Sevabertinib Approved for HER2-Mutant Nonsquamous NSCLC

A mutation-defined approval

The FDA decision dated September 9, 2026, covers sevabertinib for patients with previously treated, advanced nonsquamous non-small cell lung cancer whose tumors harbor qualifying activating mutations in the HER2, or ERBB2, tyrosine kinase domain. The indication is more narrowly defined than the phrases “HER2-positive” or “HER2-altered” might suggest.

That distinction matters at the bedside. HER2 can be altered through a pathogenic sequence variant, increased gene copy number or excess protein expression. These findings arise through different biological mechanisms, are measured with different assays and cannot be treated as interchangeable evidence of eligibility for a mutation-directed drug.

The relevant result is an activating mutation within the receptor’s tyrosine kinase domain that is covered by the FDA-approved companion diagnostic and the drug labeling. Many clinically recognized HER2 exon 20 insertions occur in this domain, but exon number alone should not replace review of the exact variant and the test’s approved intended use. A variant of uncertain significance, a mutation outside the covered domain, isolated amplification or protein overexpression does not establish eligibility under this indication.

What the regulatory decision establishes

Accelerated approval permits earlier authorization of therapies for serious conditions based on a surrogate or intermediate clinical endpoint considered reasonably likely to predict clinical benefit. In oncology, objective response rate and response durability commonly support such decisions when randomized survival data are not yet available. Continued approval may depend on confirmatory evidence verifying clinical benefit.

The FDA notification supplied for this brief identifies the regulatory action but does not provide enough extractable detail to report the supporting cohort size, objective response rate with confidence interval, median duration of response or follow-up duration reliably. Those values should come directly from the final prescribing information and multidisciplinary FDA review, rather than being reconstructed from preliminary conference reports or earlier data cuts.

This evidentiary boundary is important. Accelerated approval establishes that the FDA judged the total evidence sufficient for conditional marketing in the specified population. It does not, by itself, show improved overall survival, establish superiority over another HER2-directed therapy, or support use in biomarker groups omitted from the label.

The approval population should therefore be read as a series of linked criteria: advanced nonsquamous NSCLC; prior systemic treatment as specified in the label; a qualifying HER2 tyrosine-kinase-domain activating mutation; and detection with an FDA-approved test. Removing any one of those elements broadens the population beyond the regulatory evidence.

How testing determines eligibility

An FDA-approved companion diagnostic is not simply any next-generation sequencing panel that reports ERBB2. Its approved labeling specifies the analyte, specimen type, testing method and treatment context for which performance was reviewed. Clinicians and molecular tumor boards should compare the laboratory report with that intended-use statement and with the sevabertinib label.

A report may use terms such as pathogenic, likely pathogenic, activating, exon 20 insertion, amplification, copy-number gain, overexpression or variant of uncertain significance. These descriptions answer different questions. For this indication, the central question is whether the reported sequence alteration is a covered activating mutation in the HER2 tyrosine kinase domain.

The wording “FDA-authorized testing” is sometimes used broadly, but drug labels generally identify an FDA-approved companion diagnostic when a test is essential to safe and effective use. A laboratory-developed assay may provide clinically useful information, yet that does not automatically make it the FDA-approved test required by the indication. Laboratories should clearly state the assay’s regulatory status and limitations.

The exact variant also matters. A positive ERBB2 entry on a condensed report is not enough if the alteration, transcript, protein consequence or classification is unclear. Review of the full molecular report can prevent accidental substitution of amplification, overexpression or an uncertain sequence variant for a qualifying activating kinase-domain mutation.

What this changes in practice

The approval turns a defined subset of HER2 mutations into a treatment-linked biomarker in advanced nonsquamous NSCLC. It strengthens the rationale for broad molecular profiling that can identify ERBB2 sequence alterations, while preserving enough detail to distinguish the qualifying mutation class from other HER2 results.

For pathology and oncology services, the operational task is not merely to mark HER2 as positive or negative. Reports and electronic systems need to retain the alteration type, genomic and protein notation, domain location, pathogenicity assessment, specimen source and assay identity. That information allows the treating team to compare the result with the companion diagnostic and drug labels.

The decision also has policy implications. Coverage rules, prior authorization forms and clinical pathways should reproduce the approved biomarker language rather than reduce it to “HER2-positive NSCLC.” Overly broad wording could authorize treatment outside the evidence base; overly restrictive wording could exclude a qualifying mutation because a payer recognizes only a familiar exon-level shorthand.

Important uncertainties

The principal limitation of an accelerated approval is that response-based evidence does not resolve whether sevabertinib extends survival or improves quality of life compared with available alternatives. Confirmatory evidence remains necessary to verify clinical benefit and further characterize uncommon toxicities, resistance mechanisms and outcomes in clinically underrepresented groups.

Generalizability may also be limited by the molecular and clinical selection used in the supporting development program. Results in qualifying kinase-domain mutations should not be extrapolated to HER2 amplification, immunohistochemical overexpression, uncertain variants, other NSCLC histologies or earlier-stage disease without direct evidence.

A further limitation for this explainer is source granularity: the cited FDA notification establishes approval, but the supplied record does not include the full efficacy dataset or safety tables. Prescribing decisions require consultation of the current FDA label, including adverse reactions, warnings, drug interactions and any required dose modifications.

Questions clinicians ask

Does HER2 overexpression on immunohistochemistry qualify a patient?

No. The indication is tied to activating HER2 tyrosine-kinase-domain mutations detected by an FDA-approved test. Protein overexpression, gene amplification and an activating sequence mutation are distinct biomarker categories; overexpression or amplification alone does not satisfy a mutation-defined indication.

Which molecular result is the relevant one?

The report should identify the exact ERBB2 sequence alteration and support its classification as an activating mutation within the tyrosine kinase domain covered by the approved test and drug label. An exon label can be helpful, but it should not replace confirmation of the specific variant and its regulatory coverage.

Is any next-generation sequencing panel sufficient?

Not automatically. A panel may detect ERBB2 accurately without being the FDA-approved companion diagnostic for sevabertinib. The assay’s intended use, specimen type and covered variants should be checked against the current companion diagnostic and prescribing information before concluding that the result establishes eligibility.

What does accelerated approval leave unanswered?

It leaves the magnitude of longer-term clinical benefit to be verified. Response and durability can support earlier access, but they do not by themselves demonstrate improved overall survival, comparative superiority or benefit in broader HER2 biomarker groups. Confirmatory evidence and continued regulatory review remain important.

References

1. Oncology (Cancer)/Hematologic Malignancies Approval Notifications — U.S. Food and Drug Administration, 2026 2. Accelerated Approval Program — U.S. Food and Drug Administration, 2024 3. Companion Diagnostics — U.S. Food and Drug Administration, 2024