Mendelian randomization analyses suggest a role for cholesterol in the development of endometrial cancer

Blood lipids have been associated with the development of a range of cancers, including breast, lung and colorectal cancer. For endometrial cancer, observational studies have reported inconsistent associations between blood lipids and cancer risk. To reduce biases from unmeasured confounding, we performed a bidirectional, two-sample Mendelian randomization analysis to investigate the relationship between levels of three blood lipids (low-density lipoprotein [LDL] and high-density lipoprotein [HDL] cholesterol, and triglycerides) and endometrial cancer risk. Genetic variants associated with each of these blood lipid levels (P < 5 × 10-8 ) were identified as instrumental variables, and assessed using genome-wide association study data from the Endometrial Cancer Association Consortium (12 906 cases and 108 979 controls) and the Global Lipids Genetic Consortium (n = 188 578). Mendelian randomization analyses found genetically raised LDL cholesterol levels to be associated with lower risks of endometrial cancer of all histologies combined, and of endometrioid and non-endometrioid subtypes. Conversely, higher genetically predicted HDL cholesterol levels were associated with increased risk of non-endometrioid endometrial cancer. After accounting for the potential confounding role of obesity (as measured by genetic variants associated with body mass index), the association between genetically predicted increased LDL cholesterol levels and lower endometrial cancer risk remained significant, especially for non-endometrioid endometrial cancer. There was no evidence to support a role for triglycerides in endometrial cancer development. Our study supports a role for LDL and HDL cholesterol in the development of non-endometrioid endometrial cancer. Further studies are required to understand the mechanisms underlying these findings.

A comprehensive re-assessment of the association between vitamin D and cancer susceptibility using Mendelian randomization

Previous Mendelian randomization (MR) studies on 25-hydroxyvitamin D (25(OH)D) and cancer have typically adopted a handful of variants and found no relationship between 25(OH)D and cancer; however, issues of horizontal pleiotropy cannot be reliably addressed. Using a larger set of variants associated with 25(OH)D (74 SNPs, up from 6 previously), we perform a unified MR analysis to re-evaluate the relationship between 25(OH)D and ten cancers. Our findings are broadly consistent with previous MR studies indicating no relationship, apart from ovarian cancers (OR 0.89; 95% C.I: 0.82 to 0.96 per 1 SD change in 25(OH)D concentration) and basal cell carcinoma (OR 1.16; 95% C.I.: 1.04 to 1.28). However, after adjustment for pigmentation related variables in a multivariable MR framework, the BCC findings were attenuated. Here we report that lower 25(OH)D is unlikely to be a causal risk factor for most cancers, with our study providing more precise confidence intervals than previously possible.

Cross-Cancer Genome-Wide Association Study of Endometrial Cancer and Epithelial Ovarian Cancer Identifies Genetic Risk Regions Associated with Risk of Both Cancers

BACKGROUND

Accumulating evidence suggests a relationship between endometrial cancer and ovarian cancer. Independent genome-wide association studies (GWAS) for endometrial cancer and ovarian cancer have identified 16 and 27 risk regions, respectively, four of which overlap between the two cancers. We aimed to identify joint endometrial and ovarian cancer risk loci by performing a meta-analysis of GWAS summary statistics from these two cancers.

METHODS

Using LDScore regression, we explored the genetic correlation between endometrial cancer and ovarian cancer. To identify loci associated with the risk of both cancers, we implemented a pipeline of statistical genetic analyses (i.e., inverse-variance meta-analysis, colocalization, and M-values) and performed analyses stratified by subtype. Candidate target genes were then prioritized using functional genomic data.

RESULTS

Genetic correlation analysis revealed significant genetic correlation between the two cancers (rG = 0.43, P = 2.66 × 10-5). We found seven loci associated with risk for both cancers (P Bonferroni < 2.4 × 10-9). In addition, four novel subgenome-wide regions at 7p22.2, 7q22.1, 9p12, and 11q13.3 were identified (P < 5 × 10-7). Promoter-associated HiChIP chromatin loops from immortalized endometrium and ovarian cell lines and expression quantitative trait loci data highlighted candidate target genes for further investigation.

CONCLUSIONS

Using cross-cancer GWAS meta-analysis, we have identified several joint endometrial and ovarian cancer risk loci and candidate target genes for future functional analysis.

IMPACT

Our research highlights the shared genetic relationship between endometrial cancer and ovarian cancer. Further studies in larger sample sets are required to confirm our findings.

Pre- and Post-Diagnosis Diet Quality and Ovarian Cancer Survival

BACKGROUND

Prior studies evaluating diet quality in relation to ovarian cancer survival are sparse, and to date none have assessed diet quality or diet-quality change after diagnosis.

METHODS

In the prospective Ovarian cancer Prognosis And Lifestyle (OPAL) study, diet-quality scores were calculated using data from food frequency questionnaires completed pre-diagnosis (n = 650) and 12 months' post-diagnosis (n = 503). We used Cox proportional hazard models to estimate adjusted hazard ratios (HR) and 95% confidence intervals (CI) for the association between diet quality and survival.

RESULTS

During the median follow-up of 4.4 years, 278 women died from ovarian cancer. There was no evidence of an association between diet quality pre- or post-diagnosis and progression-free, overall, or ovarian cancer-specific survival. No survival advantage was observed for women who had either improved their diet quality or who consumed a high-quality diet both before and 12 months after diagnosis.

