The detection of MTAP loss in non-small cell lung cancer: a novel biomarker with therapeutic implications

Methylthioadenosine phosphorylase (MTAP) loss is approaching a pivotal moment in precision oncology, shifting from a biologic observation to a potentially actionable biomarker. The emergence of methylthioadenosine (MTA)-cooperative protein arginine methyltransferase 5 (PRMT5) inhibitors, along with parallel strategies such as methionine adenosyltransferase 2A (MAT2A) inhibition, has made MTAP loss an increasingly relevant therapeutic vulnerability across multiple tumor types, including non-small cell lung cancer (NSCLC) (1). As these strategies move closer to broader clinical use, however, one deceptively simple question becomes central: how should MTAP-deficient tumors be identified accurately, reproducibly, and efficiently in routine practice? Are modern next-generation sequencing (NGS) panels sufficient to detect MTAP loss?

Answering that question requires precision about language, because the readouts in play are related but not equivalent. Throughout this commentary we distinguish MTAP protein deficiency—absence of MTAP protein on immunohistochemistry (IHC), a functional readout—from MTAP (homozygous) deletion—biallelic loss of the MTAP locus demonstrated by molecular testing—and we reserve cyclin-dependent kinase inhibitor 2A (CDKN2A) loss for copy-number loss of the adjacent CDKN2A locus. That these readouts should not be used interchangeably is the central message of this commentary.

In this context, the study by Brune and colleagues (2) is both timely and important. Using a large real-world NSCLC cohort, the authors evaluated MTAP status across multiple platforms: IHC, routine NGS workflows that inferred CDKN2A copy-number variants, and The Cancer Genome Atlas (TCGA) copy-number variation (CNV) data. This design is particularly valuable because it allows comparison between direct assessment of MTAP protein expression and the genomic surrogate approaches commonly used in practice. Their findings show that MTAP status is not simply a biologic fact waiting to be uncovered, but rather a result that depends heavily on assay design, tissue quality, tumor purity, and the interpretation framework applied.

Because MTAP lies immediately adjacent to CDKN2A at 9p21.3 and the two are co-deleted in the large majority of 9p21-deleted NSCLC—and because CDKN2A copy number is already captured by many routine NGS/CNV pipelines whereas MTAP frequently is not—CDKN2A loss has been widely adopted as a convenient surrogate for MTAP status. It is precisely this assumed equivalence that Brune and colleagues interrogate. Inferred from a non-MTAP-inclusive panel (Oncomine Precision Assay-NGS), CDKN2A loss captured only 28.4% (23/81) of IHC-defined MTAP-deficient tumors, whereas the TCGA analyses—in which both loci are sequenced directly—revealed the converse error: 27.1% (58/214) of CDKN2A-deleted tumors retained MTAP. This is not merely a technical issue but a clinically meaningful one: if therapeutic eligibility depends on true MTAP loss, CDKN2A is an insufficient surrogate. Reliance on imperfect surrogate testing risks both false inclusion of biologically inappropriate patients and false exclusion of patients who may benefit from MTAP-directed therapies. These findings therefore support direct assessment of MTAP status by validated IHC or direct genomic methods, rather than assuming equivalence with CDKN2A loss alone.

The significance of this distinction is ultimately biological. The therapeutic vulnerability of interest arises from MTAP loss itself, not from CDKN2A. Mavrakis and colleagues showed that MTAP deletion results in the accumulation of MTA, partial suppression of PRMT5 activity, and a cellular state in which deeper PRMT5 inhibition becomes selectively lethal (1). This synthetic-lethal model is the mechanistic basis for MTAP-directed therapy, and it explains why assay definition is not a purely technical matter: when the drug target is MTAP loss specifically, defining MTAP status as precisely as possible becomes clinically consequential.

