Abstract 3917: A functional precision medicine assay for immunotherapy and combination therapy testing in liver and pancreatic cancers (PEAR-PAL)

Abstract The PEAR-PAL study was aimed at developing a novel ex vivo 3D immune-tumor model with integrated AI-based image analysis to optimize precision therapies in hepatocellular carcinoma (HCC) and pancreatic ductal adenocarcinoma (PDAC). Both indications currently lack effective biomarkers to guide combination treatment clinical decision-making. Fresh samples from 46 patients (30 HCC, 16 PDAC) were analyzed to evaluate the feasibility of functional precision medicine (FPM) in high-unmet-need cancers. The platform isolates viable tumor and immune cells, co-culturing them in a cancer-specific 3D hydrogel matrix, allowing subsequent testing of FDA-approved immunotherapies, chemotherapies, and tyrosine kinase inhibitors (TKIs) using a patented microphysiological immuno-oncology plate. After isolation, the 3D immune-microtumors were exposed to various drug combinations and monitored daily over five days through confocal imaging and a computer vision pipeline. Key metrics (e.g., immune cell infiltration, tumor cell viability, and microtumor volume) were evaluated to capture inter- and intra-patient variability in response to clinical treatment regimens. Therapies included chemotherapies (e.g., gemcitabine, nab-paclitaxel), TKIs (e.g., cabozantinib, sorafenib), and immunotherapies (e.g., pembrolizumab). In PDAC, further optimization focused on testing multi-drug chemotherapeutic combinations such as FOLFIRINOX. Tumor processing was optimized, achieving mean viability of 73±19% for HCC and 80±17% for PDAC (N=36 and 12). Live/Dead and CyQuant assays determined the optimal hydrogels to maintain 3D cell viability over the five-day assay. PBMC infiltration in the optimized hydrogel formulation increased PBMC counts from 272 to 1, 485 by day 4 compared to Matrigel. Dosing for each therapeutic class was optimized via 2D and 3D dose-response assays in cell lines, through molecular readouts (in-cell western, pRTK arrays, Live/Dead, CTG) and validated in primary samples. A microphysiological plate was developed for sequential treatment addition/removal and imaging without disrupting the 3D immune-microtumor. Functional assays showed patient-specific variability in responses to treatments, with pembrolizumab decreasing microtumor volume, enabling CD8- PBMC infiltration, and increasing T-cell infiltration 3-fold compared to controls (n=2). Future trials will correlate ex vivo findings with clinical outcomes, positioning the PEAR-PAL platform as a promising tool for individualized treatment in liver and pancreatic cancers. This approach exemplifies the potential of FPM to address current limitations in precision oncology for complex cancers lacking actionable biomarkers. Citation Format: Eleonora Peerani, Farah Sangkolah, Niamh A. Leonard, Yann Bouremel, Demi Wiskerke, Thomas D. Richardson, Magda B. Martinez Espuga, Diana Visan, George R. de Fraine, Jay Kearney, Francesco Iori, Duleek N. Ranatunga. A functional precision medicine assay for immunotherapy and combination therapy testing in liver and pancreatic cancers (PEAR-PAL) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3917.