Open Collections

UBC Faculty Research and Publications

Characteristics and outcome of the COEUR Canadian validation cohort for ovarian cancer biomarkers Le Page, Cécile; Rahimi, Kurosh; Köbel, Martin; Tonin, Patricia N; Meunier, Liliane; Portelance, Lise; Bernard, Monique; Nelson, Brad H; Bernardini, Marcus Q; Bartlett, John M S; Bachvarov, Dimcho; Gotlieb, Walter H; Gilks, Blake; McAlpine, Jessica N; Nachtigal, Mark W; Piché, Alain; Watson, Peter H; Vanderhyden, Barbara; Huntsman, David G; Provencher, Diane M; Mes-Masson, Anne-Marie Mar 27, 2018

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

Item Metadata

Download

Media
52383-12885_2018_Article_4242.pdf [ 1.34MB ]
Metadata
JSON: 52383-1.0364561.json
JSON-LD: 52383-1.0364561-ld.json
RDF/XML (Pretty): 52383-1.0364561-rdf.xml
RDF/JSON: 52383-1.0364561-rdf.json
Turtle: 52383-1.0364561-turtle.txt
N-Triples: 52383-1.0364561-rdf-ntriples.txt
Original Record: 52383-1.0364561-source.json
Full Text
52383-1.0364561-fulltext.txt
Citation
52383-1.0364561.ris

Full Text

RESEARCH ARTICLE Open AccessCharacteristics and outcome of the COEURCanadian validation cohort for ovariancancer biomarkersCécile Le Page1, Kurosh Rahimi1,2, Martin Köbel3, Patricia N. Tonin4, Liliane Meunier1, Lise Portelance1,Monique Bernard1, Brad H. Nelson5, Marcus Q. Bernardini6, John M. S. Bartlett7, Dimcho Bachvarov8,Walter H. Gotlieb9, Blake Gilks10,11, Jessica N. McAlpine10, Mark W. Nachtigal12, Alain Piché13, Peter H. Watson5,Barbara Vanderhyden14,15, David G. Huntsman10,11, Diane M. Provencher1,16 and Anne-Marie Mes-Masson1,17,18*AbstractBackground: Ovarian carcinoma is the most lethal gynecological malignancy due to early dissemination and acquiredresistance to platinum-based chemotherapy. Reliable markers that are independent and complementary to clinicalparameters are needed to improve the management of patients with this disease. The Canadian Ovarian ExperimentalUnified Resource (COEUR) provides researchers with biological material and associated clinical data to conduct biomarkervalidation studies. Using standards defined by the Canadian Tissue Repository Network (CTRNet), we have previouslydemonstrated the quality of the biological material from this resource. Here we describe the clinical characteristics of theCOEUR cohort.Methods: With support from 12 Canadian ovarian cancer biobanks in Canada, we created a central retrospective cohortcomprised of more than 2000 patient tissue samples with associated clinical data, including 1246 high-grade serous, 102low-grade serous, 295 endometrioid, 259 clear cell and 89 mucinous carcinoma histotypes. A two-step reclassificationprocess was applied to assure contemporary histological classification (histotyping). For each histotypes individually, weevaluated the association between the known clinico-pathological parameters (stage, cytoreduction, chemotherapytreatment, BRCA1 and BRCA2 mutation) and patient outcome by using Kaplan-Meier and Cox proportional hazardregression analyses.Results: The median follow-up time of the cohort was 45 months and the 5-year survival rate for patients with high-grade serous carcinomas was 34%, in contrast to endometrioid carcinomas with 80% at 5 years. Survival profiles differedby histotype when stratified by stage or cytoreduction. Women with mucinous or clear cell carcinomas at advancedstage or with non-optimally debulked disease had the worst outcomes. In high-grade serous carcinoma, we observedsignificant association with longer survival in women harboring BRCA1 or BRCA2 mutation as compared to patientswithout detectable mutation.Conclusions: Our results show the expected survival rates, as compared with current literature, in each histotypesuggesting that the cohort is an unbiased representation of the five major histotypes. COEUR, a one stop comprehensivebiorepository, has collected mature outcome data and relevant clinical data in a comprehensive manner allowingstratified analysis.Keywords: Epithelial ovarian cancer, Histotype, Biomarker, BRCA, Survival, Treatment response* Correspondence: anne-marie.mes-masson@umontreal.ca1Centre de recherche du Centre hospitalier de l’Université de Montréal(CRCHUM) and Institut du cancer de Montréal, Montreal, QC, Canada17Department of Medicine, Université de Montréal, Montreal, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Le Page et al. BMC Cancer  (2018) 18:347 https://doi.org/10.1186/s12885-018-4242-8BackgroundAlthough epithelial ovarian cancer (EOC) has an inci-dence rate ten times less frequent than breast cancer, isthe fifth-leading cause of cancer-related deaths amongwomen in the Western world [1]. EOC encompasses atleast five distinct diseases, represented by five histo-pathological types, herein designate as histotypes, withunique characteristics, different molecular features, dif-ferent clinical behavior, and sites of origin: high-gradeserous carcinoma (HGSC), low-grade serous carcinoma(LGSC), endometrioid carcinoma (EC), mucinous car-cinoma (MC), and clear cell carcinoma (CCC) [2, 3].The less common histotypes LGSC and MC, could beconsidered rare diseases with approximately 1000 casesper year in the US [1]. HGSC is the most common histo-type and is often diagnosed at an advanced stage (stageIII or IV) purportedly due its rapid growth rate, origin inthe fallopian tube, its associated with local spread intothe peritoneal cavity, and late onset of symptoms. This isin contrast to other histotypes, which often present as alocalized pelvic mass. The standard treatment for EOC,particularly for HGSC, is debulking surgery andplatinum-based chemotherapy. However, resistance tochemotherapy often develops, contributing to a low 5-year survival rate for patients with advanced stage dis-ease [1]. Assessment of clinical parameters, such as his-totype, disease stage and residual disease (RD), areimportant factors in determining the management of pa-tients, but they cannot always predict response to treat-ment and survival. Although the length of survival hassignificantly improved in the last decade with recent dis-coveries of new therapeutic targets and drug develop-ments, survival rates and the quality of life of patientsremains poor [4]. Therefore, reliable markers that are in-dependent and complementary to clinical parametersare needed to improve patient management.Diseases as heterogeneous as EOC require the supportof large research networks to provide access to enoughspecimens to investigate the association between linkedclinical data and deep molecular analysis for the devel-opment of precision medicine. For this reason, the Aus-tralian Ovarian Cancer Study was launched in 2001 andhas accrued more than 2000 cases in an attempt to iden-tify genetic variants that can be associated with the de-velopment of ovarian cancer (http://www.aocstudy.org).In UK, BriToc is collecting tumor samples at diagnosisand recurrence (http://ovarian.org.uk/our-research/the-research-we-fund/) in four academic centers. Likewise,the Canadian Ovarian Experimental Unified Resource(COEUR) program, the first Canadian project to be initi-ated by the Ovarian Cancer Consortium, formed in2009, with the collaboration of the GOC (Society of Gy-necologic Oncology of Canada), and a group of investi-gators and biobank scientists established the program todevelop a high-quality biospecimen repository with asso-ciated clinical data for large-scale collaborations thatcould advance research in molecular biomarker valid-ation in EOC (http://www.tfri.ca/en/research/transla-tional-research/coeur.aspx). Importantly, this resource isavailable to researchers through a managed reviewprocess for approved ovarian cancer studies at: http://www.tfri.ca/en/research/translational-research/coeur/coeur_access.aspx. The main goal of the COEUR focuseson biomarker validation for more accurate classificationof pathological specimens and better prediction of theclinical outcomes. Taking advantage of the pre-existingovarian cancer biobanks in Canada, we created a centralretrospective cohort of patient tissue samples. The feasi-bility of this pan-Canadian ovarian cancer repository andbiomarker program was assessed in the pilot phase ofthe project that was completed in 2010. The main activ-ities were centralized at the Centre de recherche duCentre hospitalier de l’Université de Montréal(CRCHUM, Montreal, QC), and involved the small col-lection of specimens with associated data, and includedprocessing and quality control of samples for use in re-search assays [5]. We demonstrated that an adequate re-source was available to create a national cohort for thevalidation of molecular biomarkers in ovarian cancer.Moreover, this pilot study allowed