Heterogeneous Quantile Treatment Effect Estimation for Longitudinal Data with High-Dimensional Confounding

2025-08-22

AI Summary

This study addresses the heterogeneous causal inference challenge regarding chemotherapy’s dynamic impact on circulating tumor DNA (ctDNA) in non-small cell lung cancer (NSCLC), confronting three key issues: high-dimensional confounding, unknown correlation structures in longitudinal repeated measurements, and heavy-tailed, non-Gaussian ctDNA trajectory distributions. To this end, we propose a novel framework integrating convolutional smoothing quantile regression with orthogonal random forests, enabling robust, heterogeneous estimation of quantile treatment effects in high dimensions—while supporting covariate-driven individualized subgroup identification and maintaining insensitivity to misspecification of nuisance parameters. We establish theoretical consistency and asymptotic normality of the estimator. Simulation studies demonstrate superior finite-sample performance. Applied to a real NSCLC cohort, our method uncovers subgroup-specific effects of chemotherapy on ctDNA clearance rates, delivering an interpretable and generalizable methodological foundation for precision efficacy assessment.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Causality
• Search and Optimization: Non-convex Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Causal inference plays a fundamental role in various real-world applications. However, in the motivating non-small cell lung cancer (NSCLC) study, it is challenging to estimate the treatment effect of chemotherapy on circulating tumor DNA (ctDNA). First, the heterogeneous treatment effects vary across patient subgroups defined by baseline characteristics. Second, there exists a broad set of demographic, clinical and molecular variables act as potential confounders. Third, ctDNA trajectories over time show heavy-tailed non-Gaussian behavior. Finally, repeated measurements within subjects introduce unknown correlation. Combining convolution-smoothed quantile regression and orthogonal random forest, we propose a framework to estimate heterogeneous quantile treatment effects in the presence of high-dimensional confounding, which not only captures effect heterogeneity across covariates, but also behaves robustly to nuisance parameter estimation error. We establish the theoretical properties of the proposed estimator and demonstrate its finite-sample performance through comprehensive simulations. We illustrate its practical utility in the motivated NSCLC study.

Problem

• Estimating heterogeneous treatment effects across patient subgroups
• Addressing high-dimensional confounding from demographic and clinical variables
• Modeling heavy-tailed non-Gaussian longitudinal data with correlations

Innovation

• Convolution-smoothed quantile regression
• Orthogonal random forest framework
• High-dimensional confounding robust estimation