Six accepted abstracts showcase proprietary platforms for refining tumor classification, identifying resistance mechanisms, and guiding personalized immunotherapy across diverse cancers

WALTHAM, Mass.: BostonGene, developer of the leading AI model for tumor and immune biology, announced today the acceptance of six abstracts at the ESMO Congress 2026. As a premier global platform for oncology hosted by the European Society for Medical Oncology, the ESMO Congress brings together international experts to showcase practice-changing research and clinical breakthroughs across cancer care. Demonstrating the power of biologically grounded AI for drug development and precision clinical decision-making, BostonGene will showcase clinical research conducted in collaboration with world-class healthcare institutions. The studies feature advanced multimodal analytics and proprietary technologies to refine ambiguous diagnostics, identify actionable biomarkers, and map secondary resistance mechanisms.

BostonGene session details are as follows:

FPN: 4316P
Title: Mutational signature profiling identifies bona fide POLE/POLD1 loss-of-proofreading mutations in high-TMB tumors
Date & time: October 23 | 3:15 – 4:00 PM
Presenter: Polina Oschepkova, MS, BostonGene

Using a proprietary mutational signature profiling approach, BostonGene analyzed POLE/POLD1 exonuclease domain variants in high tumor-mutational-burden cases to accurately differentiate true driver loss-of-proofreading mutations from non-functional bystander variants. These findings demonstrate the utility of signature profiling in refining biomarker interpretation, helping biopharma and clinical research teams better stratify patients for immune checkpoint inhibitors and design targeted clinical trials.

FPN: 1065P
Title: DNG64-CAR-V Chimeric Amphotropic RNA Vector For CCNG1 Expressing Cancers (NCT04091295)
Date & time: October 23 | 3:15 – 4:00 PM
Presenter: Erlinda Gordon, MD, Sarcoma Oncology Center

BostonGene’s multimodal analytics platform evaluated transcriptomic CCNG1 expression and clinical outcomes in patients with advanced solid tumors treated with DNG64-CAR-V, an RNA vector targeting CCNG1 (NCT04091295). Elevated CCNG1 expression was detected in over 95% of 332 tumor samples, where combination therapy achieved a 50% objective response rate and a 90% clinical benefit rate. These data validate CCNG1 as an actionable biomarker and support the continued clinical development of DNG64-CAR-V combination regimens.

Research done in collaboration with the Sarcoma Oncology Center

FPN: 4863P
Title: RNA‑seq phasing spotlights secondary oncogenic mutations for clinical interpretation
Date & time: October 24 | 12:00 – 12:45 PM
Presenter: Omar Alhalabi, MD, UT MD Anderson Cancer Center

Conventional short-read DNA sequencing often fails to determine whether multiple gene mutations in a tumor occur on the same or on opposite alleles, obscuring how cancers become resistant to treatment. BostonGene applied an RNA-seq variant phasing approach across over 8,000 tumor samples to map allele-specific expression and secondary oncogenic alterations. This study provides critical insights into complex mutational architecture, helping biopharma partners uncover secondary resistance mechanisms and accelerate next-generation drug design.

Research done in collaboration with UT MD Anderson Cancer Center

FPN: 2431P
Title: Predicting immunotherapy and platinum therapy response in extrapulmonary neuroendocrine carcinomas by tumor microenvironment transcriptomics
Date & time: October 26 | 12:00 – 12:45 PM
Presenter: Joseph Kim, MD, PhD, Mayo Clinic

In a collaboration with Mayo Clinic, BostonGene profiled 15 extrapulmonary neuroendocrine carcinoma (EP-NEC) samples using the BostonGene Tumor Portrait® test, demonstrating that RNA-seq-derived MKI67 expression outperformed traditional Ki-67 immunohistochemistry in predicting chemotherapy response. By mapping distinct tumor microenvironment subtypes and delivering precise RNA-based metrics, this platform can help optimize clinical trial design, refine patient selection, and de-risk drug development across EP-NEC and other solid tumors.

Research done in collaboration with the Mayo Clinic

FPN: 3302eP (e-poster)
Title: Using RNA-seq-based gene fusion identification to refine tumor classification and diagnostics

To address challenges in diagnosing pathologically ambiguous malignancies, BostonGene utilized the BostonGene Tumor Portrait® test to evaluate the clinical utility of comprehensive RNA-seq transcriptomic profiling. Through low-dimensional embedding and unsupervised clustering, the analysis reclassified discordant cases and identified novel alterations, including a SLC38A1::GLI1 fusion. These results underscore the power of RNA-seq fusion detection and transcriptomic subtyping to refine diagnostic accuracy and reveal novel therapeutic targets.

Research done in collaboration with Mount Sinai

FPN: 4722eP (e-poster)
Title: Integrated analysis of tissue, plasma and PBMCs in triple negative inflammatory breast cancer patients by response to Keynote 522 regimen

BostonGene performed longitudinal multimodal profiling of tissue and blood samples to uncover mechanisms driving response and resistance to the KEYNOTE-522 regimen in triple-negative inflammatory breast cancer. Utilizing BostonGene’s Kassandra® cell deconvolution tool, the study identified robust T-cell activation in responding patients compared to persistent naive signals and immunosuppressive cell populations in non-responders. The findings highlight actionable, minimally-invasive biomarker signatures to guide therapy personalization and overcome early disease progression.

Research done in collaboration with the UT MD Anderson Cancer Center Inflammatory Breast Cancer Team

To learn more or to schedule a meeting with BostonGene during the event, please contact Lili DeTraglia at events@bostongene.com. For more information, please visit the ESMO Congress 2026 website.

About BostonGene Corporation

BostonGene powers an AI model of tumor and immune biology to deliver disease-level insights and enable precision decision-making across oncology and immune-mediated diseases, spanning drug development and clinical care. By integrating multimodal data, including genomic, transcriptomic, immune, and clinical signals, BostonGene generates biologically grounded intelligence to optimize patient selection, trial design, and therapeutic strategy. This creates a scalable AI decision layer that improves development outcomes and clinical management. BostonGene partners with leading biopharmaceutical organizations and academic institutions to accelerate the delivery of precision therapies while continuously expanding its capabilities across new diseases and complex biological systems. For more information, visit www.BostonGene.com.

Fonte: Business Wire