Sevabertinib Makes HER2 Testing Actionable in NSCLC

What the approval changes

Sevabertinib is now a targeted treatment option for adults with locally advanced or metastatic nonsquamous non-small cell lung cancer whose tumors harbor activating mutations in the HER2 tyrosine kinase domain. The FDA decision moves this molecular finding from a descriptive genomic result to one that can determine eligibility for an approved therapy.

That distinction matters. HER2 can be abnormal in several biologically different ways, including mutation, amplification and protein overexpression. The sevabertinib indication is defined by activating mutations in the HER2 tyrosine kinase domain, not by any positive HER2 result. HER2 is also commonly reported under the gene name ERBB2.

A pathology report therefore needs to identify the alteration at sufficient resolution to establish that it is an activating kinase-domain mutation covered by the indication. A report that lists only “HER2 positive,” amplification or an immunohistochemistry score does not, by itself, establish eligibility for a mutation-directed kinase inhibitor.

The FDA also tied patient selection to an FDA-authorized test. Clinicians and laboratories should use the companion diagnostic specified in the current prescribing information and FDA device records, including the authorized specimen type and variant-reporting criteria. Broad next-generation sequencing may detect the alteration during routine profiling, but an analytically valid finding is not automatically interchangeable with the test required by the drug label.

Identifying the eligible population

The indication applies to adults with locally advanced or metastatic nonsquamous NSCLC. Histology, disease stage and molecular result all matter. The relevant biomarker is an activating HER2 tyrosine kinase domain mutation; HER2 protein expression alone is a different measurement and should not be treated as a proxy.

Testing is most useful when performed early enough to inform systemic treatment planning. Tissue-based comprehensive genomic profiling can assess ERBB2 alongside other actionable alterations, although tissue quantity and quality may limit analysis. Plasma-based circulating tumor DNA testing can sometimes provide faster or complementary information, but a negative plasma result does not reliably exclude a tumor alteration when circulating DNA is sparse. The specimen and platform must remain consistent with the FDA-authorized testing pathway in the label.

Variant interpretation deserves particular attention. Not every ERBB2 sequence change is activating, and variants of uncertain significance should not be presumed to satisfy the indication. Reports should specify the nucleotide or protein-level alteration, its location and its classification. When terminology is unclear, review with the molecular laboratory or a molecular tumor board can distinguish a qualifying kinase-domain mutation from an unrelated or uncertain finding.

Repeat testing may be reasonable when earlier testing was incomplete, used a limited panel that did not adequately cover ERBB2, or produced an uninformative result because of low tumor content. Reassessment can also be relevant if only HER2 immunohistochemistry or copy-number testing was performed. The approval does not make all forms of HER2 testing equivalent.

How to interpret the evidence

Accelerated approval allows the FDA to authorize a drug for a serious condition on the basis of an endpoint considered reasonably likely to predict clinical benefit. In oncology, that endpoint is often tumor response and the durability of response rather than a demonstrated improvement in overall survival.

This pathway provides earlier access while preserving an important evidentiary obligation. Continued approval may depend on confirmatory evidence that verifies clinical benefit. A regulatory finding based on response cannot establish the same conclusions as a sufficiently powered randomized trial showing longer survival, better quality of life or superior outcomes against an active comparator.

The FDA decision should therefore be read as evidence that sevabertinib has sufficient activity in the biomarker-defined population to support accelerated authorization, not as proof of superiority over every available treatment sequence. Cross-trial comparisons are particularly unreliable because prior therapies, disease burden, central nervous system involvement, response assessment and follow-up can differ substantially.

The full FDA review and prescribing information remain the appropriate sources for the number of patients evaluated, study design, response rate, confidence interval, response duration, follow-up and safety findings supporting the decision. Those details should guide interpretation of precision and generalizability rather than the approval status alone.

What this means in practice

The immediate operational implication is that advanced nonsquamous NSCLC workflows should preserve the ability to identify qualifying HER2 mutations. A broad report that clearly distinguishes mutation, amplification and expression can reduce ambiguity and avoid treatment decisions based on the wrong biomarker category.

The result must then be interpreted in clinical context. Eligibility under the indication does not remove the need to consider performance status, organ function, previous treatment, comorbidities, disease tempo, central nervous system disease and patient preferences. Prescribing, dosing, dose modification, interaction management and adverse-event monitoring should follow the current FDA label rather than extrapolation from another HER2-directed therapy.

Health systems and payers will also need to align coverage with the companion diagnostic requirement. Policies that recognize only HER2 immunohistochemistry, or that treat all ERBB2 abnormalities as equivalent, risk both false inclusion and false exclusion. Conversely, overly narrow testing policies can miss a small biomarker-defined population for whom the result now has a direct treatment consequence.

The approval also reinforces a broader policy point: biomarker access is part of drug access. Turnaround time, tissue stewardship, insurance coverage and the availability of validated sequencing can determine whether an eligible patient is identified while treatment decisions are still open.

What remains uncertain

Accelerated approval carries limitations. Evidence supporting such decisions may come from a nonrandomized or otherwise limited development program, leaving uncertainty about comparative benefit, optimal sequencing and long-term outcomes. Small molecular subgroups can also produce wide confidence intervals and limited representation of older adults, people with poor performance status or patients with active brain metastases.

Confirmatory evidence will be important for determining whether response translates into durable clinical benefit and how sevabertinib compares with other systemic options. Additional data are also needed to clarify intracranial activity, mechanisms of acquired resistance, outcomes after other HER2-directed therapies and the role of tissue versus plasma testing across different clinical settings.

Safety interpretation should remain anchored to the FDA label and accumulated postmarketing evidence. Toxicities associated with one HER2-directed drug cannot automatically be assigned to another, even when the agents target the same pathway.

Questions clinicians ask

Does any HER2-positive result establish eligibility?

No. The indication is for an activating mutation in the HER2 tyrosine kinase domain. HER2 amplification, protein overexpression and a variant of uncertain significance are distinct findings and should not be assumed to qualify without confirmation that the reported alteration matches the label-defined biomarker.

Is broad next-generation sequencing sufficient?

Broad sequencing can identify ERBB2 mutations and may be efficient when several actionable drivers are being assessed. For treatment selection, however, the result must meet the FDA-authorized diagnostic framework specified in the current label, including relevant specimen, assay and variant criteria. An unauthorized assay should not be assumed equivalent solely because it reports ERBB2.

What does accelerated approval tell us about clinical benefit?

It indicates that the FDA judged the available endpoint reasonably likely to predict benefit in a serious disease with unmet need. It does not necessarily show longer overall survival or superiority to an active comparator. Continued approval may depend on confirmatory evidence that verifies the anticipated benefit.

Should a negative plasma result end the search for HER2 mutations?

Not necessarily. Plasma testing can be informative, but low circulating tumor DNA may produce a false-negative result. If suspicion remains and adequate tissue is available, tissue testing with an appropriate validated assay can provide another opportunity to identify a qualifying HER2 kinase-domain mutation.

References

1. FDA grants accelerated approval to sevabertinib for locally advanced or metastatic non-squamous non-small cell lung cancer — U.S. Food and Drug Administration, 2026
2. Accelerated Approval Program — U.S. Food and Drug Administration, 2024
3. List of Cleared or Approved Companion Diagnostic Devices — U.S. Food and Drug Administration, 2026