{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T05:22:35Z","timestamp":1775798555197,"version":"3.50.1"},"reference-count":47,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T00:00:00Z","timestamp":1708300800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T00:00:00Z","timestamp":1708300800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Large language models (LLMs) have been shown to have significant potential in few-shot learning across various fields, even with minimal training data. However, their ability to generalize to unseen tasks in more complex fields, such as biology and medicine has yet to be fully evaluated. LLMs can offer a promising alternative approach for biological inference, particularly in cases where structured data and sample size are limited, by extracting prior knowledge from text corpora. Here we report our proposed few-shot learning approach, which uses LLMs to predict the synergy of drug pairs in rare tissues that lack structured data and features. Our experiments, which involved seven rare tissues from different cancer types, demonstrate that the LLM-based prediction model achieves significant accuracy with very few or zero samples. Our proposed model, the CancerGPT (with\u2009~\u2009124M parameters), is comparable to the larger fine-tuned GPT-3 model (with\u2009~\u2009175B parameters). Our research contributes to tackling drug pair synergy prediction in rare tissues with limited data, and also advancing the use of LLMs for biological and medical inference tasks.<\/jats:p>","DOI":"10.1038\/s41746-024-01024-9","type":"journal-article","created":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T18:02:11Z","timestamp":1708365731000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":94,"title":["CancerGPT for few shot drug pair synergy prediction using large pretrained language models"],"prefix":"10.1038","volume":"7","author":[{"given":"Tianhao","family":"Li","sequence":"first","affiliation":[]},{"given":"Sandesh","family":"Shetty","sequence":"additional","affiliation":[]},{"given":"Advaith","family":"Kamath","sequence":"additional","affiliation":[]},{"given":"Ajay","family":"Jaiswal","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9933-2205","authenticated-orcid":false,"given":"Xiaoqian","family":"Jiang","sequence":"additional","affiliation":[]},{"given":"Ying","family":"Ding","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7815-6310","authenticated-orcid":false,"given":"Yejin","family":"Kim","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2024,2,19]]},"reference":[{"key":"1024_CR1","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1038\/s41586-023-05881-4","volume":"616","author":"M Moor","year":"2023","unstructured":"Moor, M. et al. Foundation models for generalist medical artificial intelligence. Nature 616, 259\u2013265 (2023).","journal-title":"Nature"},{"key":"1024_CR2","unstructured":"Brown, T.B. et al. Language Models are Few-Shot Learners. Preprint at https:\/\/arxiv.org\/abs\/2005.14165 (2020)."},{"key":"1024_CR3","unstructured":"OpenAI: GPT-4 Technical Report. Preprint at https:\/\/arxiv.org\/abs\/2303.08774 (2023)."},{"key":"1024_CR4","doi-asserted-by":"publisher","first-page":"2215907120","DOI":"10.1073\/pnas.2215907120","volume":"120","author":"M Mitchell","year":"2023","unstructured":"Mitchell, M. & Krakauer, D. C. The debate over understanding in AI\u2019s large language models. Proc. Natl. Acad. Sci. 120, 2215907120 (2023).","journal-title":"Proc. Natl. Acad. Sci."},{"key":"1024_CR5","unstructured":"Radford, A. et al. Language Models are Unsupervised Multitask Learners. Preprint at https:\/\/d4mucfpksywv.cloudfront.net\/better-language-models\/language-models.pdf (2018)."},{"key":"1024_CR6","doi-asserted-by":"crossref","unstructured":"Veit, A. et al. Learning from noisy large-scale datasets with minimal supervision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 839\u2013847 (2017).","DOI":"10.1109\/CVPR.2017.696"},{"key":"1024_CR7","doi-asserted-by":"crossref","unstructured":"Wertheimer, H. Few-Shot learning with localization in realistic settings. In: 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), vol. 0, pp. 6551\u20136560 (2019).","DOI":"10.1109\/CVPR.2019.00672"},{"key":"1024_CR8","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1016\/j.drudis.2016.05.015","volume":"21","author":"W Sun","year":"2016","unstructured":"Sun, W., Sanderson, P. E. & Zheng, W. Drug combination therapy increases successful drug repositioning. Drug Discov. Today 21, 1189\u20131195 (2016).","journal-title":"Drug Discov. Today"},{"key":"1024_CR9","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1093\/nar\/gkz337","volume":"47","author":"B Zagidullin","year":"2019","unstructured":"Zagidullin, B. et al. DrugComb: an integrative cancer drug combination data portal. Nucleic Acids Res. 47, 43\u201351 (2019).","journal-title":"Nucleic Acids Res."