CONCLUSIONS

Higher pre- and post-diagnosis diet quality was not associated with better survival outcomes in this cohort of women with ovarian cancer.

IMPACT

Diet quality is important for a range of health outcomes but may not improve survival after a diagnosis of ovarian cancer.

Hysterectomy with and without oophorectomy and all-cause and cause-specific mortality

BACKGROUND

Hysterectomy is one of the most commonly performed gynecologic surgeries, with an estimated 30% of women in Australia undergoing the procedure by age of 70 years. In the United States, about 45% of women undergo hysterectomy in their lifetime. Some studies have suggested that this procedure increases the risk of premature mortality. With many women making the decision to undergo hysterectomy for a benign indication each year, additional research is needed to clarify whether there are long-term health consequences of hysterectomy.

OBJECTIVE

This study aimed to examine the association between hysterectomy for benign indications, with or without removal of the ovaries, and cause-specific and all-cause mortality.

STUDY DESIGN

Our cohort of 666,588 women comprised the female population of Western Australia with linked hospital and health records from 1970 to 2015. Cox regression models were used to assess the association between hysterectomy and all-cause, cardiovascular disease, cancer, and other mortality by oophorectomy type (categorized as none, unilateral, and bilateral), with no hysterectomy or oophorectomy as the reference group. We repeated these analyses using hysterectomy without oophorectomy as the reference group. We also investigated whether associations varied by age at the time of surgery, although small sample size precluded this analysis in women who underwent hysterectomy with unilateral salpingo-oophorectomy. In our main analysis, women who underwent hysterectomy or oophorectomy as part of cancer treatment were retained in the analysis and considered unexposed to that surgery. For a sensitivity analysis, we censored procedures performed for cancer.

RESULTS

Compared with no surgery, hysterectomy without oophorectomy before 35 years was associated with an increase in all-cause mortality (hazard ratio, 1.29; 95% confidence interval, 1.19-1.40); for surgery after 35 years of age, there was an inverse association (35-44 years: hazard ratio, 0.93; 95% confidence interval, 0.89-0.97). Similarly, hysterectomy with bilateral salpingo-oophorectomy before 45 years of age was associated with increased all-cause mortality (35-44 years: hazard ratio, 1.15; 95% confidence interval, 1.04-1.27), but decreased mortality rates after 45 years of age. In our sensitivity analysis, censoring gynecologic surgeries for cancer resulted in many cancer-related deaths being excluded for women who did not have surgery for benign indications and thus increased the hazard ratios for the associations between both hysterectomy without oophorectomy and hysterectomy with bilateral salpingo-oophorectomy and risk of all-cause and cancer-specific mortality. The sensitivity analysis therefore potentially biased the results in favor of no surgery.

CONCLUSION

Among women having surgery for benign indications, hysterectomy without oophorectomy performed before 35 years of age and hysterectomy with bilateral salpingo-oophorectomy performed before 45 years of age were associated with an increase in all-cause mortality. These procedures are not associated with poorer long-term survival when performed at older ages.

Offspring sex and risk of epithelial ovarian cancer: a multinational pooled analysis of 12 case-control studies

While childbearing protects against risk of epithelial ovarian cancer (EOC), few studies have explored the impact on maternal EOC risk of sex of offspring, which may affect the maternal environment during pregnancy. We performed a pooled analysis among parous participants from 12 case-controls studies comprising 6872 EOC patients and 9101 controls. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multivariable logistic regression for case-control associations and polytomous logistic regression for histotype-specific associations, all adjusted for potential confounders. In general, no associations were found between offspring sex and EOC risk. However, compared to bearing only female offspring, bearing one or more male offspring was associated with increased risk of mucinous EOC (OR = 1.45; 95% CI = 1.01-2.07), which appeared to be limited to women reporting menarche before age 13 compared to later menarche (OR = 1.71 vs 0.99; P-interaction = 0.02). Bearing increasing numbers of male offspring was associated with greater risks of mucinous tumors (OR = 1.31, 1.84, 2.31, for 1, 2 and 3 or more male offspring, respectively; trend-p = 0.005). Stratifying by hormonally-associated conditions suggested that compared to bearing all female offspring, bearing a male offspring was associated with lower risk of endometrioid cancer among women with a history of adult acne, hirsutism, or polycystic ovary syndrome (OR = 0.49, 95% CI = 0.28-0.83) but with higher risk among women without any of those conditions (OR = 1.64 95% CI = 1.14-2.34; P-interaction = 0.003). Offspring sex influences the childbearing-EOC risk relationship for specific histotypes and conditions. These findings support the differing etiologic origins of EOC histotypes and highlight the importance of EOC histotype-specific epidemiologic studies. These findings also suggest the need to better understand how pregnancy affects EOC risk.

