# OncoReason: Structuring Clinical Reasoning in LLMs for Robust and Interpretable Survival Prediction

arXiv: https://arxiv.org/abs/2510.17532 · HTML: https://mlsanity.com/paper/2510.17532/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 18 of its 81 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)

## Builds on (most cited first, 18 of 18)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [DeepSeek-VL: Towards Real-World Vision-Language Understanding](https://mlsanity.com/paper/2403.05525/) (2024)
- [MEDITRON-70B: Scaling Medical Pretraining for Large Language Models](https://mlsanity.com/paper/2311.16079/) (2023)
- [Med42-v2: A Suite of Clinical LLMs](https://mlsanity.com/paper/2408.06142/) (2024)
- [Training Chain-of-Thought via Latent-Variable Inference](https://mlsanity.com/paper/2312.02179/) (2023)
- [Parameter-Efficient Fine-Tuning of LLaMA for the Clinical Domain](https://mlsanity.com/paper/2307.03042/) (2023)
- [Prompting Large Language Models for Zero-Shot Clinical Prediction with Structured Longitudinal Electronic Health Record Data](https://mlsanity.com/paper/2402.01713/) (2024)
- [A Hopfieldian View-based Interpretation for Chain-of-Thought Reasoning](https://mlsanity.com/paper/2406.12255/) (2024)
- [Focus on Your Question! Interpreting and Mitigating Toxic CoT Problems in Commonsense Reasoning](https://mlsanity.com/paper/2402.18344/) (2024)
- [Fine-Tuning a Local LLaMA-3 Large Language Model for Automated Privacy-Preserving Physician Letter Generation in Radiation Oncology](https://mlsanity.com/paper/2408.10715/) (2024)
- [Large Language Models for Medical Forecasting -- Foresight 2](https://mlsanity.com/paper/2412.10848/) (2024)

## Built on by (most cited first, 1 of 1)

- [TwinWeaver: An LLM-Based Foundation Model Framework for Pan-Cancer Digital Twins](https://mlsanity.com/paper/2601.20906/) (2026)

## Abstract (arXiv)

Predicting cancer treatment outcomes requires models that are both accurate and interpretable, particularly in the presence of heterogeneous clinical data. While large language models (LLMs) have shown strong performance in biomedical NLP, they often lack structured reasoning capabilities critical for high-stakes decision support. We present a unified, multi-task learning framework that aligns autoregressive LLMs with clinical reasoning for outcome prediction on the MSK-CHORD dataset. Our models are trained to jointly perform binary survival classification, continuous survival time regression, and natural language rationale generation. We evaluate three alignment strategies: (1) standard supervised fine-tuning (SFT), (2) SFT with Chain-of-Thought (CoT) prompting to elicit step-by-step reasoning, and (3) Group Relative Policy Optimization (GRPO), a reinforcement learning method that aligns model outputs to expert-derived reasoning trajectories. Experiments with LLaMa3-8B and Med42-8B backbones demonstrate that CoT prompting improves F1 by +6.0 and reduces MAE by 12%, while GRPO achieves state-of-the-art interpretability and predictive performance across BLEU, ROUGE, and BERTScore. We further show that existing biomedical LLMs often fail to produce valid reasoning traces due to architectural constraints. Our findings underscore the importance of reasoning-aware alignment in multi-task clinical modeling and set a new benchmark for interpretable, trustworthy LLMs in precision oncology.