Estimating individualized treatment effects using an individual participant data meta-analysis.

BACKGROUND: One key aspect of personalized medicine is to identify individuals who benefit from an intervention. Some approaches have been developed to estimate individualized treatment effects (ITE) with a single randomized control trial (RCT) or observational data, but they are often underpowered for the ITE estimation. Using individual participant data meta-analyses (IPD-MA) might solve this problem. Few studies have investigated how to develop risk prediction models with IPD-MA, and it remains unclear how to combine those methods with approaches used for ITE estimation. In this article, we compared different approaches using both simulated and real data with binary and time-to-event outcomes to estimate the individualized treatment effects from an IPD-MA in a one-stage approach. METHODS: We compared five one-stage models: naive model (NA), random intercept (RI), stratified intercept (SI), rank-1 (R1), and fully stratified (FS), built with two different strategies, the S-learner and the T-learner constructed with a Monte Carlo simulation study in which we explored different scenarios with a binary or a time-to-event outcome. To evaluate the performance of the models, we used the c-statistic for benefit, the calibration of predictions, and the mean squared error. The different models were also used on the INDANA IPD-MA, comparing an anti-hypertensive treatment to no treatment or placebo ( N = 40 237 , 836 events). RESULTS: Simulation results showed that using the S-learner led to better ITE estimation performances for both binary and time-to-event outcomes. None of the risk models stand out and had significantly better results. For the INDANA dataset with a binary outcome, the naive and the random intercept models had the best performances. CONCLUSIONS: For the choice of the strategy, using interactions with treatment (the S-learner) is preferable. For the choice of the method, no approach is better than the other.

Do machine learning methods lead to similar individualized treatment rules? A comparison study on real data.

Identifying patients who benefit from a treatment is a key aspect of personalized medicine, which allows the development of individualized treatment rules (ITRs). Many machine learning methods have been proposed to create such rules. However, to what extent the methods lead to similar ITRs, that is, recommending the same treatment for the same individuals is unclear. In this work, we compared 22 of the most common approaches in two randomized control trials. Two classes of methods can be distinguished. The first class of methods relies on predicting individualized treatment effects from which an ITR is derived by recommending the treatment evaluated to the individuals with a predicted benefit. In the second class, methods directly estimate the ITR without estimating individualized treatment effects. For each trial, the performance of ITRs was assessed by various metrics, and the pairwise agreement between all ITRs was also calculated. Results showed that the ITRs obtained via the different methods generally had considerable disagreements regarding the patients to be treated. A better concordance was found among akin methods. Overall, when evaluating the performance of ITRs in a validation sample, all methods produced ITRs with limited performance, suggesting a high potential for optimism. For non-parametric methods, this optimism was likely due to overfitting. The different methods do not lead to similar ITRs and are therefore not interchangeable. The choice of the method strongly influences for which patients a certain treatment is recommended, drawing some concerns about their practical use.

What should be done and what should be avoided when comparing two treatments?

The preferred approach to compare two treatments is a randomized controlled trial (RCT). Indeed, randomization ensures that the groups compared are similar. Well-designed and well-conducted RCTs thus allow to draw causal conclusions on the relative efficacy and safety of treatments compared. However, it is not always possible to conduct RCTs for all clinical questions of interest, and observational data may also be used to infer on the relative effectiveness of treatments. In this review, we present different approaches that allow statistically valid comparisons of the effectiveness of treatments using observational data under some assumptions. Those are based on regression modelling or the propensity score. We also present the principles of target trial emulation.

Study designs for clinical trials applied to personalised medicine: a scoping review.

OBJECTIVE: Personalised medicine (PM) allows treating patients based on their individual demographic, genomic or biological characteristics for tailoring the 'right treatment for the right person at the right time'. Robust methodology is required for PM clinical trials, to correctly identify groups of participants and treatments. As an initial step for the development of new recommendations on trial designs for PM, we aimed to present an overview of the study designs that have been used in this field. DESIGN: Scoping review. METHODS: We searched (April 2020) PubMed, Embase and the Cochrane Library for all reports in English, French, German, Italian and Spanish, describing study designs for clinical trials applied to PM. Study selection and data extraction were performed in duplicate resolving disagreements by consensus or by involving a third expert reviewer. We extracted information on the characteristics of trial designs and examples of current applications of these approaches. The extracted information was used to generate a new classification of trial designs for PM. RESULTS: We identified 21 trial designs, 10 subtypes and 30 variations of trial designs applied to PM, which we classified into four core categories (namely, Master protocol, Randomise-all, Biomarker strategy and Enrichment). We found 131 clinical trials using these designs, of which the great majority were master protocols (86/131, 65.6%). Most of the trials were phase II studies (75/131, 57.2%) in the field of oncology (113/131, 86.3%). We identified 34 main features of trial designs regarding different aspects (eg, framework, control group, randomisation). The four core categories and 34 features were merged into a double-entry table to create a new classification of trial designs for PM. CONCLUSIONS: A variety of trial designs exists and is applied to PM. A new classification of trial designs is proposed to help readers to navigate the complex field of PM clinical trials.