Statin use and survival following a diagnosis of ovarian cancer: A prospective observational study

Most women with ovarian cancer have a poor prognosis, but studies have reported an association between statin use and improved survival. We investigated the potential survival benefit of statins in women with ovarian cancer using data from the Ovarian cancer Prognosis and Lifestyle study, a prospective study of Australian women aged 18 to 79 years, diagnosed with ovarian cancer from 2012 to 2015 and followed for 5 to 8 years. We obtained information from patient-completed questionnaires and medical records. We defined exposure based on prediagnosis use, as most women used statins continuously (prediagnosis and postdiagnosis) and few started using statins postdiagnosis. We measured survival from date of first treatment (surgery or neoadjuvant chemotherapy) until date of death or last follow-up. We used Cox regression to calculate hazard ratios (HR) and 95% confidence intervals (CI), adjusting for potential confounders. To reduce bias due to confounding by indication, we also applied inverse probability of treatment weighting (IPTW). Of 955 eligible women, 21% reported statin use before diagnosis. Statin users had a slightly better survival (HR = 0.90, 95% CI = 0.70-1.15) that was driven by lipophilic statin use (HR = 0.82, 95% CI = 0.61-1.11), with no association for hydrophilic statins (HR = 1.04, 95% CI = 0.72-1.49). The IPTW model weighted to all women with ovarian cancer also suggested a possible reduction in mortality associated with lipophilic statins (HR = 0.80, 95% CI = 0.54-1.21). In analyses restricted to women with hyperlipidaemia, the HRs were further from the null. Our findings are consistent with previous evidence, suggesting that lipophilic statins might improve ovarian cancer survival. Further investigation, in larger cohorts, or preferably in a randomised trial, is required.

Menopausal hormone therapy prior to the diagnosis of ovarian cancer is associated with improved survival

PURPOSE

Prior studies of menopausal hormone therapy (MHT) and ovarian cancer survival have been limited by lack of hormone regimen detail and insufficient sample sizes. To address these limitations, a comprehensive analysis of 6419 post-menopausal women with pathologically confirmed ovarian carcinoma was conducted to examine the association between MHT use prior to diagnosis and survival.

METHODS

Data from 15 studies in the Ovarian Cancer Association Consortium were included. MHT use was examined by type (estrogen-only (ET) or estrogen+progestin (EPT)), duration, and recency of use relative to diagnosis. Cox proportional hazards models were used to estimate the association between hormone therapy use and survival. Logistic regression and mediation analysis was used to explore the relationship between MHT use and residual disease following debulking surgery.

RESULTS

Use of ET or EPT for at least five years prior to diagnosis was associated with better ovarian cancer survival (hazard ratio, 0.80; 95% CI, 0.74 to 0.87). Among women with advanced stage, high-grade serous carcinoma, those who used MHT were less likely to have any macroscopic residual disease at the time of primary debulking surgery (p for trend <0.01 for duration of MHT use). Residual disease mediated some (17%) of the relationship between MHT and survival.

CONCLUSIONS

Pre-diagnosis MHT use for 5+ years was a favorable prognostic factor for women with ovarian cancer. This large study is consistent with prior smaller studies, and further work is needed to understand the underlying mechanism.

Abstract 3498: Ovarian cancer risk factors multiply to create high risk in pre-menopausal women

Background: Ovarian cancer is the eighth-most common cancer among women. Although a number of established risk factors exist for this disease, analyses of these exposures have largely not taken into account potential interactions. We present an interaction analysis of eight ovarian cancer risk factors, followed by the creation of modifiable and non-modifiable risk scores for the disease.

Methods: Data used in this analysis come from nine studies in the Ovarian Cancer Association Consortium (OCAC): one from Australia, one from Europe and seven from the US. Analyses were restricted to pre-menopausal women without first-degree family histories of ovarian cancer, leading to a final study population of 1504 cases and 3515 controls. Five potentially modifiable predictors (oral contraceptive (OC) use duration, body mass index, tubal ligation, hysterectomy and breastfeeding) and three non-modifiable predictors (parity, incomplete pregnancies and history of endometriosis) were considered. A complete analysis of interactions between the eight variables as well as with age, race/ethnicity, education level and study site was conducted. No statistically significant pairwise interactions were found with likelihood ratio statistics, thus suggesting no departure from multiplicativity in the odds ratios. The modifiable and non-modifiable risk scores were calculated as the product of the odds ratios for each group of exposures. Logistic regression was used to model the association between ovarian cancer and the risk scores. Age, race/ethnicity, education level and study site were included in the model.

Results: The associations between the second, third and fourth quartiles of the modifiable risk score compared to the lowest quartile were 1.67 (95% CI 1.35-2.09), 2.18 (95% CI 1.76-2.71) and 3.72 (95% CI 3.04-4.57), respectively. The odds ratios for the non-modifiable risk score were 1.17 (95% CI 0.95-1.44), 1.24 (95% CI 0.99-1.54) and 2.06 (95% CI 1.68-2.53), respectively for the second, third and fourth quartiles compared to the first. Women in the highest quartiles of both the modifiable and non-modifiable risk scores, representing 14.6% of the women, had a 7.67-fold increased risk of ovarian cancer (95% CI 4.68-11.39) compared to women in the lowest quartiles for both risk scores (3.6% of the women).

Conclusions: Pre-menopausal women in the highest risk quartiles have a substantially elevated risk of ovarian cancer based on these eight exposures. Modifiable factors contribute more than non-modifiable factors to ovarian cancer risk among pre-menopausal women with no family history in our study. Women in the highest risk group (fourth quartile of both risk scores) could reduce their risk by intervening on the modifiable exposures; for example, 63% of women in this group had not used oral contraceptives and 93% had not had tubal ligation. However, careful consideration of the risks and benefits of such prevention measures is needed.