This need for accurate MTAP assessment is already becoming a practical issue in clinical development. Enrollment criteria across MTAP-directed trials remain heterogeneous: some studies accept biallelic MTAP deletion by NGS or loss of MTAP protein by IHC, whereas others require homozygous MTAP deletion with central confirmation, use broader terminology such as “MTAP deletion or loss”, or even permit CDKN2A “null” status in certain trial components as shown in Table 1 (3-9). This variability is clinically meaningful because different biomarker definitions may alter both patient selection and the apparent activity of MTAP-directed therapies across studies. In that context, the findings from Brune et al. are particularly important: CDKN2A loss inferred from non–MTAP-inclusive CNV pipelines should not be treated as interchangeable with MTAP loss. As these therapies move forward, greater precision and standardization in MTAP testing will be essential.

Does this mean that IHC is the preferred method? Not necessarily. IHC is closer to the functional endpoint, namely protein presence or absence, and is often faster and more scalable than fluorescence in situ hybridization (FISH) or expanded molecular testing. However, the reliability of IHC depends on rigorous standardization, including antibody clone selection, internal controls, minimum tumor cell requirements, and clear protocols for interpreting weak, patchy, or heterogeneous staining. Recent evidence from other 9p21-driven cancers highlights these issues. For example, in pleural mesothelioma, MTAP IHC demonstrates strong diagnostic performance and reproducibility. Still, the choice of antibody can significantly affect the rate of equivocal results and concordance with molecular copy-number findings (10).

The interpretive definition of MTAP “loss” is itself unstandardized. Brune et al. define MTAP deficiency as complete absence of cytoplasmic tumor-cell staining against an internal positive control, with a minimum of 100 evaluable tumor cells. Yet the proportion of tumor cells that must show loss before a case is called MTAP-deficient differs across studies. Heterogeneous or focal MTAP loss further complicates classification. Because MTAP-directed trials likewise differ in how they define eligibility (Table 1), this absence of a common threshold has direct consequences for patient selection and cross-study comparison, and argues for explicit, harmonized IHC scoring criteria as MTAP testing scales into routine practice.

The question with which we began—whether NGS panels suffice to detect MTAP loss—thus resolves not into a choice between IHC and NGS, but into one between MTAP-inclusive, deletion-sensitive testing and surrogate inference: a panel suffices only insofar as it directly interrogates MTAP and reliably distinguishes homozygous from heterozygous loss. A more mature framing is fit-for-purpose testing, in which each assay is deployed with explicit understanding of its analytic strengths and limitations. CNV detection from targeted NGS panels, in particular, is a distinct analytical challenge that requires dedicated validation and may be less robust than single-nucleotide variant (SNV) or insertion/deletion detection in settings of low tumor purity or limited locus coverage (11). Any such framework also demands reporting language that clearly separates “MTAP protein deficiency by IHC” from “MTAP homozygous deletion by molecular testing”, rather than collapsing them into a single biomarker category.

The remaining standardization gaps shape how IHC is used, not whether; the entry point follows the workflow already in motion. Comprehensive genomic profiling (CGP) that interrogates MTAP gives the richest readout—a DNA-level deletion call with co-mutation context—but is obtained in only a minority of advanced NSCLC, roughly one in five in recent cohorts (12). Where it has been run and covers MTAP, its call is the primary readout, with IHC as a low-cost orthogonal check; an “MTAP-intact” molecular call that conflicts with unequivocal IHC deficiency should defer to IHC, since bulk CNV calling is prone to false negatives at low tumor purity or limited locus coverage. For the majority without CGP—or with a small, low-purity biopsy, or needing rapid triage—MTAP IHC, read with a validated clone, an internal control, and a defined minimum tumor-cell content, is the appropriate primary screen: a single inexpensive, rapid stain, robust to the low purity that degrades bulk CNV calling. Ordering CGP solely to determine MTAP status is hard to justify against IHC on cost or time, though where CGP is already indicated the MTAP call comes at no marginal cost. Discordant or equivocal cases reflex to MTAP-directed FISH (or an MTAP-inclusive molecular assay), which alone separates homozygous from heterozygous loss; CDKN2A copy number from a non-MTAP-inclusive panel must never substitute for an MTAP call.

