American Journal of Epidemiology

Frandsen et al. respond to “A note on variable adjustments”

Abstract In response to M. Naylor’s valuable critique of our study “Polycystic Ovary Syndrome and Endometrial Cancer Risk: Results from a Nationwide Cohort Study,” we recognize the merit in the suggested areas. We recognize that including information on hormone therapy may have added to the understanding of the relationship between polycystic ovary syndrome (PCOS) and endometrial cancer risk. We further acknowledge the importance of the frequency of gynecological visits, which we have treated as a mediator rather than a confounder, given its role in reflecting PCOS severity and management. This approach, however, may introduce surveillance bias by influencing early cancer detection rates. Although the Danish health care system is fully subsidized, our exclusive use of hospital records may miss some PCOS cases managed in primary or specialized care settings outside the hospital, potentially leading to underestimation of the true association. Incorporating these variables in future studies could enhance analytical scope, though it would involve complex methodologies. We appreciate the comprehensive feedback, which underscores the necessity for further studies to elucidate the links between PCOS and endometrial cancer. These insights will inform future research and advance understanding in this area.

Childhood social and economic disadvantage and the risk of uterine fibroids among Black women

Abstract The association of childhood social and economic disadvantage (“disadvantage”) and uterine fibroid risk is understudied. We examined the association between disadvantage and fibroid incidence using standardized ultrasound exams at repeated visits, among 1230 participants 23 to 35 years of age in the Study of Environment, Lifestyle and Fibroids. Six disadvantage variables collected at baseline (ie, food insecurity, neighborhood safety, childhood income, mother’s educational attainment, household composition, and quiet bedroom for sleep) were evaluated separately, and using a latent class dichotomous (high/low) disadvantage variable. We also looked at possible modifying effects of a supportive childhood environment. Using Cox models to estimate incidence rate ratios (RR) and 95% confidence intervals (CI), with age as the time scale, we found little evidence for an increased risk of fibroids for any of the disadvantage variables or with the latent class construct. Having a supportive social environment in childhood had little impact on the associations between disadvantage and fibroid incidence. These findings are consistent with and expand upon prior findings from large studies with more limited data on social and economic disadvantage and less accurate data on timing of fibroid incidence.

Rare germline genetic variation in PAX8 transcription factor binding sites and susceptibility to epithelial ovarian cancer

Abstract Common genetic variation throughout the genome and rare coding variants identified to date explain about half of the inherited genetic component of epithelial ovarian cancer risk. It is likely that rare variation in the noncoding genome will explain some of the unexplained heritability, but identifying such variants is challenging. The primary problem is a lack of statistical power to identify individual risk variants by association, as power is a function of sample size, effect size, and allele frequency. Power can be increased by using burden tests, which test for the association of carriers of any variant in a specified genomic region. This has the effect of increasing the putative effect allele frequency. PAX8 is a transcription factor that plays a critical role in tumor progression, migration, and invasion. Furthermore, regulatory elements proximal to target genes of PAX8 are enriched for common ovarian cancer risk variants. We hypothesized that rare variation in PAX8 binding sites is also associated with ovarian cancer risk but unlikely to be associated with risk of breast, colorectal, or endometrial cancer. We have used publicly available, whole-genome sequencing data from the UK 100,000 Genomes Project to evaluate the burden of rare variation in PAX8 binding sites across the genome. Data were available for 522 ovarian cancers, 2984 breast cancers, 2696 colorectal cancers, 836 endometrial cancers, and 2253 noncancer controls. Active binding sites were defined using data from multiple PAX8 and H3K27 chromatin immunoprecipitation sequencing experiments. We found no association between the burden of rare variation in PAX8 binding sites (defined in several ways) and risk of ovarian, breast, or endometrial cancer. An apparent association with colorectal cancer was likely to be a technical artifact as a similar association was also detected for rare variation in random regions of the genome. Despite the null result, this study provides a proof-of-principle for using burden testing to identify rare, noncoding germline genetic variation associated with disease. Larger sample sizes available from large-scale sequencing projects, together with improved understanding of the function of the noncoding genome, will increase the potential of similar studies in the future.

Ovarian cancer risk prediction: a clinical epidemiology perspective

Abstract Ovarian cancer is a rare and highly heterogeneous disease usually detected at late stages when outcomes are poor. Population-based screening approaches have not been successful at reducing ovarian cancer mortality, but preventive bilateral salpingo-oophorectomy is highly effective at preventing ovarian cancer in high-risk populations. Ovarian cancer risk prediction models may allow identification of populations at increased risk of ovarian cancer for preventive interventions or targeted early detection. We propose a life-course approach to ovarian cancer risk prediction based on the time at which a risk model should be applied and the risk factors that are available. The discriminative ability of ovarian cancer risk prediction models published so far is limited, with areas under the curve ranging from 0.58 to 0.65 for different combinations of risk factors and genetic susceptibility markers. Currently proposed absolute risk thresholds for preventive surgery are around 4% lifetime risk. The absolute risk predicted by ovarian cancer risk models ranges from 0.6% to 2.5% lifetime risk in the general population, highlighting the need to improve ovarian cancer risk prediction models and evaluating new preventive approaches that can be offered to individuals at lower risk.

