MET Exon 14 Skipping in Non–Small-Cell Lung Cancer: A Predictive Biomarker Primer

What the test measures

MET is a receptor tyrosine kinase that, under normal conditions, sends carefully rationed growth and survival signals. One of the ways the cell keeps that signaling in check is by tagging the activated receptor for degradation — a built-in "off switch." Exon 14 of the MET gene encodes the juxtamembrane domain that houses the binding site (a tyrosine residue, Y1003, recruiting the Cbl E3 ubiquitin ligase) responsible for that degradation. When splice-site mutations cause exon 14 to be skipped during mRNA processing, the resulting receptor is missing that regulatory handle. The protein still works — it just doesn't get turned off on schedule, so MET signaling runs longer and louder than it should [2].

That's the essential biology: MET exon 14 skipping (often written METex14) isn't a mutation that creates a new fusion protein or a hotspot amino-acid change. It's a splicing defect that removes a degradation signal, and the practical consequence is a stabilized, oncogenically active receptor [2].

METex14 turns up in roughly 3–4% of non-squamous NSCLC, and it's notably enriched — reported in some series at 20% or more — in pulmonary sarcomatoid carcinoma, an aggressive and otherwise hard-to-treat subtype [2]. The patient population skews older than the typical EGFR- or ALK-driven cohort, which matters clinically because it shapes who you should have a high pretest suspicion for [2].

How it's tested

Assay. Next-generation sequencing is the workhorse here, and the RNA-versus-DNA question is, frankly, more consequential for this biomarker than for most. RNA-based NGS is preferred because it reads the mature transcript directly: if exon 14 has been skipped, the RNA shows exon 13 spliced straight onto exon 15, and the skipping event is unambiguous [1,2].

DNA-based NGS, by contrast, has to infer skipping from the underlying splice-site mutations — and those mutations are scattered across a large, diverse stretch of intronic and exonic sequence flanking exon 14. A DNA panel that doesn't tile that region densely enough will miss real events. That's the crux of the problem: a "negative" DNA result can mean the variant isn't there, or it can mean the assay simply couldn't see it. RNA sidesteps that ambiguity by capturing the functional readout [1,2].

Specimen and preanalytics. The standard specimen is formalin-fixed, paraffin-embedded (FFPE) tissue. RNA is more fragile than DNA and degrades readily in older blocks, with prolonged fixation, or in decalcified bone specimens — so RNA-based testing demands attention to tissue quality and adequate tumor cellularity. Circulating tumor DNA (ctDNA) from plasma is an alternative when tissue is scarce or the patient can't tolerate a repeat biopsy. It's a useful tool, but in practice I'd treat a negative ctDNA result with real skepticism unless tissue has genuinely been attempted, because low tumor shedding produces false negatives and a plasma "absent" doesn't reliably rule the alteration out.

Scoring. Interpretation is refreshingly binary: METex14 skipping is either present or absent. There's no percentage-positive threshold, no staining intensity to grade, no allele-fraction cutoff baked into the call itself. Either the transcript shows the skip or it doesn't. That simplicity is genuinely helpful for trainees who've cut their teeth on the graded, cutoff-laden world of immunohistochemistry — but it's worth remembering that the clean binary output rests on a nuanced assay choice underneath.

What each result means

METex14 skipping present. The tumor carries an activating alteration that stabilizes MET signaling and behaves as an oncogenic driver [2]. This is the actionable result.

METex14 skipping absent. No skipping event was detected. Given everything above, the interpretation depends on the assay: a negative RNA-based result in a well-preserved specimen is reassuring, while a negative DNA-based or ctDNA result carries more residual uncertainty, particularly when clinical suspicion is high — say, an older patient with sarcomatoid histology and no other identified driver.

One point that trips people up: METex14 skipping and MET amplification are distinct entities, even though both dysregulate the same receptor. Amplification is an increase in gene copy number; skipping is a splicing defect. They can co-occur, and MET amplification can also emerge as a resistance mechanism, but they're detected differently and shouldn't be conflated on a report. Co-occurring alterations — a concurrent MET amplification, or the rare overlap with other driver mutations — complicate interpretation and are worth flagging explicitly rather than burying in a variant table.

What it determines for treatment eligibility

A present METex14 result informs eligibility for the MET tyrosine kinase inhibitor (TKI) class — the targeted agents developed and tested in patients whose tumors carry this alteration [1,2]. The evidence base comes from clinical trials that enrolled METex14-positive NSCLC and demonstrated meaningful activity for MET-directed TKIs [1].

Let me be precise about what that does and doesn't mean. A positive result tells you a patient's tumor belongs to the biologic group for which this drug class was designed. It does not, by itself, dictate that any individual should receive any specific drug — that's a treatment-selection decision made by the treating oncologist, weighing the patient's overall clinical picture, comorbidities, prior therapies, and preferences. The biomarker opens a door; walking through it is a separate, individualized conversation. Conversely, a confidently negative result generally means the tumor won't be routed toward a MET TKI on the basis of this marker, and attention turns to other drivers.

Caveats and what's evolving

The dominant methodological controversy isn't a scoring cutoff — it's assay sensitivity. Because DNA-based panels can miss splice variants they don't adequately cover, discordance between DNA and RNA testing remains a live concern, and it's why RNA-based detection is favored where feasible [1,2]. Labs validating METex14 detection on DNA-only platforms need to demonstrate adequate coverage of the flanking splice regions; clinicians reading those reports should know which method generated the call.

Acquired resistance is the frontier that's genuinely unsettled. Tumors that respond to MET TKIs eventually progress, and the mechanisms — secondary MET kinase-domain mutations, bypass-pathway activation, the MET amplification I mentioned earlier — are still being characterized [1,2]. What that means practically is that a single baseline METex14 result captures one moment in a moving target; the biology at progression may look quite different, and repeat molecular profiling at that point is an area of active investigation rather than settled practice.

If there's one forward-looking thing worth watching, it's how the companion diagnostic landscape consolidates around RNA-based detection and how resistance-directed testing gets standardized — because the day-to-day diagnostic question is shifting from "is METex14 present at diagnosis?" toward "what's driving this tumor now that it's escaped?" We don't have tidy answers to the second question yet, and that gap is where much of the interesting work in this space is happening.

References

1. Lung Cancer: Targeted Therapy in 2025. Current Oncology. 2025. PMC11941068 (identifier to be verified).
2. MET Exon 14 Skipping Mutations in Non–Small-Cell Lung Cancer. 2021. PMCID: PMC8140815.