Objective To evaluate the prognostic value of intratumoral metabolic heterogeneity parameters derived from ¹⁸F-fluorodeoxyglucose (FDG) PET/CT in patients with metastatic malignant melanoma (MM).

Methods A retrospective analysis was conducted on 96 patients with MM who underwent ¹⁸F-FDG PET/CT at the Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School between March 2014 and December 2023. The cohort included 51 males and 45 females, with a mean age of (62.0±13.4) years. LIFEx software was used to automatically delineate the region of interest (ROI) of metastatic lesion with the highest ¹⁸F-FDG uptake in each patient. Traditional PET/CT metabolic parameters, including maximum standardized uptake value (SUV_max), metabolic tumor volume (MTV), and total lesion glycolysis (TLG), were measured. Intratumoral metabolic heterogeneity parameters, such as the area under the cumulative standardized uptake value volume histogram curve (AUC-CSH), linear regression slope, SUV_max/mean standardized uptake value (SUV_mean), and coefficient of variation (CV), were also calculated. MTV was delineated using thresholds of SUV_max=2.5, 3.0, 3.5, or 40%, 60%, and 80% of SUV_max, and the absolute values of the regression slopes were defined as heterogeneity index (HI)-1 and HI-2. AUC-CSH, SUV_max/SUV_mean, and CV were calculated using a threshold of SUV=2.5. Using receiver operating characteristic (ROC) curve to determine the critical value of measurement data. Kaplan-Meier survival curves was plotted, and a Log-rank test was performed to compare differences in overall survival (OS) and progression-free survival (PFS) based on different metabolic parameter thresholds. Univariate and multivariate Cox proportional hazard model analyses (forward stepwise method) were conducted to identify prognostic factors, including clinical characteristics, traditional PET/CT metabolic parameters, and PET-based metabolic heterogeneity parameters. Collinearity analysis was performed prior to multivariate Cox proportional hazard model analysis.

Results The median OS of the 96 patients was 15.50 (1.00–103.00) months. Disease progression occurred in 68 patients (70.8%), and 46 patients (47.9%) died. Multivariate Cox proportional hazard model analysis identified gender (HR=0.49, 95%CI: 0.29–0.81, P=0.005), LDH level (HR=1.99, 95%CI: 1.10–3.60, P=0.024), short diameter of metastatic lesions (HR=2.85, 95%CI: 1.14–7.09, P=0.025), SUV_max (HR=2.30, 95%CI: 1.33–3.96, P=0.003), and AUC-CSH (HR=0.43, 95%CI: 0.22–0.85, P=0.016) as independent prognostic factors of PFS. Age (HR=2.08, 95%CI: 1.10–3.93, P=0.024), primary tumor location (HR=2.20, 95%CI: 1.16–4.17, P=0.016), and SUV_max (HR=2.91, 95%CI: 1.14–7.42, P=0.025) were independent prognostic factors of OS.

Conclusions The HI-1, SUV_max/SUV_mean derived from ¹⁸F-FDG PET/CT has potential value in predicting PFS in patients with MM. The AUC-CSH is an independent prognostic factor for PFS , and AUC-CSH has certain value in predicting OS in MM patients.