Abstract B116: Integrative proteogeonomic profiling of PDAC reveals updated epithelial subtypes and cross-omic predictors of survival
Bibliographic record
Abstract
Abstract Pancreatic ductal adenocarcinoma (PDAC) chemotherapy regimens are largely uniform, tailored mainly to patient fitness or rare genomic alterations. This strategy fails to capitalize on the disease’s substantial molecular heterogeneity, including variations in copy number alterations, immune infiltration, stromal composition, and cancer cell phenotypes. These features are not independent; they reflect interconnected biological processes and co-evolving pathways that shape tumour progression. Multi-omic integration is therefore essential—not only to capture the complexity of PDAC biology, but also to clarify the molecular basis of tumours classified as ‘intermediate’ by current stromal and cancer taxonomies, thereby enabling rational therapeutic targeting. To this end, we performed imaging mass cytometry on three serial sections of a PDAC tissue microarray (221 resected tumours, ∼4 cores each), generating >800 multiplexed images (40–43 channels) focused on epithelial, immune, or stromal biomarkers. We profiled 76 immune and stromal cell types and states (hypoxic, proliferative, apoptotic, under tension), as well as six tumour phenotypes defined by expression of epithelial transcription factors (GATA6, FOXA2, PDX1), classical markers (AGR2, TFF1, CEACAM6), basal markers (TP63, KRT5, S100A2, CAV1), and other PDAC-associated proteins (S100A4, MMP7, MUC16). These six cancer cell types captured the classical and basal PDAC signatures, along with four discrete “intermediate” states with distinct associations to stromal heterogeneity, RNA subtype (n = 92), tumour ploidy (n = 182), and patient outcome. These phenotypes are also detectable in unmatched single-cell RNA-seq data (n = 163), though with more overlap in marker expression. Using matched 30X whole genome sequencing (n = 182), we identified mutations and copy number alterations linked to shifts in cancer and stromal cell phenotypes, raising the question of which molecular axis best informs clinical prognosis. To address this, we applied a modified version of Stabl, a Lasso-based machine learning approach, to compare across omic layers and identify features most strongly associated with overall survival. Per-modality analysis showed that models incorporating omics outperformed those based on clinical features alone, with imaging data slightly outperforming genomics. Cross-omic integration revealed several prognostically relevant copy number aberrations, fibroblast phenotypes, and cancer cell states. By consolidating information on tumour phenotypes, stromal niches, and genomic alterations, this work aims to focus future drug discovery on the most clinically impactful molecular features in PDAC. Citation Format: Ferris Nowlan, Noor Shakfa, Tiak Ju Tan, Sibyl Drissler, Elizabeth Sunnucks, Jennifer L. Gorman, Chengxin Yu, Sheng-Ben Liang, Barbara Gruenwald, Ayelet Borgida, Edward L. Chen, Golnaz Abazari, Miralem Mrkonjic, Julie M. Wilson, Kieran R. Campbell, Robert C. Grant, Anne-Claude Gringas, Grainne M. O'Kane, Faiyaz Notta, Steve Gallinger, Hartland W. Jackson. Integrative proteogeonomic profiling of PDAC reveals updated epithelial subtypes and cross-omic predictors of survival [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research—Emerging Science Driving Transformative Solutions; Boston, MA; 2025 Sep 28-Oct 1; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl_3):Abstract nr B116.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".