies that could reduce the ovarian cancer such as OCP use.  -Referral to gynecologic oncologist for biopsy -Treatment: Risk reducing surgery (bilateral oophorectomy, hysterectomy etc) and/or chemotherapy  109  Underlying causes of death and their ICD-10 codes Causes of Death ICD-10 codes Ovarian cancer or ovarian cancer related C510-C58 Breast cancer C500-C509 Colorectal cancer C180-C218 Other cancer Lung: C33, C340-C349, C384, C450;  GI: C150-C179, C220-C269;  Blood lymph: C810-C969, C463;  Other malignancy: C000-C148, C300-C449, C451-C462, C467-C499, C609, C620-C768, C5091, C80;  Non malignant and unspecified: D000-D489 Cardiovascular  Rheumatic/Valvular: I050-I099, I340-I38;  Hypertension: I10-I159; Ischemic: I200-I259;  Conductive & Dysrhythmic: I440-I499;  Heart failure:  I500-I509;  Congenital: Q200-Q249;  Pulmonary: I260-I289;  Cardiomyopathy: I420-I429;  Unspecified: I312-I318, I510-I513, I515-I519 Other chronic Diabetes: E100-E149;  COPD: J440-J449;  AIDS/HIV: B200-B24;  Pneumonia: J120-J181, J188-J189;  110  Other infectious and parasitic disease: A000-B199, B250-B999, U049;  Asthma: J450-J459, J46;  Cerebro and other vascular disease: I600-I698, I700-I879, I950-I959, I880-I899;  Liver disease: K700-K7699;  Pulmonary fibrosis: J841; ALS/MS: G122, G1221, G35;  Lung disease due to external agents: J60-J709 External causes of death Motor vehicle accidents (MVA): V020-V049, V090-V099, V120-V149, V190-V196, V200-V799, V803-V805, V820-V821, V823-V839, V840-V878, V880-V888, V8900-V8909,V8920-V8929, V8990-V8999, Y850;  Poisoning: X40-X49;  Falls: W00-W19; Suicide: X60-X84, Y870;  Other external: Y10-Y369, Y890-Y899;  Other unintentional injuries: V010, V019, V050-V069, V091, V099, V100-V119, V150-V189, V198-V199, V250-V259, V350-V359, V450-V459, V550-V559, V650-V659, V750-V759, V800-V802, V806-V819, V822, V879, V889, V910-V919, V930-V949, V950-V978, V98-V99, W20-W64, W75-W99, X20-X39, X50-X59, Y40-Y849, Y859, Y86, Y880-Y883  Unclassified cause of death Codes that did not meet criterion above      111  1- year, 3- year and 5- year age standardised overall survival rates among women with ‘others and unspecified’ EOCs, 1980 – 2015                                   112  1- year, 3- year and 5- year age standardised disease specific survival rates among women with ‘others and unspecified’ EOCs, 1980 – 2015                  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0372092/manifest

Comment

Related Items