The independent and joint associations of hysterectomy and uterine fibroids or endometriosis with ovarian cancer incidence: results from a US-based cohort

Abstract Uterine fibroids and endometriosis may be associated with an increased risk of ovarian cancer. Less is known about the role of hysterectomy in these associations. We estimated the independent and joint associations of hysterectomy, fibroids, and endometriosis with ovarian cancer incidence in the prospective Sister Study cohort (2003-2009). We used time-varying Cox proportional hazards models to estimate covariate-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs). By the end of follow-up, 34% of 40 928 eligible participants had fibroids, 13% had endometriosis, and 7% had both. A total of 274 women developed ovarian cancer during follow-up (median = 12.3 years). In mutually adjusted models, fibroids (HR, 1.65; 95% CI, 1.28-2.12) and possibly endometriosis (HR, 1.16; 95% CI, 0.82-1.63) were positively associated with ovarian cancer. Hysterectomies (20% of participants) were also positively associated with ovarian cancer (HR, 1.29; 95% CI, 0.95-1.74). There was some evidence that hysterectomies may mitigate ovarian cancer risk among women with fibroids (HR, 0.83; 95% CI, 0.56-1.24) but not among women with endometriosis (HR, 1.20; 95% CI, 0.65-2.22). Identifying these joint associations adds to our understanding of ovarian cancer etiology and may help inform decisions about how women with fibroids, endometriosis, and hysterectomies are treated and surveilled for ovarian cancer. This article is part of a Special Collection on Gynecological Cancer.

Prediagnostic whole-blood cadmium and molybdenum associated with pancreatic cancer in an American cohort

Abstract Environmental exposures to elements such as cadmium might be contributing to the increasing incidence of pancreatic cancer. Few prospective studies have examined the association between trace elements and pancreatic ductal adenocarcinoma (PDAC). We conducted a nested case-control study in participants aged 55-74 years at baseline from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohort to examine the association between 12 trace elements measured in prediagnostic whole-blood samples and PDAC. From May 1998 through December 2014, 318 incident PDAC cases were identified during follow-up to 16.7 years. Of 636 control participants, 2 who were alive when each case patient was diagnosed were selected and matched by age (±5 years), sex, calendar date of blood sample collection (2-month blocks), and race and ethnic group. We used multivariable adjusted conditional logistic regression to calculate odds ratios (ORs) and 95% CIs. Cadmium and molybdenum were associated with PDAC (highest compared with lowest quintile: for cadmium, OR = 1.81 [95% CI, 01.12-2.95], P = .03 for trend; for molybdenum, OR = 0.50 [95% CI, 0.32-0.80], P = .02 for trend). The inverse molybdenum association was only observed among ever smokers (OR = 0.31 [95% CI, 0.17-0.58]; P = .003 for trend, P = .03 for interaction) with no association in never smokers. Lead, arsenic, and other trace elements were not associated with PDAC. Our results support that an increasing prediagnostic whole-blood level of cadmium is associated with increased PDAS risk, whereas that for molybdenum reduces PDAC risk.

Efficient risk-based collection of biospecimens in cohort studies: designing a prospective study of diagnostic performance for multicancer detection tests

Abstract In cohort studies, it can be infeasible to collect specimens on an entire cohort. For example, to estimate sensitivity of multiple multi-cancer detection (MCD) assays, we desire an extra 80 mL of cell-free DNA (cfDNA) blood, but this much extra blood is too expensive for us to collect on everyone. We propose a novel epidemiologic study design that efficiently oversamples those at highest baseline disease risk from whom to collect specimens, to increase the number of future cases with cfDNA blood collection. The variance reduction ratio from our risk-based subsample versus a simple random (sub)sample (SRS) depends primarily on the ratio of risk model sensitivity to the fraction of the cohort selected for specimen collection subject to constraining the risk model specificity. In a simulation where we chose 34% of the Prostate, Lung, Colorectal, and Ovarian Screening Trial cohort at highest risk of lung cancer for cfDNA blood collection, we could enrich the number of lung cancers 2.42-fold. The standard deviation of lung-cancer MCD sensitivity was 31%-33% reduced versus SRS. Risk-based collection of specimens on a subsample of the cohort could be a feasible and efficient approach to collecting extra specimens for molecular epidemiology.

Invited Commentary: Relationship Between Ovulation and Markers of Systemic Inflammation Versus Markers of Localized Inflammation

AbstractIn this issue of the American Journal of Epidemiology, Huang et al. (Am J Epidemiol. 2020;189(7):660–670) report an inverse relationship between lifetime ovulatory years (LOY) and circulating levels of C-reactive protein (CRP), counter to their hypothesis. CRP is a nonspecific systemic marker of chronic inflammation, and there is evidence that there are other drivers of systemic inflammation as measured by CRP. There also is a body of evidence supporting the possibility that ovulation may be responsible for a localized inflammatory response. Because ovarian cancer is the fifth most common cause of cancer death in women, preventive strategies are urgently needed. The findings of this report underscore the need for new research initiatives to determine the relationship between incessant ovulation and ovarian cancer risk in order to identify mechanisms of carcinogenesis.