the COEUR toproceed with confidence in collecting 2000 retrospectivecases and enter the next phase of the program.In the second phase of the COEUR program we col-lected DNA from normal fallopian tube or buffy coat,FFPE tissues, frozen tissues, ascites, plasma and serumwhen available. The initial goal was to collect samplesfrom the five main histotypes: HGSC, EC, MC, LGSCand CCC. HGSC represented the majority of collectedcases, whereas MC and LGSC cases were limited to atleast 100 cases. Because before 2015, local ovarian can-cer diagnosis showed up to 15% of misclassification, heretwo pathologists, one at CRCHUM and one at the Uni-versity of Calgary, performed a double central review ofFFPE blocks.The present study aimed to describe the characteristicsand clinico-pathological prognostic factors of the pan-Canadian COEUR cohort. Because different histopathol-ogies of ovarian cancer are now recognized as distinctdiseases, we analyzed these parameters separately foreach histotype. Combined with associated clinical data,we show that the survival rates reflect the different clin-ical behavior of these diseases [2].MethodsParticipating biobanksThe participating biobanks of the COEUR program aredescribed in Table 1. A material transfer agreement wassigned in 2010 by all participating biobanks, the TerryLe Page et al. BMC Cancer  (2018) 18:347 Page 2 of 18Table1DescriptionofbiobanksparticipatingintheCOEURprogramBankName*Location(Canada)*EstablishedinFundingorsupport*Partner*WebaddressSegalCancerCenterLadyDavisInstitutebiobankLadyDavisInstitute-JewishGeneralHospital,Montreal,QC2003FondsdelarechercheensantéQuébec(networkRRCancer),Montreal-IsraelCancerResearchFoundation,theeGloria’sGirlsFund,theSusanandJonathnWenerfundandtheGarberfund.CTRNet,OvarianCancerCanadawww.rrcancer.caBanquedetissusetdedonnéesduRéseauderecherchesurlecancerduFRQSCRCHUM-Montreal,QC2000FondsdelarechercheensantéQuébec(networkRRCancer),OvarianCancerCanadaCTRNetwww.rrcancer.caOttawaOvarianCancerTissueBankOttawaHospitalResearchInstitute,Ottawa,ON1995OCC,OttawaHospitalOvarianCancerCanadaN/AOVCAREGynecologicTissueBankVancouverGeneralHospital,Vancouver,BC2001OVCAREBC,MichaelSmithFoundationforHealthResearchandOvarianCancerCanadaBCCancerAgency,VancouverCoastalHealthResearchInstitute,UniversityBC,VancouvverGeneralHospitalwww.ovcare.ca/research/platforms.phpOvariancancerspecimenanddatabankatCHUQHopitalHotel-DieudeQuebec,Quebec,QC2000FondsNationaldeRechercheduQuébec(networkRRCancer)FondsdelarechercheensantéQuébec(networkRRCancer),CTRNetwww.rrcancer.caBanqued’échantillonsbiologiquesetdedonnéesdeSherbrookeFacultéderechercheetdessciencesdelasanté,UniversitédeSherbrooke,Sherbrooke,QC1998FondsdelarechercheensantéQuébec(RRCancer)andQuebecbreastcancerfoundationCTRNet,OvarianCancerCanadawww.rrcancer.caAlbertaCancerResearchBiobankUniversityofAlberta,Edmonton,AB;andUnivesityofCalgary,Calgary,AB2001AlbertaCancerFoundation,CanadianBreastCancerfoundationCTRNet,CanadianBreastCancerfoundationhttp://www.acrb.ca/BCCancerAgencyTumourTissueRepositoryBCCancer,Agency,VancouverIslandCenter,BC2004BCCancerAgencyandProvinicalHealthServicesAuthorityCTRNetwww.bccrc.ca/dept/ttrUHNBioBankTorontoGeneralHospital2013UHNResearchInstitute,UHNFoundationN/Ahttps://www.uhnresearch.ca/service/uhn-biospecimen-sciences-program-bspUHNprograminBiospecimenSciencesSunnybrookHospital,Toronto,ON2001UHNResearchInstitute,UHNFoundationN/AN/AOntarioTumourBankOntarioInstituteforCancerResearchMaRSCentre,Toronto,ON2004OntarioInstituteforCancerResearchN/Aontariotumourbank.caManitobaOvarianBiobankingProgramUniversityofManitoba,Winnipeg,MN2010N/ACTRNetN/A*Abbreviations:CHUQCentreHospitalierUniversitairedeQuebec,CRCHUMCentredeRechercheduCHUM,CRTNetCanadianTissueRepositoryNetwork;rechercheduQuebec,UHNUniversityHealthNetworkLe Page et al. BMC Cancer  (2018) 18:347 Page 3 of 18Fox Research Institute and the CRCHUM, to centralizeand distribute samples and their associated clinical datafor approved research. Between 2010 and 2017, a total of2045 retrospective cases were collected and shipped tothe central location, the CRCHUM.Selection criteria and specimen collectionThe initial selection criteria included patients with nochemotherapy treatment before oophorectomy (neoad-juvent treatment) to avoid potential effects on proteinexpression by the chemotherapeutic treatment and sub-sequent bias towards biomarker validation. However,100 cases with neoadjuvant treatment were collectedbefore full clinical data was assessed and passedthrough this filter and were included in the cohort. In-clusion criteria also required a minimum 12-monthfollow-up if there was no event of recurrence or death.This criteria was not verified by all biobanks: we re-ceived 55 samples with little (< 6 months) or no follow-up time after diagnosis. For biospecimens, selection cri-teria included either frozen or FFPE tissues, and ifavailable, DNA from non-cancerous tissues, serum, as-cites, or plasma. The last criteria involved the exclusionof the following histotypes: borderline tumors, mixedhistotypes, Brenner or undifferentiated carcinomas.Specimens (n = 2045) that were received at the CHUMcentral location were re-evaluated under the selectioncriteria described above. In addition, the COSPv3 8-immunohistochemical marker panel in conjunctionwith re-review arbitration (when necessary) was appliedto confirm or reclassify the histotype diagnosis of eachcase. Twenty-nine cases (2.4%) could not be assigned toone of the five histotypes (i.e Brenner tumor, not pri-mary ovarian or undifferentiated). Twenty-five casesdid not satisfy the selection criteria because only bor-derline tumor tissue was found on the sample received.Six cases could not be assigned to a histotype becauseFFPE blocks were not available for additional analysis,and the original diagnosis did not clearly indicatewhether they were LGSC or HGSC cases.Clinical data collection and definition of parametersIn accordance with the ethics approval and materialtransfer agreement, only anonymized data was trans-ferred from the source biobank to the central location atthe CRCHUM (Montreal, QC, Canada). A templatespreadsheet standardized the collection of data and min-imized variation in data submission across participatinginstitutions. Data were reviewed by a scientist in consult-ation with the clinical data monitoring committee, com-prised of two gynecologic oncologists who identifiedpotentially inaccurate or non-standardized data. Spread-sheets were reviewed for discrepancies or missing data,and participating biobanks were contacted to correctinconsistant data.Clinical data were updated annually for follow-up. InDecember 2017, the 5-year survival rate for the entirecohort was 45%, which corresponds with the overall sur-vival rate reported for ovarian carcinoma [1]. Cases withno follow-up were not included in the overall survivalanalysis. Long-term survivors were defined as patientswho survived at least 10 years after diagnosis. Patientswith less than 10 years follow-up and no record of deathspecific to ovarian cancer were not included in the long-term survivor proportion indicated in Table 2. Theamount of residual disease (RD) or cytoreduction effi-ciency was difficult to collect due to lack of consistentreporting in the biobank data (66%). To standardize thisdata we translated the RD amount as non-optimal cytor-eduction when RD was > 1 cm as per 2012 clinical stan-dards [6]. When only ‘optimal’ cytoreduction wascollected from biobanks, we could not translate this in-formation into the measurable variable (> 1 cm or 1–2 cm) since the standard has changed over time and alsomay not be the same in all institutions.In particular, data on BRCA1 or BRCA2 mutationspertained to the presence of a mutation in either ofthese genes or both, or was recorded as non-carrierswhen no pathogenic mutation was found after targetedor full gene testing. Testing was done locally, often ingenetic clinic and counselling clinics, although the ori-ginal file were not available only the transcription fromthe hospital clinical files. Type of genetic testing or spe-cific type of mutation was not requested. Cases carryinga variant of unknown significance and cases carryingdouble BRCA1 and BRCA2 mutations were excludedfrom survival analysis due the limited number of casesin these 2 categories.Radiotherapy treatment and imaging results werecoded as positive or negative. Progression was definedaccording to the Gynecologic Cancer InterGroup(GCIG), the earliest date between CA125 rise and ob-jective clinical progression. The definition of CA125-associated recurrence was not the same in all institu-tions. When CA125 data were available, we defined theCA125 