# Prognostic Value of Systemic Inflammatory Markers in Locally Advanced or Metastatic Melanoma: A Real-World Analysis

## TL;DR

This study shows that blood-based inflammation and nutrition markers can help predict survival in advanced melanoma patients, but traditional factors like treatment response remain the strongest predictors.

## Contribution

The study evaluates the real-world prognostic value of multiple dynamic and baseline inflammatory and nutritional indices in advanced melanoma.

## Key findings

- Elevated baseline SII and SIRI were significantly associated with shorter overall survival.
- 6-month SIRI and dynamic SIRI showed strong prognostic value for survival outcomes.
- Response to first-line therapy was the only independent predictor of disease progression in multivariate analysis.

## Abstract

Malignant melanoma is an aggressive cancer with highly variable clinical outcomes, even among patients with similar disease stages. Reliable and easily accessible biomarkers are needed to improve prognostic assessment and guide treatment decisions. Systemic inflammatory and nutritional indices, such as the systemic immune–inflammatory index (SII), systemic inflammatory response index (SIRI), and the Controlling Nutritional Status (CONUT) score, can be calculated from routine blood tests and reflect the host’s immune and nutritional status. In this real-world study, we evaluated the prognostic value of baseline and longitudinal inflammatory indices, including 6-month and dynamic SIRI, in patients with locally advanced or metastatic melanoma. We found that poor nutritional status and elevated inflammatory markers were associated with worse survival outcomes. However, established clinical factors—particularly disease stage and treatment response—remained the strongest independent predictors of prognosis. Our findings suggest that inflammation- and nutrition-based indices may serve as complementary tools alongside conventional clinical parameters to support risk stratification and long-term monitoring in melanoma patients using readily available laboratory data.

Background/Objectives: Malignant melanoma remains a highly aggressive malignancy with substantial mortality despite advances in systemic therapy. Identifying simple and reproducible prognostic biomarkers is essential for improving risk stratification. Inflammation- and nutrition-based indices—including the Systemic Immune–Inflammation Index (SII), Systemic Inflammatory Response Index (SIRI), dynamic SIRI, and the Controlling Nutritional Status (CONUT) score—have shown prognostic value in various cancers. This study assessed the prognostic significance of these indices in patients with locally advanced or metastatic melanoma using real-world data. Methods: A retrospective cohort of 138 patients treated between 2010 and 2023 was analyzed. Baseline demographic, clinical, nutritional, and inflammatory parameters were collected. Optimal cut-off values for SII, SIRI, 6-month SIRI, and dynamic SIRI were determined using receiver operating characteristic analysis. Overall survival (OS) and progression-free survival (PFS) were evaluated using the Kaplan–Meier method, and independent predictors were identified with multivariate Cox regression. Results: Elevated baseline SII and SIRI were significantly associated with shorter overall survival. Both 6-month SIRI and dynamic SIRI demonstrated strong prognostic value, emphasizing the importance of longitudinal inflammatory changes. In multivariate analysis, response to first-line therapy emerged as the only independent predictor of disease progression. Patients with a CONUT score ≥ 3 showed significantly shorter OS and PFS in univariate analyses, underscoring the prognostic relevance of nutritional status. Conclusions: SII, SIRI, 6-month SIRI, dynamic SIRI, and CONUT are practical, accessible, and reproducible biomarkers with meaningful prognostic value in advanced melanoma. Incorporating these indices into routine clinical assessment may enhance risk stratification and support more personalized treatment decision-making.