Lunit Presents AI-Powered Pathology Research in Ovarian and Bladder Cancers at SITC 2026

Studies with Labcorp and CellCarta identify proliferation-related tissue differences in HRD ovarian cancer and explore spatial immune features associated with immunotherapy response in bladder cancer

SEOUL, South Korea, Nov. 4, 2026 /PRNewswire/ -- Lunit (KRX: 328130), a leading provider of AI for cancer diagnostics and precision oncology, today announced that it will present two collaborative studies with Labcorp and CellCarta using its AI-powered pathology platform, Lunit SCOPE, at the Society for Immunotherapy of Cancer (SITC) 41st Annual Meeting, taking place November 4–8 in Phoenix, Arizona.

The presentations mark Lunit's sixth consecutive year of presenting research at SITC since 2021. The ovarian cancer study examines histologic characteristics associated with homologous recombination deficiency (HRD), while the bladder cancer study explores spatial immune patterns associated with immunotherapy response.

The first study, conducted with Labcorp, examined histologic and immune characteristics associated with HRD in ovarian carcinomas using AI-powered pathology analysis. HRD is an important biomarker used to guide treatment with PARP inhibitors, but whether HRD-associated biological differences can be identified through routine histopathology remains an open research question.

Researchers analyzed 339 ovarian carcinoma whole-slide images, including 88 HRD and 251 HRP cases, using Lunit SCOPE IO to quantify 22 immune and morphological features, including immune phenotypes, cell densities, and characteristics of tumor and stromal compartments.

Molecular profiling using Labcorp's OmniSeq INSIGHT assay showed that HRD tumors had higher tumor mutational burden (TMB) than HRP tumors. In the pathology analysis using Lunit SCOPE IO, HRD tumors showed approximately 1.5-fold higher mitotic cell density (p=0.002), along with significantly higher tumor cell density in cancer areas (p=0.031) and combined cancer and stromal compartments (p=0.016).

Across the cohort, immune phenotypes classified by Lunit SCOPE IO were also associated with genomic features, with inflamed tumors showing higher TMB than immune-desert and immune-excluded tumors (median 4.7 vs. 3.9 mutations per megabase; p=0.038).

The second study, conducted with CellCarta, examined associations between spatial immune organization and response to neoadjuvant atezolizumab, an anti-PD-L1 immunotherapy, in patients with muscle-invasive bladder cancer (MIBC).

Using tissue samples from the Phase II ABACUS study, researchers combined multiplex immunohistochemistry and spatial analysis to characterize immune cell populations and their interactions. As part of the study, tertiary lymphoid structures (TLS) in H&E-stained tissue were assessed by CellCarta pathologists and with Lunit SCOPE IO, while CellCarta's spatial analysis technologies were used to examine tumor and immune cell populations and their spatial relationships.

Responders showed higher immune cell infiltration, greater immune cell diversity, and closer proximity between tumor cells and T cells. Significant colocalization of antigen-presenting cells and CD4-positive T cells was also observed, along with a trend toward more TLS being located closer to tumor cell nests. In contrast, nonresponders generally exhibited lower immune infiltration, with T cells located farther from tumor cells and predominantly displaying an exhausted phenotype.

These results suggest that spatial immune organization, beyond immune cell abundance alone, may offer insights into immunotherapy response and warrant further investigation as a potential source of predictive biomarkers.

Together, the studies illustrate how AI-powered pathology can contribute to different aspects of tumor microenvironment research, from identifying proliferation-related tissue features associated with HRD in ovarian cancer to supporting the assessment of immune structures alongside complementary spatial analysis in bladder cancer.

"AI-powered pathology is expanding how researchers can analyze tumor tissue and investigate cancer biology," said Brandon Suh, CEO of Lunit. "Through collaborations with global research partners such as Labcorp and CellCarta, we aim to deepen our understanding of the tumor microenvironment and advance the role of AI in precision oncology research."

Research Presentations at SITC 2026: 

1. Poster Presentation [#1024 / Nov. 7]: AI-based spatial profiling reveals histologic differences between homologous recombination-deficient and -proficient ovarian carcinomas

2. Poster Presentation [#476 / Nov. 7]: Spatial immune architecture and tumor–T cell proximity are associated with response to neoadjuvant atezolizumab in muscle-invasive bladder cancer

* Lunit SCOPE products are for research use only. Not for use in diagnostic procedures or decisions.

About Lunit

Founded in 2013, Lunit (KRX: 328130) is a global leader on a mission to conquer cancer through AI. Our clinically validated solutions span medical imaging, breast health, and biomarker analysis, empowering earlier detection, smarter treatment decisions, and more precise outcomes across the cancer care continuum.

Lunit offers a comprehensive suite spanning risk prediction and early detection to precision oncology. Our FDA-cleared Lunit INSIGHT Breast Suite and breast health solutions support cancer screening in thousands of medical institutions worldwide, while the Lunit SCOPE platform is used in research partnership with global pharma and laboratory leaders for biomarker research, and companion diagnostic development.

Trusted by over 10,000 sites in more than 65 countries, Lunit combines deep medical expertise with continuously evolving datasets to deliver measurable impact for patients, clinicians, and researchers alike. Headquartered in Seoul with global offices, Lunit is driving the worldwide fight against cancer. Learn more at lunit.io.

SOURCE Lunit