Novel genomic risk stratification model for primary high-grade malignant peripheral nerve sheath tumor (MPNST)

Risk stratification across the three main clinical subsets of malignant peripheral nerve sheath tumor (MPNST), neurofibromatosis type I (NF1-related), sporadic, and prior radiation therapy (RT), is based mainly on clinicopathologic parameters, such as size, grade, stage, and NF1 status. Moreover, no prior study investigated the additional impact of genomic alterations in the prognosis of high-grade MPNST using clinically validated DNA targeted next-generation sequencing (NGS) panels. Our goal was to integrate clinicopathologic and genomic parameters using an elastic-net penalized Cox proportional hazards machine learning model using OncoCast for risk prediction. Herein we perform comprehensive mutational and copy number profiling using Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) on 81 primary localized high-grade MPNSTs (51% NF1-related, 38% sporadic, 11% RT-associated). The most common genomic alterations included NF1 (51% germline, 59% somatic), CDKN2A/B (62%), PRC2 components (SUZ12, EED) (53%), and TP53 (25%). Variables selected by OncoCast as significantly associated with survival were used to construct a three-tier risk stratification model for progression-free survival (PFS) and disease-specific survival (DSS). For PFS, patients with chr16 deletion were classified as high-risk, those with concurrent germline and somatic NF1 alterations as low-risk, and the remainder was assigned to the intermediate-risk group. For DSS, cases with fraction genome altered (FGA) > 50% were defined as high-risk, those with PRC2 abnormalities, CDKN2A deletion, TERT promoter mutation, or chr16 deletion as intermediate-risk, and cases lacking all the aforementioned alterations as low-risk. The high-risk group showed significantly inferior survival compared to the low-risk group (both PFS and DSS p < 0.001). Subgroup analysis showed that among NF1-related MPNST, co-occurring somatic NF1 mutation or WT TP53 was associated with superior PFS. Collectively, genomic alterations detected by clinical NGS panels provide potential new biomarkers for risk stratification that can be integrated with conventional parameters to provide improved prognostication and guide therapeutic strategies. © 2026 The Pathological Society of Great Britain and Ireland.