News & Updates

New publication on AI-based pathology in Hodgkin lymphoma

Our paper on using hierarchical vision transformers to predict Epstein–Barr virus status and histological subtype from Hodgkin lymphoma whole-slide images has been published in the Journal of Pathology Informatics. The study explores how information relevant to lymphoma classification can be extracted directly from routine histopathology images.

More information: Zsolt Bedőházi et al. Hierarchical vision transformers for Epstein-Barr virus status and histological subtype prediction in Hodgkin lymphoma whole-slide images, Journal of Pathology Informatics, 2026. DOI: 10.1016/j.jpi.2026.100689

New preprint on therapy-associated polyposis

Our new preprint investigates the genomic basis of therapy-associated polyposis, a rare condition that can develop many years after treatment for childhood or young-adult cancer. We identified a mutational signature associated with previous alkylating-agent treatment, providing insight into how earlier cancer therapy may contribute to the development of polyposis later in life.

More information: Yashika Parashar et al. Genomic characterization of therapy-associated polyposis reveals an alkylating mutational signature from prior treatment, medRxiv, 2026. Preprint.

AI combines histopathology tissue analysis and protein data to improve ovarian cancer treatment predictions

Our international research team has developed a novel approach that significantly improves predictions of how patients with High-Grade Serous Ovarian Cancer (HGSOC) will respond to platinum-based chemotherapy using pathological tissue images and proteomics data.

More information: Oz Kilim et al. Histopathology and proteomics are synergistic for high-grade serous ovarian cancer platinum response prediction, NPJ Precision Oncology DOI: 10.1038/s41698-025-00808-w

New Publication in Nature Cancer

Our paper "ALK peptide vaccination restores the immunogenicity of ALK-rearranged non-small cell lung cancer" was published in Nature Cancer. The study demonstrates that ALK-derived peptide vaccination can reverse immune escape in ALK-positive NSCLC, restoring T-cell recognition of tumor cells and providing a potential therapeutic strategy for patients with ALK-rearranged lung cancer.

Best Poster Award

Received the Best Poster Award at the DF/HCC Breast and Gynecologic Cancer Symposium, presented by Dana-Farber/Harvard Cancer Center. The award recognized our work on identifying DNA repair deficiency biomarkers with clinical relevance for ovarian and breast cancer treatment stratification.

Conference Presentation at Dana-Farber

Presented our work on DNA repair deficiencies at the Annual Symposium on Cancer Research at Dana-Farber Cancer Institute, Boston. The presentation covered novel computational approaches for detecting homologous recombination deficiency (HRD) and nucleotide excision repair (NER) defects across multiple cancer types using whole-genome sequencing data.

Research Milestone: 160+ Citations for HRD Method

Our method for estimating homologous recombination deficiency — implemented in the scarHRD R package — has now been cited in over 160 publications. The method is used by research groups worldwide to identify HRD-positive tumors, supporting companion diagnostic development for PARP inhibitor therapy in ovarian, breast, and prostate cancers.