progression date as described by the GCIG: thefirst date when CA125 levels rose two times above theupper limit of the reference (35 U/ml) or nadir [7]. Forpatients with post-operative measurable RD, response totreatment was defined as first progression if it occurredwithin 6 months of the end of first-line treatment(debulking surgery and chemotherapy). Information onchemotherapy and radiation treatments included thetype of drug and date of treatment initiation and com-pletion. The cohort showed a relatively homogenoustreatment type with 83% treated with a combination ofplatinum-taxane (carboplatinum or cisplatinum andLe Page et al. BMC Cancer  (2018) 18:347 Page 4 of 18Table 2 Patient Characteristics of the COEUR cohortHistotypes HGSC EC LGSC CCC MC TotalNumber of samplesper histotypes1246 296 102 259 88 1991Age diagnosis (years) Mean 62 58 53 56 53 60Age end of follow-up (years) Mean 66 63 59 61 58 64Follow-up time (months) Mean 47 64 60 58 52 51Min-Max 0–211 0–268 0–270 0–247 1–228 0–270Total N 1240 294 102 258 87 1983Stage 1 70(6%) 145(54%) 13(14%) 127(52%) 61(74%) 4162 137(12%) 74(28%) 7(7%) 54(22%) 11(13%) 2833 835(72%) 45(17%) 67(72%) 58(23%) 8(9%) 10134 123(11%) 5(2%) 6(6%) 7(3%) 3(3%) 144Total N 1165 269 93 246 83 1856Residual Disease none 207(28%) 120(75%) 16(22%) 93(68%) 36(81%) 472(36%)< 1 cm 231(31%) 29(18%) 21(32%) 26(19%) 4(10%) 310(24%)suboptimal 387(47%) 11(7%) 27(41%) 17(12%) 4(10%) 110(8%)Total 824 160 64 136 44 1311BRCA1 and BRCA2mutation statusMutation status knownbut not identified (%)1(< 1%)) 0 1(3%) 0 0 2(< 1%)BRCA1 Carriers (%) 68(14%) 0 2(6%) 0 0 70(12%)double BRCA1/BRCA2carrier (%)4(< 1%) 0 0 0 0 4(< 1%)BRCA2 carriers (%) 35(7%) 0 1(3%) 0 0 36(6%)Non carriers (%) 370(76%) 44(100%) 31(89%) 25(100%) 6(100%) 476(80%)Carriers of variants ofunknown significance10(2%) 0 0 0 0 10(< 2%)Total BRCA mutationcarrier108 0 4 0 0 112Survival rate Alive 386(32%) 227(83%) 45(45%) 148(59%) 63(73%) 869(45%)Deceased 830 47 56 107 23 1063Total 1216 274 101 255 86 19325-years survival rate Alive 353(34%) 138(80%) 46(54%) 118(57%) 34(65%) 689(44%)Deceased 678 35 39 89 18 859Total 1031 173 85 207 52 1548Response to treatmenta sensitive 421(68%) 27(73%) 29(60%) 16(44%) 4(57%) 497(66%)resistant 201 10 19 20 3 253Total 622 37 48 36 7 750Chemotherapy beforefirst progressionnone 22(2%) 35(14%) 8(67%) 15(6%) 37(53%) 117(7%)carbo+taxol 955(81%) 184(74%) 66(67%) 195(82%) 31(40%) 1431(78%)cispl+taxol 71(62%) 8(3%) 10(11%) 6(3%) 0 95(5%)platinum 63(5%) 14(6%) 7(7%) 13(5%) 3(4%) 100(6%)platinum+other 38(3%) 6(2%) 5(6%) 6(2.5%) 1(1%) 56(3%)taxol 3(< 1%) 0% 0 0 0 3(< 1%)other 17(1%) (< 1%) 0 3(1%) 1(< 1%) 22(1%)Total 1169 248 96 238 73 1824Le Page et al. BMC Cancer  (2018) 18:347 Page 5 of 18docetaxel or paclitaxel). Platinum alone was used in 6%of cases and other agents in 4% of cases and includedgemcitabine, cyclophosphamide, doxorubicine, topote-can or procytox (Table 2). The remaining 7% did notreceive chemotherapy treatment. Radiotherapy was re-corded in 3.4% of all patients, but mainly restricted toHGSC and CCC patients.TMA and immunohistochemistry-supported histotypingThe histotype of FFPE samples was reviewed by a cen-tral pathologist at the CRCHUM and confirmed on theconstructed TMA by a second pathologist using theCOSPv3 8-marker panel [8]. A process further referredto a two-step central pathology review. Cases with dis-cordant original and predicted histotype were subjectedto arbitration by biomarker-assisted review [8]. Whenno FFPE sample was available, the original histotypewas used for the study. Tumor stage and cytoreductionwere scored according to criteria from the Federationof Gynecologists and Obstetricians [9].We also noticed that more than 10% of collected FFPEblocks of LGSC and MC did not contain representativeproportions of malignant tumor cells but rather borderlinetumor cells. Because only one block per case was availablefor review, we could not confirm the malignant diagnosisof the patient and did not include these cases in this study.The TMA with mucinous tumors was additionally stainedwith the Müllerian cell lineage marker PAX8 and the in-testinal marker SATB2 [10]. Twenty-three cases of MCstained positively for PAX8 (11 focal and 12 diffuse) and3.3% (n = 2/61) stained positive with SATB2. No case wasdeemed to be a metastatic intestinal carcinoma, whichwould have been indicated by absence of PAX8 expressionin combination with presence of SATB2 expression.Statistical analysisKruskal-Wallis and Chi-square tests were used respectivelyto assess the association between continuous and categor-ical variables or two categorical variables. Disease-specificsurvival analyses were evaluated with Kaplan-Meier curvescoupled with the log-rank test. To estimate the hazard ra-tio (HR) we used Cox proportional regression analyses. Tofacilitate comparisons with previous analyses, we usedKaplan-Meier or Cox regression analyses recognizing thatcumulative incidence estimates and competing risk regres-sion analyses may be more appropriate. Response to treat-ment and long-term survival were evaluated by logisticregression and odds ratio (OR). Significance level was setat p < 0.05. All statistical analyses were performed usingSPSS, version 21.ResultsHistopathology review and reclassification of the COEURcohortSince EOC encompasses distinct histopathologies, weorganized the analysis of the ovarian carcinoma cohortby histotype. As previously described by Köbel et al.[8], the most common histotype revision were casesreclassified from EC to HGSC (n = 43) and vice versa(n = 14), and from original HGSC to LGSC (n = 30)and vice versa (n = 19). A relatively significant propor-tion of MC cases were also reclassified to EC (n = 10).This histotyping exercise confirmed that CCC andHGSC were the two histotypes with the lowest reclas-sification rates at 5% (Table 3).The reclassified histotype distribution followed thepopulation frequency of ovarian cancer: HGSC (61%),LGSC (3%), EC (15%), MC (4%) and CCC (13%) (Table 2)[11]. A significant difference in patient age at diagnosiswas observed among histotypes with HGSC patients be-ing the oldest and MC the youngest of the non-serouscases (Kruskal-Wallis p < 0.001, test). The 5-year survivaland the long-term survivor rates were also quite differ-ent among histopathologies (Table 2) and in accordancewith the rates described in the literature [12]. The ex-ceptions were EC and CCC, which showed higher sur-vival rates (80% and 57%) compared to reports of 60%and 40%, respectively. This discrepancy could be ex-plained by the correct reclassification of EC to HGSC, aspreviously reported [8, 13].As expected based on independent reports [2, 14],HGSC cases have the lowest 5-yr and overall diseasespecific survival (Fig. 1a and Table 2). MC and CCCshare similar survival profile (Fig. 1a) and EC casesshowed the best outcome (83% survival).Table 2 Patient Characteristics of the COEUR cohort (Continued)Histotypes HGSC EC LGSC CCC MC TotalNumber of samplesper histotypes1246 296 102 259 88 1991Long term survivors(> 10 years)yes 73(8%) 38(45%) 10(16%) 23(18%) 7(23%) 151(13%)no 818 47 54 102 23 1044Total 891 85 64 125 30 1195Borderline ovarian tumor cases were excludedN number of case, carbo carboplatinum, cispl cisplatinum, BOT Borderline Ovarian Tumor, HGSC High-Grade Serous Carcinomaapatients with platinum-based treatmentLe Page et al. BMC Cancer  (2018) 18:347 Page 6 of 18Cohort characteristics—Disease stage and cytoreductionThe proportion of non-serous histotypes with advancedstage disease (stage III-IV) was lower than other histo-types (< 25%) particularly for EC and MC histotypes aspreviously reported [15]. When patients were restrictedto those with advanced disease (stages III and IV), thedifference in overall disease-specific survival rates be-tween HGSC and EC (30% and 53%, HR = 0.19 (0.14–0.26, 95% CI)) was drastically decreased as compared tothe rate between these two histopathologies across allstages (32% and 81%, respectively; HR = 0.51 (0.34–0.77)95% CI) Table 4 and Fig. 1b). At advanced stage, CCCand MC patients had the highest risk of death with amedian survival of 17 and 13 months respectively(Table 4 and Fig. 1b). However, when patients were eval-uated with presence of residual disease in advanced stagedisease (further referred as “poor prognosis” disease),the difference in survival between HGSC and EC was noFig. 