Invited Commentary: Rethinking Cervical Cancer Elimination in Terms of Lifetime Risk Rather Than Arbitrarily Defined Age-Standardized Incidence Rates

Abstract In their accompanying article, Vänskä et al. (Am J Epidemiol. 2021;190(4):506–514) provide us with cohort lifetime risks of cervical cancer attributable to different types of human papillomavirus in Sweden. We argue that a standardized lifetime risk such as those calculated by Vänskä et al. might be a more appropriate public health target for cervical cancer elimination than age-standardized incidence rates. Age standardization to an arbitrary standard age distribution implies an implicit value choice regarding the weight of different age groups for which we find little moral justification. Conversely, a standardized lifetime risk uses standard life expectancy as a weight, corresponding to the likelihood that cervical cancer would impact a woman and prevent her from pursuing opportunities within a standard life span. Based on the data from Vänskä et al., a standardized lifetime risk of 129–250 cervical cancers per 100,000 women born could be an aspirational alternative public health target for cervical cancer elimination as a public health problem, complementary to the World Health Organization’s arbitrary draft target of 4 cervical cancers per 100,000 age-standardized woman-years.

Racial and ethnic differences in epithelial ovarian cancer risk: an analysis from the Ovarian Cancer Association Consortium

Abstract Limited estimates exist on risk factors for epithelial ovarian cancer (EOC) in Asian, Hispanic, and Native Hawaiian/Pacific Islander women. Participants in this study included 1734 Asian (n = 785 case and 949 control participants), 266 Native Hawaiian/Pacific Islander (n = 99 case and 167 control participants), 1149 Hispanic (n = 505 case and 644 control participants), and 24 189 White (n = 9981 case and 14 208 control participants) from 11 studies in the Ovarian Cancer Association Consortium. Logistic regression models estimated odds ratios (ORs) and 95% CIs for risk associations by race and ethnicity. Heterogeneity in EOC risk associations by race and ethnicity (P ≤ .02) was observed for oral contraceptive (OC) use, parity, tubal ligation, and smoking. We observed inverse associations with EOC risk for OC use and parity across all groups; associations were strongest in Native Hawaiian/Pacific Islander and Asian women. The inverse association for tubal ligation with risk was most pronounced for Native Hawaiian/Pacific Islander participants (odds ratio (OR) = 0.25; 95% CI, 0.13-0.48) compared with Asian and White participants (OR = 0.68 [95% CI, 0.51-0.90] and OR = 0.78 [95% CI, 0.73-0.85], respectively). Differences in EOC risk factor associations were observed across racial and ethnic groups, which could be due, in part, to varying prevalence of EOC histotypes. Inclusion of greater diversity in future studies is essential to inform prevention strategies. This article is part of a Special Collection on Gynecological Cancers.

Updated CP*Trends: an online tool to compare cohort and period trends across cancer sites

Abstract CP*Trends is a widely used Surveillance, Epidemiology, and End Results Program (SEER) website used to explore temporal effects of period and cohort on cancer incidence and mortality. It provides a graphical display of smoothed rates, and a C-P score that helps to assess the magnitude of the effect of cohort and period. This update provides results for African Americans and Whites. The C-P score has an intrinsic bias favoring cohort because there are many more cohorts than periods. An adjusted C-P score removes some of this advantage. Bootstrap confidence intervals are given, which allow one to see the effects of different sample sizes on the model results. Finally, users may control window size used in the smoothing algorithm, which helps to avoid oversmoothing or masking of trends. The method is illustrated using data on cervical cancer incidence trends for African Americans and Whites, 1975-2018. Rates are higher for African Americans, and both races have contributions for cohort. However, the period effect is only strongly evident in Whites. Visual inspection of White trends suggests possible differences for those older and younger than age 50. These methods are applied in an interactive website displaying incidence and mortality trends for over 20 cancer sites in the United States.

Polycystic ovary syndrome and endometrial cancer risk: results from a nationwide cohort study

Abstract Most previous studies found an elevated risk of endometrial cancer among women with polycystic ovary syndrome (PCOS). However, these had highly varying methods for ascertainment of PCOS diagnoses and limitations such as few exposed women and short follow-up. In this cohort study, we investigated the association between PCOS and endometrial cancer among women born in Denmark between January 1, 1940, and December 31, 1993 (n = 1 719 121). Data in this study, including PCOS and endometrial cancer diagnoses and covariates, were derived from nationwide registers. We used Cox proportional hazards regression models to calculate hazard ratios (HRs) and 95% CIs. A total of 7862 endometrial cancer cases were identified during 23.7 years of follow-up (IQR, 37.7-61.9). We found an increased risk of endometrial cancer among women with PCOS compared with women without PCOS (HR = 3.02; 95% CI, 2.03-4.49). The risk was increased for premenopausal women (HR = 5.82; 95% CI, 3.64-9.30), whereas no marked association was seen for postmenopausal women. However, for postmenopausal women, results were limited by few cases and young age at the end of follow-up. Mounting evidence of an increased risk for endometrial cancer among women with PCOS reinforces the need for prevention and early detection. This article is part of a Special Collection on Gynecological Cancers.

