Michele Waters

Michele Waters

Auto-generated profile, based on publicly available information

Michele Waters

Computational Oncology Service, Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, USA

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2024 in Nature
Verified
Journal Article
Open access
PDF file
JJJustin Jee
Justin Jee
CFChristopher Fong
Christopher Fong
KPKarl Pichotta
Karl Pichotta
TTThinh Ngoc Tran
Thinh Ngoc Tran
ALAnisha Luthra
Anisha Luthra
XBXinran Bi
Xinran Bi
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.

2 March 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
KBKevin M. Boehm
Kevin M. Boehm
ONOmar S. M. El Nahhas
Omar S. M. El Nahhas
AMAntonio Marra
Antonio Marra
MWMichele Waters
Michele Waters
JJJustin Jee
Justin Jee
JKJakob Nikolas Kather
Jakob Nikolas Kather
The Oncotype DX® Recurrence Score (RS) is an assay for hormone receptor-positive early breast cancer with extensively validated predictive and prognostic value. However, its cost and lag time have limited global adoption, and previous attempts to estimate it using clinicopathologic variables have had limited success. To address this, we assembled 6172 cases across three institutions and developed Orpheus, a multimodal deep learning tool to infer the RS from H&E whole-slide images. Our model identifies TAILORx high-risk cases (RS > 25) with an area under the curve (AUC) of 0.89, compared to a leading clinicopathologic nomogram with 0.73. Furthermore, in patients with RS ≤ 25, Orpheus ascertains risk of metastatic recurrence more accurately than the RS itself (0.75 vs 0.49 mean time-dependent AUC). These findings have the potential to guide adjuvant therapy for high-risk cases and tailor surveillance for patients at elevated metastatic recurrence risk.

2 July 2025 in npj Digital Medicine
Verified
Journal Article
Open access
PDF file
MZMenglei Zhu
Menglei Zhu
HLHui Lin
Hui Lin
JJJue Jiang
Jue Jiang
AJAbbas J. Jinia
Abbas J. Jinia
JJJustin Jee
Justin Jee
ALAnyi Li
Anyi Li
Subspecialty knowledge barriers have limited the adoption of large language models (LLMs) in oncology. We introduce Woollie, an open-source, oncology-specific LLM trained on real-world data from Memorial Sloan Kettering Cancer Center (MSK) across lung, breast, prostate, pancreatic, and colorectal cancers, with external validation using University of California, San Francisco (UCSF) data. Woollie surpasses ChatGPT in medical benchmarks and excels in eight non-medical benchmarks. Analyzing 39,319 radiology impression notes from 4002 patients, it achieved an overall area under the receiver operating characteristic curve (AUROC) of 0.97 for cancer progression prediction on MSK data, including a notable 0.98 AUROC for pancreatic cancer. On UCSF data, it achieved an overall AUROC of 0.88, excelling in lung cancer detection with an AUROC of 0.95. As the first oncology specific LLM validated across institutions, Woollie demonstrates high accuracy and consistency across cancer types, underscoring its potential to enhance cancer progression analysis.

12 November 2024 in Nature Communications
Verified
Journal Article
Open access
PDF file
KKKenneth L. Kehl
Kenneth L. Kehl
JJJustin Jee
Justin Jee
KPKarl Pichotta
Karl Pichotta
MPMorgan A. Paul
Morgan A. Paul
PTPavel Trukhanov
Pavel Trukhanov
NSNikolaus Schultz
Nikolaus Schultz
Databases that link molecular data to clinical outcomes can inform precision cancer research into novel prognostic and predictive biomarkers. However, outside of clinical trials, cancer outcomes are typically recorded only in text form within electronic health records (EHRs). Artificial intelligence (AI) models have been trained to extract outcomes from individual EHRs. However, patient privacy restrictions have historically precluded dissemination of these models beyond the centers at which they were trained. In this study, the vulnerability of text classification models trained directly on protected health information to membership inference attacks is confirmed. A teacher-student distillation approach is applied to develop shareable models for annotating outcomes from imaging reports and medical oncologist notes. ‘Teacher’ models trained on EHR data from Dana-Farber Cancer Institute (DFCI) are used to label imaging reports and discharge summaries from the Medical Information Mart for Intensive Care (MIMIC)-IV dataset. ‘Student’ models are trained to use these MIMIC documents to predict the labels assigned by teacher models and sent to Memorial Sloan Kettering (MSK) for evaluation. The student models exhibit high discrimination across outcomes in both the DFCI and MSK test sets. Leveraging private labeling of public datasets to distill publishable clinical AI models from academic centers could facilitate deployment of machine learning to accelerate precision oncology research.

28 October 2024 in Nature Genetics
Verified
Journal Article
Open access
PDF file
CTCerise Tang
Cerise Tang
VCVenise Jan Castillon
Venise Jan Castillon
MWMichele Waters
Michele Waters
CFChris Fong
Chris Fong
TPTricia Park
Tricia Park
EREd Reznik
Ed Reznik
Obesity is a risk factor for cancer, but whether obesity is linked to specific genomic subtypes of cancer is unknown. We examined the relationship between obesity and tumor genotype in two clinicogenomic corpora. Obesity was associated with specific driver mutations in lung adenocarcinoma, endometrial carcinoma and cancers of unknown primaries, independent of clinical covariates, demographic factors and genetic ancestry. Obesity is therefore a driver of etiological heterogeneity in some cancers.

26 September 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
TTThinh N. Tran
Thinh N. Tran
CFChris Fong
Chris Fong
KPKarl Pichotta
Karl Pichotta
ALAnisha Luthra
Anisha Luthra
RSRonglai Shen
Ronglai Shen
JJJustin Jee
Justin Jee
Characterizing and validating which mutations influence development of cancer is challenging. Artificial intelligence (AI) has delivered significant advances in protein structure prediction, but its utility for identifying cancer drivers is less explored. We evaluate multiple computational methods for identifying cancer driver mutations. For re-identifying known drivers, methods incorporating protein structure or functional genomic data outperform methods trained only on evolutionary data. We validate variants of unknown significance (VUSs) annotated as pathogenic by testing their association with overall survival in two cohorts of patients with non-small cell lung cancer (N = 7965 and 977). VUSs identified as pathogenic drivers by AI in KEAP1 and SMARCA4 are associated with worse survival, unlike “benign” VUSs. “Pathogenic” VUSs also exhibit mutual exclusivity with known oncogenic alterations at the pathway level, further suggesting biological validity. AI predictions thus contribute to a more comprehensive understanding of tumor genetics as validated by real-world data.