},{"key":"1024_CR10","doi-asserted-by":"crossref","unstructured":"Chen, H., & Li, J. DrugCom: Synergistic Discovery of Drug Combinations Using Tensor Decomposition. In 2018 IEEE International Conference on Data Mining (ICDM), 899\u2013904 (2018).","DOI":"10.1109\/ICDM.2018.00108"},{"key":"1024_CR11","doi-asserted-by":"publisher","first-page":"4483","DOI":"10.1093\/bioinformatics\/btaa287","volume":"36","author":"Z Sun","year":"2020","unstructured":"Sun, Z., Huang, S., Jiang, P. & Hu, P. DTF: Deep tensor factorization for predicting anticancer drug synergy. Bioinformatics 36, 4483\u20134489 (2020).","journal-title":"Bioinformatics"},{"key":"1024_CR12","doi-asserted-by":"publisher","first-page":"5446","DOI":"10.1158\/0008-5472.CAN-18-0740","volume":"78","author":"H Li","year":"2018","unstructured":"Li, H., Li, T., Quang, D. & Guan, Y. Network propagation predicts drug synergy in cancers. Cancer Res. 78, 5446\u20135457 (2018).","journal-title":"Cancer Res."},{"key":"1024_CR13","doi-asserted-by":"publisher","first-page":"2334","DOI":"10.1109\/TCBB.2021.3086702","volume":"19","author":"HI Kuru","year":"2022","unstructured":"Kuru, H. I., Tastan, O. & Cicek, A. E. MatchMaker: A deep learning framework for drug synergy prediction. IEEE\/ACM Trans. Comput. Biol. Bioinform. 19, 2334\u20132344 (2022).","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform."},{"key":"1024_CR14","doi-asserted-by":"publisher","first-page":"4782","DOI":"10.1093\/bioinformatics\/btac579","volume":"38","author":"X Liu","year":"2022","unstructured":"Liu, X. et al. Multi-way relation-enhanced hypergraph representation learning for anti-cancer drug synergy prediction. Bioinformatics 38, 4782\u20134789 (2022).","journal-title":"Bioinformatics"},{"key":"1024_CR15","doi-asserted-by":"publisher","first-page":"1538","DOI":"10.1093\/bioinformatics\/btx806","volume":"34","author":"K Preuer","year":"2018","unstructured":"Preuer, K. et al. DeepSynergy: predicting anti-cancer drug synergy with Deep Learning. Bioinformatics 34, 1538\u20131546 (2018).","journal-title":"Bioinformatics"},{"key":"1024_CR16","doi-asserted-by":"publisher","first-page":"1008653","DOI":"10.1371\/journal.pcbi.1008653","volume":"17","author":"Q Liu","year":"2021","unstructured":"Liu, Q. & Xie, L. TranSynergy: Mechanism-driven interpretable deep neural network for the synergistic prediction and pathway deconvolution of drug combinations. PLoS Comput. Biol. 17, 1008653 (2021).","journal-title":"PLoS Comput. Biol."},{"key":"1024_CR17","doi-asserted-by":"publisher","first-page":"bbac588","DOI":"10.1093\/bib\/bbac588","volume":"24","author":"S-R Hosseini","year":"2023","unstructured":"Hosseini, S.-R. & Zhou, X. CCSynergy: an integrative deep-learning framework enabling context-aware prediction of anti-cancer drug synergy. Brief. Bioinform. 24, bbac588 (2023).","journal-title":"Brief. Bioinform."},{"key":"1024_CR18","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1093\/jamia\/ocaa212","volume":"28","author":"Y Kim","year":"2021","unstructured":"Kim, Y. et al. Anticancer drug synergy prediction in understudied tissues using transfer learning. J. Am. Med. Inform. Assoc. 28, 42\u201351 (2021).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"1024_CR19","unstructured":"Hegselmann, S. et al. TabLLM: Few-shot classification of tabular data with large language models. AISTATS abs\/2210.10723, (2022)."},{"key":"1024_CR20","unstructured":"Phan, L. N. et al. SciFive: a text-to-text transformer model for biomedical literature. Preprint at https:\/\/arxiv.org\/abs\/2106.03598 (2021)."},{"key":"1024_CR21","doi-asserted-by":"publisher","first-page":"528","DOI":"10.1124\/dmd.110.035873","volume":"39","author":"L Guo","year":"2011","unstructured":"Guo, L. et al. Similarities and differences in the expression of drug-metabolizing enzymes between human hepatic cell lines and primary human hepatocytes. Drug Metab. Dispos. 