A Bayesian approach to sharing information on sensitivity of a Multi-Cancer Early Detection test across and within tumour types and stages

Sofia Dias; Yiwen Liu; Stephen Palmer; Marta O Soares

Home/Archive/arXiv

arXiv · 2504.21517

A Bayesian approach to sharing information on sensitivity of a Multi-Cancer Early Detection test across and within tumour types and stages

Sofia Dias·Yiwen Liu·Stephen Palmer·Marta O Soares

View original source ↗Read source document ↗Export metadata ↓

Abstract

The Galleri (R) (GRAIL) multi-cancer early detection test measures circulating tumour DNA (ctDNA) to predict the presence of more than 50 different cancers, from a blood test. If sensitivity of the test to detect early-stage cancers is high, using it as part of a screening programme may lead to better cancer outcomes, but available evidence indicates there is heterogeneity in sensitivity between cancer types and stages. We describe a framework for sharing evidence on test sensitivity between cancer types and/or stages, examining whether models with different sharing assumptions are supported by the evidence and considering how further data could be used to strengthen inference. Bayesian hierarchical models were fitted, and the impact of information sharing in increasing precision of the estimates of test sensitivity for different cancer types and stages was examined. Assumptions on sharing were informed by evidence from a review of the literature on the determinants of ctDNA shedding and its detection in a blood test. Support was strongest for the assumption that sensitivity can be shared only across stage 4 for all cancer types. There was also support for the assumption that sensitivities can be shared across cancer types for each stage, if cancer types expected to have low sensitivity are excluded which increased precision of early-stage cancer sensitivity estimates and was considered the most appropriate model. High heterogeneity limited improvements in precision. For future research, elicitation of expert opinion could inform more realistic sharing assumptions.

Explore related subjects

stat.ME

Keep this discovery

BibTeX RIS

Sofia Dias, Yiwen Liu, Stephen Palmer, Marta O Soares. 2025-04-30. A Bayesian approach to sharing information on sensitivity of a Multi-Cancer Early Detection test across and within tumour types and stages. https://arxiv.org/abs/2504.21517

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Surprise Reduction and Nullification in Bayesian and Inverse Bayesian Inference under Ambiguous Prediction-Error Attribution

In non-stationary environments, prediction errors may signal environmental change or transient outliers, and adaptive systems must track such changes without overreacting to outliers. We distinguish surprise reduction, which updates beliefs to fit observations, from surprise nullification, which weakens constraints imposed by the predictive structure, and formalize both within Bayesian and inverse Bayesian (BIB) inference. Belief and likelihood updates are derived from variational objectives sharing a nullification strength, determined endogenously by minimizing surprise under the candidate post-update predictive distribution. In the Gaussian case, nullification expands belief and likelihood variances by a common factor relative to standard Bayesian updating, leaving the ratio unchanged. BIB thus defers attribution of the prediction error, committing to neither latent-state change nor observation-process uncertainty. The nullification strength is carried over as a candidate and is maintained or released according to the predictive surprise of the next observation. In a mean estimation task with outliers and changepoints, no scanned parameter setting of a Sage-Husa-type adaptive Kalman filter, fixed-strength BIB variant, or belief-forgetting-only variant outperforms BIB in both changepoint tracking and post-outlier stability. An oracle-informed reduced Bayesian model tracks changepoints better but is less stable after outliers. Although BIB maintains no explicit hypotheses about changepoints or outliers, it generates event-dependent dynamics. The learning rate increases after changepoints, whereas after outliers, nullification is released, and this increase is suppressed. Deferring attribution and letting subsequent observations differentiate the responses may constitute a principle of adaptive inference in non-stationary environments.

stat.ME↗

arXiv2026

Generalized Ridge Refitting for the Lasso and Prediction Improvement Bounds

We study a class of Lasso based estimators obtained by applying a quadratic correction on the Lasso equicorrelation set. The penalty matrix determines both the magnitude and geometry of the correction and contains, among other cases, the isotropic Lasso--Ridge correction, least squares refitting, Gram proportional interpolation between the Lasso and least squares, and coordinate specific penalties. We first derive a closed form representation and isolate the positive gain component of the resulting prediction improvement. We then control the remaining stochastic linear term in expectation by localizing the random signed equicorrelation model around a deterministic reference support. This yields a finite sample expectation bound that explicitly accounts for the randomness induced by Lasso model selection. The resulting decomposition provides a unified framework for understanding when Lasso based quadratic corrections can improve prediction.

stat.ME↗

arXiv2026

Discretization in covariate-adaptive randomization: gains and losses

Covariate-adaptive randomization(CAR) is widely implemented in clinical trials to balance prognostic covariates across treatment arms. Continuous covariates are often discretized into strata in practice, yet their consequences are not clearly understood. This paper provides a comprehensive study of the impact of discretization on both the CAR design process and the inferential results thereafter. We establish the asymptotic properties of both imbalance measures and treatment effect estimators under discretized and non-discretized settings. Practical recommendations are given on when and how discretization should be employed. We show that discretization in design is generally recommended, as it enhances robustness against model misspecification. However, if the true model is known, the most efficient strategy is to balance covariates according to that model in the design. The theoretical results are corroborated by extensive simulation studies and an empirical application to a diabetes trial dataset. Together, the results clarify the gains and losses of discretization in CAR and pave the way for learning impact of discretization to other designs and beyond.

stat.ME↗