Deep learning is a subdiscipline of artificial intelligence that uses a machine learning technique called artificial neural networks to extract patterns and make predictions from large data sets. The increasing adoption of deep learning across healthcare domains together with the availability of highly characterised cancer datasets has accelerated research into the utility of deep learning in the analysis of the complex biology of cancer. While early results are promising, this is a rapidly evolving field with new knowledge emerging in both cancer biology and deep learning. In this review, we provide an overview of emerging deep learning techniques and how they are being applied to oncology. We focus on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference, and provide perspectives on how the different data types can be integrated to develop decision support tools. We provide specific examples of how deep learning may be applied in cancer diagnosis, prognosis and treatment management. We also assess the current limitations and challenges for the application of deep learning in precision oncology, including the lack of phenotypically rich data and the need for more explainable deep learning models. Finally, we conclude with a discussion of how current obstacles can be overcome to enable future clinical utilisation of deep learning.

High-grade epithelial ovarian carcinomas containing mutated BRCA1 or BRCA2 (BRCA1/2) homologous recombination (HR) genes are sensitive to platinum-based chemotherapy and PARP inhibitors (PARPi), while restoration of HR function due to secondary mutations in BRCA1/2 has been recognized as an important resistance mechanism. We sequenced core HR pathway genes in 12 pairs of pretreatment and postprogression tumor biopsy samples collected from patients in ARIEL2 Part 1, a phase II study of the PARPi rucaparib as treatment for platinum-sensitive, relapsed ovarian carcinoma. In 6 of 12 pretreatment biopsies, a truncation mutation in BRCA1, RAD51C, or RAD51D was identified. In five of six paired postprogression biopsies, one or more secondary mutations restored the open reading frame. Four distinct secondary mutations and spatial heterogeneity were observed for RAD51C. In vitro complementation assays and a patient-derived xenograft, as well as predictive molecular modeling, confirmed that resistance to rucaparib was associated with secondary mutations. Significance: Analyses of primary and secondary mutations in RAD51C and RAD51D provide evidence for these primary mutations in conferring PARPi sensitivity and secondary mutations as a mechanism of acquired PARPi resistance. PARPi resistance due to secondary mutations underpins the need for early delivery of PARPi therapy and for combination strategies. Cancer Discov; 7(9); 984–98. ©2017 AACR. See related commentary by Domchek, p. 937. See related article by Quigley et al., p. 999. See related article by Goodall et al., p. 1006. This article is highlighted in the In This Issue feature, p. 920

Accurately identifying patients with high-grade serous ovarian carcinoma (HGSOC) who respond to poly(ADP-ribose) polymerase inhibitor (PARPi) therapy is of great clinical importance. Here we show that quantitative BRCA1 methylation analysis provides new insight into PARPi response in preclinical models and ovarian cancer patients. The response of 12 HGSOC patient-derived xenografts (PDX) to the PARPi rucaparib was assessed, with variable dose-dependent responses observed in chemo-naive BRCA1/2-mutated PDX, and no responses in PDX lacking DNA repair pathway defects. Among BRCA1-methylated PDX, silencing of all BRCA1 copies predicts rucaparib response, whilst heterozygous methylation is associated with resistance. Analysis of 21 BRCA1-methylated platinum-sensitive recurrent HGSOC (ARIEL2 Part 1 trial) confirmed that homozygous or hemizygous BRCA1 methylation predicts rucaparib clinical response, and that methylation loss can occur after exposure to chemotherapy. Accordingly, quantitative BRCA1 methylation analysis in a pre-treatment biopsy could allow identification of patients most likely to benefit, and facilitate tailoring of PARPi therapy.

Acquired resistance to PARP inhibitors (PARPi) is a major challenge for the clinical management of high grade serous ovarian cancer (HGSOC). Here, we demonstrate CX-5461, the first-in-class inhibitor of RNA polymerase I transcription of ribosomal RNA genes (rDNA), induces replication stress and activates the DNA damage response. CX-5461 co-operates with PARPi in exacerbating replication stress and enhances therapeutic efficacy against homologous recombination (HR) DNA repair-deficient HGSOC-patient-derived xenograft (PDX) in vivo. We demonstrate CX-5461 has a different sensitivity spectrum to PARPi involving MRE11-dependent degradation of replication forks. Importantly, CX-5461 exhibits in vivo single agent efficacy in a HGSOC-PDX with reduced sensitivity to PARPi by overcoming replication fork protection. Further, we identify CX-5461-sensitivity gene expression signatures in primary and relapsed HGSOC. We propose CX-5461 is a promising therapy in combination with PARPi in HR-deficient HGSOC and also as a single agent for the treatment of relapsed disease.

KRAS is the most frequently mutated oncogene in cancer. Activating mutations in codon 12, especially G12D, have the highest prevalence across a range of carcinomas and adenocarcinomas. With inhibitors to KRAS-G12D now entering clinical trials, understanding the biology of KRAS-G12D cancers, and identifying biomarkers that predict therapeutic response is crucial. In this Review, we discuss the genomics and biology of KRAS-G12D adenocarcinomas, including histological features, transcriptional landscape, the immune microenvironment, and how these factors influence response to therapy. Moreover, we explore potential therapeutic strategies using novel G12D inhibitors, leveraging knowledge gained from clinical trials using G12C inhibitors.

In high-grade serous ovarian carcinoma (HGSC), deleterious mutations in DNA repair gene RAD51C are established drivers of defective homologous recombination and are emerging biomarkers of PARP inhibitor (PARPi) sensitivity. RAD51C promoter methylation (meRAD51C) is detected at similar frequencies to mutations, yet its effects on PARPi responses remain unresolved. In this study, three HGSC patient-derived xenograft (PDX) models with methylation at most or all examined CpG sites in the RAD51C promoter show responses to PARPi. Both complete and heterogeneous methylation patterns were associated with RAD51C gene silencing and homologous recombination deficiency (HRD). PDX models lost meRAD51C following treatment with PARPi rucaparib or niraparib, where a single unmethylated copy of RAD51C was sufficient to drive PARPi resistance. Genomic copy number profiling of one of the PDX models using SNP arrays revealed that this resistance was acquired independently in two genetically distinct lineages. In a cohort of 12 patients with RAD51C-methylated HGSC, various patterns of meRAD51C were associated with genomic “scarring,” indicative of HRD history, but exhibited no clear correlations with clinical outcome. Differences in methylation stability under treatment pressure were also observed between patients, where one HGSC was found to maintain meRAD51C after six lines of therapy (four platinum-based), whereas another HGSC sample was found to have heterozygous meRAD51C and elevated RAD51C gene expression (relative to homozygous meRAD51C controls) after only neoadjuvant chemotherapy. As meRAD51C loss in a single gene copy was sufficient to cause PARPi resistance in PDX, methylation zygosity should be carefully assessed in previously treated patients when considering PARPi therapy. Significance: Homozygous RAD51C methylation is a positive predictive biomarker for sensitivity to PARP inhibitors, whereas a single unmethylated gene copy is sufficient to confer resistance.