Favoring the hierarchical constraint in penalized survival models for randomized trials in precision medicine.

BACKGROUND: The research of biomarker-treatment interactions is commonly investigated in randomized clinical trials (RCT) for improving medicine precision. The hierarchical interaction constraint states that an interaction should only be in a model if its main effects are also in the model. However, this constraint is not guaranteed in the standard penalized statistical approaches. We aimed to find a compromise for high-dimensional data between the need for sparse model selection and the need for the hierarchical constraint. RESULTS: To favor the property of the hierarchical interaction constraint, we proposed to create groups composed of the biomarker main effect and its interaction with treatment and to perform the bi-level selection on these groups. We proposed two weighting approaches (Single Wald (SW) and likelihood ratio test (LRT)) for the adaptive lasso method. The selection performance of these two approaches is compared to alternative lasso extensions (adaptive lasso with ridge-based weights, composite Minimax Concave Penalty, group exponential lasso and Sparse Group Lasso) through a simulation study. A RCT (NSABP B-31) randomizing 1574 patients (431 events) with early breast cancer aiming to evaluate the effect of adjuvant trastuzumab on distant-recurrence free survival with expression data from 462 genes measured in the tumour will serve for illustration. The simulation study illustrates that the adaptive lasso LRT and SW, and the group exponential lasso favored the hierarchical interaction constraint. Overall, in the alternative scenarios, they had the best balance of false discovery and false negative rates for the main effects of the selected interactions. For NSABP B-31, 12 gene-treatment interactions were identified more than 20% by the different methods. Among them, the adaptive lasso (SW) approach offered the best trade-off between a high number of selected gene-treatment interactions and a high proportion of selection of both the gene-treatment interaction and its main effect. CONCLUSIONS: Adaptive lasso with Single Wald and likelihood ratio test weighting and the group exponential lasso approaches outperformed their competitors in favoring the hierarchical constraint of the biomarker-treatment interaction. However, the performance of the methods tends to decrease in the presence of prognostic biomarkers.

Escherichia coli-Specific CXCL13-Producing TFH Are Associated with Clinical Efficacy of Neoadjuvant PD-1 Blockade against Muscle-Invasive Bladder Cancer.

Biomarkers guiding the neoadjuvant use of immune-checkpoint blockers (ICB) are needed for patients with localized muscle-invasive bladder cancers (MIBC). Profiling tumor and blood samples, we found that follicular helper CD4+ T cells (TFH) are among the best therapeutic targets of pembrolizumab correlating with progression-free survival. TFH were associated with tumoral CD8 and PD-L1 expression at baseline and the induction of tertiary lymphoid structures after pembrolizumab. Blood central memory TFH accumulated in tumors where they produce CXCL13, a chemokine found in the plasma of responders only. IgG4+CD38+ TFH residing in bladder tissues correlated with clinical benefit. Finally, TFH and IgG directed against urothelium-invasive Escherichia coli dictated clinical responses to pembrolizumab in three independent cohorts. The links between tumor infection and success of ICB immunomodulation should be prospectively assessed at a larger scale. SIGNIFICANCE: In patients with bladder cancer treated with neoadjuvant pembrolizumab, E. coli-specific CXCL13 producing TFH and IgG constitute biomarkers that predict clinical benefit. Beyond its role as a biomarker, such immune responses against E. coli might be harnessed for future therapeutic strategies. This article is highlighted in the In This Issue feature, p. 2221.

Accounting for grouped predictor variables or pathways in high-dimensional penalized Cox regression models.

BACKGROUND: The standard lasso penalty and its extensions are commonly used to develop a regularized regression model while selecting candidate predictor variables on a time-to-event outcome in high-dimensional data. However, these selection methods focus on a homogeneous set of variables and do not take into account the case of predictors belonging to functional groups; typically, genomic data can be grouped according to biological pathways or to different types of collected data. Another challenge is that the standard lasso penalisation is known to have a high false discovery rate. RESULTS: We evaluated different penalizations in a Cox model to select grouped variables in order to further penalize variables that, in addition to having a low effect, belong to a group with a low overall effect; and to favor the selection of variables that, in addition to having a large effect, belong to a group with a large overall effect. We considered the case of prespecified and disjoint groups and proposed diverse weights for the adaptive lasso method. In particular we proposed the product Max Single Wald by Single Wald weighting (MSW*SW) which takes into account the information of the group to which it belongs and of this biomarker. Through simulations, we compared the selection and prediction ability of our approach with the standard lasso, the composite Minimax Concave Penalty (cMCP), the group exponential lasso (gel), the Integrative L1-Penalized Regression with Penalty Factors (IPF-Lasso), and the Sparse Group Lasso (SGL) methods. In addition, we illustrated the methods using gene expression data of 614 breast cancer patients. CONCLUSIONS: The adaptive lasso with the MSW*SW weighting method incorporates both the information in the grouping structure and the individual variable. It outperformed the competitors by reducing the false discovery rate without severely increasing the false negative rate.

An alternative trial-level measure for evaluating failure-time surrogate endpoints based on prediction error.

To validate a failure-time surrogate for an established failure-time clinical endpoint such as overall survival, the meta-analytic approach is commonly used. The standard correlation approach considers two levels: the individual level, with Kendall's τ measuring the rank correlation between the endpoints, and the trial level, with the coefficient of determination R 2 measuring the correlation between the treatment effects on the surrogate and on the final endpoint. However, the estimation of R 2 is not robust with respect to the estimation error of the trial-specific treatment effects. The alternative proposed in this article uses a prediction error based on a measure of the weighted difference between the observed treatment effect on the final endpoint and a model-based predicted effect. The measures can be estimated by cross-validation within the meta-analytic setting or external validation on a set of trials. Several distances are presented, with varying weights, based on the standard error of the observed treatment effect and of its predicted value. A simulation study was conducted under different scenarios, varying the number and the size of the trials, Kendall's τ and R 2 . These measures have been applied to individual patient data from a meta-analysis of trials in advanced/recurrent gastric cancer (20 randomized trials of chemotherapy, 4069 patients). The distance-based measures appeared to be robust with respect to different values of simulation parameters in several scenarios (such as Kendall's τ, size and number of clinical trials). The absolute prediction error can be an alternative to the trial-level R 2 for evaluation of candidate time-to-event surrogates.