TwinWeaver: An LLM-Based Foundation Model Framework for Pan-Cancer Digital Twins

TL;DR

TwinWeaver is a novel framework that leverages large language models to improve forecasting and risk stratification in pan-cancer clinical data, enabling more accurate and interpretable digital twins for precision oncology.

Contribution

It introduces TwinWeaver, a unified LLM-based framework for modeling sparse, multi-modal clinical time series, and demonstrates its effectiveness in building Genie Digital Twin for cancer prognosis.

Findings

Significantly reduces forecasting error with median MASE of 0.87.

Improves risk stratification with average C-index of 0.703.

Generalizes well to out-of-distribution clinical trials, outperforming baselines.

Abstract

Precision oncology requires forecasting clinical events and trajectories, yet modeling sparse, multi-modal clinical time series remains a critical challenge. We introduce TwinWeaver, an open-source framework that serializes longitudinal patient histories into text, enabling unified event prediction as well as forecasting with large language models, and use it to build Genie Digital Twin (GDT) on 93,054 patients across 20 cancer types. In benchmarks, GDT significantly reduces forecasting error, achieving a median Mean Absolute Scaled Error (MASE) of 0.87 compared to 0.97 for the strongest time-series baseline (p<0.001). Furthermore, GDT improves risk stratification, achieving an average concordance index (C-index) of 0.703 across survival, progression, and therapy switching tasks, surpassing the best baseline of 0.662. GDT also generalizes to out-of-distribution clinical trials, matching trained baselines at zero-shot and surpassing them with fine-tuning, achieving a median MASE of 0.75-0.88 and outperforming the strongest baseline in event prediction with an average C-index of 0.672 versus 0.648. Finally, TwinWeaver enables an interpretable clinical reasoning extension, providing a scalable and transparent foundation for longitudinal clinical modeling.