1 Kaplan-Meier survival curves of disease-specific survival by EOC histotypes. a All five histotypes (HGSC = 1246, EC = 274, LGSC = 101, CCC = 255,MC= 86). b Comparison of all histotypes of advanced stage (stage III and IV) (HGSC = 940, EC = 49, LGSC = 73, CCC = 65, MC= 11). c Comparison of allhistotypes of advanced stage cases with residual disease (RD) (poor prognosis) (HGSC = 621, EC = 18, LGSC = 49, CCC = 27, MC = 7). d Comparison ofall histotypes of low stage diseases (stage I and II). Significance (p) is indicated by log-rank (HGSC = 206, EC = 214, LGSC = 20, CCC = 178, MC= 72)Table 3 Re-classification of histotypesOriginal classification: Revised Classification TotaloriginalBOT HGSC EC LGSC CCC MC other unknownHGSC 1 1131 (95%) 14 30 9 2 3 1 1191EC 1 43 267 (84%) 0 0 2 3 1 317LGSC 9 19 0 69 (72%) 0 0 0 0 96CCC 1 4 4 0 248 (96%) 0 0 0 257MC 10 6 10 0 0 84 (73%) 6 0 116other 1 7 1 1 1 0 11(55%) 0 20unknown 2 36 0 3 1 0 0 6 (NA) 48Total after reclassification 25 1246 296 102 259 88 21 8 2045(%) represent the % of original cases correctly classified, BOT Borderline Ovarian Tumor, HGSC High-Grade Serous Carcinoma, EC Endometrioid Carcinoma, LGSCLow-Grade Serous Carcinoma, CCC Clear Cell Carcinoma, MC Mucinous carcinomaLe Page et al. BMC Cancer  (2018) 18:347 Page 7 of 18Table4UnivariateCoxregressionmodelofhistotypesondiseasespecificsurvivalHistotypeAllcasesAdvancedstagePoorprognosisLowstageHazard95.0%CInHazard95.0%CInHazard95.0%CInHazard95.0%CIRatioLowerUpperpvalueRatioLowerUpperpvalueRatioLowerUpperpvalueRatioLowerUppernpvalueHGSC1216reference940reference621reference206referenceEC0.190.140.262740.0000.510.340.77490.0010.850.481.51180.5750.200.120.342140.000LGSC0.620.470.811010.0010.670.500.91730.0090.720.501.03490.0690.680.311.48200.332CCC0.510.410.622550.0001.631.242.16650.0011.541.002.36270.0500.620.430.891780.01MC0.350.230.53860.0002.951.585.52110.00113.826.4129.7970.0000.480.260.88720.017BoldvalueshighlightstheadjustedstatisticallysignificantdifferencewiththereferencecategoryEOCepithelialovariancancer,nnumberofpatientswithfollow-up,CIconfidenceinterval,lowstage:patientswithstageIandII,advancedstage:patientswithstageIIIandIV,HGSCHigh-GradeSerousCarcinoma,ECEndometrioidCarcinoma,LGSCLow-GradeSerousCarcinomaLe Page et al. BMC Cancer  (2018) 18:347 Page 8 of 18longer significant (21% and 33% respectively, p = 0.575;Table 4 and Fig. 1c). CCC and MC remained as the mostdeadly diseases in this category (Fig. 1c and Table 4).However, the MC cases were too few (n = 7) and theanalysis of the MC “poor prognosis” cases should beconsidered with caution.Among low stage diseases (stage I-II), there are no sig-nificant survival differences between HGSC and LGSC(p = 0.332, Table 4). HGSC and LGSC histotypes werethe most deadly (62% and 65% survival, respectively, Fig.1d) and EC remained the histopathology with the bestprognostic outcome with a median survival of 15 yearsand a rate of 93% at 5-years (Fig. 1d).The prognostic value of cytoreduction alone as a riskfactor for disease specific survival was more pronouncedin non-serous cases (Table 5 and Fig. 2). Residual miliarytumor was predominantly observed in serous cases(80%) and categorized as a suboptimal cytoreduction.The survival rate for EC cases with no RD was 92%compared to 25% for suboptimal cytoreduction, whereasthe CCC cases with suboptimal cytoreduction showedonly 7% survival rate compared to 72% with no RD. Bycontrast, the survival of patients in LGSC cases with noRD versus those with cytoreduction to ≤1 cm was sur-prisingly similar (Fig. 2d). A significant difference wasonly seen between LGSC patients with optimal and sub-optimal cytoreduction (HR = 2.40 (1.08–5.36, 95% CI), p= 0.033, Table 5). The MC cases with RD were too rarefor a rigorous analysis and the only 4 cases with subopti-mal RD died within a year.Cohort characteristics—BRCAAs 20–25% of HGSC are associated with BRCA1/BRCA2germline and somatic mutations [16, 17] we requestedinformation from each participating site on mutationstatus of these genes. This variable was not available forthe majority of patients, particularly in non-serous casessince mutation testing is mostly undertaken when a her-editary factor was suspected. In addition, BRCA muta-tion testing is sometimes targeted to only a fewmutation sites found to recur in specific populations,and thus not always fully sequenced or assessed for lon-ger deletions. For this reason, cases identified as a BRCAnon-carrier genotype in our cohort were likely over-estimated. Nonetheless, we collected 112 cases of knownmutation BRCA mutation carriers and 476 nominallynon-carrier cases (Table 2). The proportion of BRCAmutation carriers was 19% of all tested cases, and 22% ofHGSC. Indeed, 96% of BRCA mutations (n = 108) weredetected in HGSC, while four cases were classified asLGSC.As expected, the age of HGSC patients with BRCAmutation was, on average, lower than the age of non-carrier HGSC patients (53 and 60 years, respectively, p< 0001, Mann-Whitney U test). Patients with a BRCA2mutation also presented more often with advanced dis-ease compared to patients with a BRCA1 mutation (69%compared to 80%, p = 0.012, Chi-Square) as previouslyobserved by others in larger cohorts [18, 19]. In HGSCcases treated with a platinum-based regimen, we ob-served a significant association between the presence ofany BRCA mutation and survival, with a median survivalof 96 months in the mutated group as compared to61 months in the non-carrier group (HR = 0.56(0.42–0.76 95%CI), p < 0.001, Table 6; Fig. 3a). When stratified toadvanced stage disease, the BRCA mutation carrier sta-tus remained a significant prognostic variable (HR = 0.57(%95 CI 0.41–0.80), p = 0.001, Table 6 and Fig. 3b) butnot when stratified to low stage disease (p = 0.82,Table 6). The improved outcome of mutation carrierwomen was slightly higher in patients with mutatedBRCA2 than with mutated BRCA1 (Fig. 3a and Table 6).This trend was even stronger in “poor prognosis” cases,where only BRCA2 carriers remained of significant sur-vival advantage (HR = 0.39 (%95 CI 0.20–0.77), p = 0.006) with a median survival of 77 months compared to58 months for BRCA1 carriers and 45 months in thenon-carrier group.Interestingly, we also observed a better long-term sur-vival rate (> 10 years) among BRCA1 or − 2 mutationcarriers compared to the non-carrier group (21% vs 11%,OR = 0.432, 95%CI (0.21–0.90), p = 0.024, n = 319). Inadvanced stages patients, BRCA1 or BRCA2 mutationwas still associated with long-term survival (22% vs 10%,OR = 0.41, 95% CI (0.18–5.78), p = 0.94, n = 283) but nomore among poor prognosis patients (p = 0.27).Cohort characteristics—Chemotherapeutic treatmentsWe compared the disease-specific survival rates amongthe most common types of chemotherapeutic regimens.Only a few patients were treated with a regimen otherthan platinum as their first adjuvant chemotherapy(Table 2) and, therefore, were excluded from this ana-lysis. Treatment regimens that were analyzed includedcarboplatinum/taxane, cisplatinium/taxane, platinumalone or platinum + another non-taxane agent. HGSCpatients subjected to both platinum plus taxane treat-ments had better survival compared to patients treatedwith platinum alone or platinum + another agent (p < 0.001 and Table 6). In “poor-prognosis” HGSC cases, plat-inum + another agent regimen was no more significantlyless effective than a platinum/taxane regimen (p = 0.093,Table 6 and Fig. 3c).Treatment resistance in patients with RD was definedas progression within 6 months of the end of first treat-ment, that is following surgery and treated with achemotherapy agent as per the standard clinical defin-ition [20]. We restricted the analysis to platinum-basedLe Page et al. BMC Cancer  (2018) 18:347 Page 9 of 18Table5UnivariateCoxregressionmodelofclinicalparametersondiseasespecificsurvivalindifferentEOCResidualdiseaseHGSCECCCCLGSCHazard95.0%CInHazard95.0%CInHazard95.0%CInHazard95.0%CInRatioLowerUpperpvalueRatioLowerUpperpvalueRatioLowerUpperpvalueRatioLowerUpperpvalueNone206reference119reference93reference16reference<1cm1.721.312.232300.0003.801.638.84290.0022.241.174.29260.0151.100.452.72210.835Suboptimal2.902.283.703470.00015.035.7239.47100.0008.304.3815.76170.0002.401.085.36270.033StageI70reference140reference124reference13referenceII1.250.772.031360.3761.910.804.61730.1472.361.324.24540.0041.320.295.9170.719III3.172.094.818200.0006.483.0413.82440.0008.154.9313.48580.0002.730.987.63680.056IV4.612.937.261200.000NANANA5NA19.357.5749.4370.0007.111.8727.0060.004BoldvalueshighlightsthestatisticallysignificantdifferencewiththereferencecategoryNAnonapplicablenumberofcasestoolow,nnumberofpatientswithfollow-up,HRHazardratio,EOCepithelialovariancancer,CIconfidenceinterval,HGSCHigh-GradeSerousCarcinoma,ECEndometrioidCarcinomaLe Page et al. BMC Cancer  (2018) 18:347 Page 10 of 18treated patients because of the limited number of non-platinum treated cases. There was no difference intreatment response rate between