Neighborhood disorder and ovarian cancer survival in Black women

Abstract Ovarian cancer (OC) is the fifth leading cause of cancer mortality among women in the US. Black women experience significantly lower OC survival than White women. Evidence suggests that this disparity is not solely the result of barriers to healthcare access but may also be impacted by other factors, including neighborhood social characteristics. To investigate this further, this study developed an approach for remotely estimating the degree of physical disorder (PD) in neighborhoods using structured audits of Google Street View imagery for participants in the AACES, a multi-site population-based study of Black women newly diagnosed with OC. We then assessed whether neighborhood disadvantage (ND) and PD were associated with overall survival. We fit Weibull accelerated failure time models to assess the association of both PD and ND with survival among 471 Black women with OC (n = 317 deaths). Both PD (event time ratio (ETR), 0.99; 95% CI, 0.98, 1.00) and Area Deprivation Index (ETR: 0.96, 95% CI: 0.94, 1.00) were associated with shorter survival. The results suggest that both physical and social neighborhood characteristics may impact survival in woman with OC, but further research is warranted.

Health care access domains and treatment as mediators of ovarian cancer racial disparities in survival: a structural equation modeling analysis in SEER-Medicare

Abstract Racial differences in health care access (HCA) may contribute to disparities in ovarian cancer (OC) survival. We used structural equation models (SEMs) to examine associations between race and HCA domains (affordability, availability, accessibility) in relation to overall and OC-specific mortality. Non-Hispanic (NH)–Black and non-Black (Hispanic, NH-White) women diagnosed with OC in 2008-2015 were identified from Surveillance, Epidemiology, and End Results–Medicare. Cox proportional hazards regression was used to conduct mediation analysis for associations between race and HCA domains with overall and OC-specific mortality. SEM models adjusting for demographic and clinical covariates were used to estimate hazard ratios (HRs) and 95% CIs. A total of 4629 eligible patients with OC were identified, including 255 (5.5%) patients who were NH-Black. In SEM adjusting for demographic, clinical, and HCA latent variables, there was a total effect of NH-Black race on overall (HR, 1.11, 95% CI, 1.03-1.19) and OC-specific mortality (HR, 1.16; 95% CI, 1.08, 1.24), which was primarily driven by a direct effect. There was a modest indirect association between NH-Black race and mortality through decreased treatment receipt, though not through HCA. There is a need for studies investigating additional social and biological mechanisms that contribute to worse cancer survival among NH-Black patients. This article is part of a Special Collection on Gynecological Cancer.

Endometrial cancer survival in populations of African descent

Abstract To examine whether the endometrial cancer (EC) survival disadvantage among Black populations is US-specific, a comparison between African-descent populations from different countries with a high development index is warranted. We analyzed 28 213 EC cases from cancer registries in Florida (2005-2018) and the French Caribbean islands of Martinique (2005-2018) and Guadeloupe (2008-2018) combined. Kaplan-Meier and all-cause Cox proportional hazards models were used to compare survival. Models were stratified by EC histology type and the main predictor examined was race/ethnicity (non-Hispanic White [NHW] and no-Hispanic Black [NHB] women in the United States versus Black women residing in the Caribbean). For endometrioid and nonendometrioid EC, after adjusting for age, histology, stage at diagnosis, receipt of surgery, period of diagnosis, and poverty level, US NHB women and Caribbean Black women had a higher risk of death relative to US NHW women. There was no difference between US NHB and Caribbean Black women (hazard ratio [HR] = 1.07; 95% CI, 0.88-1.30) with endometrioid EC. However, Caribbean Black women with nonendometrioid carcinomas had a 40% higher risk of death (HR = 1.40; 95% CI, 1.13-1.74) than US NHB women. The low EC survival among US Black women extends to foreign populations of African descent. For the aggressive nonendometrioid ECs, survival among Caribbean Black women outside of the United States is considerably worse. This article is part of a Special Collection on Gynecological Cancers.

Associations between common contraceptive use and circulating inflammatory biomarkers

Abstract Ovarian cancer incidence has declined in recent decades, due in part to oral contraceptive (OC) use and tubal ligation. However, intrauterine device (IUD) use has increasingly replaced OC use. As ovarian cancer is an inflammation-related disease, we examined the association of OC use, IUD use, and tubal ligation with plasma levels of C-reactive protein (CRP), interleukin 6, and soluble tumor necrosis factor α receptor 2 in the Nurses’ Health Study (NHS) and NHSII. After adjusting for reproductive, hormonal, and lifestyle factors and mutual adjustment for other methods of contraception, there were no differences in inflammatory markers between ever and never use of each method. However, CRP levels decreased from an average of 30.4% (95% CI, –53.6 to 4.4) with every 5 years since initial IUD use (P-trend = .03), while CRP increased an average of 9.9% (95% CI, 5.7, 14.3) with every 5 years of use of OC (P-trend < .0001) as well as differences by body mass index and menopausal status. Our results suggest IUD use and tubal ligation are not associated with higher circulating inflammatory markers long term, although long duration of OC use may increase generalized inflammation, which may in part explain why its protective effect wanes over time. This article is part of a Special Collection on Gynecological Cancer.

