Genomics of MTAP Loss in \>500,000 Solid Tumor Specimens Profiled Using Comprehensive Genomic Profiling Platforms

Our plain-English summary of Genomics of MTAP Loss in >500,000 Solid Tumor Specimens Profiled Using Comprehensive Genomic Profiling Platforms, by D.C. Pavlick, S. Rengarajan, J.K. Lee, et al (JCO Precision Oncology, 2026).

The study

• Authors: D.C. Pavlick, S. Rengarajan, J.K. Lee, D.X. Jin, G.M. Frampton, B.J. Fendler, A. Savol, O. Gjoerup, H. Tukachinsky, R.S.P. Huang, A.B. Schrock, J.S. Ross, A. Necchi
• Journal: JCO Precision Oncology
• Published: 2026;10(5):e2600037
• Read it: PubMed · Publisher · Free full text

Abstract

Purpose: Methylthioadenosine phosphorylase (MTAP) genomic loss is an emerging biomarker for PRMT5 and MAT2A inhibitors based on synthetic lethality. The MTAP gene is located on chromosome 9p21.3 near CDKN2A/B. MTAP homozygous loss across tumor types, specific exons lost, MTAP expression, and the landscape of coalterations were assessed.

Methods: 409,755 tissue biopsies (TBx) and 85,801 liquid biopsies (LBx) were sequenced using hybrid capture-based NGS (FoundationOne CDx or FoundationOne Liquid CDx), evaluating all classes of genomic alterations and circulating tumor DNA (ctDNA) tumor fraction (TF). Additionally, 6,740 TBx were subjected to both RNA and DNA sequencing to compare MTAP gene expression with genomic MTAPloss.

Results: The prevalence of pan-tumor MTAPloss was 11.4% using TBx, with the highest prevalence observed in mesothelioma (32.8%), glioma (32.7%), pancreatic (28.9%), and bladder cancers (26.4%). Prevalence of MTAPloss on LBx approximated that of TBx when ctDNA TF ≥20% (positive percent agreement [PPA] = 86.2%). CDKN2Aloss was not a reliable proxy for MTAPloss; 33.7% of CDKN2Aloss cases did not have MTAPloss. MTAPloss was enriched with CDKN2A/B losses pan-tumor, EGFR in NSCLC, ERBB2 in colorectal cancer, and PTEN in prostate cancer, while MTAPno_loss was enriched for RB1 pan-tumor, APC in colorectal cancer, CCNE1 in breast and ovarian cancer, and SPOP in prostate cancer. Complete loss (exons 1-8) was observed in 82.9%, multiple exons in 16.5% and exon 8 alone in <1%. MTAPloss, specifically the number of exons lost, was correlated with reduced RNA expression.

Conclusion: MTAPloss is a frequent pan-tumor alteration that predicts potential sensitivity to PRMT5 or MAT2A inhibitors. Although most tumors exhibit complete MTAPloss, 16.5% are characterized by partial MTAPloss where clinical benefit remains uncertain. The genomic coalteration landscape reported here may inform future PRMT5 or MAT2A clinical trials, potentially in combination with other agents.