9p21 loss also bears on immunotherapy. NSCLC with 9p21 deletion trends toward an immune-cold phenotype and poorer programmed cell death protein 1 (PD-1)/programmed death-ligand 1 (PD-L1) outcomes (13), and one candidate mechanism is MTAP-specific: its loss raises intratumoral MTA, which can suppress tumor-infiltrating T cells (14), and blunt interferon signaling. But 9p21.3 deletions also co-remove CDKN2A and type I interferon cluster, so the responsible gene cannot be read from “9p21 loss” or a CDKN2A surrogate alone. Dual-cohort data make the confounding concrete: the lower tumor mutational burden and worse checkpoint-inhibitor outcomes of MTAP-deleted tumors largely vanished once oncogenic drivers were excluded, and the residual survival signal tracked co-deleted CDKN2A as closely as MTAP (15). Because MTAP deletion almost never occurs without CDKN2A loss—behaving as a near-subset of it—MTAP’s independent effect is unresolvable rather than excluded; the same asymmetry that makes CDKN2A a poor surrogate for MTAP also prevents isolating MTAP’s role. The implication is asymmetric: accurate MTAP testing already defines eligibility for MTAP-directed therapy, which exploits PRMT5 hyper-dependence in MTAP-deleted NSCLC, reported at approximately 8–10% in the US Genomics Evidence Neoplasia Information Exchange (GENIE) cohort and 16–20% in the Japanese Center for Cancer Genomics and Advanced Therapeutics (C-CAT) cohort (15). In this sense, MTAP status should be viewed not merely as a tumor-classification marker, but as an upstream determinant for how targeted therapy and immunotherapy strategies may ultimately be integrated.

In summary, Brune and colleagues introduce no new biologic mechanism; their contribution is to the translational problem of implementing an emerging biomarker reliably. By exposing the bidirectional limitations of CDKN2A surrogate testing and establishing MTAP IHC as a direct functional readout, the work moves this biomarker closer to routine use—what remains is to standardize it, so that the reliability seen here translates beyond a single experienced center. Within a fit-for-purpose framework, MTAP IHC, MTAP-directed FISH, and MTAP-inclusive CGP are complementary tools ordered by clinical context, cost, and turnaround, not subordinate steps ranked once and for all. The broader lesson is one of discipline: as MTAP-directed therapy moves from rationale to routine eligibility—and as the same readout is increasingly invoked in immunotherapy decisions—what we call “MTAP loss” must be defined as deliberately as it is detected. Naming precisely what is measured, rather than collapsing protein deficiency, deletion, and surrogate loss into one label, is what will determine whether the right patients reach these therapies.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Translational Lung Cancer Research. The article has undergone external peer review.

