Abstract 4075: GEM-X Flex: scalable and cost-effective whole transcriptome single cell analysis for cancer marker detection from FFPE tissue section slides

Abstract Formalin-fixed paraffin-embedded (FFPE) tissue blocks are the most widely used sample type in clinical research for preserving tissue morphology and supporting long-term storage. These samples can be particularly valuable for studying rare cancers or when sample procurement is challenging. The growing demand for comprehensive single cell gene profiling calls for methods that require less sample input, reduce costs and enable high-throughput analysis for large clinical studies. Large-cohort studies are often limited by financial constraints. To overcome this challenge, we performed a cost-effective, high-throughput experiment with eight different FFPE blocks (glioblastoma, colorectal, lung, breast, skin, endocervical cancers, normal kidney, and reactive lymph node), with two replicates per sample. Using the 16-barcode multiplexing approach of GEM-X Flex, we processed 320, 000 cells pooled from all 16 samples in a single lane (20, 000 cells per sample). The distinct barcodes allowed us to obtain high quality single cell sequencing data from each individual sample with ∼< 8% undetected multiples. We identified tumor markers specific to each cancer, including PROM1 (glioblastoma), BRCA1 (colorectal), KRT19 (lung), MUC1 (breast), GPC3 (breast), and MUC16 (endocervical). This method allows processing up to 2.56 million cells per chip across 8 lanes, reducing the cost per sample to just one cent per cell, making large-scale studies more cost-efficient. We also extended this approach to tissue sections already mounted on slides and subjected to histological analysis, enabling the extraction of the most information from precious clinical samples. Using the GEM-X Flex assay, we generated whole transcriptome profiles from as few as 25, 000 nuclei isolated from 5 μm FFPE tissue sections on standard superfrost slides. Gene expression profiles, cell type proportions, and assay sensitivity from these thin sections were comparable to those from 25 μm tissue scrolls from the same FFPE block, demonstrating that high-quality single cell RNA sequencing is achievable with minimal sample input. The ability to work with 5 μm tissue sections expands options for universal sample preparation across various assays. It also broadens sample procurement opportunities by using tissue sections instead of whole FFPE blocks, significantly lowering costs. In summary, the GEM-X Flex assay enables whole-transcriptome gene profiling at single cell resolution from 5 μm FFPE tissue sections. By combining H&E image analysis with RNA sequencing data from sequential sections of the same cancer block, pathologists can achieve more precise diagnostics. The assay’s multiplexing capability significantly lowers costs and supports high-throughput studies, enabling large-scale tumor marker screening for early detection, progression monitoring, and therapeutic target development. Citation Format: Tingsheng Y. Drennon, Jawad Abousoud, Ryan T. Stott, Juan Pablo Romero, Paul Lund, Nico Sponer, Sarah Taylor, Peter Smibert, Andrew Kohlway. GEM-X Flex: scalable and cost-effective whole transcriptome single cell analysis for cancer marker detection from FFPE tissue section slides [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 4075.