Editors: Donna E. Hansel, MD, PhD, chair of pathology, Oregon Health and Science University, Portland; Richard D. Press, MD, PhD, professor and director of molecular pathology, OHSU; James Solomon, MD, PhD, assistant professor, Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York; Erica Reinig, MD, assistant professor and medical director of molecular diagnostics, University of Wisconsin-Madison; Marcela Riveros Angel, MD, molecular genetic pathology fellow, Department of Pathology, OHSU; and Andrés G. Madrigal, MD, PhD, assistant professor, clinical, Ohio State University Wexner Medical Center, Columbus.
Predictive tumor mutation biomarkers from cancer patients with historical clinicogenomics data
September 2022—The widespread availability of next-generation sequencing-based somatic mutation analysis of solid tumors has led to the routine identification of patients eligible for FDA-approved targeted therapies, including immunotherapies. However, validated targeted therapies are available for only a small number of mutations, thereby preventing many patients from realizing the benefits of these life-extending modalities. With a focus on reducing the gap in eligibility for targeted therapy, the authors conducted a large-scale retrospective analysis of interactions between mutations, drug responses, and long-term cancer survival outcomes. They performed the analysis using the nationwide U.S.-based Flatiron Health–Foundation Medicine clinicogenomic database. The authors analyzed data on more than 40,000 patients from approximately 280 cancer clinics. They obtained data on tumor mutations, treatments, long-term disease progression, and survival outcomes, as well as from other information contained in patients’ electronic health records. The patients chosen for the authors’ analysis had one of eight common cancer types. The study population included people with advanced nonsmall cell lung cancer (aNSCLC; 12,934), metastatic colorectal cancer (8,590), metastatic breast cancer (7,877), ovarian cancer (3,899), metastatic pancreatic cancer (3,505), advanced bladder cancer (1,531), advanced melanoma (1,522), and metastatic renal cell carcinoma (1,045). Genomic alterations had been identified via comprehensive genomic profiling of 499 cancer-related genes using Foundation Medicine’s next-generation sequencing tests. The authors’ complex clinicogenomics analysis revealed 458 mutation biomarkers predicting significant differences in overall survival outcomes in patients receiving certain cancer treatments. The authors identified 42 genes that are prognostic markers of survival in at least one of the eight common cancers. Mutations in such driver genes as TP53, MYC, and CDKN2A were associated with poor prognosis, consistent with findings previously reported in the literature. The authors also found, for the first time, that mutations in the NF1, MLL3, NBN, ASXL1, and SRC genes predict positive response to immunotherapy in aNSCLC patients, while mutations in the APC gene predict better response to immunotherapy in patients with advanced bladder cancer. The authors analyzed the association between patient outcomes and the co-occurrence of the targeted mutations and other mutations. They extracted all of the genes with FDA-approved targeted therapies for the corresponding cancer types from Memorial Sloan Kettering Cancer Center’s OncoKB human genetic variant database and used them as anchor genes. The authors identified 61 significant anchor gene–modifier gene interactions in real-world patients who had been treated with a targeted therapy based on an anchor gene mutation in such well-studied genes as ALK, BRAF, EGFR, MET, RET, ROS1, ERBB2, and PIK3CA. Mutation–mutation interactions that significantly predicted positive clinical outcomes were found in 25 co-mutations, and mutation–mutation interactions that predicted negative clinical outcomes were found in 36 co-mutations. To independently verify these predictive biomarkers, the authors used a data set from the American Association for Cancer Research Project Genomics Evidence Neoplasia Information Exchange Biopharma Collaborative (GENIE BPC) that included genomic results combined with specific cancer treatments and clinical outcomes. The data set included 1,411 patients with aNSCLC, 1,359 with metastatic colorectal cancer, and 1,101 with metastatic breast cancer, all of whom were treated at any of four academic cancer centers. All of the statistically significant gene–treatment interactions and mutation–mutation interactions identified in this independent GENIE BPC data set were also found in the original Flatiron Health–Foundation Medicine clinicogenomic database. The novel predictive biomarkers identified in the authors’ analysis of a large real-world cancer cohort need further investigation to verify their prognostic capabilities in patients treated with specific modalities. A long-term goal of research in this area is to use real-world genomics data in novel ways to expand the medical community’s ability to offer efficacious personalized therapies to a larger percentage of cancer patients.
Liu R, Rizzo S, Waliany S, et al. Systematic pan-cancer analysis of mutation—treatment interactions using large real-world clinicogenomics data. Nat Med. 2022. https://doi.org/10.1038/s41591-022-01873-5
Correspondence: Dr. James Zou at jamesz@stanford.edu
Detection and quantification of SARS-CoV-2 variants using wastewater sequencing
As the SARS-CoV-2 virus continues to evolve, it is critical to detect and quantify novel variants on a local level to support public health containment measures. Toward that goal, the authors conducted a sophisticated sequencing analysis of toilet waste, termed wastewater, which showed that it can be a powerful tool to detect novel SARS-CoV-2 viral variants weeks before they are seen as spikes in conventional PCR-based diagnostic test data or clinical case counts. Previous wastewater surveillance efforts have been limited in their ability to comprehensively capture, detect, and quantitate SARS-CoV-2 in sewage samples containing myriad other viruses, bacteria, and pathogens. To overcome this problem, the authors developed a method that uses nanobeads to capture and sequence nearly 95 percent of viral RNA from a wastewater sample. They also developed a bioinformatics tool, called Freyja, to quantitatively identify the COVID variants present, including novel variants not previously seen, in each sample. To test their tool, the authors collected samples over more than 11 months from a sewage-treatment plant in San Diego that treats the wastewater from approximately 2.3 million people. They also collected 20,000 wastewater samples from maintenance holes and sewage pipes at more than 130 sites collectively draining 360 buildings on the University of California San Diego campus over 10 months. For comparison purposes, the authors also analyzed traditional diagnostic nasal swab samples they collected at testing sites and diagnostic clinics in the San Diego area. They detected a wider variety of viral variants circulating in the UCSD community through wastewater samples than from traditional nasal swab samples, providing a more accurate snapshot of the kinetics of viral evolution. Knowing the precise location where positive virus samples were collected, such as a specific building, would allow public health experts to more easily pinpoint the early spread of infection and quickly initiate traditional testing and isolation measures. In addition, during a time when the SARS-CoV-2 epsilon, alpha, and delta variants were dominant, the authors were able to detect these and other variants of concern, by analyzing wastewater, up to two weeks before the variants were detected using traditional swab-based surveillance methods. They detected the omicron variant, which began spreading in the United States in late 2021, in San Diego more than 10 days before it was first detected clinically. Therefore, the lag time from sample collection to detection of viral variants can be reduced from weeks to days, providing public health experts with earlier access to actionable information about local infection trends and the emergence of new, potentially more pathogenic variants. Because wastewater is a universal measure of viral burdens in a community and does not have the inherent socio-economic and scientific biases associated with symptom-based sample collection, its utility as a more accurate public health early warning system for emerging infectious diseases, and not just for COVID, appears promising.
Karthikeyan S, Levy JI, De Hoff P, et al. Wastewater sequencing reveals early cryptic SARS-CoV-2 variant transmission. Nature. 2022. https://doi.org/10.1038/s41586-022-05049-6
Correspondence information not provided.