LGSC and EC(Table 2). Only CCC showed a significantly lower re-sponse to treatment compared to type other histotypes(OR = 2.73 (95%CI 1.42–5.26), p = 0.003). Treatment re-sistance could not be accurately analyzed in the MChistotype (n = 7). The small number of non-serous caseswith RD limited this analysis, and interpretation of ourfindings will require a larger cohort.Among the known prognostic factors, BRCA1 orBRCA2 mutation carrier status was a favorable factorrelative the non-carrier status for response to treatmentin HGSC cases treated with surgery and platinum-based chemotherapy. We observed that 35% of caseshad progressed within 6 months in the non-carriergroup (n = 248) compared to 21% in the BRCA muta-tion carrier group (n = 59) (OR = 0.44, 95% CI 0.22–0.89, p = 0.022, Table 7). A similar result was obtained inevaluating advanced stage patients although less signifi-cant (36% vs 16%, OR = 0.49, (95%CI 0.24–1.00), p = 0.042, Table 7). When BRCA mutation was stratified inBRCA1 or BRCA2 mutation status, BRCA2 alone wasno more associated with better response to treatment(Table 7). However, due to the small number of BRCA2carriers with RD, this result should be confirmed in alarger cohort.Participating biobanks of the COEUR cohortTwelve Canadian biobanks representing the five mostpopulated provinces contributed samples to the COEURrepository (Table 8). Due to sample availability in eachbiobank, the five Canadian provinces are not equallyrepresented in the COEUR cohort. The cohort may notnecessarily represent the Canadian demography of eachprovince. For example, Quebec cases represent almosthalf of the entire cohort, whereas cases from the mostpopulous Canadian province, Ontario, represent only16% of the cohort. Also, the relatively new establishmentof the Manitoba Ovarian Biobank Program (MOBP) in2011 resulted in fewer cases from this province.The distribution of death risk factors (age, histotype,response to treatment, BRCA mutation status) amongthe different contributory banks for each case byFig. 2 Kaplan-Meier survival curves of disease-specific survival stratified by residual disease. a Comparison of high-grade serous carcinoma (HGSC)with different amounts of RD (none = 206, < 1 cm = 230, suboptimal = 347). b Comparison of endometrioid carcinoma (EC) with different amountsof RD. (none = 119, < 1 cm = 29, suboptimal = 10) (c). Comparison of clear cell carcinoma (CCC) with different amounts of RD (none = 93, < 1 cm= 26, suboptimal = 17). d Comparison of low-grade serous carcinoma (LGSC) with different amounts of RD (none = 16, < 1 cm = 21, suboptimal =27). Significance (p) is indicated by log-rankLe Page et al. BMC Cancer  (2018) 18:347 Page 11 of 18Table6UnivariateCoxregressionmodelofclinicalparametersondiseasespecificsurvivalinstratifiedhigh-gradeserouscarcinoma(HGSC)AllHGSCAdvancedstageHGSCPoorprognosisHGSCLowStageHGSCHazard95.0%CIHazard95.0%CIHazard95.0%CIHazard95.0%CIRatioLowerUppernpvalueRatioLowerUppernpvalueRatioLowerUppernpvalueRatioLowerUppernpvalueBRCAmutationcarrierstatusaNoncarrier351reference293reference236reference61referenceBRCA1carrier0.620.440.88670.0070.650.440.95450.0290.790.511.20320.2680.920.402.11210.849BRCA2carrier0.420.240.72340.0020.410.220.76260.0040.390.200.77210.0060.800.193.4470.766Allmutatedvsnomutation0.560.420.764720.0000.570.410.803750.0010.640.440.932940.020.820.371.81850.78ChemotherapytreatmenttypeCarboplatinum+taxane940reference727reference469reference167referenceCisplatinum+taxane1.140.871.50710.3480.630.440.89670.7281.110.831.49600.495NANANA4NAPlatinum1.861.372.52630.0000.420.240.73460.0002.651.833.83380.0001.650.713.84130.245Platinum+other1.931.352.75380.0000.560.420.76330.0051.440.942.19280.093NANANA3NAProvincesQuebec584reference472reference275referenceLowerUpperNAOntario0.800.660.981950.0340.690.540.881190.0030.770.581.03890.074NANANANANAAlberta0.710.560.911200.0070.730.560.96880.0250.510.340.78410.002NANANANANAManitoba0.460.151.45100.1850.200.031.4570.111NANANA6NANANANANANABritishColumbia0.990.841.173070.9171.020.861.212480.8411.210.991.472100.064NANANANANABoldvalueshighlightsthestatisticallysignificantdifferencewiththereferencecategoryNAnonapplicable,numberofcasestoolow,nnumberofpatientswithfollow-up,HRHazardratioaplatinum-basedpatientsonly,lowstage:diseasestageIandII,advancedstage:diseasestageIIIandIVLe Page et al. BMC Cancer  (2018) 18:347 Page 12 of 18Table 7 Univariate logistic regression model of clinical parameters on treatment response in stratified high—grade serouscarcinoma (HGSC)aAll HGSC Advanced stage HGSCHazard 95.0% CI Hazard 95.0% CIRatio Lower Upper n p value Ratio Lower Upper n p valueBRCA mutation carrier statusNon carrier 247 reference 228 referenceBRCA1 carrier 0.29 0.11 0.77 38 0.013 0.34 0.13 0.91 32 0.031BRCA2 carrier 0.72 0.27 1.89 22 0.499 0.73 0.27 1.94 21 0.524All mutated vs no mutation 0.44 0.22 0.89 307 0.022 0.49 0.24 1.00 281 0.042ProvincesQuebec 256 reference 239 referenceOntario 0.47 0.27 0.82 93 0.008 0.40 0.21 0.74 80 0.00Alberta 0.22 0.08 0.58 46 0.002 0.19 0.07 0.55 40 0.002Manitoba NA NA NA 8 0.999 NA NA NA 5 NABritish Columbia 0.95 0.63 1.44 160 0.808 0.92 0.60 1.40 155 0.689Bold values highlights the statistically significant difference with the reference categorya platinum–based patients onlya bc dFig. 3 Kaplan-Meier survival curves of disease-specific survival stratified by BRCA status (a-d). a Comparison of HGSC patients by BRCA mutation. Mutationnon-carriers (black, n = 351), BRCA1 mutation positive (dark grey, n = 67) and BRCA2 mutation positive (dashed light grey, n = 34). b Comparison of HGSCpatients by BRCA mutation in advanced stage disease. Mutation non-carriers (black, n = 293), BRCA1 mutation positive (dark grey, n = 45) and BRCA2mutation positive (dashed light grey, n = 26). c Kaplan-Meier survival curves of disease-specific survival stratified by chemotreatment type. Comparison ofpoor prognosis HGSC patients by chemo-treatment: carboplatinum/taxane (blue, n = 469); cisplatinum/taxane (green, n = 60); platinum alone (gold,n = 38); platinum + another agent (violet, n = 28). d Kaplan-Meier survival curves of disease-specific survival stratified by source site. Comparison ofprovincial origins of poor prognosis HGSC patients in the COEUR cohort (Quebec = 275, Ontario = 89, Alberta n = 41, British Columbia = 210). Significance(p) is indicated by log-rankLe Page et al. BMC Cancer  (2018) 18:347 Page 13 of 18province was heterogeneous (Table 8). In Quebec, we alsonoticed more cases treated with platinum and another agentas an alternative to platinum + taxane, while other provincesprovided more cases treated with platinum as a single agent.To date, the literature has no reports on the ovarian can-cer survival rate among Canadian provinces compared withthe province of Quebec due to a lack of a Quebec-basedregistry. Moreover, current statistics often do not considerpatient group according to metric such as stage, cytoreduc-tion, and histopathology, when surveying survival of ovariancancer patients. To compare EOC death rates among prov-inces, we limited our analysis to cases with either advancedstage HGSC or the “poor prognosis” patients. Manitoba wasnot included due to the small number of cases (n = 7 foradvanced HGSC patients). The 5-year and overall survivalwas better in Ontario and Alberta compared to Quebec(OR = 0.56 (95% CI 0.36–0.87) p = 0.010, and Table 8),although less significant when Ontario were restricted to“poor prognosis” patients (p = 0.074, Table 6 and Fig. 3d).DiscussionThe COEUR has enabled the assembly of many retro-spective cases (> 2000) in a relatively short period oftime (7 years). This network has ensured thestandardization of the clinical data after collection anda long-term follow-up. Moreover, to facilitate this en-deavor, a single material transfer agreement was estab-lished and signed by all institutions, which provided aconsistent legal and ethics policy for subsequent usersTable 8 Distribution and clinical characteristcs of the COEUR cases per Canadian provinceLocation of tumor bank Quebec Ontario Manitoba Alberta British Columbia TotalTotal cases 976 330 242 17 479 2046Histotype BOT 9 4 7 0 3 23HGSC 603 199 120 10 314 1246EC 132 56 31 3 73 295LGSC 52 24 19 0 8 103CCC 123 37 36 1 62 259MC 43 1 28 2 14 88other 14 8 1 1 4 29Age (mean years)a at diagnosis 63 60 61 61 62 62at end of follow-up 66 64 65 66 66 65Year of diagnosis (range) 1992–2015 1996–2013 2012–2015 1998–2015 1998–2013 1992–2015Follow-up (mean, months) 54(0–270) 51(0–168) 32(9–49) 49(0–181) 47(0–173) 48(0–270)Survival ratea Disease specific survival 30% 41% 73% 36% 29% 33%5-years survival rate 31% 40% NA 42% 33% 34%Long Term Survival rate[> 10 