Personalized survival prediction in primary urinary tract melanoma (PMUT) using a nomogram.

Abstract

e21550 Background: PMUT is an extremely rare subtype of mucosal melanoma that accounts for 1% of all melanomas. A prognostic index for estimating the clinical outcomes of PMUT would be attractive in the current clinical practice. We aimed to develop and validate an effective nomogram for predicting the survival of patients with PMUT. Methods: We analyzed data from the SEER database of patients diagnosed with PMUT between 2004 and 2021. Patients who were not diagnosed based on histology, previous history of cancer or other concurrent malignancies, or unknown data were excluded. The clinicopathological features of each group were compared using the chi-square test. Kaplan-Meier estimation, log-rank tests, and Cox regression analyses were used to identify the prognostic factors for OS and CSS. A nomogram was developed to predict 5-year OS, and a calibration curve was plotted to evaluate the agreement between the actual (observed) outcomes and expected probabilities. Results: In total, 1,914 patients were included, of whom 1,829 were female. Most patients (86.4%) were White, followed by Asians or Pacific Islanders (6.6%). The most common primary site in females was the vulva (72.2%), followed by the vagina (24.3%), whereas in males, it was the penis (57.6%) and scrotum (27.1%). Tumor stage was most frequently categorized as T4 (21.8%) and T3 (7.9%). Additionally, 42.3% of the patients had no nodal involvement, while metastasis was observed in 10.5% of the cases. Surgery was the most common treatment (88.5%), followed by radiation (19.6%) and chemotherapy (7.3%). Patients who did not receive radiation had higher 5-year survival rates (OS: 42.55%, CSS: 50.01%) than those who received radiation (OS: 20.29%, CSS: 23.26%). Surgery significantly improved survival (OS: 41.52%, CSS: 48.08%) compared with patients who did not undergo surgery (OS: 10.0%, CSS: 17.14%). Younger patients ( < 67 years) had better survival rates (OS: 52.84%, CSS: 54.86%) than older patients (OS: 46.8%, CSS: 55.10%). Patients with metastasis had worse outcomes (OS, 7.61%; CSS, 8.2%) than those without metastasis (OS, 47.22%; CSS, 55.18%). Multivariate Cox regression analysis identified older age, chemotherapy, and metastasis as poor prognostic factors, while surgery was a good prognostic factor. We developed a prognostic nomogram that integrates significant variables, including metastasis, surgery, chemotherapy, radiation, and age. The model assigned the highest weight to metastatic status, followed by surgery, age, and chemotherapy. The AUC was 0.61 (95% CI: 0.59–0.63) and the calibration curves exhibited a mean error of 0.023. Conclusions: This study highlights the potential of nomograms in advancing personalized medicine for PMUT. Further research is needed to assess the impact of radiotherapy, as it was not identified as a significant prognostic factor, and to better understand the role of chemotherapy, which was associated with poorer outcomes.