Methodological challenges in translational drug response modeling in cancer: A systematic analysis with FORESEE

Abstract

Translational models directly relating drug response specific processes that can be observed in vitro to their in vivo role in cancer patients constitute a crucial part of the development of personalized medication. Unfortunately, current studies often focus on the optimization of isolated model characteristics instead of examining the overall modeling workflow and the interplay of the individual model components. Moreover, they are often limited to specific data sets only. Therefore, they are often confined by the irreproducibility of the results and the non-transferability of the approaches into other contexts. In this study, we present a thorough investigation of translational models and their ability to predict the drug responses of cancer patients originating from diverse data sets using the R-package FORESEE. By systematically scanning the modeling space for optimal combinations of different model settings, we can determine models of extremely high predictivity and work out a few modeling guidelines that promote simplicity. Yet, we identify noise within the data, sample size effects, and drug unspecificity as factors that deteriorate the models' robustness. Moreover, we show that cell line models of high accuracy do not necessarily excel in predicting drug response processes in patients. We therefore hope to motivate future research to consider in vivo aspects more carefully to ultimately generate deeper insights into applicable precision medicine.

Artificial intelligence, machine learning, and drug repurposing in cancer

“…clinical in vivo responses to treatments), rather than merely drug efficacies in established cell lines (in vitro responses), as the former enables straightforward translational precision oncology applications and drug repurposing opportunities. A recent systematic analysis investigated the importance of a number of modeling components for the clinical treatment response prediction of cancer patients [115]. As expected, the sample size of the patient response data was found as an important determinant for the predictive modeling, along with experimental noise within the data that can easily deteriorate the models' robustness.…”

“…Rather surprisingly, the in vitro drug treatment profile was not among the most predictive feature when predicting the clinical response of the same drug in actual cancer patients. These results indicate that even cell line models of high accuracy do not necessarily translate to accurate predictions of drug response processes in cancer patients in vivo [115].…”

Two-step multi-omics modelling of drug sensitivity in cancer cell lines to identify driving mechanisms

“…Concluding that the most important modelling factor is the choice of features and agreeing with Costello et al on the dominance of gene expression data, they rate the choice of algorithm as the third most important modelling factor. Expanding on this concept of systematically identifying optimal choices in distinct steps of the modelling pipeline and applying it to translational modelling, Turnhoff et al have published an R package that can be used to perform even more intricate analyses in the context of predicting clinical responses while training on cell line data [ 10 , 11 ].…”

Facilitating Drug Discovery in Breast Cancer by Virtually Screening Patients Using In Vitro Drug Response Modeling

“…There are likely several contributing factors to explain this skew. First, regarding the translational models, it has been shown that well performing computational models are not necessarily specific to a drug of interest [ 38 ]. That is, general mechanisms of drug response, such as factors associated with multi-drug resistance, are likely encoded in many of the independent models and as such might skew the imputed distribution of all the drugs in a similar direction.…”