Citation Format: Minh Tung Phung, Bhramar Mukherjee, Alice W. Lee, Penelope M. Webb, Harvey A. Risch, Jennifer Anne Doherty, Holly R. Harris, Marc T. Goodman, Roberta B. Ness, Francesmary Modugno, Allan Jensen, Susanne K. Kjaer, Kathryn L. Terry, Daniel W. Cramer, Argyrios Ziogas, Hoda Anton-Culver, Malcolm C. Pike, Anna H. Wu, Celeste Leigh Pearce. Ovarian cancer risk factors multiply to create high risk in pre-menopausal women [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3498.

Insomnia and its association with quality of life in women with ovarian cancer

OBJECTIVES

Insomnia is common in women with ovarian cancer but there are limited prospective data on the frequency and degree of impact on patients. Our objective was to determine the prevalence of insomnia over the first three years after a diagnosis of ovarian cancer; and the relationship between insomnia and quality of life.

METHODS

OPAL (Ovarian cancer, Prognosis And Lifestyle) is a prospective study of Australian women with epithelial ovarian cancer; 894 provided information on insomnia symptoms, medications and quality of life at three, six, nine, 12, 24 and 36 months after diagnosis. Generalised linear mixed models were used to determine the relationship between insomnia and quality of life measured at the same time and three months later.

RESULTS

One-quarter of women reported symptoms consistent with clinical insomnia within three years after diagnosis and an additional 13% regularly used sleep medication (total 36% affected). Excluding 7% who reported insomnia symptoms before diagnosis, 22% reported new insomnia, which reduced to 17% when also excluding women on chemotherapy. The proportion of women with clinical (14%) or subclinical (28%) insomnia symptoms was highest at three months after diagnosis. Compared to women with no insomnia, those with clinical insomnia had significantly lower quality of life measured at the same time (8.4 points lower, 95% CI: 7.2-9.5), and three months later (5.5 points lower, 95% CI: 3.4-7.6).

CONCLUSIONS

Over a third of women with ovarian cancer likely experience insomnia after diagnosis; this may persist and is associated with poorer quality of life.

A healthy lifestyle and survival among women with ovarian cancer

Ovarian cancer has a poor survival rate and, understandably, women often want to know whether there is anything they can do to improve their prognosis. Our goal was to investigate the association between a healthy lifestyle prediagnosis and postdiagnosis and survival in a cohort of Australian women with invasive epithelial ovarian cancer. We calculated a healthy lifestyle index (HLI) based on women's self-reported smoking status, height, weight, physical activity, diet and alcohol consumption before diagnosis (n = 678) and after completing primary treatment (n = 512). Clinical data and vital status for each woman were ascertained through medical records. Cox proportional hazards regression was conducted to calculate hazard ratios (HR) and 95% confidence interval (CI) for all-cause mortality. There was a suggestive association between a more healthy lifestyle before diagnosis and better survival (HR 0.79, 95% CI: 0.59-1.04), however, the association was stronger for lifestyle after diagnosis, with women in the highest tertile having significantly better survival than women in the lowest tertile (HR 0.61, 95% CI: 0.40-0.93; P-trend = .02). Current smoking, particularly postdiagnosis, was associated with higher mortality (HR 1.68, 95% CI: 1.17-2.42; HR 2.82, 95% CI: 1.29-6.14, for prediagnosis and postdiagnosis smoking, respectively), but women who quit after diagnosis had survival outcomes similar to nonsmokers (HR 0.99, 95% CI: 0.57-1.72). Higher physical activity after diagnosis was associated with better survival (HR 0.60, 95% CI: 0.39-0.92; P-trend = .02). A healthy lifestyle after diagnosis, in particular not smoking and being physically active, may help women with ovarian cancer improve their prognosis.

Abstract B38: Use of progestin-only injectable contraceptive is associated with reduced risk of ovarian cancer in the Ovarian Cancer Association Consortium

Background: Combined oral contraceptive use is associated with a decreased risk of ovarian carcinoma (cancer). However, the relationship between progestin-only contraceptives and ovarian cancer risk is unclear. Two previous studies have suggested a protective effect whereas another reported a non-statistically significant increased risk of ovarian cancer. The current study examined the association between use of depot medroxyprogesterone acetate (DMPA), a progestin-only injectable contraceptive, and ovarian cancer risk, using data from seven case-control studies participating in the Ovarian Cancer Association Consortium (OCAC).

Methods: A pooled analysis examining the relationship between DMPA use and ovarian cancer risk was conducted using 7,679 invasive ovarian cancer cases and 11,136 controls from six studies from the United States and one from Australia. Combined oral contraceptive use, parity, education level, age, and race/ethnicity were considered important a priori confounders and were included in all statistical models. OCAC study site was also included in all models. The impact of additional exposures, including a personal history of endometriosis, first-degree family history of ovarian cancer, tubal ligation, breastfeeding, body mass index, and menopausal hormonal therapy use on the association between DMPA use and ovarian cancer were considered. None of these variables was found to impact the DMPA use-ovarian cancer association by &gt;10% and thus not included in the final models. Odds ratios (OR) and 95% confidence intervals (CI) were generated from logistic regression models. The association between duration of DMPA use, categorized as never use, &lt;2 years of use, and 2+ years of use, and ovarian cancer risk was examined in the two studies that had this information available.