39, 528\u2013538 (2011).","journal-title":"Drug Metab. Dispos."},{"key":"1024_CR22","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-20294-x","volume":"12","author":"A Warren","year":"2021","unstructured":"Warren, A. et al. Global computational alignment of tumor and cell line transcriptional profiles. Nature Communications 12, 22 (2021).","journal-title":"Nature Communications"},{"key":"1024_CR23","doi-asserted-by":"crossref","unstructured":"Deng, J. et al. ImageNet: A large-scale hierarchical image database. In IEEE Conference on Computer Vision and Pattern Recognition, 248\u2013255 (2009).","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"1024_CR24","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-019-09186-x","volume":"10","author":"F Cheng","year":"2019","unstructured":"Cheng, F., Kov\u00e1cs, I. A. & Barab\u00e1si, A.-L. Network-based prediction of drug combinations. Nat. Commun. 10, 1197 (2019).","journal-title":"Nat. Commun."},{"key":"1024_CR25","doi-asserted-by":"publisher","first-page":"762","DOI":"10.1158\/1541-7786.MCR-21-0735","volume":"20","author":"Y-C Tang","year":"2022","unstructured":"Tang, Y.-C. & Gottlieb, A. SynPathy: Predicting drug synergy through Drug-Associated pathways using deep learning. Mol. Cancer Res. 20, 762\u2013769 (2022).","journal-title":"Mol. Cancer Res."},{"key":"1024_CR26","doi-asserted-by":"crossref","unstructured":"He, L. et al. Methods for high-throughput drug combination screening and synergy scoring. In: Cancer Systems Biology: Methods and Protocols, pp. 351\u2013398 (2018).","DOI":"10.1007\/978-1-4939-7493-1_17"},{"key":"1024_CR27","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1038\/s42256-022-00458-8","volume":"4","author":"P Schramowski","year":"2022","unstructured":"Schramowski, P., Turan, C., Andersen, N., Rothkopf, C. A. & Kersting, K. Large pre-trained language models contain human-like biases of what is right and wrong to do. Nat. Mach. Intell. 4, 258\u2013268 (2022).","journal-title":"Nat. Mach. Intell."},{"key":"1024_CR28","doi-asserted-by":"publisher","first-page":"50","DOI":"10.14778\/3421424.3421431","volume":"14","author":"Y Li","year":"2020","unstructured":"Li, Y., Li, J., Suhara, Y., Doan, A. & Tan, W.-C. Deep entity matching with pre-trained language models. Proc. VLDB Endowment 14, 50\u201360 (2020).","journal-title":"Proc. VLDB Endowment"},{"key":"1024_CR29","doi-asserted-by":"publisher","first-page":"738","DOI":"10.14778\/3574245.3574258","volume":"16","author":"A Narayan","year":"2022","unstructured":"Narayan, A., Chami, I., Orr, L. & R\u00e9, C. Can Foundation Models Wrangle Your Data? Proc VLDB Endow 16, 738\u2013746 (2022).","journal-title":"Proc VLDB Endow"},{"key":"1024_CR30","unstructured":"OpenAI. Introducing ChatGPT. https:\/\/openai.com\/blog\/chatgpt (2022)."},{"key":"1024_CR31","unstructured":"OpenAI. Fine-Tuning - OpenAI API. https:\/\/platform.openai.com\/docs\/guides\/fine-tuning (2021)."},{"key":"1024_CR32","doi-asserted-by":"publisher","first-page":"3907","DOI":"10.1093\/bioinformatics\/bty452","volume":"34","author":"C Suphavilai","year":"2018","unstructured":"Suphavilai, C., Bertrand, D. & Nagarajan, N. Predicting cancer drug response using a recommender system. Bioinformatics 34, 3907\u20133914 (2018).","journal-title":"Bioinformatics"},{"key":"1024_CR33","doi-asserted-by":"crossref","unstructured":"Chen, T. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 785\u2013794 (2016).","DOI":"10.1145\/2939672.2939785"},{"key":"1024_CR34","unstructured":"Huang, X., Khetan, A., Cvitkovic, M., & Karnin, Z. TabTransformer: Tabular Data Modeling Using Contextual Embeddings. Preprint at https:\/\/arxiv.org\/abs\/2012.06678 (2020)."},{"key":"1024_CR35","doi-asserted-by":"publisher","first-page":"509","DOI":"10.3389\/fchem.2019.00509","volume":"7","author":"P Sidorov","year":"2019","unstructured":"Sidorov, P., Naulaerts, S., Ariey-Bonnet, J., Pasquier, E. & Ballester, P. J. Predicting synergism of cancer drug combinations using NCI-ALMANAC data. Front Chem. 7, 509 (2019).","journal-title":"Front Chem"},{"key":"1024_CR36","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-45236-6","volume":"9","author":"R Celebi","year":"2019","unstructured":"Celebi, R., Bear Don\u2019t Walk IV, O., Movva, R., Alpsoy, S. & Dumontier, M. In-silico prediction of synergistic Anti-Cancer drug combinations using multi-omics data. Sci. Rep. 9, 8949 (2019).","journal-title":"Sci. Rep."