Rudolf Kaaks

Prospective evaluation of 92 protein biomarkers for early detection of endometrial cancer

Abstract The human epididymis protein 4 (HE4) remains the best available endometrial cancer (EC) biomarker; however, its discrimination between cases and cancer‐free individuals is limited and might be improved when combined with other protein markers. We evaluated the discrimination capacity of 92 proteins as potential early detection biomarkers for EC in nested case–control studies in the European Prospective Investigation into Cancer and Nutrition (EPIC) (63 cases, 123 controls) and Janus (75 cases, 146 controls) cohorts, evaluating blood samples taken ≤2 years prior to diagnosis. Proteins were measured with the Olink Target 96 Oncology II panel assays. Areas under the receiver operating characteristic curves (AUCs) were calculated using logistic regression. The discrimination between cases and controls of top‐performing proteins was modest (EPIC: HE4, CA125, CAIX, and S100A4; Janus: HE4, CA125, FURIN, CXCL13, and IL6; AUC range: 0.65 [S100A4], 0.76 [HE4, EPIC] within 0 to <12 months of blood collection) and decreased as the time between blood draw and cancer diagnosis increased (12–24 months AUC range: 0.49 [S100A4], 0.69 [CA125, Janus]). The combination of these other markers with HE4 did not improve discrimination. HE4 and other candidate proteins had limited discrimination between EC cases and controls and hence do not appear to be useful for early detection of this disease in women at average population risk.

Metabolically Defined Body Size Phenotypes and Risk of Endometrial Cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC)

Abstract Background: Obesity is a risk factor for endometrial cancer but whether metabolic dysfunction is associated with endometrial cancer independent of body size is not known. Methods: The association of metabolically defined body size phenotypes with endometrial cancer risk was investigated in a nested case–control study (817 cases/ 817 controls) within the European Prospective Investigation into Cancer and Nutrition (EPIC). Concentrations of C-peptide were used to define metabolically healthy (MH; <1st tertile) and metabolically unhealthy (MU; ≥1st tertile) status among the control participants. These metabolic health definitions were combined with normal weight (NW); body mass index (BMI)<25 kg/m2 or waist circumference (WC)<80 cm or waist-to-hip ratio (WHR)<0.8) and overweight (OW; BMI≥25 kg/m2 or WC≥80 cm or WHR≥0.8) status, generating four phenotype groups for each anthropometric measure: (i) MH/NW, (ii) MH/OW, (iii) MU/NW, and (iv) MU/OW. Results: In a multivariable-adjusted conditional logistic regression model, compared with MH/NW individuals, endometrial cancer risk was higher among those classified as MU/NW [ORWC, 1.48; 95% confidence interval (CI), 1.05–2.10 and ORWHR, 1.68; 95% CI, 1.21–2.35] and MU/OW (ORBMI, 2.38; 95% CI, 1.73–3.27; ORWC, 2.69; 95% CI, 1.92–3.77 and ORWHR, 1.83; 95% CI, 1.32–2.54). MH/OW individuals were also at increased endometrial cancer risk compared with MH/NW individuals (ORWC, 1.94; 95% CI, 1.24–3.04). Conclusions: Women with metabolic dysfunction appear to have higher risk of endometrial cancer regardless of their body size. However, OW status raises endometrial cancer risk even among women with lower insulin levels, suggesting that obesity-related pathways are relevant for the development of this cancer beyond insulin. Impact: Classifying women by metabolic health may be of greater utility in identifying those at higher risk for endometrial cancer than anthropometry per se.