years]9% 7% NA 12% 6% 8%Treatment type none 47 5 55 0 23 130carboplatin+taxol 690 238 160 15 351 1454cisplatin+taxol 40 28 9 0 20 97platinum 21 21 15 0 45 102platinum+other 47 2 1 0 6 56taxol 3 0 0 0 0 3other 15 6 0 0 1 22Total 863 300 240 15 446 1864% response to treatmenta 69% 84% 90% 91% 71% 75%BRCA mutation carrier status non carriers 306 27 6 9 137 485carriers 58 16 2 4 33 113total 370 33 8 13 170 594% carriers 16% 48% 25% 31% 19% 19%NA non applicable, BOT borderline tumor, HGSC High-Grade Serous Carcinoma, EC Endometrioid Carcinoma, LGSC Low-Grade Serous Carcinoma, CCC Clear CellCarcinoma, MC Mucinous Carcinomaa HGSC patients onlyLe Page et al. BMC Cancer  (2018) 18:347 Page 14 of 18to access the centralized collection of specimens andclinical data. To date 25 national and internationalstudies got access to the cohort [5, 8, 21–23]. The co-hort contains more than 100 LGSC and 80 MC cases,the rarest of all EOC histotypes. Although these num-bers are small, most biomarker studies analyzing thesehistotypes have rarely described a cohort of this sizecontaining the least common histotypes, particularlyafter contemporary histopathology review. Indeed, ofsignificant value is the two-step central pathology re-view of all FFPE cases with application of current diag-nostic immunohistochemical markers adding anobjective molecular layer. This rigorous review of allcases by the same pathologists standardized and cor-rected the inaccuracies of previous or out-dated path-ology records and resulted in reclassifying a significantnumber of EC cases to HGSC, supported by improvedoutcome stratification. A significant proportion of MCcases were also reclassified to EC or metastatic disease,which in turn highlighted the rarity of advanced stageMC cases. The possibility of contamination of the co-hort by cases with metastatic intestinal adenocarcin-omas from the colon and appendix amongst presumedMC cases, was minimized by immunohistochemistryphenotyping. This qualitative revision of cases ensuresa high standard for a shared research resource.The COEUR cohort has some limitations. One draw-back is the collection of cases based on specimen avail-ability several years after initial collection, which maynot represent the overall biobank population, nor eventhe provincial ovarian cancer population. Another limi-tation of the COEUR cohort, even if it represents thedemographic situation of uncommon histotypes, is theunder-representation of rare histotypes at advancedstages. This will limit the ability to reach sufficient statis-tical power in biomarker studies. For example, we werenot able to analyze the cytoreduction impact on survivalin MC cases or the treatment resistance in patients withRD. An alternative to this limitation could be the spe-cific collection of rare cases. Another important point isthat only one sample per patient was collected, indicat-ing that review of the complete case was not possible forthe majority of cases. Another disadvantage of a multi-center repository is the different standards applied forsubjective clinico-pathological parameters, such as dis-ease stage, cytoreduction amount, and treatment regi-men. For example, some institutions did not includerising CA125 levels as an indicator of disease progres-sion, and perhaps allowed a less accurate progressiontime and less regular follow-up than other institutions.In this report, we have described the clinical character-istics of the COEUR cohort, which show similarities tothe characteristics of the general ovarian cancer popula-tion as reported in the literature. The proportion ofnon-serous histotypes with advanced stage disease waslower than other histotypes (< 25%), particularly withMC cases, as previously reported [15, 24]. The 5-yearsurvival and the long-term survival rates of each histo-type are in accordance with the rates described in the lit-erature, particularly when histotypes were diagnosedwith the current histological criteria [2, 8, 13, 14].Different clinical outcomes were observed among his-totypes; when not adjusted to prognostic clinicopara-meters such as stage and cytoreduction, HGSC casesdemonstrate the worst outcome compared to the otherhistotypes. This data is particularly noteworthy as indi-vidual biomarker studies, that have been limited in stat-istical power due to low incidence of non-serous EOC,are prone to consider risk factors and prognostic bio-markers for all ovarian tumors as one group. Sincemany biomarkers have been associated with specifichistotypes, studies evaluating outcome in cohorts com-prising all histotypes are likely confounded by inherentdifferences in the outcomes of specific histotype ratherthan the biomarker itself [2]. Furthermore, and as seenhere, stratification of histotypes by stage and amount ofRD should also be considered in biomarker prognosticstudies (Fig. 1b-h). Indeed, when adjusted for advancedstage or suboptimal cytoreduction, MC and CCC ratherthan HGSC appears to be the deadliest disease with25% survival consistent with independent reports [13,25–28]. Although EC is reported to have a similar sur-vival outcome in advanced disease as compared toHGSC [11], an analysis of the COEUR cohort revealeda strong survival advantage for EC and this observationcould be attributed to the reclassification of some ECto HGSC cases.The COEUR cohort has a good representation of thefive major histotypes without major bias, and hence, isa useful resource for biomarker validation. Indeed,some studies have already used these specimens to val-idate a diagnostic biomarker algorithm and have usedsamples as an inter-laboratory quality control for suchmarkers, which are now routinely used in Canadianand British pathology laboratories [8, 29]. In one study,COEUR samples were used to optimize p53 immuno-histochemistry and achieved a 97% accuracy in predict-ing TP53 mutation status in ovarian carcinomas, whichis now widely adopted in pathology departments [21].The COEUR cohort has also validated markers of prog-nostic value that are currently under further investiga-tion for their clinical utility [22].Due to its role in the repair of double strand DNAbreaks by homologous recombination, patients harboringa BRCA mutation and consequently a defect in homolo-gous recombination in their tumors are more sensitive toDNA crosslinking caused by DNA-damaging-platinum-based chemotherapeutic treatments. In the COEURLe Page et al. BMC Cancer  (2018) 18:347 Page 15 of 18cohort, all of the BRCA mutations were exclusively re-ported in serous cases. After histotype reclassification anddata monitoring, we did not observe any BRCA mutationsin non-serous cases, although these diseases are rarelytested for familial risk by DNA sequencing, HGSC havebeen shown to be enriched for BRCA1/2 carriers [30]. Toconfirm the paucity of BRCA mutations in other non-HGSC histotypes in the COEUR cohort, further geneticstests are necessary for the remaining cases not sequenced.Borderline tumors and LGSC tumors are usually notassociated with BRCA familial risk except in rare cases[31–33]. These cases have often been identified in pa-tients of young age (≤50 yrs) within the Jewish popula-tion and show aggressive characteristics [32, 33]. In theCOEUR cohort, among the four LGSC cases withBRCA mutation, three cases underwent the two-steppathology review and the LGSC diagnosis was con-firmed. The last case was originally diagnosed as HGSCbut had only one FFPE specimen sent to the COEUR.Therefore, we were not able to review and confirm theLGSC diagnosis of this patient. Since here we collectedonly data for the BRCA germline mutation carrier sta-tus but not details about the mutation type, we are un-certain whether the same mutations are present in thetumor tissue or whether an unrelated histotype evolvedin these BRCA mutation carriers. Lastly, we do notknow the specific significance of these BRCA mutationsand whether they unambiguously represent a pathogno-monic mutation. Our data suggest that BRCA muta-tions are largely restricted to HGSC consistent withrecent independent reports [30, 34].The presence of a germline and/or somatic BRCA mu-tation has been recognized as a good prognostic factor[16, 18, 19, 35–37] in small cohorts and confirmed byBolton et al. in a pooled analysis of 909 BRCA1 and 304BRCA2 patients [18]. Our data are consistent with thesereports (Fig. 2). Yang et al. have also reported a bettersurvival for 29 BRCA2 cases as compared to 37 BRCA1cases found in the TCGA database and attributed this toa longer treatment-free interval after chemotherapytreatment [38]. We also observed a better overall andlong-term survival of advanced stage BRCA mutationcarriers, particularly