Risk factors for second primary cancer in a prospective cohort of endometrial cancer survivors: an Alberta Endometrial Cancer Cohort Study

Abstract We examined associations between modifiable and nonmodifiable cancer-related risk factors measured at endometrial cancer diagnosis and during early survivorship (~3 years postdiagnosis) with second primary cancer (SPC) risk among 533 endometrial cancer survivors in the Alberta Endometrial Cancer Cohort using Fine and Gray subdistribution hazard models. During a median follow-up of 16.7 years (IQR, 12.2-17.9), 89 (17%) participants developed an SPC; breast (29%), colorectal (13%), and lung (12%) cancers were the most common. Dietary glycemic load before endometrial cancer diagnosis (≥90.4 vs < 90.4 g/day: subhazard ratio [sHR] = 1.71; 95% CI, 1.09-2.69), as well as older age (≥60 vs < 60 years: sHR = 2.48; 95% CI, 1.34-4.62) and alcohol intake (≥2 drinks/week vs none: sHR = 3.81; 95% CI, 1.55-9.31) during early survivorship, were associated with increased SPC risk. Additionally, reductions in alcohol consumption from prediagnosis to early survivorship significantly reduced SPC risk (sHR = 0.34; 95% CI, 0.14-0.82). With 1 in 6 survivors developing an SPC, further investigation of SPC risk factors and targeted surveillance options for high-risk survivors could improve long-term health outcomes in this population. Reductions in dietary glycemic load and alcohol intake from prediagnosis to early survivorship showed promising risk reductions for SPCs and could be important modifiable risk factors to target among endometrial cancer survivors. This article is part of a Special Collection on Gynecological Cancer.

Derivation of three occupational status measures in young Black women: the study of environment, lifestyle, and fibroids

Abstract Socioeconomic Status (SES), a complex, multidimensional construct, is commonly assessed using education or combinations of education, income, and/or occupation. Though epidemiological studies tend to collect occupational data, focus on occupation-based SES measures has been limited. We describe a method to process and derive occupational status measures and then compare them for application in epidemiological research. Using US Census Bureau’s Industry and Occupational Codes to standardize self-reported baseline occupational and industry data for 1053 Black/African American participants in the Study of Environment, Lifestyle & Fibroids, we derived 3 established occupational status measures: Hauser Warren Socioeconomic Index (HWSEI), Nam-Powers-Boyd Occupational Status Score (NPBOSS90) and Nakao-Treas Prestige Score (PRENT). All 3 scores range from 0 to 100 where higher scores reflect higher status. Participants averaged in age of 29 years with 62% employed at baseline, most commonly in Customer Service. Median scores varied across Hauser Warren Socioeconomic Index (30.3), Nam-Powers-Boyd Occupational Status Score (43.2), and PRENT (41.7). The occupational status derivation process contributes to epidemiological methods by providing guidance to those seeking to elucidate the SES-health relationship beyond merely income or education. If education and income are already being collected, PRENT, which does not incorporate these factors, may offer unique insights.

Trends in endometrial cancer incidence in the United States by race/ethnicity and age of onset from 2000 to 2019

Abstract Endometrial cancer is one of few cancers that has continued to rise in incidence over the past decade, with disproportionate increases in adults younger than 50 years old. We used data from the Surveillance, Epidemiology, and End Results Registry (2000-2019) to examine endometrial cancer incidence trends by race/ethnicity and age of onset among women in the United States. Case counts and proportions, age-adjusted incidence rates (per 100 000), and average annual percent changes were calculated by race/ethnicity, overall and stratified by age of onset (early vs late). We found a disproportionate increase in endometrial cancer incidence among women of color, for both early and late onset endometrial cancer. The highest increases in early onset endometrial cancer (<50 years old) were observed among American Indian/Alaska Native women (4.8), followed by Black (3.3), Hispanic/Latina (3.1), and Asian and Pacific Islander women (2.4), whereas White women (0.9) had the lowest increase. Late onset (≥50 years old) endometrial cancer incidence followed a similar pattern, with the greatest increases for women of color. The increasing burden of endometrial cancer among women of color, particularly those younger than 50 years old, is a major public health problem necessitating further research and clinical efforts focused on health equity. This article is part of a Special Collection on Gynecological Cancer.

Using electronic health record data to identify incident uterine fibroids and endometriosis within a large, urban academic medical center: a validation study

Abstract Electronic health records (EHRs) present opportunities to study uterine fibroids and endometriosis within diverse populations. When using EHR data, it is important to validate outcome classification via diagnosis codes. We performed a validation study of 3 approaches ([1] International Classification of Diseases-10 (ICD-10) code alone, [2] ICD-10 code + diagnostic procedure, and [3] ICD-10 code + all diagnostic information) to identify incident uterine fibroids and endometriosis patients among n = 750 NYU Langone Health 2016-2023. Chart review was used to determine the true diagnosis status. When using a binary classification system (incident vs nonincident patient), Approaches 2 and 3 had higher positive predictive values (PPVs) for uterine fibroids (0.86 and 0.87 vs 0.78) and for endometriosis (0.70 and 0.73 vs 0.66), but Approach 1 outperformed the other 2 in negative predictive values (NPVs) for both outcomes. When using a 3-level classification system (incident vs prevalent vs disease-free patients), PPV for prevalent patients was low for all approaches, while PPV/NPV of disease-free patients was generally above 0.8. Using ICD-10 codes alone yielded higher NPVs but resulted in lower PPVs compared with the other approaches. Continued validation of uterine fibroids/endometriosis EHR studies is warranted to increase research into these understudied gynecologic conditions.

