Synthetic Lethality and Genomics
At a high level, mutations in cancer genes may be classified into loss-of-function (mostly in tumor suppressor genes) or gain-of-function defects (predominantly in oncogenes). While synthetic lethality interactions are more commonly associated with loss-of-function alleles, it can also apply to gain-of-function alleles as well. As stated earlier, the most common example cited for synthetic lethality is between the loss-of-function tumor suppressor BRCA genes and the PARP enzyme. In this case, patients harboring these mutations benefit from inhibition of PARP. However, there is also a need to investigate gain-of-function treatments. Initial synthetic lethality studies involved RNAi or chemical library screening, yet they often have shown limited translational value. Recently, Lee et al. (Lee, J. S. et al. Nat Commun 9, 2546, doi:10.1038/s41467-018-04647-1 (2018)), demonstrated the effective use of a genome-wide computational predcative model, ISLE (Identification of clinically-relevant Synthetic Lethality). The ISLE model showed that only a small subset of synthetic lethal gene candidates that were identified from in vitro screens were also associated with patient prognosis in a TCGA cohort. Moreover, the same group (Figure taken from Sahu, A. D. et al. Mol Syst Biol 15, e8323, doi:10.15252/msb.20188323 (2019)) also demonstrated that genomic and clinical survival data can be combined to identify synthetic rescue (SR) interactions (Figure 2). SR interactions denote a specific type of genetic interaction where a change in the activity of one gene reduces the cell’s fitness but an alteration of another gene (termed the SR partner) rescues cell viability (e.g., the rescue of Myc alterations by BCL2 activation in lymphomas). When a gene is targeted by a small molecule inhibitor or an antibody, the tumor may respond by changing the activity of the rescuer gene(s), conferring resistance to therapies.
Figure 2: Mol Syst Biol, Vol: 15, Issue: 3, First published: 12 March 2019, DOI: (10.15252,msb.20188323)
Recently Lee’s team performed a retrospective analysis of the WINTHER trial data. The WINTHER trial was the first large-scale basket clinical trial to incorporate transcriptomics data for cancer therapy in patients with advanced solid tumors. The patients were assigned to therapy on the basis of DNA sequencing as well as RNA expression. Overall, 303 patients consented; but only 107 (35%; 69 in arm A and 38 in arm B) were evaluable for therapy. The most common diagnoses were colon, head and neck, and lung cancers. Among the 107 patients, the rate of stable disease was ≥6 months while partial or complete response rate was 26.2%. They then applied the same genetic interaction models as described in the previously-mentioned publications and showed that by targeting synthetic lethal vulnerabilities in patients’ tumors identified by transcriptomic profiling, beneficial response rates could be as high as ~85%. The authors also applied this synthetic lethality-based approach on different targeted therapies across various cancer indications from publicly available data that was inclusive of pre-treatment transcriptomic data and clinical outcomes information. They concluded that synthetic lethality genetic interactions of the drug target genes can serve as effective biomarkers for predicting drug response in many cancer types. The authors also evaluated response to checkpoint inhibitors and used protein expression levels of PD-1 and CTLA-4 in TCGA data to identify the synthetic lethality and synthetic rescue partners of the respective checkpoint genes. They were able to identity statistically-significant synthetic rescue interactions, but not for synthetic lethality partners of PD-1 and CTLA-4.
The power of both DNA and RNA data
These recent studies demonstrate that whole exome sequencing (WES) and whole transcriptome sequencing (WTS) are powerful tools that can significantly benefit the field of precision medicine. As we know, tumors are quite complicated and typically have multiple changes at both the DNA and RNA level helping them to grow and metastasize. Thus, it is essential to investigate both. Furthermore, identifying clinically-relevant synthetic lethality or rescue interactions can play an important role in identifying acquired drug resistance mechanisms, especially in checkpoint inhibitor treatments. WES can identify germline and somatic mutations in genes involved in DNA repair pathways to identify the ‘BRCAness’ phenotype or mutations in MMR (MSI status). WES can also be utilized to study copy number variation and loss of heterozygosity which can be combined with mutations in BRCA (either germline or somatic) to identify tumors that may benefit from PARP inhibitors. Similarly, WTS can help create precise gene expression signatures that can be attributed to both primary and adaptive resistance in patients.
At Personalis, our ImmunoID NeXT Platform® represents an end-to-end solution for immuno-oncology and all precision oncology applications. It combines the pioneering ImmunoID NeXT Platform® assay, which employs enhanced WES and WTS, with sophisticated analytical engines to provide researchers with comprehensive genomic/transcriptomic data to accelerate their biomarker discovery programs. As shown in Figure 3, the unique design of the assay and analytical algorithms deliver critical tumor- and immune-related information including, but not limited to:
- T-cell receptor (TCR) alpha & beta repertoire
- Neoantigen detection and neoantigen load
- Tumor mutational burden (TMB)
- Microsatellite instability (MSI) characterization
- Human leukocyte antigens (HLA) typing, HLA and beta-2 microglobulin (B2M) somatic mutations, and HLA loss of heterozygosity (LOH)
- Tumor escape and resistance mechanisms
- Oncoviral detection