Aims Salivary duct carcinoma (SDC) is an aggressive salivary malignancy that results in high mortality rates and is often resistant to chemotherapy. Anti‐programmed death‐1 (PD‐1)/programmed death ligand‐1 (PD‐L1) checkpoint inhibitors have led to dramatic improvements in patients with various cancers. Other immunotherapeutic approaches, e.g. cancer vaccines, have shown promising results. Cancer testis antigens, e.g. preferentially expressed antigen in melanoma (PRAME), are regarded as promising vaccine targets because of their tumour‐specific expression pattern. Methods and results We analysed the immunoexpression of PD‐L1, PD‐1, major histocompatibility complex class I (MHC I) and PRAME in 53 SDCs. The immunoexpression levels of PD‐L1 in tumour cells (TCs) and immune cells (ICs), PD‐1 in ICs, PRAME in TCs and MHC I in TCs were analysed, and were correlated with outcome. PRAME expression was seen in 83% of SDCs. No PRAME staining was present in normal salivary gland tissue. With the three established diagnostic algorithms proposed for head and neck squamous cell carcinoma, the criteria being a combined positive score of ≥1, TC% ≥1%, and TC% ≥25%, 35 (66%), 17 (32%) and three cases (6%), respectively, were deemed to be positive for PD‐L1. PD‐1‐positive ICs were seen in 35 (66%) cases. MHC I down‐regulation was seen in 82% of SDCs. There was a significant correlation among PD‐L1 expression in ICs, PD‐1 expression in ICs, and PRAME expression in TCs. PD‐L1 expression in TCs and lack of PD‐1 expression in ICs were associated with decreased disease‐specific survival in SDC patients. Conclusions Alterations of the tumour immune microenvironment are common in SDCs, including expression of PD‐1/PD‐L1 and PRAME, which opens the way to potential novel immune therapies, such as cancer vaccination and PD‐1/PD‐L1 blockade, in these tumours.

Purpose: In 2016, non-invasive encapsulated follicular variant of papillary thyroid carcinoma (NI-EFVPTC) was renamed as noninvasive thyroid follicular neoplasm with papillary-like nuclear features (NIFTP). However, as the study cohort did not mention tumors with oncocytic features,

Defining cellular and subcellular structures in images, referred to as cell segmentation, is an outstanding obstacle to scalable single-cell analysis of multiplex imaging data. While advances in machine learning-based segmentation have led to potentially robust solutions, such algorithms typically rely on large amounts of example annotations, known as training data. Datasets consisting of annotations which are thoroughly assessed for quality are rarely released to the public. As a result, there is a lack of widely available, annotated data suitable for benchmarking and algorithm development. To address this unmet need, we release 105,774 primarily oncological cellular annotations concentrating on tumor and immune cells using over 40 antibody markers spanning three fluorescent imaging platforms, over a dozen tissue types and across various cellular morphologies. We use readily available annotation techniques to provide a modifiable community data set with the goal of advancing cellular segmentation for the greater imaging community.

2657 Background: Immune-related colitis (irColitis) is associated with significant morbidity and mortality among patients treated with immune checkpoint inhibition (ICI). The gut microbiota has been implicated in the pathophysiology of irColitis. We hypothesized that abnormal fecal microbiome features would be present at the time of irColitis onset, and that restoring the microbiome to a healthy state would mitigate disease severity. Methods: We present fecal microbiota profiles from N = 18 patients with irColitis and describe our clinical experience of N = 5 patients treated with healthy donor fecal microbial transplantation (FMT). Recipients of ICI between May 2019 and June 2021 at Memorial Sloan Kettering Cancer Center whose stool samples were investigated for diarrhea were included in our dataset (N = 100). N = 18 patients with a diagnosis of irColitis were identified. Patients that had other causes of diarrhea were classified as part of the “no-colitis” comparator group (N = 32). All patients with infectious causes of diarrhea ( C. difficile, other GI pathogen) were excluded from the analysis. Stool samples were profiled by whole metagenomic shotgun sequencing. FMT products from individual healthy donors were procured from the OpenBiome fecal bank and delivered via colonoscopy. Results: There was no difference in alpha diversity between the irColitis and no-colitis groups. Patients in the irColitis group had significantly higher relative abundance of Proteobacteria at the time of symptom onset compared to the no-colitis group (p = 0.04). Escherichia spp. and Klebsiella spp. were associated with irColitis, also described in multivariate analyses. Five patients with irColitis refractory to steroids and biologics received healthy-donor FMT, with initial clinical improvement in irColitis symptoms observed in four of five (80%) patients. Two out of five subsequently exhibited recurrence of irColitis symptoms following a course of antibiotics for pneumonia and urinary tract infection, respectively. Both received a second “salvage” FMT that was, again, followed by clinical improvement of irColitis. In these two cases, fecal profiles demonstrated dysbiotic shifts in the gut microbiome post-antibiotics characterized by Proteobacteria expansion, which was salvaged post salvage FMT, mirroring clinical resolution of irColitis symptoms. Conclusions: We observed that Proteobacteria expansion is characteristic of irColitis at time of symptom onset. FMT was followed by clinical improvements in several cases of refractory irColitis. Strategies to restore or prevent microbiome injury—with a particular focus on salvage FMT after antibiotic injury—in the context of immunotherapy toxicities could be further explored in prospective clinical trials.