MOSAIC: A Multi-Feature Genomic Instability Score Independently Predicts Sarcoma Survival in 2138 Patients.

Abstract

Objectives: While genomic profiling has identified prognostic markers in sarcoma, clinical risk stratification remains largely histology-based. The combined impact of multiple genomic instability features on survival remains poorly understood. This study evaluated the prognostic utility of a novel Measure Of Sarcoma Aggregate Instability Complex (MOSAIC) Score.

Methods: We conducted a secondary analysis of 2138 sarcoma patients from the Sarcoma (MSK, Nat Commun. 2022) data set in cBioPortal. The MOSAIC score integrated 6 genomic instability markers: mutation burden, whole-genome doubling, copy number alterations, ploidy, microsatellite instability, and a fraction of genome altered. Cox proportional hazards models adjusted for clinical covariates assessed survival associations.

Results: Higher MOSAIC scores correlated with worse survival (HR=1.11 per unit, 95% CI: 1.07-1.15, P=4.67 × 10-8). The first principal component explained 47.6% of the total variance. A threshold effect emerged between quartiles, with higher quartiles showing nearly doubled mortality risk (adjusted HRs: Q2=1.89, Q3=1.88, Q4=1.96; all P<0.001). Chromosomal instability markers (copy number alterations: 0.51, whole-genome doubling: 0.49) contributed more than point mutations (mutation burden: 0.15). MOSAIC's prognostic impact varied by anatomic site, with strong effects in thoracic (HR=1.66, P=4.74 × 10-4) and trunk (HR=1.50, P=0.018) sarcomas.

Conclusions: MOSAIC provides independent prognostic value across sarcoma subtypes, surpassing conventional clinical factors. Its integration of routine genomic features broadens applicability and may inform immunotherapy response prediction. The observed threshold effect and dominance of chromosomal instability markers offer novel insights into sarcoma biology and therapeutic targeting. Prospective validation is warranted for clinical implementation.