Theoretical potential for endometrial cancer prevention through primary risk factor modification: Estimates from the EPIC cohort

Endometrial cancer (EC) incidence rates vary ~10‐fold worldwide, in part due to variation in EC risk factor profiles. Using an EC risk model previously developed in the European EPIC cohort, we evaluated the prevention potential of modified EC risk factor patterns and whether differences in EC incidence between a European population and low‐risk countries can be explained by differences in these patterns. Predicted EC incidence rates were estimated over 10 years of follow‐up for the cohort before and after modifying risk factor profiles. Risk factors considered were: body mass index (BMI, kg/m2), use of postmenopausal hormone therapy (HT) and oral contraceptives (OC) (potentially modifiable); and, parity, ages at first birth, menarche and menopause (environmentally conditioned, but not readily modifiable). Modeled alterations in BMI (to all ≤23 kg/m2) and HT use (to all non‐HT users) profiles resulted in a 30% reduction in predicted EC incidence rates; individually, longer duration of OC use (to all ≥10 years) resulted in a 42.5% reduction. Modeled changes in not readily modifiable exposures (i.e., those not contributing to prevention potential) resulted in ≤24.6% reduction in predicted EC incidence. Women in the lowest decile of a risk score based on the evaluated exposures had risk similar to a low risk countries; however, this was driven by relatively long use of OCs (median = 23 years). Our findings support avoidance of overweight BMI and of HT use as prevention strategies for EC in a European population; OC use must be considered in the context of benefits and risks.

A Prospective Study Consortium for the Discovery and Validation of Early Detection Markers for Ovarian Cancer – Baseline Findings for CA125

Abstract Purpose: Epithelial ovarian cancer (EOC) is a lethal malignancy. Cancer antigen 125 (CA125), the “best” available marker for detecting EOC, has insufficient sensitivity and specificity for earlier-stage disease and is not a meaningful screening tool, motivating the search for further biomarkers. Cancer biomarker discovery is enhanced by “omics” technologies. Discovery studies for EOC biomarkers should be conducted in prediagnosis blood samples from prospective cohorts to maximize the likelihood of identifying markers that can detect disease before usual diagnosis and in earlier disease stage while reducing methodologic biases. Experimental Design: Individual cohorts with prediagnosis blood samples have insufficient sample size for such studies. Thus, we established “Prospective Early Detection Consortium for Ovarian Cancer” (“PREDICT”)—a collaboration of nine prospective studies—to assemble a sufficient number of EOC cases with blood samples collected ≤18 months before diagnosis plus controls. The 457 cases and 1,687 controls have circulating CA125 measured using a clinical assay. Results: The discrimination capacity for single CA125 measurements in samples collected <6 months prior to diagnosis was high (AUC; PREDICT overall = 0.92; range across cohorts of nonpregnant individuals = 0.89–0.98) and declined with extended time between blood collection and diagnosis. Between-cohort variability in CA125 levels and predictive performance was observed. Conclusions: Ongoing investigations in PREDICT are evaluating the early detection potential of tumor-associated autoantibodies and miRNAs using CA125 as a benchmark. PREDICT is a well-characterized resource for identifying and validating detection markers for EOC that may then be used in multimodal screening as a complement to CA125 and combined with imaging.

Lifetime ovulatory years and risk of epithelial ovarian cancer: a multinational pooled analysis

Abstract Background The role of ovulation in epithelial ovarian cancer (EOC) is supported by the consistent protective effects of parity and oral contraceptive use. Whether these factors protect through anovulation alone remains unclear. We explored the association between lifetime ovulatory years (LOY) and EOC. Methods LOY was calculated using 12 algorithms. Odds ratios (ORs) and 95% confidence intervals (CIs) estimated the association between LOY or LOY components and EOC among 26 204 control participants and 21 267 case patients from 25 studies. To assess whether LOY components act through ovulation suppression alone, we compared beta coefficients obtained from regression models with expected estimates assuming 1 year of ovulation suppression has the same effect regardless of source. Results LOY was associated with increased EOC risk (OR per year increase = 1.014, 95% CI = 1.009 to 1.020 to OR per year increase = 1.044, 95% CI = 1.041 to 1.048). Individual LOY components, except age at menarche, also associated with EOC. The estimated model coefficient for oral contraceptive use and pregnancies were 4.45 times and 12- to 15-fold greater than expected, respectively. LOY was associated with high-grade serous, low-grade serous, endometrioid, and clear cell histotypes (ORs per year increase = 1.054, 1.040, 1.065, and 1.098, respectively) but not mucinous tumors. Estimated coefficients of LOY components were close to expected estimates for high-grade serous but larger than expected for low-grade serous, endometrioid, and clear cell histotypes. Conclusions LOY is positively associated with nonmucinous EOC. Differences between estimated and expected model coefficients for LOY components suggest factors beyond ovulation underlie the associations between LOY components and EOC in general and for non-HGSOC.