Results: The frequency of DMPA use among controls ranged from 1.25% to 3.53% across the seven studies. DMPA use was more common in controls than in cases in all of the studies. Overall, ever use of DMPA was associated with a 26% decreased risk of ovarian cancer (95% CI 0.58-0.94), after taking into account combined oral contraceptive use, parity, education level, age, race/ethnicity, and OCAC study site. A significant trend with duration of use was observed in the two studies with these data (p=0.02).

Conclusions: DMPA use appears to be associated with a decreased risk of ovarian cancer. The finding provides additional evidence that progestins may be protective for ovarian cancer. Further evaluation of the role of DMPA as a potential primary prevention strategy for ovarian cancer, especially in women for whom combined oral contraceptive use is contraindicated due to concerns about estrogen-induced thromboembolic events, is warranted.

Citation Format: Minh Tung Phung, Penelope M. Webb, Jennifer Anne Doherty, Holly R. Harris, Pamela J. Thompson, Marc T. Goodman, Kirsten Moysich, Francesmary Modugno, Roberta B. Ness, Joellen M. Schildkraut, Andrew Berchuck, Daniel W. Cramer, Kathryn L. Terry, Linda Titus, Alice W. Lee, Malcolm C. Pike, Anna H. Wu, Celeste Leigh Pearce. Use of progestin-only injectable contraceptive is associated with reduced risk of ovarian cancer in the Ovarian Cancer Association Consortium [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr B38.

Abstract B37: Germline mutations in new susceptibility genes for non-high-grade serous ovarian cancer

Improving risk prediction and prevention strategies through identifying new susceptibility genes for non-high-grade serous ovarian cancer (non-HGS) would represent an important advance in reducing incidence and mortality. Epithelial ovarian cancer (EOC) has five main histotypes that have distinct pathologies, molecular changes, clinical characteristics, and tissues of origin. They are classified as high-grade serous (HGS), low-grade serous (LGS), clear-cell (CCC), endometrioid (END), and mucinous (MUC). A family history of breast or ovarian cancer is the strongest single risk factor for EOC. Germline mutations in the high-penetrance susceptibility genes BRCA1 and BRCA2 confer EOC risks of 44% and 17% by age 80, respectively. Only 40% of the excess familial risk is known, suggesting there are many other susceptibility genes yet unidentified. The non-HGS histotypes are poorly studied due to limited sample size within individual studies. Thus, families of cases with non-HGS tumors have few risk prediction options. Whole-exome sequencing (WES) analysis was performed on 251 non-HGS cases (56 LGS, 55 CCC, 117 END, 23 MUC). Cases were screened negative for BRCA1/2 mutations and selected for a family history of ovarian or breast cancer, or young onset (&lt;45 years) of ovarian cancer. WES identified 1,278 genes with rare predicted truncating mutation in at least one case. Predicted truncating mutations in cases were compared to 171,584 controls from the Exome Aggregation Consortium (ExAC) and the Genome Aggregation Database (gnomAD). Gene-set enrichment analysis showed enrichment for genes involved in the DNA repair pathway (p=3.08 × 10−7). Twenty-five candidate genes (including 4 DNA repair genes) were selected for validation by targeted sequencing. Five genes important in HGS (BRIP1, FANCM, PALB2, RAD51C, RAD51D) as well as two additional candidates (XRCC2 and XRCC3) were also sequenced. Targeted sequencing was performed on 1,779 cases (669 END, 356 CCC, 327 LGS, 427 MUC) and 1,863 controls from studies in the Ovarian Cancer Association Consortium (OCAC). Library preparation and target enrichment was performed using the Fluidigm Juno System and Illumina sequencing was performed on a NovaSeq 6000 Sequencing System. A case-control analysis of predicted truncating variants in the 32 genes was performed. Using a minimum alternate allele frequency of 30%, we identified a higher frequency of mutations in non-HGS cases than controls in ERCC6 (p=0.034) and IL31RA (p=0.034). ERCC6 is involved in DNA repair and IL31RA is a cytokine receptor. Sanger sequencing validation of variants to confirm these results is ongoing. Several other genes showed suggestive higher frequencies of mutations in cases than controls. A larger case control analysis will be performed to confirm those findings. Identifying novel susceptibility genes for non-HGS may have clinical impact by reducing disease-associated mortality through improving risk prediction, identifying prevention strategies, and developing new targeted treatments.

Citation Format: Marina Pavanello, Ed Dicks, Honglin Song, Amir Ariff, Adelyn Bolithon, Maria P. Intermaggio, Mark Pinese, Kirsten Moysich, Kunle O. Odunsi, Ellen Goode, David D. Bowtell, Peter Fasching, Jennifer A. Doherty, Francesmary Modugno, Susanne K. Kjær, Penelope M. Webb, Anna Wu, Anna deFazio, Ovarian Cancer Association Consortium, Paul James, Deepak Subramanian, Ian Campbell, Simon A. Gayther, Paul D.P. Pharoah, Susan J. Ramus. Germline mutations in new susceptibility genes for non-high-grade serous ovarian cancer [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr B37.

Abstract B32: Are ovarian cancer risk factors different for women with endometriosis?

Background: Endometriosis is a common gynecologic disorder that affects approximately 10% of women of reproductive age. It is also a well-established risk factor for ovarian cancer. Whether hormonal-related and other risk factors for ovarian cancer (e.g., parity, oral contraceptive use) are the same for women with and without endometriosis is currently unknown.