Epigenome-wide differential methylation and differential variability as predictors of high-grade cervical intraepithelial neoplasia (CIN2+)

Abstract CpG site methylation patterns have potential to improve differentiation of high-grade screening-detected cervical abnormalities. We assessed CpG differential methylation (DM) and differential variability (DV) in high-grade (CIN2+) vs low-grade (≤ CIN1) lesions. In ≤ CIN1 (n = 117) and CIN2+ (n = 31) samples, cervical sample DNA underwent testing with Illumina HumanMethylation arrays. We assessed DM and DV of CpG methylation M-values among 9 cervical cancer–associated genes. We fit CpG-specific linear models and estimated empirical Bayes standard errors and false discovery rates (FDRs). An exploratory epigenome-wide association study (EWAS) aimed to detect novel DM and DV CpGs (FDR < 0.05) and Gene Ontology (GO) term enrichment. Compared to ≤ CIN1, CIN2+ exhibited greater methylation at CCNA1 cluster 1 (M-value difference 0.24; 95% CI, 0.04-0.43) and RARB cluster 2 (0.16; 95% CI, 0.05-0.28), and lower methylation at CDH1 cluster 1 (–0.15; 95% CI, –0.26 to –0.04). CIN2+ exhibited lower variability at CDH1 cluster 2 (variation difference –0.24; 95% CI, –0.41 to –0.05) and FHIT cluster 1 (–0.30; 95% CI, –0.50 to –0.09). EWAS detected 3534 DM and 270 DV CpGs. Forty-four GO terms were enriched with DM CpGs related to transcriptional, structural, developmental, and neuronal processes. Methylation patterns may help triage screening-detected cervical abnormalities and inform US screening algorithms. This article is part of a Special Collection on Gynecological Cancer.

Adjuvant therapy in early uterine serous carcinoma

Abstract Uterine serous carcinoma (USC) is a rare diagnosis but is associated with high mortality. There are limited data to guide adjuvant treatment decisions in early-stage disease. The purpose of this study is to evaluate the impact of adjuvant therapy on recurrence-free survival (RFS) and overall survival (OS) in early-stage USC. Patients with stage I and II USC treated at a single institution from January 2006 to December 2019 were identified. Demographic, clinicopathologic, treatment, and outcome data were collected. Data were compared using descriptive statistics. Survival analyses were performed using Kaplan-Meier and Cox proportional hazards methods. Ninety-four patients were identified. Median follow-up time was 33.5 months. The median age was 68 years (range, 49-87), the majority of patients were White (n = 78, 83.0%), and the median BMI was 30.7 (range, 14.2-57.3). Minimally invasive surgical staging was performed in 59.6% of cases (n = 56). Most patients had stage IA disease (n = 70, 74.5%). Most patients (n = 79, 84.0%) received adjuvant therapy, and a majority of patients received a combination of systemic chemotherapy and radiation therapy (n = 55, 58.5%), with the most common combination being chemotherapy plus vaginal brachytherapy (n = 42, 44.7%). Most patients (n = 77, 81.9%) remain without evidence of disease, while 17 patients (18.1%) have recurred. Patients receiving 6 cycles of adjuvant chemotherapy experienced improved OS (P = .004) and improved RFS (P = .02) compared to those receiving no adjuvant chemotherapy. Patients with early-stage USC who received 6 cycles of adjuvant chemotherapy had significantly improved OS and RFS when compared to those patients who did not receive adjuvant chemotherapy. This article is part of a Special Collection on Gynecological Cancer.

The impact of adjusting for hysterectomy prevalence on cervical cancer incidence rates and trends among women aged 30 years or older—United States, 2001-2019

Abstract Hysterectomy protects against cervical cancer when the cervix is removed. However, measures of cervical cancer incidence often fail to exclude women with a hysterectomy from the population-at-risk denominator, underestimating and distorting disease burden. In this study, we estimated hysterectomy prevalence from the Behavioral Risk Factor Surveillance System surveys to remove the women who were not at risk of cervical cancer from the denominator and combined these estimates with the US Cancer Statistics data. From these data, we calculated age-specific and age-standardized incidence rates for women aged >30 years from 2001-2019, adjusted for hysterectomy prevalence. We calculated the difference between unadjusted and adjusted incidence rates and examined trends by histology, age, race and ethnicity, and geographic region using joinpoint regression. The hysterectomy-adjusted cervical cancer incidence rate from 2001-2019 was 16.7 per 100 000 women—34.6% higher than the unadjusted rate. After adjustment, incidence rates were higher by approximately 55% among Black women, 56% among those living in the East South Central division, and 90% among women aged 70-79 and ≥80 years. These findings underscore the importance of adjusting for hysterectomy prevalence to avoid underestimating cervical cancer incidence rates and masking disparities by age, race, and geographic region. This article is part of a Special Collection on Gynecological Cancers.

Differing Age-Specific Cervical Cancer Incidence Between Different Types of Human Papillomavirus: Implications for Predicting the Impact of Elimination Programs

Abstract The elimination of cervical cancer rests on high efficacy of human papillomavirus (HPV) vaccines. The HPV type distribution among cases of invasive cervical cancer (ICC) is used to make predictions about the impact of eliminating different types of HPV, but accumulating evidence of differences in age-specific cancer incidence by HPV type exists. We used one of the largest population-based series of HPV genotyping of ICCs (n = 2,850; Sweden, 2002–2011) to estimate age-specific ICC incidence by HPV type and obtain estimates of the cancer-protective impact of the removal of different HPV types. In the base case, the age-specific ICC incidence had 2 peaks, and the standardized lifetime risk (SLTR, the lifetime number of cases per birth cohort of 100,000 females) for HPV-positive ICC was 651 per 100,000 female births. In the absence of vaccine types HPV 16 and HPV 18, the SLTR for ICC was reduced to 157 per 100,000 female births (24% of HPV-positive SLTR). Elimination of all 9 types that can currently be vaccinated against reduced the remaining SLTR to 47 per 100,000 female births (7%), the remaining ICC incidence only slowly increasing with age. In conclusion, after elimination of vaccine-protected HPV types, very few cases of ICC will be left, especially among fertile, reproductive-age women.