Sexually Transmitted Infections and Risk of Epithelial Ovarian Cancer: Results From the Finnish Maternity Cohort

Abstract Background Sexually transmitted infections, specifically Chlamydia trachomatis (CT), may be associated with epithelial ovarian cancer (EOC) risk. The association between CT and EOC subtypes is unclear. Our aim was to investigate whether history of CT and other infections (Mycoplasma genitalium [MG], herpes simplex virus type 2 [HSV-2], and human papillomavirus [HPV]) are associated with EOC risk by histotype. Methods We measured antibodies (Abs) to CT, MG, HSV-2, and HPV-16/18 in serum samples in a nested case-control study in the Finnish Maternity Cohort (N = 484 cases 1:1 matched to controls). Logistic regression was used to calculate relative risks (RRs) and 95% confidence intervals (CIs) in seropositive versus seronegative individuals in all cases, as well as serous (n = 249), clear cell and endometrioid (n = 91), and mucinous (n = 144) EOC. Results CT seropositivity was not associated with EOC risk (eg, CT pGP3-Ab: RR, 0.92 [95% CI, .72–1.19]), regardless of disease subtype. We observed a positive association between MG seropositivity and mucinous EOC (RR, 1.66 [95% CI, 1.09–2.54]; P for heterogeneity by histotype ≤ .001), but not other subtypes. No associations were observed with seropositivity to multiple STIs. Conclusions CT infection was not associated with EOC risk, with associations observed only for MG and mucinous EOC. Mechanisms linking MG to mucinous EOC remain to be elucidated.

Serologic markers of Chlamydia trachomatis and other sexually transmitted infections and subsequent ovarian cancer risk: Results from the EPIC cohort

AbstractA substantial proportion of epithelial ovarian cancer (EOC) arises in the fallopian tube and other epithelia of the upper genital tract; these epithelia may incur damage and neoplastic transformation after sexually transmitted infections (STI) and pelvic inflammatory disease. We investigated the hypothesis that past STI infection, particularly Chlamydia trachomatis, is associated with higher EOC risk in a nested case‐control study within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort including 791 cases and 1669 matched controls. Serum antibodies against C. trachomatis, Mycoplasma genitalium, herpes simplex virus type 2 (HSV‐2) and human papillomavirus (HPV) 16, 18 and 45 were assessed using multiplex fluorescent bead‐based serology. Conditional logistic regression was used to estimate relative risks (RR) and 95% confidence intervals (CI) comparing women with positive vs. negative serology. A total of 40% of the study population was seropositive to at least one STI. Positive serology to C. trachomatis Pgp3 antibodies was not associated with EOC risk overall, but with higher risk of the mucinous histotype (RR = 2.30 [95% CI = 1.22‐4.32]). Positive serology for chlamydia heat shock protein 60 (cHSP60‐1) was associated with higher risk of EOC overall (1.36 [1.13‐1.64]) and with the serous subtype (1.44 [1.12‐1.85]). None of the other evaluated STIs were associated with EOC risk overall; however, HSV‐2 was associated with higher risk of endometrioid EOC (2.35 [1.24‐4.43]). The findings of our study suggest a potential role of C. trachomatis in the carcinogenesis of serous and mucinous EOC, while HSV‐2 might promote the development of endometrioid disease.

Cohort Profile: The Ovarian Cancer Cohort Consortium (OC3)