Predicting patient response with models trained on cell lines and patient-derived xenografts by nonlinear transfer learning.

Preclinical models have been the workhorse of cancer research, producing massive amounts of drug response data. Unfortunately, translating response biomarkers derived from these datasets to human tumors has proven to be particularly challenging. To address this challenge, we developed TRANSACT, a computational framework that builds a consensus space to capture biological processes common to preclinical models and human tumors and exploits this space to construct drug response predictors that robustly transfer from preclinical models to human tumors. TRANSACT performs favorably compared to four competing approaches, including two deep learning approaches, on a set of 23 drug prediction challenges on The Cancer Genome Atlas and 226 metastatic tumors from the Hartwig Medical Foundation. We demonstrate that response predictions deliver a robust performance for a number of therapies of high clinical importance: platinum-based chemotherapies, gemcitabine, and paclitaxel. In contrast to other approaches, we demonstrate the interpretability of the TRANSACT predictors by correctly identifying known biomarkers of targeted therapies, and we propose potential mechanisms that mediate the resistance to two chemotherapeutic agents.

SpaGE: Spatial Gene Enhancement using scRNA-seq.

Single-cell technologies are emerging fast due to their ability to unravel the heterogeneity of biological systems. While scRNA-seq is a powerful tool that measures whole-transcriptome expression of single cells, it lacks their spatial localization. Novel spatial transcriptomics methods do retain cells spatial information but some methods can only measure tens to hundreds of transcripts. To resolve this discrepancy, we developed SpaGE, a method that integrates spatial and scRNA-seq datasets to predict whole-transcriptome expressions in their spatial configuration. Using five dataset-pairs, SpaGE outperformed previously published methods and showed scalability to large datasets. Moreover, SpaGE predicted new spatial gene patterns that are confirmed independently using in situ hybridization data from the Allen Mouse Brain Atlas.

PRECISE: a domain adaptation approach to transfer predictors of drug response from pre-clinical models to tumors.

MOTIVATION: Cell lines and patient-derived xenografts (PDXs) have been used extensively to understand the molecular underpinnings of cancer. While core biological processes are typically conserved, these models also show important differences compared to human tumors, hampering the translation of findings from pre-clinical models to the human setting. In particular, employing drug response predictors generated on data derived from pre-clinical models to predict patient response remains a challenging task. As very large drug response datasets have been collected for pre-clinical models, and patient drug response data are often lacking, there is an urgent need for methods that efficiently transfer drug response predictors from pre-clinical models to the human setting. RESULTS: We show that cell lines and PDXs share common characteristics and processes with human tumors. We quantify this similarity and show that a regression model cannot simply be trained on cell lines or PDXs and then applied on tumors. We developed PRECISE, a novel methodology based on domain adaptation that captures the common information shared amongst pre-clinical models and human tumors in a consensus representation. Employing this representation, we train predictors of drug response on pre-clinical data and apply these predictors to stratify human tumors. We show that the resulting domain-invariant predictors show a small reduction in predictive performance in the pre-clinical domain but, importantly, reliably recover known associations between independent biomarkers and their companion drugs on human tumors. AVAILABILITY AND IMPLEMENTATION: PRECISE and the scripts for running our experiments are available on our GitHub page (https://github.com/NKI-CCB/PRECISE). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Paradoxical Activation of Oncogenic Signaling as a Cancer Treatment Strategy.

Cancer homeostasis depends on a balance between activated oncogenic pathways driving tumorigenesis and engagement of stress response programs that counteract the inherent toxicity of such aberrant signaling. Although inhibition of oncogenic signaling pathways has been explored extensively, there is increasing evidence that overactivation of the same pathways can also disrupt cancer homeostasis and cause lethality. We show here that inhibition of protein phosphatase 2A (PP2A) hyperactivates multiple oncogenic pathways and engages stress responses in colon cancer cells. Genetic and compound screens identify combined inhibition of PP2A and WEE1 as synergistic in multiple cancer models by collapsing DNA replication and triggering premature mitosis followed by cell death. This combination also suppressed the growth of patient-derived tumors in vivo. Remarkably, acquired resistance to this drug combination suppressed the ability of colon cancer cells to form tumors in vivo. Our data suggest that paradoxical activation of oncogenic signaling can result in tumor-suppressive resistance. Significance: A therapy consisting of deliberate hyperactivation of oncogenic signaling combined with perturbation of the stress responses that result from this is very effective in animal models of colon cancer. Resistance to this therapy is associated with loss of oncogenic signaling and reduced oncogenic capacity, indicative of tumor-suppressive drug resistance.

Few-shot learning creates predictive models of drug response that translate from high-throughput screens to individual patients.

Cell-line screens create expansive datasets for learning predictive markers of drug response, but these models do not readily translate to the clinic with its diverse contexts and limited data. In the present study, we apply a recently developed technique, few-shot machine learning, to train a versatile neural network model in cell lines that can be tuned to new contexts using few additional samples. The model quickly adapts when switching among different tissue types and in moving from cell-line models to clinical contexts, including patient-derived tumor cells and patient-derived xenografts. It can also be interpreted to identify the molecular features most important to a drug response, highlighting critical roles for RB1 and SMAD4 in the response to CDK inhibition and RNF8 and CHD4 in the response to ATM inhibition. The few-shot learning framework provides a bridge from the many samples surveyed in high-throughput screens (n-of-many) to the distinctive contexts of individual patients (n-of-one).