Using Radiomics and Machine Learning Applied to MRI to Predict Response to Neoadjuvant Chemotherapy in Locally Advanced Cervical Cancer.

Neoadjuvant chemotherapy plus radical surgery could be a safe alternative to chemo-radiation in cervical cancer patients who are not willing to receive radiotherapy. The response to neoadjuvant chemotherapy is the main factor influencing the need for adjunctive treatments and survival. In the present paper we aim to develop a machine learning model based on cervix magnetic resonance imaging (MRI) images to stratify the single-subject risk of cervical cancer. We collected MRI images from 72 subjects. Among these subjects, 28 patients (38.9%) belonged to the "Not completely responding" class and 44 patients (61.1%) belonged to the 'Completely responding' class according to their response to treatment. This image set was used for the training and cross-validation of different machine learning models. A robust radiomic approach was applied, under the hypothesis that the radiomic features could be able to capture the disease heterogeneity among the two groups. Three models consisting of three ensembles of machine learning classifiers (random forests, support vector machines, and k-nearest neighbor classifiers) were developed for the binary classification task of interest ("Not completely responding" vs. "Completely responding"), based on supervised learning, using response to treatment as the reference standard. The best model showed an ROC-AUC (%) of 83 (majority vote), 82.3 (mean) [79.9-84.6], an accuracy (%) of 74, 74.1 [72.1-76.1], a sensitivity (%) of 71, 73.8 [68.7-78.9], and a specificity (%) of 75, 74.2 [71-77.5]. In conclusion, our preliminary data support the adoption of a radiomic-based approach to predict the response to neoadjuvant chemotherapy.

Risk Analysis Stratification of CIN 2+ Development in a Cohort of Women Living with HIV with Negative Pap Smear and HR-HPV Test during a Long-Term Follow-Up.

BACKGROUND: Women living with Human Immunodeficiency Virus are at higher risk of cervical cancer and precancer compared to women without HIV infection. The aim of the study is to evaluate the risk factors for the development of CIN2+ in a cohort of WLWH with negative colposcopy and cytology during a long follow-up period. METHODS: We enrolled, in a multicentric retrospective cohort study, WLWH who attended the colposcopic services from 1999 to 2019. Patients with a normal Pap smear, a negative HR-HPV test, and at least one year of follow-up were considered for the anlysis. RESULTS: The five-year cumulative incidence of histologically confirmed HSIL was 8.3% (95% CI = 2.6-13.6) among subjects with a CD4+ cell count of <200 cells/µL at any visit and 2.1% (95% CI = 0.7-3.4, p = 0.001) in women with a CD4+ cell count of persistently >200 cells/µL. In women with persistent HR-HPV infection, the five-year cumulative incidence of CIN 2+ was 6% (95% CI = 1.6-10.2) versus 2% (95% CI = 0.4-3.6, p = 0.012) in women without HPV infection. An HIV viremia of >200 copies/mL, a CD4+ cell count of <200 cells/µL, persistent HR-HPV infection, and smoking ≥10 cigarettes/day were all independent and statistically significant risk factors associated with the development of CIN2+ during follow-up. CONCLUSIONS: WLWH with good immune status and negative Pap smear and HR-HPV test have a low risk for CIN2+.

Ki67 as a Predictor of Response to PARP Inhibitors in Platinum Sensitive BRCA Wild Type Ovarian Cancer: The MITO 37 Retrospective Study.

BACKGROUND: There is compelling need for novel biomarkers to predict response to PARP inhibitors (PARPi) in BRCA wild-type (WT) ovarian cancer (OC). METHODS: MITO 37 is a multicenter retrospective study aiming at correlating Ki67 expression at diagnosis with a clinical outcome following platinum treatment and PARPi maintenance. Clinical data were collected from high grade serous or endometroid BRCAWT OC treated with niraparib or rucaparib maintenance between 2010-2021 in 15 centers. Ki67 expression was assessed locally by certified pathologists on formalin-fixed paraffin embedded (FFPE) tissues. Median Ki67 was used as a cut-off. RESULTS: A total of 136 patients were eligible and included in the analysis. Median Ki67 was 45.7% (range 1.0-99.9). The best response to platinum according to median Ki67 was 26.5% vs. 39.7% complete response (CR), 69.1% vs. 58.8% partial response (PR), 4.4% vs. 1.5% stable disease (SD). The best response to PARPi according to median Ki67 was 19.1% vs. 36.8% CR, 26.5% vs. 26.5% PR, 26.5 vs. 25% SD, 27.9% vs. 16.2% progressive disease (PD). No statistically significant differences in progression free survival (PFS) and overall survival (OS) were identified between low and high Ki67. PFS and OS are in line with registration trials. CONCLUSIONS: Ki67 at diagnosis did not discriminate responders to PARPi.