MET Protein Overexpression by Immunohistochemistry: A Predictive Biomarker in Non–Small Cell Lung Cancer

What the Test Measures

MET is a receptor tyrosine kinase encoded by the MET gene, and its ligand, hepatocyte growth factor, drives signaling that governs cell proliferation, survival, and motility. In a subset of non–small cell lung cancers (NSCLC), that pathway becomes a dependency. But there are several biologically distinct ways this happens, and pathologists need to keep them separate.

Here's the crucial distinction. c-MET protein overexpression — measured by immunohistochemistry (IHC) — refers to an abundance of MET protein on the tumor cell surface and cytoplasm. It is not the same as a MET exon 14 skipping mutation (METex14) or MET gene amplification, both of which are genomic events detected by nucleic acid–based or in situ methods. A tumor can overexpress MET protein without carrying either genomic alteration, and the converse is also true. Don't conflate the three; they occupy different columns in the diagnostic algorithm and increasingly gate different therapies. This article concerns the protein-level readout only.

How It's Tested

The assay is immunohistochemistry performed on formalin-fixed, paraffin-embedded (FFPE) tissue — the same material generated in routine diagnostic workup. That's an advantage: no separate specimen type is required, and archival blocks can often be used.

Preanalytic factors matter more here than clinicians sometimes appreciate. Fixation time, tissue age, and section thickness all influence membranous and cytoplasmic staining intensity, and IHC for a membrane receptor is sensitive to these variables. Overexposed or under-fixed tissue can shift a case across a threshold. Scoring assesses the intensity and proportion of tumor cells with MET staining, and results are dichotomized for clinical purposes into c-MET high versus c-MET low/negative overexpression states [3]. As with other predictive IHC assays, the specific antibody clone, platform, and scoring algorithm are tied to the validated companion or complementary assay used to establish eligibility — a reminder that IHC is not a generic, interchangeable stain.

What Each Result State Means

Two result states are reported, and they are best understood in relation to one another rather than in isolation. A c-MET high result indicates MET protein overexpression at or above the validated threshold and identifies tumors more likely to respond to MET-directed antibody–drug conjugate therapy. A c-MET low/negative result falls below that threshold and does not identify eligibility for this class on the basis of protein overexpression. Neither result speaks to METex14 or MET amplification status — a c-MET low tumor may still harbor a targetable genomic MET alteration detectable only by orthogonal testing, which is precisely why the protein result cannot stand alone in a complete MET workup.

What It Determines for Treatment Eligibility

The predictive value of this assay is specific and recent. In May 2025, the FDA granted accelerated approval to telisotuzumab vedotin-tllv, a MET-directed antibody–drug conjugate (ADC), for previously treated NSCLC with high c-MET protein overexpression [3]. A c-MET high IHC result therefore informs eligibility for MET-directed ADC therapy in that clinical setting. This is an educational description of how the biomarker functions within an eligibility framework — it is not treatment advice, and no individual should infer a therapeutic course from a biomarker result.

What's notable is that this approval introduces an IHC arm into a MET diagnostic algorithm that has been dominated by next-generation sequencing (NGS) and FISH [1]. Historically, MET testing meant hunting for METex14 or amplification by molecular methods. Protein overexpression by IHC is a different predictive axis serving a different drug mechanism — an ADC that delivers cytotoxic payload to MET-expressing cells rather than a kinase inhibitor blocking a mutated pathway. The honest answer is that the field is still working out exactly where this IHC assay sits alongside NGS panels, and that uncertainty is worth naming rather than papering over. In practice, an oncologist may now need both molecular and protein-level MET data to fully characterize a tumor's therapeutic options.

Caveats and What's Evolving

Accelerated approval is a conditional pathway. It permits earlier access based on endpoints reasonably likely to predict benefit, with confirmatory data expected to follow. Readers should regard the eligibility linkage as established but subject to ongoing validation [3].

Threshold instability is the sharper concern. Predictive IHC cutoffs for protein overexpression have a history of moving as data mature, and the field has been here before. The HER2-low breast and lung experience is instructive: introducing a lower expression category redefined a population that older, binary scoring never captured, and it exposed how much the answer depends on where the line is drawn. MET IHC may face the same pressure as evidence accumulates, and a threshold that looks settled today could be recalibrated.

Two practical problems deserve emphasis. Interobserver variability means that two competent pathologists scoring the same slide can disagree near the cutoff — and near a dichotomous threshold, a one-step difference in intensity call can flip a patient's eligibility, which is not a trivial consequence for a borderline case. Interlaboratory concordance is the related question at scale: whether a c-MET high call in one laboratory reliably reproduces in another depends on harmonized antibody clones, platforms, staining protocols, and scoring training. Without that harmonization, the same tumor could be classified differently depending on where its block is sent — an outcome that undermines the whole point of a predictive assay. Proficiency testing and external quality assessment will be central to keeping this stain trustworthy across the many laboratories now expected to perform it.

The broader NSCLC therapeutic landscape is shifting quickly, with multiple molecularly and protein-defined approvals arriving in short succession [1][2]. In that environment, a biomarker's clinical meaning can change on a timescale of months rather than years, and diagnostic teams should treat the current algorithm as a snapshot.

The question the field most needs to answer next is a concordance one: how well c-MET IHC status tracks — or fails to track — with MET genomic alterations across real-world cohorts, and whether the three MET readouts are best used sequentially, in parallel, or reflexively. Pathologists rolling out this assay should watch their own near-threshold cases closely, push for participation in external quality schemes as they emerge, and resist the temptation to treat a c-MET high result as interchangeable with a molecular MET finding. Getting that discipline right early is what will keep this new arm of the algorithm clinically credible as the evidence catches up.

---

References

1. Review. Seven new FDA approvals in 2025 across molecular targets in NSCLC. 2025. (PMC12873738 — citation to be verified prior to publication.)

2. Review. Research progress in oncogene-addicted NSCLC. Cancer Biol Med. 2025. doi:10.20892/j.issn.2095-3941.2025.0153 (citation to be verified prior to publication.)

3. U.S. Food and Drug Administration. FDA grants accelerated approval to telisotuzumab vedotin-tllv for NSCLC with high c-MET protein overexpression. FDA Drugs@FDA / fda.gov approval page; 2025. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-grants-accelerated-approval-telisotuzumab-vedotin-tllv-nsclc-high-c-met-protein-overexpression

Note: References 1 and 2 cite review sources whose identifiers (PMC number and DOI) require confirmation before publication. The core claims concerning telisotuzumab vedotin's accelerated approval, the c-MET high eligibility linkage, and the distinction between protein overexpression and genomic MET alterations are supported by reference 3; the surrounding NSCLC-approval-landscape context is attributed to references 1 and 2 and should be adjusted if those identifiers cannot be verified.