Beyond variant detection: Mismatch repair gene alterations in plasma cell-free DNA from 1,226 glioma samples, and plasma-confirmed functional deficiency in an unbiopsied pontine glioma.
AI interpretation is pending for this paper.
Open original publication →What the AI sees
Not AI summarized yet.
Research significance
Pending deeper interpretation.
Source abstract
BACKGROUND: Mismatch repair (MMR) gene alterations are uncommon in glioma but can produce hypermutation and sensitivity to checkpoint inhibition. Profiling is often unavailable in diffuse intrinsic pontine glioma (DIPG), where biopsy carries substantial risk. We assessed whether plasma cell-free DNA (cfDNA) sequencing detects these alterations in gliomas and report an unbiopsied DIPG case in which plasma findings were functionally corroborated. METHODS: Retrospective analysis of an index DIPG case profiled in plasma, together with sample-level analysis of 1,226 glioma plasma cfDNA profiles from a clinical digital-sequencing assay and 5,369 glioma tissue samples from public datasets. We assessed MSH6, MSH2, MLH1, and PMS2 alterations, tumor mutational burden (TMB), and microsatellite instability (MSI). Plasma-tissue comparisons used Fisher's exact tests, Bonferroni threshold p < 0.0125. RESULTS: In the index case, plasma cfDNA showed a truncating MSH6 variant at 100% variant allele fraction, MSI-high status, and TMB of 132.3 mut/Mb in a child never given an alkylating agent; all other somatic-range variants lay below 1.7%, indicating constitutional origin. Pembrolizumab was given across 21 doses with radiographic stability, with progression at approximately 19 months; at 24 months the patient was alive, performance status 3 to 4. In the cohort, 680 of 1,226 samples (55.5%) had at least one non-synonymous alteration. Within these, MSH6 was altered in 8 (1.2%), MSH2 and MLH1 in 6 each (0.9%) and PMS2 in 2 (0.3%). MSH6 was the most altered MMR gene in both plasma and tissue, at 1.2% each, and no gene differed significantly after correction. Median TMB was higher in MMR-altered samples in both cohorts. CONCLUSIONS: Plasma cfDNA detected MMR gene alterations at frequencies indistinguishable from tissue, and in one unbiopsied case gave functional evidence of deficiency through concurrent MSI-high status and hypermutation. Where tissue is unobtainable a positive result identifies a subgroup with a specific therapeutic hypothesis, whereas a microsatellite-stable result is weak evidence against deficiency. Plasma complements rather than replaces tissue testing.