Prognostic Significance of Systemic Immune-Inflammation Index in Patients with de Novo non‑M3 Acute Myeloid Leukemia

The systemic immune-inflammation index (SII), integrating peripheral neutrophil, platelet, and lymphocyte counts, demonstrates proven prognostic value across malignancies. Its clinical utility in de novo non-M3 acute myeloid leukemia (AML) remains underexplored.

This retrospective study established the prognostic significance of SII in 262 non-M3 AML patients (diagnosed April 2016-April 2021) and developed a validated survival nomogram. Optimal SII stratification thresholds were determined by X-tile analysis. Clinical characteristics, induction response, overall survival (OS) and event-free survival (EFS) were analyzed. Multivariable Cox regression identified independent prognostic factors for nomogram construction. Model validation included C-index, calibration curves, time-dependent ROC analysis, and decision curve assessment (DCA).

X-tile identified 43.3×10⁹/L as the optimal SII cutoff. High SII (≥43.3×10⁹/L) correlated with inferior median OS (11 vs 66 months; P<0.001) and EFS (8 vs 20 months; P<0.001) versus low SII. Subgroup analysis revealed the adverse prognostic impact of high SII was particularly evident in patients aged <60 years, those with favorable 2022 ELN risk, and non-transplant recipients. The multivariate analysis pinpointed SII≥43.3×10⁹/L (Hazard Ratio [HR]=1.781, 95% Confidence Interval [CI]:1.264-2.509, P=0.001) as a significant independent predictor of OS. Additionally, age, white blood cell (WBC) count ≥100×10⁹/L, 2022 ELN risk classification, and receiving HSCT were also independent predictors of OS. Based on these independent predictors, we developed a prognostic nomogram. The nomogram demonstrated good discriminatory ability (C-index: 0.715), accurate calibration, and provided clinical net benefit in DCA. Moreover, the time-dependent ROC AUCs for 1-, 2-, and 3-year OS were 0.781, 0.752 and 0.781, respectively.

SII is an independent prognostic marker in de novo non-M3 AML. The proposed nomogram, incorporating both SII and established prognostic variables, enables precise and individualized survival prediction. This tool has the potential to inform risk-adapted treatment strategies and guide decision-making in clinical practice.