Although cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) have been proven to improve outcomes for patients with breast cancer, choosing when to initiate CDK4/6i therapy often presents challenges. While the treatment can be initiated either in the first- or second-line setting, or with an aromatase inhibitor or fulvestrant, studies have not directly compared these 2 options, nor have they determined the benefit of continuation of CDKi therapy with changes in hormonal therapy backbone.
Researchers attempted to solve this problem with an algorithm-based efficacy prediction, according to an exploratory analysis presented at the 2019 San Antonio Breast Cancer Symposium.
By utilizing patient-level data from 8 randomized controlled pivotal trials totaling 4580 breast cancer patients, the researchers performed exploratory analyses of the predictive power of the baseline patient and disease characteristic data. The variables included age, race, ethnicity, country and continent of origin, height, weight, body mass index, menopause status, estrogen receptor status, progesterone receptor status, the Eastern Cooperative Oncology Group performance score, histological type and grade, and initial disease stage.
After gathering the data, researchers created baseline random survival forest models for patients that received a CDKi plus hormonal therapy and for patients that received hormonal therapy alone.
The preliminary findings produced prediction accuracies of 69.2% for the CDKi model, and 70.6% for the hormonal therapy alone model. Median predicted survival times for patients were calculated in both groups and used to identify patients that were more likely to benefit from initial use of CDKi therapy.
Through further analysis, the researchers also identified patient characteristics that are prognostic of survival.
The authors noted that future efforts are necessary to extend the prediction capabilities for adverse event development.
Mason J, Gong Y, Amiri-Kordestani L, Wedam S, et al. Prediction of CDK inhibitor efficacy in ER+/HER2- breast cancer using machine learning algorithms. Accessed Dec 4, 2019. https://plan.core-apps.com/sabcs2019/abstract/c3d31d6ffb8feb46fe802df1c90511a7.