Predicting Cohort-Specific Cervical Cancer Incidence From Population-Based Surveys of Human Papilloma Virus Prevalence: A Worldwide Study

Abstract Predictions of cervical cancer burden and the impact of measures taken to control this cancer are usually data-demanding and based on complex assumptions. We propose a predictive method (called PANDORA) based on human papillomavirus (HPV) prevalence, measured 1993–2008, and cervical cancer incidence (CCI), measured 1993–2012, in the same birth cohorts from different worldwide locations, informed by data on age at detection of high-risk HPV and sexual debut. The model can predict CCI among high-risk HPV–positive women and predict CCI up to 14 years following high-risk HPV detection. We found CCI to increase during the 14 years following high-risk HPV detection in unscreened women aged <35 years but to remain mainly constant among women ≥35 years. Age at sexual debut was a significant modifier of CCI. Using our model, we accurately reproduced CCI among high-risk HPV–positive women as observed in cohort studies and in the general population of multiple countries. We also predicted the annual number of cervical cancer cases and CCI in locations with HPV prevalence data but no cancer registry. These findings could inform cervical cancer control programs in settings without cancer registries, as they can be used to predict future cervical cancer burden from population-based surveys of HPV prevalence.

Fiber-type prebiotics and gynecological and breast cancers risk: the PrebiotiCa study

Abstract Prebiotics may influence the risk of hormone-related female cancers by modulating the gut microbiota involved in estrogen metabolism. We evaluated the association of fiber-type prebiotic intake with breast, endometrial, and ovarian cancers. Data derived from a network of Italian hospital-based case-control studies (1991-2006), including 2560 cases of cancer of the breast (n = 2588 control participants), 454 of the endometrium (n = 908 control participants), and 1031 of the ovary (n = 2411 control participants). Inulin-type fructans and selected fructo-oligosaccharides (namely, nystose, kestose, and 1F-β-fructofuranosylnystose) and galacto-oligosaccharides (namely, raffinose and stachyose) were quantified in food products via laboratory analyses. Prebiotic intake was estimated by multiplying intake according to food frequency questionnaire responses by the foods’ prebiotic content. Odds ratios (ORs) and the corresponding 95% CIs were derived by multiple logistic regression models. Nystose intake was marginally directly associated with breast (for quartile 4 vs quartile 1: OR = 1.20; 95% CI, 1.00-1.45), ovarian (OR = 1.39; 95% CI, 1.04-1.84), and endometrial (OR = 1.32; 95% CI, 0.85-2.03) cancer risk. High amounts of 1F-β-fructofuranosylnystose intake were inversely associated with ovarian cancer (OR = 0.67; 95% CI, 0.52-0.85). Inulin-type fructans, kestose, raffinose, and stachyose were not associated with the 3 cancers. The intake of most fiber-type prebiotics was not appreciably and consistently associated with breast, endometrial, and ovarian cancer risks. This article is part of a Special Collection on Gynecological Cancer.

Indicators of cure for women living after uterine and ovarian cancers: a population-based study

Abstract This study aims to estimate long-term survival, cancer prevalence, and several cure indicators for Italian women with gynecological cancers. Thirty-one cancer registries, representing 47% of the Italian female population, were included. Mixture cure models were used to estimate net survival, cure fraction, time to cure (when 5-year conditional net survival becomes > 95%), cure prevalence (women who will not die of cancer), and already cured (living longer than time to cure). In 2018, 0.4% (121 704) of Italian women were alive after diagnosis of corpus uteri cancer, 0.2% (52 551) after cervical cancer, and 0.2% (52 153) after ovarian cancer. More than 90% of patients with uterine cancers and 83% with ovarian cancer will not die from their neoplasm (cure prevalence). Women with gynecological cancers have a residual excess risk of death <5% at 5 years after diagnosis. The cure fraction was 69% for corpus uteri, 32% for ovarian, and 58% for cervical cancer patients. Time to cure was ≤10 years for women with gynecological cancers aged <55 years; 74% of patients with cervical cancer, 63% with corpus uteri cancer, and 55% with ovarian cancer were already cured. These results can contribute to improving follow-up programs for women with gynecological cancers and supporting efforts against discrimination of already cured ones. This article is part of a Special Collection on Gynecological Cancers.

