Junyi Gao

Background: While maintenance rituximab therapy (MT) for follicular lymphoma improves progression-free survival, individual benefit varies, creating a critical need to avoid overtreatment and its associated burdens. Methods: We developed a double machine learning framework using a retrospective 404-patient cohort to estimate the individualized treatment effect of MT on the risk of disease progression within 24 months. Results: Our model revealed heterogeneity in treatment benefit, successfully distinguishing patients most likely to benefit from those who are not. A retrospective simulated application identified a “low-risk, low-benefit” subgroup that might potentially forgo MT. Furthermore, aligning historical clinical decisions with our model’s recommendations was associated with a lower progression rate compared to nonalignment (14.8% vs. 38.0%). Extensive sensitivity analyses confirmed the robustness of this treatment heterogeneity against potential unmeasured confounding and temporal practice shifts. Conclusions: Supported by an updated interactive clinical decision tool, this data-driven framework provides a robust strategy to personalize MT by separating prognostic risk from predictive benefit. It serves as a hypothesis-generating template to guide future prospective risk-adapted trials and reduce unnecessary treatment in oncology.

Clinical trial-site selection is often inefficient, leading to low enrolment, poor participant diversity and costly delays. We developed DocTr, a cross-modal deep learning framework to optimize this process. DocTr uniquely integrates patient encounter data from medical claims, unstructured trial documents and historical enrolment relationships from OpenPayments data to recommend clinician investigators, specifically optimizing for recommendation accuracy, demographic fairness and operational efficiency. Evaluated on 24,984 clinicians and 5,210 trials, DocTr achieved 58% higher match similarity than leading baselines. A genetic optimization algorithm further refines recommendations, improving fairness scores related to patient race and ethnicity by up to 25% compared with the ground-truth enrolment while minimizing competing trials to near zero. DocTr also provides accurate recruitment cost estimations. By making site selection substantially more efficient, accurate and fair, this model offers a powerful method to accelerate patient access to new therapies.

Objectives: Large language models’ (LLMs’) performance in high-stakes, compliance-driven settings such as drafting clinical research documents remains underexplored. This study aims to build a benchmark and an evaluation framework for assessing LLMs’ compliance and factuality in generating informed consent forms (ICFs) from clinical trial protocols. Materials and Methods: We introduce InformBench, a benchmark comprising 900 clinical trial documents, and propose an evaluation framework grounded in regulatory guidelines and site-specific consent templates. We assess LLM performance on transforming trial protocols, often hundreds of pages, into concise, patient-facing ICFs. Additionally, we design InformGen, a retrieval-augmented, human-in-the-loop pipeline aimed at improving generation quality. Results: Baseline LLMs such as GPT-4o achieved only 70%-80% compliance and exhibited factual errors in 18%-43% of cases. In contrast, InformGen substantially improved outputs, achieving nearly 100% regulatory compliance and over 90% factual accuracy, as validated by 5 domain-expert annotators. Discussion: The study reveals critical limitations in current LLMs for clinical research document drafting, particularly in regulatory sensitivity and factual grounding. Our results highlight the need for domain-specific benchmarks and structured evaluations to support safe deployment in real-world clinical research workflows. Conclusion: LLMs offer value in clinical research document generation but must be adapted and rigorously evaluated for high-stakes applications. Our benchmark and framework provide a foundation for improving and assessing LLM-generated outputs in compliance-critical domains.