Validation of MSIntuit as an AI-based pre-screening tool for MSI detection from colorectal cancer histology slides.

Mismatch Repair Deficiency (dMMR)/Microsatellite Instability (MSI) is a key biomarker in colorectal cancer (CRC). Universal screening of CRC patients for MSI status is now recommended, but contributes to increased workload for pathologists and delayed therapeutic decisions. Deep learning has the potential to ease dMMR/MSI testing and accelerate oncologist decision making in clinical practice, yet no comprehensive validation of a clinically approved tool has been conducted. We developed MSIntuit, a clinically approved artificial intelligence (AI) based pre-screening tool for MSI detection from haematoxylin-eosin (H&E) stained slides. After training on samples from The Cancer Genome Atlas (TCGA), a blind validation is performed on an independent dataset of 600 consecutive CRC patients. Inter-scanner reliability is studied by digitising each slide using two different scanners. MSIntuit yields a sensitivity of 0.96-0.98, a specificity of 0.47-0.46, and an excellent inter-scanner agreement (Cohen's κ: 0.82). By reaching high sensitivity comparable to gold standard methods while ruling out almost half of the non-MSI population, we show that MSIntuit can effectively serve as a pre-screening tool to alleviate MSI testing burden in clinical practice.

Severe placental lesions due to maternal SARS-CoV-2 infection associated to intrauterine fetal death.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection can cause severe placental lesions leading rapidly to intrauterine fetal death (IUFD). From August 2020 to September 2021, in the pathology department of Toulouse Oncopole, we analyzed 50 placentas from COVID-19-positive unvaccinated mothers. The purpose of our study is to describe the clinicopathological characteristics of these placental damages and to understand the pathophysiology. Ten of them (20%) showed placental lesions with positive immunohistochemistry for SARS-CoV-2 in villous trophoblasts. In five cases (10%), we observed massive placental damage associating trophoblastic necrosis, fibrinous deposits, intervillositis, as well as extensive hemorrhagic changes due to SARS-CoV-2 infection probably responsible of IUFD by functional placental insufficiency. In five other cases, we found similar placental lesions but with a focal distribution that did not lead to IUFD but live birth. These lesions are independent of maternal clinical severity of COVID-19 infection because they occur despite mild maternal symptoms and are therefore difficult to predict. In our cases, they occurred 1-3 weeks after positive SARS-CoV-2 maternal real-time polymerase chain reaction testing and were observed in the 2nd and 3rd trimesters of pregnancies. When these lesions are focal, they do not lead to IUFD and can be involved in intrauterine growth restriction. Our findings, together with recent observations, suggest that future pregnancy guidance should include stricter pandemic precautions such as screening for a wider array of COVID-19 symptoms, enhanced ultrasound monitoring, as well as newborn medical surveillance.

[Improving practice in breast pathology: 34-months experience of the regional SENOPATH network and webinars as a tool for diagnosis of difficult lesions of the breast]. / Amélioration des pratiques en pathologie mammaire : bilan à 34 mois du groupe régional de relecture SENOPATH et télépathologie pour les lésions de diagnostic difficile.

Pathologists commonly face breast lesions that are difficult to diagnose. To reduce second opinion delay, erase geographical barrier and provide continuing education, we aimed to develop a telepathology-based regional network of pathologists. With the support of ONCOMIP network, we founded a peer-group named SENOPATH, composed of experienced breast pathologists practising in private laboratories, university hospitals or comprehensive cancer center in the region of Midi-Pyrénées in France. Submitted cases are digitalized at the University Hospital, stored in a shared space with a possible access via Internet prior to the SENOPATH sessions. The group meets monthly, via a synchronized webinar and multihead microscope session. A consensual diagnosis and final pathology report is issued for each case, and sent to the referring clinician via the patient medical file securely hosted by ONCOMIP. Between 2012 and 2014, 142 cases were reviewed, for either diagnostic 'routine' difficulty or rare histological type. The SENOPATH group, also regularly called by oncologists to solve difficult cases, has considerably improved the pathologist network in Southern France. Supported by the webinar tool, its educational impact is prominent, with a considerable progress in the region with regards to standardization of pathology processes, literature review and knowledge sharing.