Cellular State Transformations Using Deep Learning for Precision Medicine Applications.

We introduce the Transcriptome State Perturbation Generator (TSPG) as a novel deep-learning method to identify changes in genomic expression that occur between tissue states using generative adversarial networks. TSPG learns the transcriptome perturbations from RNA-sequencing data required to shift from a source to a target class. We apply TSPG as an effective method of detecting biologically relevant alternate expression patterns between normal and tumor human tissue samples. We demonstrate that the application of TSPG to expression data obtained from a biopsy sample of a patient's kidney cancer can identify patient-specific differentially expressed genes between their individual tumor sample and a target class of healthy kidney gene expression. By utilizing TSPG in a precision medicine application in which the patient sample is not replicated (i.e., n = 1 ), we present a novel technique of determining significant transcriptional aberrations that can be used to help identify potential targeted therapies.

Recommendations for the Management of Rare Kidney Cancers.

CONTEXT: The European Association of Urology Renal Cell Carcinoma Guideline Panel recently conducted a systematic review of treatment options for patients with advanced non-clear-cell renal cell carcinomas (RCCs), which showed a substantial lack of evidence for management recommendations. OBJECTIVE: To improve the outcomes of patients with rare kidney cancers (RKCs), we performed a subsequent unstructured review to determine current treatment strategies and druggable pathways, involving key stakeholders with a global perspective to generate recommendations. EVIDENCE ACQUISITION: Based on the systematic review, literature was queried in Pubmed, Medline, and abstracts from proceedings of European Society for Medical Oncology and American Society of Clinical Oncology, in addition to consulting key opinion leaders and stakeholders. A conventional narrative review strategy was adopted to summarize the data. EVIDENCE SYNTHESIS: The systematic review showed an absence of evidence for treating RKCs, with data only supporting sunitinib or MET inhibitors for some specific subtypes. However, a growing body of evidence implicates druggable pathways in specific RKC subtypes. To test hypotheses, the small patient numbers in each subtype require coordinated multicenter efforts. Many RKC patients are currently excluded from studies or are not analyzed using subtype-specific parameters, despite their unmet medical need. CONCLUSIONS: We recognize the need for additional multicenter studies and subtype-specific analyses; however, we present management recommendations based on the data available. Web-based tools facilitating subtype-specific global registries and shared translational research resources will help generate sufficient data to formulate evidence-based recommendations for guidelines. PATIENT SUMMARY: Patients confronted with rare kidney cancers are often treated the same way as clear-cell renal cell carcinoma patients, despite little evidence from randomized trials. Molecular characterization of tumors to stratify patients may improve outcomes. Availability of potential agents and trials remain a problem. Collaboration among medical centers is important to pool scarce data.