Automated real-world data integration improves cancer outcome prediction.

The digitization of health records and growing availability of tumour DNA sequencing provide an opportunity to study the determinants of cancer outcomes with unprecedented richness. Patient data are often stored in unstructured text and siloed datasets. Here we combine natural language processing annotations1,2 with structured medication, patient-reported demographic, tumour registry and tumour genomic data from 24,950 patients at Memorial Sloan Kettering Cancer Center to generate a clinicogenomic, harmonized oncologic real-world dataset (MSK-CHORD). MSK-CHORD includes data for non-small-cell lung (n = 7,809), breast (n = 5,368), colorectal (n = 5,543), prostate (n = 3,211) and pancreatic (n = 3,109) cancers and enables discovery of clinicogenomic relationships not apparent in smaller datasets. Leveraging MSK-CHORD to train machine learning models to predict overall survival, we find that models including features derived from natural language processing, such as sites of disease, outperform those based on genomic data or stage alone as tested by cross-validation and an external, multi-institution dataset. By annotating 705,241 radiology reports, MSK-CHORD also uncovers predictors of metastasis to specific organ sites, including a relationship between SETD2 mutation and lower metastatic potential in immunotherapy-treated lung adenocarcinoma corroborated in independent datasets. We demonstrate the feasibility of automated annotation from unstructured notes and its utility in predicting patient outcomes. The resulting data are provided as a public resource for real-world oncologic research.

Nature. 2024 Nov 06 [Epub ahead of print]

Justin Jee, Christopher Fong, Karl Pichotta, Thinh Ngoc Tran, Anisha Luthra, Michele Waters, Chenlian Fu, Mirella Altoe, Si-Yang Liu, Steven B Maron, Mehnaj Ahmed, Susie Kim, Mono Pirun, Walid K Chatila, Ino de Bruijn, Arfath Pasha, Ritika Kundra, Benjamin Gross, Brooke Mastrogiacomo, Tyler J Aprati, David Liu, JianJiong Gao, Marzia Capelletti, Kelly Pekala, Lisa Loudon, Maria Perry, Chaitanya Bandlamudi, Mark Donoghue, Baby Anusha Satravada, Axel Martin, Ronglai Shen, Yuan Chen, A Rose Brannon, Jason Chang, Lior Braunstein, Anyi Li, Anton Safonov, Aaron Stonestrom, Pablo Sanchez-Vela, Clare Wilhelm, Mark Robson, Howard Scher, Marc Ladanyi, Jorge S Reis-Filho, David B Solit, David R Jones, Daniel Gomez, Helena Yu, Debyani Chakravarty, Rona Yaeger, Wassim Abida, Wungki Park, Eileen M O'Reilly, Julio Garcia-Aguilar, Nicholas Socci, Francisco Sanchez-Vega, Jian Carrot-Zhang, Peter D Stetson, Ross Levine, Charles M Rudin, Michael F Berger, Sohrab P Shah, Deborah Schrag, Pedram Razavi, Kenneth L Kehl, Bob T Li, Gregory J Riely, Nikolaus Schultz, MSK Cancer Data Science Initiative Group

Memorial Sloan Kettering Cancer Center, New York, NY, USA., Dana Farber Cancer Institute, Boston, MA, USA., Caris Life Sciences, Irving, TX, USA., Memorial Sloan Kettering Cancer Center, New York, NY, USA. .