Digital Health Innovation and Informatics

SS 16 - Digital Health Innovation & Informatics 1

116 - Applying a Machine Learning Approach to Predict Acute Toxicities During Radiation for Breast Cancer Patients

Monday, October 22
4:15 PM - 4:25 PM
Location: Room 007 A/B

Applying a Machine Learning Approach to Predict Acute Toxicities During Radiation for Breast Cancer Patients
J. Reddy1, W. D. Lindsay2, C. G. Berlind3, C. A. Ahern4, and B. D. Smith5; 1Dept. of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 2Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, 3Oncora Medical, Inc., Philadelphia, PA, 4Oncora Medical, Philadelphia, PA, 5MD Anderson Cancer Center, Houston, TX

Purpose/Objective(s): Precision oncology, an innovative paradigm employing big data and predictive analytics to predict outcomes and toxicity of cancer treatment, requires a large repository of datasources which typically reside across multiple, siloed software platforms. Our institution has developed and implemented a web-based charting tool that collects clinician-entered and validated structured data including acute RT toxicities. We merged data from this tool (known as Brocade), the institutional EHR (Epic), the treatment planning system (Pinnacle), and the record and verify system (Mosaiq) to develop predictive models of acute toxicity during RT for breast cancer patients.

Materials/Methods: From 05/2016—10/2017, 2,277 consecutive RT courses for breast cancer were administered across 5 practice sites within our Institution. For each course, >230 clinical and treatment variables were collected from the aforementioned information systems. Acute toxicity outcomes included moist desquamation and NCI CTCAEv4 grade ≥2 radiation dermatitis, breast/chest wall pain, and fatigue; all outcomes were documented by the treating physician in a structured format during RT using Brocade. Random forest, gradient boosted decision tree, and logistic regression models were trained on the initial 1,977 RT courses to predict occurrence for each outcome. Five-fold cross-validation (CV) was used to select model type and hyperparameters, using area under the ROC curve (AUC) to measure performance. The best performing model for each outcome was then evaluated on an independent validation set consisting of the subsequent 300 consecutive courses of RT for breast cancer. Models with an AUC > 0.70 were considered clinically valid.

Results: Among the 2,277 patients, 99.6% were female and median age was 58 years (IQR: 48-66). The average AUCs across CV folds of the training set for all outcomes are listed in the table, as well as incidence of acute toxicities. All AUCs in the training set were >0.70. In the validation set, the incidence of radiation dermatitis, moist desquamation, and breast/chest wall pain was 27.7%, 6.7%, and 2.7%, respectively. The AUC values for these toxicity models were 0.85, 0.82, 0.77, respectively, meeting the threshold for clinical validity. Fatigue, noted in 3.7% of patients, had an AUC of 0.56.

Conclusion: Application of this machine learning-based approach yielded clinically valid models for moist desquamation, grade ≥2 radiation dermatitis, and grade ≥2 breast pain. To our knowledge, this is the first demonstration of the ability of a precision oncology approach to accurately predict acute RT toxicities in a prospective validation dataset. This approach could help identify patients who may benefit from early interventions to avert acute RT toxicity.
Training Set (n=1977)
Dermatitis (35.6%) Moist desquamation (9.5%) Breast pain (3.8%) Fatigue (2.6%)
Random forest 0.807 0.808 0.696 0.681
Gradient boosted decision trees 0.811 0.808 0.677 0.704
Logistic regression 0.813 0.812 0.691 0.672

Author Disclosure: J. Reddy: None. W.D. Lindsay: Stock; Oncora Medical. CEO, Board member; Oncora Medical. C.A. Ahern: Stock Options; Oncora Medical. B.D. Smith: Employee; UT MD Anderson Cancer Center. Research Grant; Varian Medical Systems, Inc, MD Anderson Cancer Center. Consultant; Global Oncology One. I co-invented technology that MD Anderson has licensed to Oncora Medical. In the future, if MD Anderson chooses to develop a product with Oncora, MD Anderson may receive royalties from Oncora. If that occurs, MD Anderson may choose to pass along a fraction of these royalties to me. I have no direct financial relationships or agreements with Oncora.; Oncora Medical.

Jay Reddy, MD, PhD

Disclosure:
Employment
MD Anderson Cancer Center: Assistant Professor: Employee

Presentation(s):

Send Email for Jay Reddy


Assets

116 - Applying a Machine Learning Approach to Predict Acute Toxicities During Radiation for Breast Cancer Patients



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Applying a Machine Learning Approach to Predict Acute Toxicities During Radiation for Breast Cancer Patients