},{"key":"1024_CR37","unstructured":"Wolf, T. et al. Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, 38\u201345 (2020)."},{"key":"1024_CR38","first-page":"331","volume":"47","author":"WR Greco","year":"1995","unstructured":"Greco, W. R., Bravo, G. & Parsons, J. C. The search for synergy: a critical review from a response surface perspective. Pharmacol. Rev. 47, 331\u2013385 (1995).","journal-title":"Pharmacol. Rev."},{"key":"1024_CR39","doi-asserted-by":"publisher","first-page":"2413","DOI":"10.1093\/bioinformatics\/btx162","volume":"33","author":"A Ianevski","year":"2017","unstructured":"Ianevski, A., He, L., Aittokallio, T. & Tang, J. SynergyFinder: a web application for analyzing drug combination dose-response matrix data. Bioinformatics 33, 2413\u20132415 (2017).","journal-title":"Bioinformatics"},{"key":"1024_CR40","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1016\/j.csbj.2015.09.001","volume":"13","author":"B Yadav","year":"2015","unstructured":"Yadav, B., Wennerberg, K., Aittokallio, T. & Tang, J. Searching for Drug Synergy in Complex Dose-Response Landscapes Using an Interaction Potency Model. Comput. Struct. Biotechnol. J. 13, 504\u2013513 (2015).","journal-title":"Comput. Struct. Biotechnol. J."},{"key":"1024_CR41","unstructured":"Paszke, A. et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library. Preprint at https:\/\/arxiv.org\/abs\/1912.01703 (2019)."},{"key":"1024_CR42","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1186\/1756-9966-29-161","volume":"29","author":"S Guo","year":"2010","unstructured":"Guo, S. et al. Overexpression of pim-1 in bladder cancer. J. Exp. Clin. Cancer Res. 29, 161 (2010).","journal-title":"J. Exp. Clin. Cancer Res."},{"key":"1024_CR43","doi-asserted-by":"publisher","first-page":"1001","DOI":"10.1007\/s10637-013-9926-y","volume":"31","author":"R Jones","year":"2013","unstructured":"Jones, R. et al. Phase II study to assess the efficacy, safety and tolerability of the mitotic spindle kinesin inhibitor AZD4877 in patients with recurrent advanced urothelial cancer. Invest New Drugs 31, 1001\u20131007 (2013).","journal-title":"Invest New Drugs"},{"key":"1024_CR44","doi-asserted-by":"publisher","first-page":"1425","DOI":"10.1038\/s41416-018-0082-1","volume":"118","author":"J Cortes","year":"2018","unstructured":"Cortes, J. et al. Phase I studies of AZD1208, a proviral integration moloney virus kinase inhibitor in solid and haematological cancers. Br. J. Cancer 118, 1425\u20131433 (2018).","journal-title":"Br. J. Cancer"},{"key":"1024_CR45","unstructured":"National Cancer Institute. NCI Drug Dictionary. https:\/\/www.cancer.gov\/publications\/dictionaries\/cancer-drug\/def\/pan-pim-kinase-inhibitor-azd1208 (2011)."},{"key":"1024_CR46","doi-asserted-by":"publisher","first-page":"2793","DOI":"10.18632\/oncotarget.26876","volume":"10","author":"F Cervantes-Gomez","year":"2019","unstructured":"Cervantes-Gomez, F. et al. PIM kinase inhibitor, AZD1208, inhibits protein translation and induces autophagy in primary chronic lymphocytic leukemia cells. Oncotarget 10, 2793\u20132809 (2019).","journal-title":"Oncotarget"},{"key":"1024_CR47","first-page":"3580","volume":"27","author":"G Borthakur","year":"2009","unstructured":"Borthakur, G. et al. Clinical, pharmacokinetic (PK), and pharmacodynamic findings from a phase I trial of an eg5 inhibitor (AZD4877) in patients with refractory acute myeloid leukemia (AML). J. Clin. Orthodont. 27, 3580\u20133580 (2009).","journal-title":"J. Clin. Orthodont."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01024-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01024-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01024-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,19]],"date-time":"2024-02-19T18:22:16Z","timestamp":1708366936000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-024-01024-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,19]]},"references-count":47,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["1024"],"URL":"https:\/\/doi.org\/10.1038\/s41746-024-01024-9","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,19]]},"assertion":[{"value":"19 May 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 February 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 February 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"40"}}