Predictive performance of microarray gene signatures: impact of tumor heterogeneity and multiple mechanisms of drug resistance
A partir de données portant sur des profils d'expression de gènes dans 1 550 tumeurs du sein, cette étude suggère que la présence de multiples mécanismes de résistance thérapeutique limite la capacité des signatures basées sur l'expression de gènes à faire des prédictions pertinentes d'un point de vue clinique
Gene signatures have failed to predict responses to breast cancer therapy in patients to date. In this study, we used bioinformatic methods to explore the hypothesis that the existence of multiple drug resistance mechanisms in different patients may limit the power of gene signatures to predict responses to therapy. Additionally, we explored whether sub-stratification of resistant cases could improve performance. Gene expression profiles from 1,550 breast cancers analyzed with the same microarray platform were retrieved from publicly available sources. Gene expression changes were introduced in cases defined as sensitive or resistant to a hypothetical therapy. In the resistant group, up to five different mechanisms of drug resistance causing distinct or overlapping gene expression changes were generated bioinformatically, and their impact on sensitivity, specificity and predictive values of the signatures was investigated. We found that increasing the number of resistance mechanisms corresponding to different gene expression changes weakened the performance of the predictive signatures generated, even if the resistance-induced changes in gene expression were sufficiently strong and informative. Performance was also affected by cohort composition and the proportion of sensitive versus resistant cases or resistant cases that were mechanistically distinct. It was possible to improve response prediction by sub-stratifying chemotherapy-resistant cases from actual datasets (non-bioinformatically-perturbed datasets), and by using outliers to model multiple resistance mechanisms. Our work supported the hypothesis that the presence of multiple resistance mechanisms to a given therapy in patients limits the ability of gene signatures to make clinically-useful predictions.
MAJOR FINDINGS: If resistance to a given drug or combinatorial therapy is caused by more than one mechanism, robust and highly accurate predictive gene signatures may not be successfully derived using current bioinformatics approaches, even if the changes in gene expression are strong and informative. The detrimental impact on predictive signature performance by the existence of multiple mechanisms of resistance was found to be maximum when these resulted in distinct patterns of gene expression, but overlapping changes in gene expression mitigated this effect. We propose that the sub-stratification of resistant cancers according to the potential resistance mechanisms may improve the ability to generate clinically useful predictive signatures.
Cancer Research , résumé, 2014