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0527/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0527/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mavrakis KJ, McDonald ER 3rd, Schlabach MR, et al. Disordered methionine metabolism in MTAP/CDKN2A-deleted cancers leads to dependence on PRMT5. Science 2016;351:1208-13. [Crossref] [PubMed]
2. Brune MM, Roma L, Chijioke O, et al. MTAP Expression by Immunohistochemistry: A Novel Biomarker in NSCLC. J Thorac Oncol 2026;21:112-23. [Crossref] [PubMed]
3. University of California San Francisco Clinical Trials. A Study of Anvumetostat in participants with advanced MTAP-null solid tumors (MTAPESTRY 101) [Internet]. San Francisco: University of California San Francisco; [updated 2026 Mar 26; cited 2026 Apr 27]. Available online: https://clinicaltrials.ucsf.edu/trial/NCT05094336
4. Rodon J, Prenen H, Sacher A, et al. First-in-human study of AMG 193, an MTA-cooperative PRMT5 inhibitor, in patients with MTAP-deleted solid tumors: results from phase I dose exploration. Ann Oncol 2024;35:1138-47. [Crossref] [PubMed]
5. National Cancer Institute. Safety and tolerability of TNG462 in patients with MTAP-deleted solid tumors [Internet]. Bethesda (MD): National Cancer Institute; [cited 2026 Apr 27]. Available online: https://clinicaltrials.gov/study/NCT05732831
6. University of California San Francisco Brain Tumor Center. A Phase I/II multicenter, open-label study to evaluate the safety, tolerability, and preliminary anti-tumor activity of TNG908 in patients with MTAP-deleted advanced or metastatic solid tumors [Internet]. San Francisco: University of California San Francisco; [cited 2026 Apr 27]. Available online: https://braintumorcenter.ucsf.edu/clinical-trial/phase-iii-multicenter-open-label-study-evaluate-safety-tolerability-and-preliminary
7. Bristol Myers Squibb. A study of BMS-986504 in participants with pre-treated advanced or metastatic non-small cell lung cancer (NSCLC) with homozygous MTAP deletion (MountainTAP-9) [Internet]. Princeton (NJ): Bristol Myers Squibb; [cited 2026 Apr 27]. Available online: https://www.bmsclinicaltrials.com/us/en/clinical-trials/NCT06855771
8. Bristol Myers Squibb. A study to compare the combination of BMS-986504 with pembrolizumab and chemotherapy versus placebo plus pembrolizumab and chemotherapy in first-line metastatic non-small cell lung cancer participants with homozygous MTAP deletion [Internet]. Princeton (NJ): Bristol Myers Squibb; [cited 2026 Apr 27]. Available online: https://www.bmsclinicaltrials.com/us/en/clinical-trials/NCT07063745
9. ClinicalTrials.gov. Study of IDE397 in participants with solid tumors harboring MTAP deletion [Internet]. Bethesda (MD): National Library of Medicine (US); ClinicalTrials.gov identifier: NCT04794699; [cited 2026 Apr 27]. Available online: https://clinicaltrials.gov/study/NCT04794699
10. Chapel DB, Schulte JJ, Berg K, et al. MTAP immunohistochemistry is an accurate and reproducible surrogate for CDKN2A fluorescence in situ hybridization in diagnosis of malignant pleural mesothelioma. Mod Pathol 2020;33:245-54. [Crossref] [PubMed]
11. Jennings LJ, Arcila ME, Corless C, et al. Guidelines for Validation of Next-Generation Sequencing-Based Oncology Panels: A Joint Consensus Recommendation of the Association for Molecular Pathology and College of American Pathologists. J Mol Diagn 2017;19:341-65. [Crossref] [PubMed]
12. Wallenta Law J, Bapat B, Sweetnam C, et al. Real-World Impact of Comprehensive Genomic Profiling on Biomarker Detection, Receipt of Therapy, and Clinical Outcomes in Advanced Non-Small Cell Lung Cancer. JCO Precis Oncol 2024;8:e2400075. [Crossref] [PubMed]
13. Han G, Yang G, Hao D, et al. 9p21 loss confers a cold tumor immune microenvironment and primary resistance to immune checkpoint therapy. Nat Commun 2021;12:5606. [Crossref] [PubMed]
14. Henrich FC, Singer K, Poller K, et al. Suppressive effects of tumor cell-derived 5'-deoxy-5'-methylthioadenosine on human T cells. Oncoimmunology 2016;5:e1184802. [Crossref] [PubMed]
15. Ikushima H, Watanabe K, Shinozaki-Ushiku A, et al. Clinicogenomic Landscape of MTAP-Deleted Thoracic Malignancies Across US and Japanese Nationwide Cohorts: Impact of Concurrent CDKN2A Alterations and Driver Mutations. JCO Precis Oncol 2026;10:e2600100. [Crossref] [PubMed]