Methods: We pooled questionnaire information from 10 population-based case-control studies participating in the Ovarian Cancer Association Consortium (OCAC). Data from 8,172 women with invasive epithelial ovarian cancer and 12,464 controls were included in our analysis. Associations of body mass index (BMI), parity, oral contraceptive use, breastfeeding, menopausal hormone therapy use, first-degree family history of ovarian cancer, tubal ligation, and hysterectomy with risk of ovarian cancer were stratified by history of endometriosis and analyzed using logistic regression. All models were conditioned on age, race/ethnicity, education, and OCAC study site. We fit an endometriosis interaction term for each risk factor to evaluate statistical interactions.

Results: Among women with no history of endometriosis, those who had hysterectomies showed a 12% increased risk of ovarian cancer compared to those with intact uteri (OR=1.12, 95% CI 1.01-1.24); an increased risk was not observed among women with endometriosis (OR=0.74, 95% CI 0.53-1.05; p-interaction=0.009). Use of estrogen-progestin hormone therapy was associated with decreased risk ovarian cancer for women with histories of endometriosis (OR=0.69, 95% CI 0.47-1.02), but not for those without endometriosis (OR=0.96, 95% CI 0.87-1.07; p-interaction=0.02). We did not observe any significant statistical interactions for the other risk factors considered, and no interactions were significant after consideration of multiple comparisons.

Conclusions: The associations of hysterectomy and menopausal estrogen-progestin hormone therapy use with risk of ovarian cancer seemed to differ by endometriosis status, suggesting interactions that may need to be considered in ovarian cancer risk profiling strategies. Future investigations into the biology underlying these interactions would be relevant.

Citation Format: Aruna Muthukumar, Lilah Khoja, Penelope M. Webb, Harvey Risch, Jennifer Doherty, Holly Harris, Marc Goodman, Roberta Ness, Francesmary Modugno, Susanne K. Kjaer, Allan Jensen, Joellen Schildkraut, Andrew Berchuck, Kathryn L. Terry, Daniel Cramer, Linda J. Titus, Hoda Anton-Culver, Argyrios Ziogas, Anna H. Wu, Malcolm C. Pike, Celeste L. Pearce, Alice W. Lee. Are ovarian cancer risk factors different for women with endometriosis? [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research; 2019 Sep 13-16, 2019; Atlanta, GA. Philadelphia (PA): AACR; Clin Cancer Res 2020;26(13_Suppl):Abstract nr B32.

Generating high-quality data abstractions from scanned clinical records: text-mining-assisted extraction of endometrial carcinoma pathology features as proof of principle

OBJECTIVE

Medical research studies often rely on the manual collection of data from scanned typewritten clinical records, which can be laborious, time consuming and error prone because of the need to review individual clinical records. We aimed to use text mining to assist with the extraction of clinical features from complex text-based scanned pathology records for medical research studies.

DESIGN

Text mining performance was measured by extracting and annotating three distinct pathological features from scanned photocopies of endometrial carcinoma clinical pathology reports, and comparing results to manually abstracted terms. Inclusion and exclusion keyword trigger terms to capture leiomyomas, endometriosis and adenomyosis were provided based on expert knowledge. Terms were expanded with character variations based on common optical character recognition (OCR) error patterns as well as negation phrases found in sample reports. The approach was evaluated on an unseen test set of 1293 scanned pathology reports originating from laboratories across Australia.

SETTING

Scanned typewritten pathology reports for women aged 18-79 years with newly diagnosed endometrial cancer (2005-2007) in Australia.

RESULTS

High concordance with final abstracted codes was observed for identifying the presence of three pathology features (94%-98% F-measure). The approach was more consistent and reliable than manual abstractions, identifying 3%-14% additional feature instances.

CONCLUSION

Keyword trigger-based automation with OCR error correction and negation handling proved not only to be rapid and convenient, but also providing consistent and reliable data abstractions from scanned clinical records. In conjunction with manual review, it can assist in the generation of high-quality data abstractions for medical research studies.

Association Between Breastfeeding and Ovarian Cancer Risk

Importance

Breastfeeding has been associated with a reduced risk of epithelial ovarian cancer in multiple studies, but others showed no association. Whether risk reduction extends beyond that provided by pregnancy alone or differs by histotype is unclear. Furthermore, the observed associations between duration and timing of breastfeeding with ovarian cancer risk have been inconsistent.

Objective

To determine the association between breastfeeding (ie, ever/never, duration, timing) and ovarian cancer risk overall and by histotype.

Design, Setting, and Participants

A pooled analysis of parous women with ovarian cancer and controls from 13 case-control studies participating in the Ovarian Cancer Association Consortium was performed. Odds ratios (ORs) and 95% CIs of the overall association were calculated using multivariable logistic regression and polytomous logistic regression for histotype-specific associations. All data were collected from individual sites from November 1989 to December 2009, and analysis took place from September 2017 to July 2019.

Exposures

Data on breastfeeding history, including duration per child breastfed, age at first and last breastfeeding, and years since last breastfeeding were collected by questionnaire or interview and was harmonized across studies.

Main Outcomes and Measures

Diagnosis of epithelial ovarian cancer.

