Cervical cancer is the fourth most frequently diagnosed cancer and the fourth leading cause of cancer death in women, with an estimated 604,000 new cases and 342,000 deaths worldwide in 2020. Persistent infections with high-risk HPV (hrHPV) genotypes can lead to high-grade lesions (Cervical Intraepithelial Neoplasia Grade 2 or higher, CIN2+), that if left untreated progress to cancer. hrHPV test has high sensitivity to detect CIN2+, but it has low specificity because close to 90% of women spontaneously clear the infection. Biomarkers to stratify hrHPV+ women with cervical lesions that may progress to cancer are needed. miRNAs are small noncoding RNAs that regulate gene expression, can be detected in cervical scrapes, and are differentially expressed between high- and low-grade lesions. Aim To identify miRNAs differentially expressed between CIN3+ and ≤CIN1 lesions and evaluate their potential use as biomarkers to distinguish CIN3+ in hrHPV+ women. Methods We used miRNAseq to compare miRNAs expression patterns in cervical scrapes of hrHPV+ women: 35 with low-grade lesions (NEG= 26; CIN1= 9) and 36 with high-grade lesions (CIN3= 32; SCC= 4). The samples were collected through the ASCUS-COL Trial in Medellin, Colombia. Women with low- or high-grade cervical lesions exhibit similar sociodemographic characteristics (Chi-square p-values>0.05). The RNAseq data were processed in GeneGlobe-QIAGEN, which incorporates cut-adapt, bowtie and DeSEq2. Receiver Operating Characteristic (ROC) analyses with a 95% confidence interval of Area Under the Curve (AUC) were made to evaluate the diagnostic accuracy of each miRNA differentially expressed, to detect CIN3+. Multivariate Logistic Regression Analysis using normalized counts of mapped reads identified a combination of the differentially expressed miRNAs that best predicted CIN3+. We identified putative pathways using MetaCore. Results An average of >9 million reads by sample and around 3.5 million reads mapped to miRBase V21 was obtained. The principal component analysis did not show factors that could introduce bias to differential gene expression analysis. We identified 38 miRNAs differentially expressed. Compared to <CIN1 lesions, 9 miRNAs were overexpressed and 29 underexpressed in CIN3+ lesions. Six miRNAs presented AUC>0.60 (p-value <0.05) to detect CIN3+. The best predictive combination of 9 miRNAs exhibits an AUC of 0.89, 95% CI (0.83 - 0.97), of which 4 were overexpressed and 5 underexpressed in CIN3+ vs <CIN1. Interestingly, 5 miRNAs underexpressed in CIN3+, target VEGF, a known angiogenic mediator linked to malignancy. Conclusion We identified miRNAs differentially expressed with good diagnostic performance to distinguish high- from low-grade lesions in cervical scrapes samples. Further validation on a larger cohort of cervical scrapes samples is needed to confirm the potential role of these miRNAs to triage hrHPV+ women. Citation Format: Martha Isabel González Ramírez, Samuel Agudelo, Maria Cecilia Agudelo, Jone Garai, Li Li, Carlos Alberto Orozco Castaño, Jovanny Zabaleta, Gloria Inés Sánchez Vásquez. miRNA expression analysis in high-risk HPV-positive cervical scrapes for the detection of cervical disease [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1497.

