CO25 | P53 isoform diversity in multiple myeloma: unexplored risk factor beyond TP53 mutation and del(17p)

Abstract

Introduction. In multiple myeloma (MM), TP53 abnormalities (deletions/mutations) confer high-risk disease. Beyond full-length p53 (p53FL), alternative isoforms modulate p53 activity and cellular stress responses, but their expression patterns and clinical impact remain largely unexplored. Defining isoform-specific biology may refine risk stratification and enable personalized therapy.
Aim. To enhance MM risk stratification by integrating p53 isoforms expression profiles, improving prognostic accuracy, identify novel molecular subgroups, and ultimately supporting tailored therapeutic strategies.
Patients and Methods. RNA-sequencing data from 659 newly diagnosed MM patients (pts) (CoMMpass) were analysed for major p53 isoforms. Expression levels were classified as absent, low, or high using median-based cut-offs. Associations with genomic features, treatment response and survival outcomes were evaluated. Isoform expression was combined with cytogenetic risk to define novel prognostic categories.
Results. p53FL and p53β were widely expressed (95% and 99%), while p53γ, Δ40p53α and Δ133p53α/β showed heterogeneous patterns. High p53FL often co-occurred with high p53β and Δ133p53α, while absence of Δ133p53β was enriched in pts with high p53FL/p53β, suggesting functional interactions among isoforms. Del(17p) was detected in 10.8% of pts and associated with lower p53FL/p53β expression (p<0.0001); pts expressing Δ133p53β did not carry del(17p) (p<0.0001). Absence of p53FL (4.9%) correlated with elevated LDH, poorer treatment response, and markedly reduced progression-free survival (PFS: 7.0 vs 32.8 vs 20.6 months (mos)) and overall survival (OS: 7.0 vs 60.3 vs 36.7 mos) for absent, high, and low expression, respectively (p<0.001). p53β expression correlated with favorable biological features, while Δ133p53β identified a small subset with poor prognosis (p<0.001). Integrating isoform profile with key cytogenetic abnormalities stratified pts into favorable, poor, and intermediate risk groups. Notably, poor-risk pts had significantly higher hazards for PFS (HR 2.74, 95% CI 1.50–4.99, p=0.000987) and OS (HR 3.79, 95% CI 1.91–7.51, p<0.001) compared with favorable-risk pts. Combined with ISS, poor-risk pts with ISS III had markedly shorter PFS (median 11.5 mos) and OS (median 21.1 mos) than favorable-risk pts (PFS 46–50 mos, OS 102 mos). Isoform-based risk retained independent significance in multivariate models (interaction p=0.002), improving the prognostic accuracy.
Conclusion. p53 isoform profiling captures MM heterogeneity and identifies high-risk pts beyond conventional TP53 assessment. An isoform-based risk classifier robustly stratifies PFS and OS, particularly when combined with ISS, supporting its potential as a prognostic tool for personalized management. Prospective validation is warranted prior to clinical implementation.
Acknowledgments: AIRC19-IG22059, BolognAIL OdV and RC2025-2797269