Many patients with recurrent and metastatic cancer fail to produce a durable response to immunotherapy, highlighting the need for additional therapeutic targets to improve the immune landscape in tumors. Recent studies have highlighted the importance of B cells in the antitumor response, with memory B cells (MBCs) being prognostic in a variety of solid tumors. MBCs are a heterogenous B cell subset and can be generated through both germinal center reactions and extrafollicular (EF) responses. EF-derived MBCs have been recently linked to poor prognosis and treatment resistance in solid tumors and thus may represent candidate biomarkers or immunotherapy targets. EF-derived MBCs, termed “double-negative” (DN) MBCs may be further classified on the basis of surface expression of CD11c and CD21 into DN1, DN2, and DN3 MBCs. CD11c − CD21 + DN1 MBCs and CD11c + CD21 − DN2 MBCs have been well studied across inflammatory diseases; however, the biology and clinical relevance of CD11c − CD21 − DN3 MBCs remain unknown. Here, we report an accumulation of DN3 MBCs in the blood and tumors of patients with head and neck squamous cell carcinoma (HNSCC) and an increase in DN3 MBCs in locally advanced HNSCC tumors. Circulating and intratumoral DN3 MBCs were hyporesponsive to antigen stimulation, had low antibody production, and failed to differentiate into antibody-secreting cells. Moreover, DN3 MBCs accumulated selectively outside of tertiary lymphoid structures. Last, circulating DN3 MBCs correlated with poor therapeutic response, advanced disease, and worse outcomes in patients with HNSCC and melanoma, supporting further assessment of EF-derived MBCs as potential biomarkers and therapeutic targets.
Understanding drug–gene relationships is essential for advancing targeted cancer therapies and drug repurposing strategies. However, the vast volume of biomedical literature poses significant challenges in efficiently extracting relevant insights. In this study, we developed an automated pipeline that leverages retrieval-augmented large language models (LLM) to infer drug–gene interactions using the most up-to-date biomedical literature. By integrating PubMed and state-of-the-art LLMs, our pipeline generates accurate, evidence-based inferences while addressing the limitations of static LLMs, such as outdated knowledge and the risk of producing misleading results. We systematically validated the pipeline’s performance using curated databases and demonstrated its ability to accurately identify both well-established and emerging drug targets. Using our pipeline, we constructed a pan-cancer drug–gene interaction network among hundreds of FDA-approved drugs and key oncogenes. In a case study on liver cancer, we identified and validated an association between CTNNB1 mutations and enhanced sensitivity to sorafenib, highlighting a potential therapeutic strategy for this challenging mutation. To facilitate broad accessibility, we developed GeneRxGPT, a user-friendly web application that enables cancer researchers to utilize the pipeline without programming expertise or extensive computational resources. It provides intuitive modules for drug–gene inference and network visualization, streamlining the exploration and interpretation of drug–gene relationships. We anticipate that GeneRxGPT will empower researchers to accelerate drug discovery and development, making it a valuable resource for the cancer research community. Significance: This study presents a novel approach that integrates LLMs with real-time biomedical literature to uncover drug–gene relationships, transforming how cancer researchers identify therapeutic targets, repurpose drugs, and interpret complex molecular interactions. GeneRxGPT, our user-friendly tool, enables researchers to leverage this approach without requiring computational expertise.
Objective: To investigate the preclinical efficacy and identify predictive biomarkers of polo-like kinase 1 (PLK1) inhibitors in small cell lung cancer (SCLC) models. Methods: We tested the cytotoxicity of selective PLK1 inhibitors (rigosertib, volasertib, and onvansertib) in a panel of SCLC cell lines. We confirmed the therapeutic efficacy of subcutaneous xenografts of representative cell lines and in four patient-derived xenograft models generated from patients with platinum-sensitive and platinum-resistant SCLC. We employed an integrated analysis of genomic and transcriptomic sequencing data to identify potential biomarkers of the activity and mechanisms of resistance in laboratory-derived resistance models. Results: Volasertib, rigosertib, and onvansertib showed strong in vitro cytotoxicity at nanomolar concentrations in human SCLC cell lines. Rigosertib, volasertib, and onvansertib showed equivalent efficacy to that of standard care agents (irinotecan and cisplatin) in vivo with significant growth inhibition superior to cisplatin in PDX models of platinum-sensitive and platinum-resistant SCLC. There was an association between YAP1 expression and disruptive or inactivation TP53 gene mutations, with greater efficacy of PLK1 inhibitors. Comparison of lab-derived onvansertib-resistant H526 cells to parental cells revealed differential gene expression with upregulation of NAP1L3, CYP7B1, AKAP7, and FOXG1 and downregulation of RPS4Y1, KDM5D, USP9Y, and EIF1AY highlighting the potential mechanisms of resistance in the clinical setting. Conclusions: We established the efficacy of PLK1 inhibitors in vitro and in vivo using PDX models of platinum-sensitive and resistant relapsed SCLC. An ongoing phase II trial is currently testing the efficacy of onvansertib in patients with SCLC (NCT05450965).
Despite significant advances in tumor biology and clinical therapeutics, metastasis remains the primary cause of cancer-related deaths. While RNA-seq technology has been used extensively to study metastatic cancer characteristics, challenges persist in acquiring adequate transcriptomic data. To overcome this challenge, we propose MetGen, a generative contrastive learning tool based on a deep learning model. MetGen generates synthetic metastatic cancer expression profiles using primary cancer and normal tissue expression data. Our results demonstrate that MetGen generates comparable samples to actual metastatic cancer samples, and the cancer and tissue classification yields performance rates of 99.8 ± 0.2% and 95.0 ± 2.3%, respectively. A benchmark analysis suggests that the proposed model outperforms traditional generative models such as the variational autoencoder. In metastatic subtype classification, our generated samples show 97.6% predicting power compared to true metastatic samples. Additionally, we demonstrate MetGen’s interpretability using metastatic prostate cancer and metastatic breast cancer. MetGen has learned highly relevant signatures in cancer, tissue, and tumor microenvironments, such as immune responses and the metastasis process, which can potentially foster a more comprehensive understanding of metastatic cancer biology. The development of MetGen represents a significant step toward the study of metastatic cancer biology by providing a generative model that identifies candidate therapeutic targets for the treatment of metastatic cancer.