Results

A total of 9973 women with ovarian cancer (mean [SD] age, 57.4 [11.1] years) and 13 843 controls (mean [SD] age, 56.4 [11.7] years) were included. Breastfeeding was associated with a 24% lower risk of invasive ovarian cancer (odds ratio [OR], 0.76; 95% CI, 0.71-0.80). Independent of parity, ever having breastfed was associated with reduction in risk of all invasive ovarian cancers, particularly high-grade serous and endometrioid cancers. For a single breastfeeding episode, mean breastfeeding duration of 1 to 3 months was associated with 18% lower risk (OR, 0.82; 95% CI, 0.76-0.88), and breastfeeding for 12 or more months was associated with a 34% lower risk (OR, 0.66; 95% CI, 0.58-0.75). More recent breastfeeding was associated with a reduction in risk (OR, 0.56; 95% CI, 0.47-0.66 for <10 years) that persisted for decades (OR, 0.83; 95% CI, 0.77-0.90 for ≥30 years; P for trend = .02).

Conclusions and Relevance

Breastfeeding is associated with a significant decrease in risk of ovarian cancer overall and for the high-grade serous subtype, the most lethal type of ovarian cancer. The findings suggest that breastfeeding is a potentially modifiable factor that may lower risk of ovarian cancer independent of pregnancy alone.

The proportion of cancers attributable to social deprivation: A population-based analysis of Australian health data

BACKGROUND

Cancer is a major disease burden globally and people who are socioeconomically disadvantaged have a higher incidence of many types of cancer. We investigated the potential to reduce socioeconomic disparities in cancer incidence in Australia by lowering the prevalence of exposure to four modifiable causes: smoking, alcohol, overweight/obesity and physical inactivity.

METHODS

We used cancer incidence data from the Australian Cancer Database and risk factor prevalence data from the Australian National Health Survey to estimate the proportions of cancers attributable to the four factors, by area-level socioeconomic disadvantage. For the three risk factors where prevalence was lowest among the least disadvantaged (smoking, overweight/obesity, physical inactivity), we also estimated the potential impact of reducing prevalence in the most disadvantaged areas to that in the least disadvantaged areas.

RESULTS

The proportion of cancer attributable to the four factors in combination ranged from 22 % in the most disadvantaged areas to 14 % in the least disadvantaged areas. If the prevalence of tobacco smoking, overweight/obesity and physical inactivity in the more disadvantaged areas were the same as that in the least disadvantaged, an estimated 19,500 cancers (4 % of all cancers diagnosed) could have been prevented in Australia between 2009 and 2013.

CONCLUSIONS

Reducing the prevalence of key causal factors in areas of greater social disadvantage would prevent many cases of cancer. Strategies to achieve this in highly disadvantaged areas are needed.

Evaluating the impact of dose reductions and delays on progression-free survival in women with ovarian cancer treated with either three-weekly or dose-dense carboplatin and paclitaxel regimens in the national prospective OPAL cohort study

OBJECTIVES

To determine the impact of chemotherapy dose reductions and dose delays on progression-free survival (PFS) in women with ovarian cancer receiving first line chemotherapy in a real world prospective cohort study.

METHODS

Patients with newly diagnosed epithelial ovarian (or peritoneal, fallopian tube) cancer enrolled in a national Australian prospective study, OPAL, who commenced three-weekly carboplatin (AUC 5 or 6) and paclitaxel 175 mg/m² (CP) or carboplatin (AUC 5 or 6) and dose-dense weekly paclitaxel 80 mg/m² (DD-CP) were eligible. Primary endpoint was PFS.

RESULTS

634 evaluable patients, 309 commenced CP and 325 DD-CP. Patient's age was similar in the two groups (median 62 years, range 21-79). All planned chemotherapy doses were completed by 66% vs 40% (p < 0.001) in the CP and DD-CP groups respectively. There was at least one treatment delay in 28% vs 58% (p < 0.001) in the CP and DD-CP groups, respectively, and 29% vs 49% (p < 0.001), respectively, required at least a 15% dose reduction for either carboplatin or paclitaxel. Median PFS was 29.2 [22.9, 43.8] and 21.5 [19.4, 23.1] months in the CP and DD-CP groups respectively. Adjusting for age, histology and FIGO stage PFS did not differ between treatment groups. Median PFS was similar in patients irrespective of dose reduction or dose delay.

CONCLUSION

Patients receiving DD-CP required more dose reductions and delays due to haematological toxicities and lower completion rates than CP without significant difference in median PFS between CP and DD-CP. Median PFS was similar in patients irrespective of dose reduction or dose delay.

Estrogen Plus Progestin Hormone Therapy and Ovarian Cancer: A Complicated Relationship Explored

BACKGROUND

Menopausal estrogen-alone therapy is a risk factor for endometrial and ovarian cancers. When a progestin is included with the estrogen daily (continuous estrogen-progestin combined therapy), there is no increased risk of endometrial cancer. However, the effect of continuous estrogen-progestin combined therapy on risk of ovarian cancer is less clear.

METHODS

We pooled primary data from five population-based case-control studies in the Ovarian Cancer Association Consortium, including 1509 postmenopausal ovarian cancer cases and 2295 postmenopausal controls. Information on previous menopausal hormonal therapy use, as well as ovarian cancer risk factors, was collected using in-person interviews. Logistic regression was used to assess the association between use of continuous estrogen-progestin combined therapy and risk of ovarian cancer by duration and recency of use and disease histotype.