BRCA2, although this favourableoutcome was not associated with a better treatment re-sponse within 6 months when BRCA1 and BRCA2 pa-tients were compared. This is in contrast to a study byMcLaughlin et al. [39] that reported no differences inlong-term survival in BRCA mutation carriers. Notably,this independent report included all ovarian carcinomashistotypes, and the 5-year survival rate of the non-carrier group was greater than 50% [39]. This may haveconfounded non-HGSC outcomes, particularly LGSC orEC, which are largely unrelated to BRCA mutations. Thepresent study is the first time where BRCA mutationwas estimated in stratified groups of homogeneous prog-nostic factor in HGSC patients after double centralpathological revision, confirming previous data showinga 5-yrs survival advantage of BRCA2 patients [18, 40].Our data show the improved long-term outcome ofHGSC patients bearing germ-line BRCA mutationtreated with platinum-based chemotherapy relative toHGSC non-carriers, although this advantage restrictedto advanced stage disease patients. These data have po-tential implications and underline the need for BRCAtesting for the clinical management and personalizedtreatment of advanced stage HGSC patients.Finally, we observed a better disease-specific survivalrate in patients from Ontario and Alberta, compared toother provinces, similarly to the pattern observed fromcancer-registries in 4 Canadian provinces [41]. In fact, astudy in Canadian showed that geographic location ofpatients could impact outcomes of ovarian cancer dudifferences in patient management and treatment [42].Altogether, these observations also highlight that bio-marker studies using the COEUR cohort should includeverification if the value of biomarker is not biased to-wards case source. This is particularly important inTMA studies since pathology departments may applydifferent standards, and tissue preservation over timemay affect biomarker stability.ConclusionsOur study of a well-defined ovarian cancer cohortshows the tremendous importance of studying EOC byhistologic histotype and the strength of a two-stepcentral pathological review, allowing for observationof significant differences of prognostic factors and bio-markers in a relatively small sample size as seen withBRCA patients. In addition, the heterogeneity of riskfactors in each histotype, such as disease stage, cytore-duction, BRCA mutation carrier status and samplesource, highlights the importance of stratifying oradjusting for these factors when evaluating the prog-nostic value of biomarkers. COEUR provides a valid-ation platform for biomarker studies on a variety ofbiological materials with contemporary histotype in-formation and high quality clinical data for the na-tional and international ovarian cancer researchercommunity.AbbreviationsCCC: Clear cell carcinoma; COEUR: Canadian ovarian experimental unifiedresource; CTRnet: Canadian tissue repository network; EC: Endometrioidcarcinoma; EOC: Epithelial ovarian cancer; FFPE: Formalin-fixed paraffin-embedded; GOC: Society of gynecologic oncology of Canada; HGSC: High-grade serous carcinoma; HR: Hazard ratio; LGSC: Low-grade serouscarcinoma; LTS: Long-term survival; MC: Mucinous carcinoma; OR: Odds ratio;RD: Residual disease; TMA: Tissue microarrayLe Page et al. BMC Cancer  (2018) 18:347 Page 16 of 18AcknowledgementsThe authors are very grateful to Jacqueline Chung for critical editing of themanuscript and grateful to Pascal Lambert, Paul Bodson and Janusz Kaczorowskifor the critical review of statistical methods and thoughtful discussions. We thankthe patients, the Gynecology-Oncology and Pathology services of all participatingcenters for tumor procurement. We are also grateful to all biobank members,particularly Adrian Driga, Sindy Badinsky, Michelle Parisien, Elizabeth McDonald,Magdalena Bachvarov, Sarah Rachel Katz, Ceola Weir, Marie-Claude Beauchamp,Amber Yasmin, David Octeau, Monique Albert, Ying Ng, Isabelle Matte, ManondeLadurantaye and Véronique Barrés, for tissue processing and clinical data collec-tion. Tumor banking was supported by each provincial organization described inTable 1. Clinical specimens from the province of Quebec were provided by theBanque de tissus et de données of the Réseau de recherche sur le cancer of theFonds de recherche du Québec–Santé.FundingThis work was supported by a grant from the Terry Fox Research Institute#2009–15.Availability of data and materialsThe datasets analyzed during the current study is available from the correspondingauthor. Requests for access is subject to review by the COEUR Study Committee.Authors’ contributionsKR and MK did the central pathology revision. CLP, LP, MB and LM were involvedin the processing of samples. CLP acquired the data, performed the statisticalanalyses and drafted the manuscript. KR and MK made substantial contributionsto conception and design of the study. MQB, JB, DB, WG, JNM, BG, MWN, AP, PS,PHW, BV, DGH, DMP and AMMM provided tumor samples and clinical data. TheCOEUR Study Committee and the Management Committee (MK, KR, PNT, BHN,JB, BV and DMP, DH and AMMM) did the strategic planning and conception ofthe cohort. DMP and MQB offered support for the clinical data monitoring. Allauthors were involved in critically revising the manuscript. In addition, all authorsparticipated in interpretation of data, revised the manuscript for intellectualcontent, have given final approval of the manuscript to be published, andagreed to be accountable for all aspect of the work.Ethics approval and consent to participateThe biobanks received ethics approval from their local review boards to collectand share samples and clinical data. All subjects gave broad written consent tofuture research with their samples and data, without restriction. Additionally,the central activities of the CRCHUM, which included collection of the COEURrepository samples and data, received local ethics approval by the ‘Comitéd’éthique de la recherche du CHUM (project reference: 2010–3552, CE 09.141 -BSP (CA)).Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.Author details1Centre de recherche du Centre hospitalier de l’Université de Montréal(CRCHUM) and Institut du cancer de Montréal, Montreal, QC, Canada.2Department of Pathology du Centre hospitalier de l’Université de Montréal,Montreal, QC, Canada. 3Department of Pathology and Laboratory ofMedicine, University of Calgary, Calgary, AB, Canada. 4Departments ofMedicine and Human Genetics, McGill University; Cancer Research Program,The Research Institute of the McGill University Health Centre, Montreal, QC,Canada. 5Tumour Tissue Repository, Trev and Joyce Deeley Research Centre,BC Cancer Agency, Victoria, BC, Canada. 6Department of Obstetrics andGynaecology, University of Toronto, Toronto, ON, Canada. 7DiagnosticDevelopment and Ontario Tumour Bank, Ontario Institute for CancerResearch, Toronto, ON, Canada. 8Centre de recherche du CHU de Quebec,Quebec, QC, Canada. 9Laboratory of Gynecologic Oncology, Lady DavisResearch Institute, Jewish General Hospital, Montreal, QC, Canada.10Department of Pathology, Vancouver General Hospital and University ofBritish Columbia, Vancouver, BC, Canada. 11Department of Obstetrics andGynaecology, University of British Columbia, Vancouver, BC, Canada.12Department of Biochemistry & Medical Genetics, University of Manitoba,Winnipeg, MB, Canada. 13Centre de Recherche du CHUS. Département deMicrobiologie et Infectiologie, Faculté de Médecine, Université deSherbrooke, Sherbrooke, Canada. 14Cancer Therapeutics Program, OttawaHospital Research Institute, Ottawa, Canada. 15Department of Cellular andMolecular Medicine, University of Ottawa, Ottawa, Canada. 16Division ofGynecologic Oncology, Université de Montréal, Montreal, Canada.17Department of Medicine, Université de Montréal, Montreal, Canada. 18900rue Saint Denis, Tour R, Montreal H2X2A0, Canada.Received: 6 October 2017 Accepted: 16 March 2018References1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67(1):7–30.2. Kobel M, Kalloger SE, Boyd N, McKinney S, Mehl E, Palmer C, Leung S, BowenNJ, Ionescu DN, Rajput A, et al. Ovarian carcinoma subtypes are differentdiseases: implications for biomarker studies. PLoS Med. 2008;5(12):e232.3. Ramalingam P. Morphologic, Immunophenotypic, and molecular features ofepithelial ovarian Cancer. Oncology (Williston Park). 2016;30(2):166–76.4. Grzankowski KS, Carney M. Quality of life in ovarian cancer. Cancer Control.2011;18(1):52–8.5. Le Page C, Kobel M, de Ladurantaye M, Rahimi K, Madore J, Babinszky S,Bachvarov DR, Bachvarova M, Beauchamp MC, Cass CE, et al. Specimenquality evaluation in Canadian biobanks participating in the COEURrepository. Biopreservation Biobanking. 