Abstract
High SOX2 expression is a potent independent biomarker for poor prognosis in multiple myeloma (MM). By analyzing transcriptomic data from the CoMMpass study, researchers found that patients with elevated SOX2 levels face significantly shorter overall survival (OS) and post-relapse survival. This association persists even after adjusting for established clinical and genetic risk factors, highlighting the role of "stemness" in driving aggressive disease phenotypes and treatment resistance.
Introduction
Multiple myeloma (MM) is a hematologic malignancy characterized by the clonal proliferation of plasma cells. Despite therapeutic advances, relapse remains nearly universal and is the primary cause of mortality. Identifying biomarkers that predict aggressive disease is crucial for risk stratification and personalized therapy.
SOX2 is a well-characterized transcription factor essential for maintaining pluripotency in embryonic stem cells. In solid tumors, SOX2 expression is often associated with "stem-cell-like" properties, including self-renewal and chemoresistance. This study investigates the clinical impact of SOX2 expression in MM, a domain where its prognostic value was previously poorly defined.
Methodology
The study utilized data from the MMRF CoMMpass study (NCT145429), specifically Interim Analysis 18, which includes over 1,000 patients with a minimum 8-year follow-up.
- Patient Stratification: Bulk RNA sequencing was performed on bone marrow samples from 754 newly diagnosed patients. "High SOX2" was defined as the top 10% of expression levels, while the remaining 90% were categorized as "Low SOX2."
- Statistical Analysis: Kaplan-Meier curves and log-rank tests were used to compare overall survival (OS) and progression-free survival (PFS). Multivariate analysis using Cox proportional hazards regression and Inverse Probability Weighting (IPW) was conducted to adjust for confounding variables such as age, gender, ISS stage, and high-risk cytogenetics (e.g., del17p, amp1q).
- Biological Pathways: Differential gene expression analysis (DESeq2) and KEGG enrichment were employed to identify pathways associated with high SOX2 expression.
Results
The analysis revealed a stark contrast in clinical outcomes based on SOX2 expression levels.
Survival Outcomes
Patients in the high SOX2 group experienced significantly shorter survival. The median OS for the high SOX2 group was 57.0 months, compared to 95.3 months for the low SOX2 group ($p = 0.0061$). While median PFS was also shorter in the high SOX2 group (24.6 vs. 37.2 months), this difference did not reach traditional statistical significance ($p = 0.086$).
Independent Risk Factor
Multivariate analysis confirmed that high SOX2 expression is an independent predictor of poor OS. After adjusting for other risk factors, the Hazard Ratio (HR) for high SOX2 was 1.6 (95% CI: 1.06–2.3, $p = 0.025$). This indicates a 60% increased risk of death for patients with high SOX2 expression at diagnosis.
Impact Post-Relapse
The negative prognostic impact of SOX2 persisted beyond the initial treatment phase. Following the first disease progression, patients who were initially "High SOX2" had a median survival of 30.0 months, significantly lower than the 52.8 months observed in the "Low SOX2" group ($p = 0.014$).
Discussion
The findings suggest that high SOX2 expression defines a distinct, aggressive biological subtype of MM. Interestingly, high SOX2 expression was inversely correlated with the "amp1q" chromosomal abnormality, suggesting that SOX2-mediated "stemness" may represent an alternative pathway to malignancy.
Mechanistically, high SOX2 levels were associated with the upregulation of SOX2OT (an overlapping transcript that promotes growth) and CCNA1 (Cyclin A1), pointing toward an accelerated cell cycle and enhanced proliferative capacity.
Limitations and Future Directions
A key limitation of this study is the reliance on bulk RNA sequencing, which masks intra-tumoral heterogeneity. It remains unclear whether SOX2 is uniformly high across the tumor or restricted to a small population of "myeloma stem cells." Future research utilizing single-cell RNA sequencing (scRNA-seq) is needed to resolve this. Additionally, the potential for SOX2 as a therapeutic target—either through direct inhibition or immunotherapy—warrants further investigation.
Conclusion
SOX2 expression at diagnosis is a robust, independent predictor of poor survival in multiple myeloma. Integrating SOX2 levels into current risk-stratification models could improve the identification of high-risk patients and guide the development of more effective, targeted treatment strategies.
Summary updated