Andrea Gazzo

Purpose:The clinical implications of distinct SMARCA4 alteration classes remain incompletely defined. We performed a pan-cancer analysis to characterize SMARCA4 mutation classes and their associations with allelic status, genomic context, and therapeutic outcomes. Experimental Design:We analyzed 68,920 tumor–normal paired samples sequenced with the MSK-IMPACT next-generation sequencing assay between 2015 and 2023. We curated oncogenic or likely oncogenic SMARCA4 alterations using OncoKB and classified them based on prior functional and structural literature. Allele-specific copy number, ploidy, fraction of genome altered, loss of heterozygosity, and whole-genome doubling were calculated. Clinical outcomes were evaluated in selected tumor cohorts treated with platinum-based therapy or immune checkpoint blockade. Results:SMARCA4 alterations were most frequent in thymic epithelial tumors (10.7%), non–small cell lung cancer (7.1%), bladder cancer (5.6%), and cervical cancer (5%). Class 1 and 2 alterations occurred in 2.9% and 0.7% of tumors, respectively. Distinct patterns of genomic complexity, co-occurring alterations, and mutational signatures were observed across tumor types based on SMARCA4 alteration allelic state and class. In a phase I trial of the BRM degrader PRT3789, patients with monoallelic class 2 SMARCA4 alterations and preserved BRG1 expression saw clinical benefit, including 1 who attained a complete response. Conclusions:SMARCA4 alteration class and allelic status define biologically and clinically distinct subsets of tumors. These findings support incorporating allelic status and functional mutation class into patient selection strategies for emerging BRM-targeted therapies.

Purpose: The clinical implications of distinct SMARCA4 alteration classes remain incompletely defined. We performed a pan-cancer analysis to characterize SMARCA4 mutation classes and their associations with allelic status, genomic context, and therapeutic outcomes. Experimental Design: We analyzed 68,920 tumor–normal paired samples sequenced with the MSK-IMPACT next-generation sequencing assay between 2015 and 2023. We curated oncogenic or likely oncogenic SMARCA4 alterations using OncoKB and classified them based on prior functional and structural literature. Allele-specific copy number, ploidy, fraction of genome altered, loss of heterozygosity, and whole-genome doubling were calculated. Clinical outcomes were evaluated in selected tumor cohorts treated with platinum-based therapy or immune checkpoint blockade. Results: SMARCA4 alterations were most frequent in thymic epithelial tumors (10.7%), non–small cell lung cancer (7.1%), bladder cancer (5.6%), and cervical cancer (5%). Class 1 and 2 alterations occurred in 2.9% and 0.7% of tumors, respectively. Distinct patterns of genomic complexity, co-occurring alterations, and mutational signatures were observed across tumor types based on SMARCA4 alteration allelic state and class. In a phase I trial of the BRM degrader PRT3789, patients with monoallelic class 2 SMARCA4 alterations and preserved BRG1 expression saw clinical benefit, including 1 who attained a complete response. Conclusions: SMARCA4 alteration class and allelic status define biologically and clinically distinct subsets of tumors. These findings support incorporating allelic status and functional mutation class into patient selection strategies for emerging BRM-targeted therapies.

Motivation: Standardized phenotypic descriptions are essential for accurate diagnosis, yet clinicians and researchers face challenges in manually extracting and mapping phenotypes from scientific literature or patient clinical records to the Human Phenotype Ontology. Recent advances in deep learning offer new opportunities for automation. We developed PhenoXtract, a novel phenotype extraction approach that combines Large Language Models and Knowledge Graph embedding. PhenoXtract is a multistep pipeline that takes clinical descriptions as input, extracts candidate phenotype entities using large language models, and maps them to terms from an enriched version of the Human Phenotype Ontology, processed as a knowledge graph. Results: Evaluation against expert-curated ground-truth datasets show a recall of 0.70 and precision of 0.85 for PhenoXtract, demonstrating concordance with manually extracted phenotypes, with a computation time of 10-20 seconds for each text analyzed. Moreover, PhenoXtract surpasses rule-based and deep learning-based state-of-the-art tools in two out of the three ground-truth datasets evaluated. These results suggest that hybrid approaches combining Large Language Models and Knowledge Graph embeddings represent a promising direction for automated clinical phenotyping at scale.

Supplementary Figures S1-S5 Supplementary Tables S1-S5

Background: Malignant phyllodes tumours (MPT) are aggressive breast fibroepithelial neoplasms. Their rarity has limited their genetic characterization, and associations with genetic ancestry and progression drivers remain poorly understood. Prior studies suggest two evolutionary pathways according to MED12 mutational status. We sought to determine the repertoire of somatic genetic alterations in cancer genes in primary versus metastatic/recurrent MPTs, and according to MED12 mutational status and genetic ancestry. Materials and methods: We analysed the paired tumour‐normal targeting sequencing data (up to 505 cancer‐related genes) of 31 MPTs (primary, n = 20; metastatic/recurrent, n = 11). Results: Metastatic/recurrent MPTs harboured a numerically higher frequency of genetic alterations in CDKN2A/2B (55% vs. 25%). Moreover, analysis of an MPT case with paired primary and metastatic samples revealed a CDKN2A/2B homozygous deletion restricted to the metastatic sample, suggesting a role for CDKN2A/2B in progression. Compared with MED12 ‐wild type MPTs, MED12 ‐mutant MPTs had higher tumour mutation burden ( P = 0.002), frequency of TERT promoter (82% vs. 35%; P = 0.02) and RB1 mutations (45% vs. 5%; P = 0.01). Genetic alterations in the PI3K pathway, including PIK3CA and PTEN , were only present in MED12 ‐wild type MPTs, and absent in MED12 ‐mutant cases (15% vs. 0%; P > 0.05). Furthermore, genetic alterations in EGFR were restricted to MPTs from patients of European genetic ancestry and absent in those of Asian ancestry (43% vs. 0%; P > 0.05). Conclusions: Taken together, the repertoire of genetic alterations in primary and metastatic/recurrent MPTs shows overlap, and CDKN2A/2B homozygous deletions may play a role in progression. Additionally, molecular profiles of MPTs may vary according to genetic ancestry and MED12 mutational status.

Supplementary Data3 from RRAS and RRAS2 Mutations Are Recurrent Oncogenic Drivers in Lung Cancer and Are Sensitive to the Pan-RAS Inhibitor RMC-6236

All supplemental figures and captions