Good Rankings, Wrong Probabilities: A Calibration Audit of Multimodal Cancer Survival Models

TL;DR

This paper systematically audits the calibration of multimodal cancer survival models, revealing widespread miscalibration despite strong discriminative performance, and evaluates methods to improve calibration.

Contribution

First comprehensive fold-level calibration audit of multimodal WSI-genomics survival models, highlighting calibration issues and evaluating post-hoc calibration methods.

Findings

All tested models fail 1-calibration on most folds.

Gating-based fusion improves calibration over other methods.

Platt scaling reduces miscalibration without harming discrimination.

Abstract

Multimodal deep learning models that fuse whole-slide histopathology images with genomic data have achieved strong discriminative performance for cancer survival prediction, as measured by the concordance index. Yet whether the survival probabilities derived from these models - either directly from native outputs or via standard post-hoc reconstruction - are calibrated remains largely unexamined. We conduct, to our knowledge, the first systematic fold-level 1-calibration audit of multimodal WSI-genomics survival architectures, evaluating native discrete-time survival outputs (Experiment A: 3 models on TCGA-BRCA) and Breslow-reconstructed survival curves from scalar risk scores (Experiment B: 11 architectures across 5 TCGA cancer types). In Experiment A, all three models fail 1-calibration on a majority of folds (12 of 15 fold-level tests reject after Benjamini-Hochberg correction). Across the full 290 fold-level tests, 166 reject the null of correct calibration at the median event time after Benjamini-Hochberg correction (FDR = 0.05). MCAT achieves C-index 0.817 on GBMLGG yet fails 1-calibration on all five folds. Gating-based fusion is associated with better calibration; bilinear and concatenation fusion are not. Post-hoc Platt scaling reduces miscalibration at the evaluated horizon (e.g., MCAT: 5/5 folds failing to 2/5) without affecting discrimination. The concordance index alone is insufficient for evaluating survival models intended for clinical use.