2013;11(2):83–93.6. Tewari KS. Advanced cytoreductive surgery workshop report. Int J GynecolCancer. 2012;22(9):1604–10.7. Rustin GJ, Vergote I, Eisenhauer E, Pujade-Lauraine E, Quinn M, Thigpen T,du Bois A, Kristensen G, Jakobsen A, Sagae S, et al. Definitions for responseand progression in ovarian cancer clinical trials incorporating RECIST 1.1 andCA 125 agreed by the gynecological Cancer Intergroup (GCIG). Int JGynecol Cancer. 2011;21(2):419–23.8. Kobel M, Rahimi K, Rambau PF, Naugler C, Le Page C, Meunier L, deLadurantaye M, Lee S, Leung S, Goode EL, et al. An Immunohistochemicalalgorithm for ovarian carcinoma typing. Int J Gynecol Pathol. 2016;35(5):430–41.9. Heintz AP, Odicino F, Maisonneuve P, Quinn MA, Benedet JL, Creasman WT,Ngan HY, Pecorelli S, Beller U. Carcinoma of the ovary. FIGO 26th annualreport on the results of treatment in gynecological Cancer. Int J GynaecolObstet. 2006;95(Suppl 1):S161–92.10. Moh M, Krings G, Ates D, Aysal A, Kim GE, Rabban JT. SATB2 expressiondistinguishes ovarian metastases of colorectal and Appendiceal origin fromprimary ovarian tumors of mucinous or Endometrioid type. Am J SurgPathol. 2016;40(3):419–32.11. Rosen DG, Zhang Z, Shan W, Liu J. Morphological and molecular basis ofovarian serous carcinoma. J Biomed Res. 2010;24(4):257–63.12. Kobel M, Kalloger SE, Lee S, Duggan MA, Kelemen LE, Prentice L, Kalli KR,Fridley BL, Visscher DW, Keeney GL, et al. Biomarker-based ovariancarcinoma typing: a histologic investigation in the ovarian tumor tissueanalysis consortium. Cancer Epidemiol Biomark Prev. 2013;22(10):1677–86.13. 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):993–9.14. Steffensen KD, Waldstrom M, Grove A, Lund B, Pallisgard N, Jakobsen A.Improved classification of epithelial ovarian cancer: results of 3 danishcohorts. Int J Gynecol Cancer. 2011;21(9):1592–600.15. Kobel M, Kalloger SE, Huntsman DG, Santos JL, Swenerton KD, Seidman JD,Gilks CB, Cheryl Brown Ovarian Cancer Outcomes Unit of the British ColumbiaCancer Agency VBC. Differences in tumor type in low-stage versus high-stageovarian carcinomas. Int J Gynecol Pathol. 2010;29(3):203–11.16. Pal T, Permuth-Wey J, Betts JA, Krischer JP, Fiorica J, Arango H, LaPolla J,Hoffman M, Martino MA, Wakeley K, et al. BRCA1 and BRCA2 mutationsaccount for a large proportion of ovarian carcinoma cases. Cancer. 2005;104(12):2807–16.17. Kanchi KL, Johnson KJ, Lu C, MD ML, Leiserson MD, Wendl MC, Zhang Q,Koboldt DC, Xie M, Kandoth C, et al. Integrated analysis of germline andsomatic variants in ovarian cancer. Nat Commun. 2014;5:3156.18. Bolton KL, Chenevix-Trench G, Goh C, Sadetzki S, Ramus SJ, Karlan BY,Lambrechts D, Despierre E, Barrowdale D, McGuffog L, et al. Associationbetween BRCA1 and BRCA2 mutations and survival in women with invasiveepithelial ovarian cancer. JAMA. 2012;307(4):382–90.Le Page et al. BMC Cancer  (2018) 18:347 Page 17 of 1819. Alsop K, Fereday S, Meldrum C, deFazio A, Emmanuel C, George J, DobrovicA, Birrer MJ, Webb PM, Stewart C, et al. BRCA mutation frequency andpatterns of treatment response in BRCA mutation-positive women withovarian cancer: a report from the Australian ovarian Cancer study group. JClin Oncol. 2012;30(21):2654–63.20. Eisenhauer EA, Vermorken JB, van Glabbeke M. Predictors of response tosubsequent chemotherapy in platinum pretreated ovarian cancer: a multivariateanalysis of 704 patients [seecomments]. Ann Oncol. 1997;8(10):963–8.21. Kobel M, Piskorz AM, Lee S, Lui S, LePage C, Marass F, Rosenfeld N, MesMasson AM, Brenton JD. Optimized p53 immunohistochemistry is anaccurate predictor of TP53 mutation in ovarian carcinoma. J pathol Clin Res.2016;2(4):247–58.22. Au KK, Le Page C, Ren R, Meunier L, Clement I, Tyrishkin K, Peterson N,Kendall-Dupont J, Childs T, Francis JA, et al. STAT1-associated intratumouralTH1 immunity predicts chemotherapy resistance in high-grade serousovarian cancer. J Pathol Clin Res. 2016;2(4):259–70.23. Lee S, Piskorz AM, Le Page C, Mes Masson AM, Provencher D, Huntsman D,Chen W, Swanson PE, Gilks CB, Kobel M. Calibration and optimization ofp53, WT1, and Napsin a immunohistochemistry ancillary tests forHistotyping of ovarian carcinoma: Canadian immunohistochemistry qualitycontrol (CIQC) experience. Int J Gynecol Pathol. 2016;35(3):209–21.24. Shimada M, Kigawa J, Ohishi Y, Yasuda M, Suzuki M, Hiura M, Nishimura R,Tabata T, Sugiyama T, Kaku T. Clinicopathological characteristics ofmucinous adenocarcinoma of the ovary. Gynecol Oncol. 2009;113(3):331–4.25. Karabuk E, Kose MF, Hizli D, Taskin S, Karadag B, Turan T, Boran N, Ozfuttu A,Ortac UF. Comparison of advanced stage mucinous epithelial ovarian cancer andserous epithelial ovarian cancer with regard to chemosensitivity and survivaloutcome: a matched case-control study. J Gynecol Oncol. 2013;24(2):160–6.26. Brown J, Frumovitz M. Mucinous tumors of the ovary: current thoughts ondiagnosis and management. Curr Oncol Rep. 2014;16(6):389.27. Simons M, Massuger L, Bruls J, Bulten J, Teerenstra S, Nagtegaal I. Relativelypoor survival of mucinous ovarian carcinoma in advanced stage: asystematic review and meta-analysis. Int J Gynecol Cancer. 2017;27(4):651–8.28. Tammela J, Geisler JP, Eskew PN Jr, Geisler HE. Clear cell carcinoma of theovary: poor prognosis compared to serous carcinoma. Eur J GynaecolOncol. 1998;19(5):438–40.29. Gilks CB, Kommoss F. Ovarian carcinoma Histotypes: their emergence asimportant prognostic and predictive markers. Oncology (Williston Park).2016;30(2):178–9.30. Tonin PN, Maugard CM, Perret C, Mes-Masson AM, Provencher DM. Areview of histopathological subtypes of ovarian cancer in BRCA-relatedFrench Canadian cancer families. Familial Cancer. 2007;6(4):491–7.31. Tone AA, McConechy MK, Yang W, Ding J, Yip S, Kong E, Wong KK,Gershenson DM, Mackay H, Shah S, et al. Intratumoral heterogeneity in aminority of ovarian low-grade serous carcinomas. BMC Cancer. 2014;14:982.32. Gotlieb WH, Chetrit A, Menczer J, Hirsh-Yechezkel G, Lubin F, Friedman E,Modan B, Ben-Baruch G, National Israel Ovarian Cancer Study G.Demographic and genetic characteristics of patients with borderline ovariantumors as compared to early stage invasive ovarian cancer. Gynecol Oncol.2005;97(3):780–3.33. Matsuo K, Tierney KE, Schneider DM, Mhawech-Fauceglia P, Roman LD,Gershenson DM. Characteristics of ovarian tumors of low malignant potentialin BRCA mutation carriers: a case series. Gynecol Oncol Rep. 2015;13:36–9.34. Schrader KA, Hurlburt J, Kalloger SE, Hansford S, Young S, Huntsman DG,Gilks CB, McAlpine JN. Germline BRCA1 and BRCA2 mutations in ovariancancer: utility of a histology-based referral strategy. Obstet Gynecol. 2012;120(2 Pt 1):235–40.35. Boyd J, Sonoda Y, Federici MG, Bogomolniy F, Rhei E, Maresco DL, Saigo PE,Almadrones LA, Barakat RR, Brown CL, et al. Clinicopathologic features ofBRCA-linked and sporadic ovarian cancer. JAMA. 2000;283(17):2260–5.36. Foulkes WD. Inherited susceptibility to common cancers. N Engl J Med.2008;359(20):2143–53.37. Xu K, Yang S, Zhao Y. Prognostic significance of BRCA mutations in ovariancancer: an updated systematic review with meta-analysis. Oncotarget. 2017;8(1):285–302.38. Yang D, Khan S, Sun Y, Hess K, Shmulevich I, Sood AK, Zhang W. Association ofBRCA1 and BRCA2 mutations with survival, chemotherapy sensitivity, and genemutator phenotype in patients with ovarian cancer. JAMA. 2011;306(14):1557–65.39. McLaughlin JR, Rosen B, Moody J, Pal T, Fan I, Shaw PA, Risch HA, Sellers TA,Sun P, Narod SA. Long-term ovarian cancer survival associated withmutation in BRCA1 or BRCA2. J Natl Cancer Inst. 2013;105(2):141–8.40. Lorusso D, Cirillo F, Mancini M, Spatti GB, Grijuela B, Ditto A, Raspagliesi F.The different impact of BRCA mutations on the survival of epithelial ovariancancer patients: a retrospective single-center experience. Oncology. 2013;85(2):122–7.41. Coleman MP, Forman D, Bryant H, Butler J, Rachet B, Maringe C, Nur U,Tracey E, Coory M, Hatcher J, et al. Cancer survival in Australia, Canada,Denmark, Norway, Sweden, and the UK, 1995-2007 (the international Cancerbenchmarking partnership): an analysis of population-based cancer registrydata. Lancet. 2011;377(9760):127–38.42. Dehaeck U, McGahan CE, Santos JL, Carey MS, Swenerton KD, Kwon JS. Theimpact of geographic variations in treatment on outcomes in ovariancancer. Int J Gynecol Cancer. 2013;23(2):282–7.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Le Page et al. BMC Cancer  (2018) 18:347 Page 18 of 18

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}]}"
                            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:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0364561/manifest

Comment

Related Items