Boosted Random Forests for Predicting Treatment Failure of Chemotherapy Regimens

TL;DR

This study develops predictive models for chemotherapy treatment failure using real-world clinical data, employing boosted random forests to achieve high accuracy, interpretability, and insights into factors influencing treatment discontinuation.

Contribution

The paper introduces a novel feature engineering pipeline and a three-axis design framework, applying boosted random forests to predict treatment failure across five cancer types with improved interpretability.

Findings

Achieved 80% baseline accuracy and 75% F1 score.

Identified key clinical features influencing treatment failure.

Provided insights into treatment discontinuation factors.

Abstract

Cancer patients may undergo lengthy and painful chemotherapy treatments, comprising several successive regimens or plans. Treatment inefficacy and other adverse events can lead to discontinuation (or failure) of these plans, or prematurely changing them, which results in a significant amount of physical, financial, and emotional toxicity to the patients and their families. In this work, we build treatment failure models based on the Real World Evidence (RWE) gathered from patients' profiles available in our oncology EMR/EHR system. We also describe our feature engineering pipeline, experimental methods, and valuable insights obtained about treatment failures from trained models. We report our findings on five primary cancer types with the most frequent treatment failures (or discontinuations) to build unique and novel feature vectors from the clinical notes, diagnoses, and medications that are available in our oncology EMR. After following a novel three axes - performance, complexity and explainability - design exploration framework, boosted random forests are selected because they provide a baseline accuracy of 80% and an F1 score of 75%, with reduced model complexity, thus making them more interpretable to and usable by oncologists.