In plain English

This was a retrospective, descriptive analysis of a Boston cancer genetic-testing laboratory's database. The authors examined 409,755 tissue and 85,801 blood (liquid) biopsies from patients with solid tumors, tested in routine care through December 2024. They measured how often both copies of the MTAP gene were lost, in whole or in part. The article reports no asbestos-exposure data or analyses. Across all tissue samples, MTAP loss was found in 11.4% (46,857 of 409,755). Mesothelioma had the highest rate, 32.8% (419 of 1,279 samples). The article does not say where these mesotheliomas arose. Rates were 32.7% in glioma, 28.9% in pancreatic cancer and 26.4% in bladder cancer. In blood samples the overall rate was 1.9%. In 856 matched tissue-and-blood pairs where tumor DNA was at least 20% of cell-free DNA in blood, the blood test detected 86.2% of tissue-detected losses. Among 46,663 MTAP-loss tissue specimens with exon-level annotation, 82.9% had loss of the whole gene. The authors concluded that MTAP loss is frequent and predicts potential sensitivity to PRMT5 or MAT2A inhibitors. They noted they had no clinical outcome data.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective, descriptive analysis of genomic test results from a commercial clinical laboratory database (FoundationCORE)
• Who was studied: Specimens from patients with solid tumors who had comprehensive genomic profiling as part of routine clinical care at Foundation Medicine, Boston, MA. The set included 409,755 tissue biopsies and 85,801 liquid biopsies profiled through December 2024, plus 7,047 tissue biopsies with both DNA and RNA results profiled through March 2025. Mesothelioma accounted for 1,279 of the tissue specimens.
• Exposure: The article reports no asbestos, fiber or occupational exposure data or analyses. The feature examined was homozygous (two-copy) loss of the MTAP gene, complete or partial. It was detected by DNA sequencing of tumor tissue (FoundationOne CDx) or of cell-free DNA from blood plasma (FoundationOne Liquid CDx).
• Main results:
  • Across all tissue biopsies, MTAP loss was found in 11.4% (46,857/409,755). Mesothelioma had the highest frequency: 32.8% (419/1,279 mesothelioma specimens).
  • In tissue biopsies, MTAP loss was found in 32.7% of glioma specimens (2,918/8,917).
  • In tissue biopsies, MTAP loss was found in 28.9% of pancreatic cancer specimens (8,416/29,136).
  • In tissue biopsies, MTAP loss was found in 26.4% of bladder cancer specimens.
  • In all liquid biopsies, MTAP loss was found in 1.9% (1,630/85,801).
  • In liquid biopsies where the circulating tumor DNA tumor fraction was 20% or more, MTAP loss was found in 7.3% (1,098/15,117).
  • In 3,946 matched tissue and liquid biopsy pairs, the tissue result was taken as the reference. The liquid biopsy had a sensitivity of 22.1%, specificity of 99.8, positive predictive value of 94.4% and negative predictive value of 89.2%.
  • In the 856 matched pairs with a circulating tumor DNA tumor fraction of 20% or more, liquid biopsy sensitivity rose to 86.2% and negative predictive value to 98.3%.
  • CDKN2A loss was found in 17.2% of tissue biopsies (70,535/409,755). Of these, 66.3% (46,736/70,535) also had MTAP loss.
  • Exon-level annotation was available for 46,663 MTAP-loss tissue specimens. In these, 82.9% had loss of the whole gene (exons 1-8). The next most frequent patterns were loss of exons 5-8 (5.1%), exons 2-8 (4.4%) and exons 6-8 (4.1%).
  • Compared across all tumor types, CDKN2A alterations were present in 99.8% of MTAP-loss specimens versus 14.9% of specimens without MTAP loss.
  • In 6,740 tissue specimens with RNA and DNA results, MTAP RNA expression was highest in specimens without MTAP loss and lowest in specimens with homozygous MTAP loss (p less than .0005). Median values were 8.8 for no loss (n = 3,668), 8.4 for heterozygous loss (n = 2,313) and 6.9 for homozygous loss (n = 759).
• The authors' conclusion: The authors concluded that MTAP loss is a frequent alteration across tumor types that predicts potential sensitivity to PRMT5 or MAT2A inhibitors. They noted that clinical benefit remains uncertain for the 16.5% of tumors with partial MTAP loss.
• Limitations the authors state:
  • No clinical outcome data to corroborate associations between MTAP loss and outcomes with specific treatments.
  • Central pathology review was limited to one tissue block per patient, and samples were collected over a large time frame.
  • Potential bias toward patients of European ancestry.
  • Results were not compared with immunohistochemical (IHC) testing for MTAP loss, an alternative method used to select patients for trials.

Topics

Medical terms in this study, explained in our Lexicon: MTAP, Biomarker, MAT2A Inhibitor, CDKN2A, ctDNA, Mesothelioma, EGFR, Colorectal Cancer.

Related studies

• Increasing the therapeutic index of 5-fluorouracil and 6-thioguanine by targeting loss of MTAP in tumor cells — B. Tang, J.R. Testa, et al. (2012)
• Evaluation of MTAP immunohistochemistry loss of expression in ovarian serous borderline tumors as a potential marker for prognosis and progression — N. Nilforoushan, N.A. Moatamed (2020)
• Prevalence of S-methyl-5'-thioadenosine Phosphorylase (MTAP) Deficiency in Human Cancer: A Tissue Microarray Study on 13,067 Tumors From 149 Different Tumor Types — N. Gorbokon, N. Wößner, et al. (2024)
• Immunohistochemical detection of MTAP and BAP1 protein loss for mesothelioma diagnosis: Comparison with 9p21 FISH and BAP1 immunohistochemistry — T. Hida, M. Hamasaki, et al. (2017)
• CDKN2A and MTAP deletions in peritoneal mesotheliomas are correlated with loss of p16 protein expression and poor survival — A.M. Krasinskas, D.L. Bartlett, et al. (2010)
• Comprehensive genomic and transcriptomic analysis enables molecularly guided therapy options in peritoneal and pleural mesothelioma — L. Möhrmann, M. Werner, et al. (2025)

Browse more research in JCO Precision Oncology, or return to medical research.