RESULTS

Ever postmenopausal use of continuous estrogen-progestin combined therapy was not associated with increased risk of ovarian cancer overall (OR = 0.85, 95% CI = 0.72, 1.0). A decreased risk was observed for mucinous ovarian cancer (OR = 0.40, 95% CI = 0.18, 0.91). The other main ovarian cancer histotypes did not show an association (endometrioid: OR = 0.86, 95% CI = 0.57, 1.3, clear cell: OR = 0.68, 95% CI = 0.40, 1.2; serous: OR = 0.98, 95% CI = 0.80, 1.2).

CONCLUSIONS

Given that estrogen-alone therapy has been shown to be associated with increased risk of ovarian cancer, these findings are consistent with the hypothesis that adding a progestin each day ameliorates the carcinogenic effects of estrogen on the cells of origin for all histotypes of ovarian cancer.

Association between genetically predicted polycystic ovary syndrome and ovarian cancer: a Mendelian randomization study

BACKGROUND

Polycystic ovary syndrome (PCOS) is a complex endocrine disorder with an estimated prevalence of 4-21% in reproductive aged women. Recently, the Ovarian Cancer Association Consortium (OCAC) reported a decreased risk of invasive ovarian cancer among women with self-reported PCOS. However, given the limitations of self-reported PCOS, the validity of these observed associations remains uncertain. Therefore, we sought to use Mendelian randomization with genetic markers as a proxy for PCOS, to examine the association between PCOS and ovarian cancer.

METHODS

Utilizing 14 single nucleotide polymorphisms (SNPs) previously associated with PCOS we assessed the association between genetically predicted PCOS and ovarian cancer risk, overall and by histotype, using summary statistics from a previously conducted genome-wide association study (GWAS) of ovarian cancer among European ancestry women within the OCAC (22 406 with invasive disease, 3103 with borderline disease and 40 941 controls).

RESULTS

An inverse association was observed between genetically predicted PCOS and invasive ovarian cancer risk: odds ratio (OR)=0.92 [95% confidence interval (CI)=0.85-0.99; P = 0.03]. When results were examined by histotype, the strongest inverse association was observed between genetically predicted PCOS and endometrioid tumors (OR = 0.77; 95% CI = 0.65-0.92; P = 0.003). Adjustment for individual-level body mass index, oral contraceptive use and parity did not materially change the associations.

CONCLUSION

Our study provides evidence for a relationship between PCOS and reduced ovarian cancer risk, overall and among specific histotypes of invasive ovarian cancer. These results lend support to our previous observational study results. Future studies are needed to understand mechanisms underlying this association.

Co-existence of leiomyomas, adenomyosis and endometriosis in women with endometrial cancer

Leiomyomas, adenomyosis, and endometriosis are reported to be risk factors for endometrial carcinoma (EC), and adenomyosis and endometriosis also for ovarian carcinoma (OC). We aimed to describe the prevalence of these conditions in EC patients with or without an OC diagnosis, and to investigate their relationship with EC risk and prognostic factors in these patients. We evaluated the co-existence of these three conditions in 1399 EC patients, and compared the prevalence of epidemiological risk factors and tumor prognostic features in patients with each condition versus not. Prevalence of conditions was also assessed in the subset of patients with prior/concurrent OC. The observed coexistence of leiomyomas, adenomyosis and endometriosis significantly deviated from that expected (P = 1.2 × 10^-8). Patients were more likely to: report a younger age at menarche (P_Trend = 0.004) if they had leiomyomas; have used oral contraceptives (P = 6.6 × 10^-5) or had ≥2 full-term pregnancies (P_Trend = 2.0 × 10^-9) if they had adenomyosis; be diagnosed with EC at younger age (P = 5.0 × 10^-11) if they had endometriosis. Patients with prior/concurrent OC were more likely to be diagnosed at younger age (P = 5.0 × 10^-5), have endometriosis (P = 9.9 × 10^-7), and present with higher stage EC (P_Trend = 6.6 × 10^-5). These findings justify further consideration of these gynecologic conditions as independent risk and prognostic factors for EC.

The influence of birth cohort and calendar period on global trends in ovarian cancer incidence

Ovarian cancer is the eighth most common cancer in women worldwide and incidence rates vary markedly by world region. Our study provides a comprehensive overview of ovarian cancer incidence trends globally, examining the influence of birth cohort and period of diagnosis on changing risk. We presented current patterns and trends of ovarian cancer incidence until 2012 using data from successive volumes of Cancer Incidence in Five Contents. The incidence of ovarian cancer is highest in northern and eastern European countries and in northern America. Declining trends were observed in most countries with the exception of a few central and eastern Asian countries. Marked declines were seen in Europe and North America for women aged 50-74 where rates have declined up to 2.4% (95% CI: -3.9, -0.9) annually in Denmark (DNK) over the last decade. Additionally, declines in the incidence rate ratio (IRR) were observed for generations born after the 1930s, with an additional strong period effect seen around 2000 in United States and DNK. In contrast, IRRs increased among younger generations born after the 1950s in Japan and Belarus. Overall, the favorable trends in ovarian cancer incidence is likely due to the increase use of oral contraceptive pills, and changes in the prevalence of other reproductive risk and protective factors for ovarian cancer over the years studied. Changes in disease classifications and cancer registry practices may also partially contribute to the variation in ovarian cancer incidence rates. Thus, continuous cancer surveillance is essential to detect the shifting patterns of ovarian cancer.