Prospective Analyses of Sedentary Behavior in Relation to Risk of Ovarian Cancer

Abstract We examined the association of sedentary behavior with risk of ovarian cancer overall, by tumor subtype, and by participant characteristics in the Nurses’ Health Study (NHS) and Nurses’ Health Study II (NHS II). A total of 69,558 NHS participants (1992–2016) and 104,130 NHS II participants (1991–2015) who reported on time spent sitting at home, at work, and while watching television were included in the analysis, which included 884 histologically confirmed ovarian cancer cases. Multivariable-adjusted Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for ovarian cancer by sitting time (no mutual adjustment for individual sitting types in primary analyses). We examined potential heterogeneity by tumor histological type (type I or II), body mass index (weight (kg)/height (m)2; < 25 or ≥25), and total physical activity (<15 or ≥15 metabolic equivalent of task–hours/week). We observed an increased risk of ovarian cancer for women who sat at work for 10–19 hours/week (HR = 1.25, 95% CI: 1.04, 1.51) and ≥20 hours/week (HR = 1.40, 95% CI: 1.14, 1.71) versus <5 hours/week. This association did not vary by body mass index, physical activity, or histotype (P for heterogeneity ≥ 0.43). No associations were observed for overall sitting, sitting while watching television, or other sitting at home. Longer sitting time at work was associated with elevated risk of ovarian cancer. Further investigations are required to confirm these findings and elucidate underlying mechanisms.

Caregiver burden and risk of epithelial ovarian cancer in the Nurses’ Health Studies

Abstract Psychosocial stress may increase ovarian cancer risk and accelerate disease progression. We examined the association between caregiver burden, a common stressor, and risk of epithelial ovarian cancer. We prospectively followed 67 724 women in the Nurses’ Health Study (1992-2012) and 70 720 women in the Nurses’ Health Study II (2001-2009) who answered questions on informal caregiving (ie, caregiving outside of work). Women who reported no informal caregiving were considered noncaregivers, while, among women who provided care outside of work, caregiver burden was categorized by time spent caregiving and perceived stress from caregiving. For the 34% of women who provided informal care for ≥15 hours per week, 42% described caregiving as moderately to extremely stressful. Pooled multivariate analyses indicated no difference in ovarian cancer risk for women providing ≥15 hours of care per week compared to noncaregivers (hazard ratio [HR] = 0.96; 95% confidence interval [CI], 0.79-1.18), and no association was evident for women who reported moderate or extreme stress from caregiving compared to noncaregivers (HR = 0.96; 95% CI, 0.75-1.22). Together with prior work evaluating job strain and ovarian cancer risk, our findings suggest that, when evaluating a stressor’s role in cancer risk, it is critical to consider how the stressor contributes to the overall experience of distress. This article is part of a Special Collection on Gynecological Cancer.

Effectiveness of 1, 2, and 3 Doses of Human Papillomavirus Vaccine Against High-Grade Cervical Lesions Positive for Human Papillomavirus 16 or 18

Abstract Before 2016, human papillomavirus (HPV) vaccination was recommended on a 3-dose schedule. However, many vaccine-eligible US females received fewer than 3 doses, which provided an opportunity to evaluate the real-world vaccine effectiveness (VE) of 1, 2, and 3 doses. We analyzed data on cervical intraepithelial neoplasia (CIN) grades 2–3 and adenocarcinoma in situ (designated CIN2+) from the HPV Vaccine Impact Monitoring Project (HPV-IMPACT; 2008–2014). Archived tissue from CIN2+ lesions was tested for 37 types of HPV. Women were classified by number of doses received ≥24 months before CIN2+ detection. Using a test-negative design, VE was estimated as 1 minus the adjusted odds ratio from a logistic regression model that compared vaccination history for women whose lesions tested positive for HPV-16/18 (vaccine-type cases) with that for women who had all other CIN2+ lesions (controls). Among 3,300 women with available data on CIN2+, typing results, and vaccine history, 1,561 (47%) were HPV-16/18–positive, 136 (4%) received 1 dose of HPV vaccine, 108 (3%) received 2 doses, and 325 (10%) received 3 doses. Adjusted odds ratios for vaccination with 1, 2, and 3 doses were 0.53 (95% confidence interval (CI): 0.37, 0.76; VE = 47%), 0.45 (95% CI: 0.30, 0.69; VE = 55%), and 0.26 (95% CI: 0.20, 0.35; VE = 74%), respectively. We found significant VE against vaccine-type CIN2+ after 3 doses of HPV vaccine and lower but significant VE with 1 or 2 doses.

Associations of Height With the Risks of Colorectal and Endometrial Cancer in Persons With Lynch Syndrome

Abstract People with Lynch syndrome (LS), who carry a pathogenic mutation in a DNA mismatch repair gene, have increased risks of colorectal cancer (CRC) and endometrial cancer (EC). A high reported variability in cancer risk suggests the existence of factors that modify cancer risk for persons with LS. We aimed to investigate the associations between height and CRC and EC risk for persons with LS using data from 2 large studies. Information on 1,115 men and 1,553 women with LS from the Colon Cancer Family Registry (1998–2007) and the GEOLynch Cohort Study (2006–2017) was harmonized. We used weighted Cox proportional hazards regression models with age on the time axis to estimate adjusted hazard ratios and 95% confidence intervals for each 5-cm increment in self-reported height. CRC was diagnosed in 947 persons during 65,369 person-years of observation, and 171 women were diagnosed with EC during 39,227 person-years. Height was not associated with CRC for either men (per 5-cm increment, hazard ratio (HR) = 1.00, 95% confidence interval (CI): 0.91, 1.11) or women (per 5-cm increment, HR = 1.01, 95% CI: 0.92, 1.11), nor was height associated with EC (per 5-cm increment, HR = 1.08, 95% CI: 0.94, 1.24). Hence, we observed no evidence for an association of height with either CRC or EC among persons with LS.