Objectives: This study examined the association between reproductive factors and risk of breast cancer (BC) according to the expression of hormone receptors (HR) in women from three Colombian cities. Methodology: 316 cases were recruited between 2012 and 2019 from major general or cancer-dedicated hospitals were women diagnosed with first primary invasive breast cancer confirmed with a biopsy with a clinical staging according to the international tumor-node-metastasis (TNM) classification and residents for at least 3 years in the cities of Barranquilla, Cali, and Medellin. Controls were selected from the general population residing in the same city district as the case for at least 3 years using a multilevel sampling frame. They were matched to cases on age (± 3 years) and city district of residence. Cases and controls were recruited and interviewed by trained health staff, and provided blood samples and tumor tissues, prior to any treatment. Socioeconomic (education level and type of health insurance), anthropometric (weight, height, waist circumference, and body mass index), family history of BC, history of benign breast disease, lifestyle and reproductive (age at menarche, pregnancy ever, parity,. breastfeeding, and its duration) variables were considered. A conditional logistic regression model was used to estimate Odds Ratios (OR) and 95% confidence intervals (95% CI) and evaluate the association of reproductive factors. The final multivariate models were adjusted for factors that changed the risk estimate by more than 10%, including level of education, history of benign breast disease, and health insurance. P values <0.05 were considered statistically significant. Results: Pregnancy ever (OR = 0.48; 95% CI, 0.26-0.90), parity (OR = 0.89; 95% CI, 0.79-0.99 per child), breastfeeding (OR = 0.97; 95% CI, 0.95-0.99) and its longer duration (OR = 0.35; 95% CI, 0.18-0.71 > 12 months) were associated with a decrease in the risk of BC. When combining the reproductive variables, in women who had more than 2 children and breastfed (OR 0.25 95% CI 0.13-0.50), whose last pregnancy was less than 19 years (OR = 0.32 95% CI, 0.17-0.63) and with age at first full-term pregnancy (FFTP) older than 21 years (OR = 0.33 95% CI, 0.17-0.63), the reduction in BC risk was greater than when evaluating the variables individually. The risk of ER+ BC was inversely related to parity (OR = 0.78 95% CI, 0.63-0.95), having breastfed (OR = 0.47 95% CI, 0.26-0.87) and the use of oral contraceptives (OR = 0.56 95% CI, 0.32 -0.99). Although an increased risk for ER- and TN tumors was observed with the use of oral contraceptives, this was not significant probably due to insufficient sample size. Conclusions: As has been observed in other studies in this Colombian population, reproductive factors and the use of hormonal contraceptives were differentially associated with the BC risk. Our data can be useful to identify risk groups to design prevention measures tailored to the specific population. Conflict of interest: None of the authors declare that they have a conflict of interest. Table 1.Odds ratios and 95% confidence intervals of association between breast cancer and reproductive factors and combination of reproductive variables.FactorNMatchedap-valueaMultivariatebp-valuebCases/ControlsOR (95% CI)cOR (95% CI)cEducation levelPrimary school or less119/1371.001.00High School140/1411.34 (0.88-2.03)0.167750.90 (0.60-1.34)0.5901Above High School57/382.27 (1.24-4.04)0.007831.29 (0.74-2.26)0.3710Health Insurance (Contributive)241/1942.24(1.49, 3.38)<0.00011.72 (1.14-2.59)0.0104History of benign breast disease (Yes)158/2314.5 (7.64-27.52)<0.000112.18 (7.5-19.9)<0.0001Pregnancy ever (Yes)263/2960.34 (0.20-0.59)0.00010.48 (0.26-0.90)0.0226Parityd (per child)0.78 (0.69-0.87)<0.00010.89 (0.79-0.99)0.0414Nulliparous61/211.001.001 child68/610.43 (0.24-0.79)0.005970.44 (0.22-0.88)0.019932 children83/970.29 (0.16-0.53)<0.00010.41 (0.21-0.80)0.00849≥3 children104/1370.23 (0.12-0.42)<0.00010.34 (0.17-0.65)0.00119Breastfeeding ever (Yes)75/1200.47 (0.32-0.68)<0.00010.50 (0.34-0.75)0.00089Duration of breastfeeding (months)d0.97 (0.95-0.99)0.01380.97 (0.95-0.99)0.01263Never114/701.001.00≤12155/1610.86 (0.46-1.61)0.64200.66 (0.35-1.19)0.17976>1247/850.41 (0.20-0.84)0.01470.35 (0.18-0.71)0.00328Combination: Parity and breast feedingNulliparous61/211.001.00≥1 children, never breastfeed181/1750.37 (0.21-0.63)0.00020.47 (0.25-0.87)0.01711 child, ever breastfeed13/200.25 (0.11-0.60)0.00190.32 (0.12-0.86)0.0238≥ 2 children, ever breastfeed61/1000.19 (0.10-0.35)<0.00010.25 (0.13-0.50)<0.0001Combination: Parity and time since last birthNulliparous61/211.001.001 child, < 19 years since last birth49/410.46 (0.27-0.90)0.02250.47 (0.22-0.99)0.04611 child, ≥ 19 years since last birth19/200.31 (0.13-0.78)0.01300.36 (0.14-0.94)0.0376≥ 2 children, < 19 years since last birth79/1060.25 (0.13-0.47)<0.00010.32 (0.17-0.63)0.0009≥ 2 children, ≥ 19 years since last birth108/270.27 (0.14-0.52)<0.00010.42 (0.22-0.81)0.0097Combination: Parity and age at FFTPeNulliparous61/211.001.001 Child, age at FFTP <2115/140.48 (020-1.14)0.094440.33 (0.12-0.92)0.03471 child, age at FFTP >2152/470.40 (0.21-0.76)0.005460.47 (0.23-0.97)0.0413>2 Children, age at FFTP <21100/1160.28 (0.15-0.50)<0.00010.43 (0.22-0.82)0.0110>2Children, age at FFTP >2181/1170.25 (0.14-0.45)<0.00010.33 (0.17-0.63)0.0008aConditional logistic regression matched on age, Location, bConditional logistic regression matched on age, Location and adjusted for education level, history of benign breast disease and Health Insurance. cORs are given for both continuous (unit of change) and catrgorical variables. dParous women only; FFTP: first full-term pregnancy, eFFTP: First full-term pregnancy, median age = 21 Table 2.Association of breast cancer and reproductive factors by hormone receptor statusReproductive variablesER+ vs Control (n = 208)ER- vs Control (n = 94)TN vs Control (n = 56)ORa (95% CI)p-valuebORa (95% CI)p-valuebORa (95% CI)p-valuebPregnancy ever0.29(0.11-0.80)0.020.55 (0.16-1.94)0.360.64 (0.2-2.8)0.55Parityd0.78(0.63-0.95)0.010.86 (0.66-1.1)0.240.94 (0.66-1.34)0.72Breastfeeding everc0.47(0.26-0.87)0.010.59 (0.26-1.4)0.220.62 (0.2-1.93)0.41Hormonal contraceptive use0.54(0.29-1.01)0.051.37 (0.5-3.5)0.511.61 (0.46-5.67)0.46Pills0.56(0.32-0.99)0.041.25 (0.51-3.1)0.621.81 (0.53-6.17)0.34aORs are given for continuous variables (unit change). bConditional logistic regression matched on age, Location, and adjusted for education level, history of benign breast disease and Health Insurance. cParous women only. dParity: number of childbirths Citation Format: Juanvilson A Zambrano, SM, Maria Del Mar Idarraga, SM, Samuel Agudelo, BSc, William Arias, MSc, Yina Zambrano, BsC, Karen Cardenas, MSc, Daniel Jurado, M. Sc, Maria C Agudelo, PhD(c), Emanuel Vasco, MD, Alison Mondul, PhD, Laura Rozek, PhD, Andres Ossa, MD, Mauricio Borrero, MD, Fernando Herazo, MD, Edgar Navarro, MD, Roberto Jaramillo, MD, Isabelle Romieu, PhD, Sabina Rinaldi, PhD and Gloria I Sánchez, PhD. Reproductive factors and molecular subtypes of breast cancer in premenopausal and postmenopausal women from 3 Colombian cities: PRECAMA & POSCAMA studies [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-12-28.