NEW YORK – The European Society for Medical Oncology on Tuesday published a set of guidelines detailing how next-generation sequencing should be used as part of routine clinical practice.
In addition to discussing the use of NGS across various cancer types, the guidelines address whether broad-based NGS should replace small panels designed to test a single gene or a handful of genes.
While NGS has been widely implemented across many cancers to guide the use of molecularly matched treatments, the clinical utility and cost effectiveness of large, multigene sequencing panels varies depending on tumor type and the efficacy and availability of targeted drugs. The new guidelines, published in Annals of Oncology, focus on the eight cancers responsible for the most deaths worldwide as well as additional cancers for which routine NGS may be justified.
"These are the first recommendations from a scientific society about the use of NGS," Fernanda Mosele of Gustave Roussy in France, the first author of the paper, said in a statement. "Our intent is that they will unify decision-making about how NGS should be used for patients with metastatic cancer."
To come to their consensus, members of ESMO's Translational Research and Precision Medicine Working Group used the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT), which ranks molecular targets and respective treatments on four levels.
Using this ranking system alongside the prevalence of each genomic alteration among patients with a given cancer, the authors assessed how many patients would need to receive NGS to match a single patient with an effective drug. They also considered cost-effectiveness research regarding the use of multigene sequencing, which they pointed out is lacking in the published literature.
Routine NGS use according to cancer type
In advanced, non-squamous NSCLC, the recommendations favored the routine use of RNA- or DNA-based NGS to detect mutations and fusions with ESCAT level I matches. Broad-based NGS could also be used routinely to detect alterations with lower than level I matches, the authors wrote, but only when specific agreements are made with payors to consider the cost of testing versus small panels. ESMO only recommends the use of off-label drugs according to NGS results when a decision-making system exists on the national or regional level. As an example, they cited the Netherland's Drug Repurposing Study, or DRUP.
ESMO also noted that investigators running clinical trials should use multigene NGS, because finding lower-level ESCAT matches could be valuable for drug development. This recommendation was consistent across other cancer types considered beyond NSCLC, too.
In metastatic breast cancer, the authors wrote that there is "currently no need to perform tumor multigene NGS in the context of daily practice." Somatic testing cannot fully substitute germline testing for actionable BRCA mutations, they explained, and actionable PIK3CA mutations can be tested with polymerase chain reaction (PCR)-based assays instead. HER2 testing, they added, could be accomplished with immunohistochemistry (IHC).
For metastatic colorectal cancer, the recommendations pointed to the utility of PCR and IHC to detect level I mutations in KRAS, NRAS, and BRAF — for which targeted treatments have demonstrated efficacy — to highlight why routine multigene NGS is not necessary. It could be used as an alternative, however, "only if it does not generate extra cost compared to standard techniques," the authors wrote. If broad-based testing is performed, testing for NTRK fusions should be included, as well as for ERBB2 amplifications and microsatellite instability, for which immune checkpoint inhibitors could be an option.
For advanced prostate cancer, the recommendations support routine NGS to assess somatic BRCA mutations in countries where PARP inhibitors are accessible. If broad-based testing is used, it should also include testing for AKT inhibitors, DNA repair genes, and MSI signature, the authors wrote, but broader panels should only be used in cases where payers agree it is cost effective.
The authors did not recommend multigene NGS for patients with metastatic gastric cancer, pancreatic cancer, or hepatocellular carcinoma. For these cancer types, they wrote, "cheap standard methods" could be used to detect MSI or NTRK fusion status, for which immune checkpoint inhibitors and TRK inhibitors, respectively, may provide benefit. However, for clinical trials and drug development, NGS is encouraged.
Finally, routine multigene NGS is recommended for metastatic cholangiocarcinoma and ovarian cancer, based on the presence of level I ESCAT matches, as well as for cancers of unknown primary, despite the absence of level I matches.
In addition, the guidelines recommended NGS for TMB testing in cervical cancer, well- and moderately- differentiated neuroendocrine tumors, salivary cancers, vulvar cancers, and thyroid cancers. Beyond these tumor types, the evidence was not strong enough to justify routine TMB testing.
In regard to the tumor-agnostic approvals of several TRK inhibitors for NTRK fusion-positive cancers, the authors recommended that NGS should only be used to detect NTRK fusions, which have a very low incidence, in cancers where the technology is otherwise recommended.
Ultimately, patients and their physicians should be the ones responsible for deciding whether sequencing a large panel of genes is appropriate, the authors wrote, "pending no extra cost for the public healthcare system, and if the patient is informed about the low likelihood of benefit."