ODFormer: a Virtual Organoid for Predicting Personalized Therapeutic Responses in Pancreatic Cancer

Jing Xu, Xiaolu Yang, Yunguang Li, Huan Wang, Yikai Li, Shijie Tang, Shiwei Guo, Jiawei Zou, Xiaohan Shi, Zexi Chen, Chaoliang Zhong, Penghao Li, Wei Jing, Kailian Zheng, Xinqian Wu, Dong Gao, Luonan Chen, Gang Jin
2025 · doi:10.1101/2025.07.08.663664
preprint OA: closed CC-BY-4.0
📄 Open PDF Full text JSON View at publisher
⚙ AI-generated summary by claude@2026-06, 2026-06-09 ⓘ

ODFormer is a virtual organoid model developed to predict individual patient responses to pancreatic cancer therapies by integrating multi-omics data.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

⚙ AI-generated deep summary by claude@2026-06, 2026-06-09 · read from full text ⓘ

This paper studied the development of ODFormer, a computational framework intended to simulate pancreatic cancer patient-derived organoids by integrating transcriptomic and mutational profiles to predict patient-specific drug responses without performing physical organoid assays. Using pretrained encoders (pan-cancer bulk transcriptomics and pancreatic cancer single-cell profiles) and training on a curated drug-response dataset spanning 183 PDOs and 98 drugs, ODFormer outperformed existing methods, achieving a reported standardized drug-response prediction performance with PCC > 0.9. The authors further report multi-cohort retrospective and independent-dataset validation (including TCGA-PDAC) and high concordance with prospective clinical responses by CA19-9, while also identifying resistance biomarkers and therapy-responsive/non-responsive subgroups. This paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Full text 1,780 characters · extracted from oa-doi-fallback · click to expand
Abstract Pancreatic cancer (PC) patient-derived organoids (PDOs) faithfully recapitulate therapeutic responses but face clinical translation barriers, including high costs and technical complexity. To address these problems and the lack of frameworks for PDO-based drug-response assays, we developed ODFormer, a computational framework that simulates PC PDOs to predict clinically actionable, patient-specific drug responses by integrating transcriptomic and mutational profiles. ODFormer first employed two encoders, pretrained on 30,000 pan-cancer bulk transcriptomics and 1 million PC single-cell profiles respectively, to distil tissue-and organoid-specific representations. Then, trained on our curated 14,000 PDO drug-response assay (across 183 PDOs and 98 drugs) using a transformer–augmented hybrid contrastive network, ODFormer significantly outperformed state-of-the-art methods, notably achieving a PCC >0.9 in predicting standardized drug response. Multi-cohort retrospective analyses further demonstrated that ODFormer-guided personalized therapy significantly improves clinical outcomes, without requiring physical organoid assays. Furthermore, ODFormer identified novel clinico-biological PC subtypes and revealed therapy resistance biomarkers by stratifying predicted responders and non-responders. These were validated using independent datasets including TCGA-PDAC. Notably, ODFormer-guided treatment efficacy showed high concordance with prospective clinical responses by CA19-9. Competing Interest Statement The authors have declared no competing interest. Footnotes We have corrected some word mistakes and grammar issues Data Availability All datasets used in this study are publicly available and the usages are fully illustrated in the Supplementary Table.

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

⚙ Ask this paper AI returns verbatim quotes from the full text · source: oa-doi-fallback ⓘ

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Funding

funders
[{'doi': '10.13039/501100012166', 'name': 'National Key R&D Program of China', 'awards': ['2022YFA1004800']}, {'doi': '10.13039/501100012166', 'name': 'National Key R&D Program of China', 'awards': ['2025YFF1207900']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['T2341007']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['T2350003']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['12131020']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['42450084']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['42450135']}, {'doi': '10.13039/501100001809', 'name': 'Natural Science Foundation of China', 'awards': ['12326614,12426310']}, {'doi': None, 'name': 'Zhejiang Province Vanguard Goose-Leading Initiative', 'awards': []}, {'doi': None, 'name': 'Zhejiang Province Vanguard Goose-Leading Initiative', 'awards': ['2025C01114']}, {'doi': None, 'name': 'Hangzhou Institute for advanced study of UCAS', 'awards': ['2024HIAS-P004']}, {'doi': None, 'name': 'JST Moonshot R&D', 'awards': ['JPMJMS2021']}]

Citation neighborhood (sparse)

Too few in-corpus citations on either side for a chart; here are the lists.

Cites (2)

References (53)

Source provenance

crossref
last seen: 2026-05-23T01:00:16.635496+00:00
europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-21T05:10:58.409756+00:00
License: CC-BY-4.0