Endometrial Neoplasms

We obtained publicly available data on somatic mutations from whole-exome sequencing and on laboratory-determined MSI statuses for 526 whole exomes as follows.
For gastric adenocarcinoma, we used whole-exome somatic mutation data from reference26 (link) (14 tumors) and reference18 (link) (22 tumors). For the tumors from reference26 (link), MSI statuses had been determined by the Promega MSI Analysis System Version 1.2 (Promega Corp, USA)14 . For the tumors from reference18 (link), MSI statuses had been determined by an extended panel of markers as described. From the two references, we called somatic mutations in these tumors using the Genome Analysis Toolkit (GATK) (https://www.broadinstitute.org/gatk/) pipeline described in reference 27 (link).
For colon (216 tumors), rectal (81 tumors) and endometrial (193 tumors) carcinomas, we obtained somatic mutation data16 (link)17 (link) from the TCGA data portal (https://tcga-data.nci.nih.gov/tcga/) on 3 July 2013. At the TCGA data portal, we chose the specific cancer type (e.g. colon adenocarcinoma) and then chose the publicly available Mutation Annotation Files (MAFs) for level 2 exome data and clicked “Build Archive”. The data portal then prepared the data and sent a link for download. Because it would be difficult for other researchers to download the MAFs as they were on the date we downloaded them, we will make the specific MAFs we used available on request. The MSI-Mono-Dinucleotide Assay15 had been used to determine the MSI statuses of these tumors. Somatic mutations in the colon and rectal tumors were called by Baylor College of Medicine using GATK and Atlas2 (https://www.hgsc.bcm.edu/software/software/atlas-2). Somatic mutations in the endometrial tumors were called by the Broad Institute using GATK.

MITO 37 is a multicenter retrospective Italian study aiming at correlating Ki67 expression with a clinical outcome following platinum treatment and PARPi maintenance. Clinical data were collected from all patients with high grade serous or endometroid BRCA WT ovarian cancer treated with niraparib or rucaparib maintenance between 2010–2021. The study has been approved by the ethical committee of participating Institutions. We collected data from 15 Italian Centers. See Table S1.
Eligible patients were at least 18 years of age and had platinum-sensitive (defined as PFI ≥ 6 months) relapsed cancer of the ovary, peritoneum, or fallopian tube (collectively defined as ovarian cancer), with disease progression after at least 1 line of chemotherapy. All patients had high-grade serous or endometrioid tumors that were classified at diagnosis according to the FIGO staging criteria. Before the start of PARPi, all the patients had received four to six cycles of platinum-based chemotherapy, which had resulted in a complete response (CR) or partial response (PR), according to investigator assessment. Patients receiving treatment within clinical trials were not eligible. All patients had been tested for BRCA1/2 germline and/or somatic mutations and resulted WT.
The primary endpoint was to determine, in a real-world setting, the overall response rate (ORR) to platinum-based chemotherapy and to PARPi according to the Ki67 value expressed as %. Patients were divided into 2 groups (low and high) using the median value as the cut-off (see below).
The secondary endpoints were to describe progression free survival (PFS) and OS comparing subgroups based on the Ki67 value.
PFS was defined as the time from the first day of niraparib or rucaparib administration to the date of objective disease progression on imaging according to the response evaluation criteria in solid tumors (RECIST), version 1.1, or death from any cause. Patients who did not experience progression disease (PD) were censored on the date of the last follow-up visit. OS was defined from the first day of niraparib or rucaparib administration to death for any cause or the last follow up visit. The median follow-up (FU) was calculated according to the reverse Kaplan–Meier formula. PFS and OS were calculated according to the Kaplan–Meier formula, and groups were compared by log-rank test.
Within 8 weeks after completion of the last dose of platinum-based chemotherapy, patients were assigned to receive oral niraparib or rucaparib as per clinical practice choice (according to previous toxicities, comorbidities, and patients’ choice) in 28-day cycles.
All the patients started rucaparib at a fixed dose of 600 mg twice daily, while patients starting niraparib were assigned a full dose of 300 mg daily or an individualized dose of 200 mg daily according to platelets count and weight (a 200 mg daily dose was assigned in case of a baseline body weight of less than 77 kg, a platelet count of less than 150,000/mm³, or both) [34 (link)].
The data cut-off for the analysis was 30 April 2022.

Quantitative analysis of immunohistochemical reactions was performed using light microscopy, and the numbers of positively stained cells for each antibody were obtained. Slides were initially examined at a low magnification (10× lens) to select the areas that were most abundant in positively identified DCs (so-called hot spots). Within the chosen areas, the counts of positively stained DCs in five representative fields of view were summed up at a high magnification (40× lens) and expressed per 1 mm². An assessment of cells’ morphology preceded establishing their numerosity: a cell was considered a positively stained DC if it had a visible nucleus, dendritic appearance (e.g., cytoplasmic processes), and intensely colored cytoplasm at 40× magnification. Analyses of each monoclonal antibody’s reactivity were performed separately for the tumor stroma, glandular endothelium, and invasive border (Figure S2 in the Supplementary Materials). The latter term refers to the field of view comprising the cancerous margin and the adjacent healthy tissues in an even ratio. Samples were also evaluated for the presence of co-occurrent non-neoplastic endometrium. Numbers of positively identified DCs per mm² in stroma, glands, and margins of healthy tissue were utilized to create an internal positive control group (Figure S3 in the Supplementary Materials). Two authors performed the pathological analyses separately and solved any disagreements via discussion.