SIR 2025
Interventional Oncology
Scientific Session
Wei Tian, MD, PhD (he/him/his)
Research Fellow
Memorial Sloan Kettering Cancer Center, United States
Joseph P. Erinjeri, MD PhD
Attending
Memorial Sloan Kettering Cancer Center, United States
Liwei Jiang, MD
Assistant Attending
Memorial Sloan Kettering Cancer Center, United States
Debkumar Sarkar, DO
Associate Attending
Memorial Sloan Kettering Cancer Center, United States
Etay Ziv, MD PhD
Associate Attending
Memorial Sloan Kettering Cancer Center, United States
Alexander Shoushtari, MD
Assistant Attending
Memorial Sloan Kettering Cancer Center, United States
Chenyang Zhan, MD, PhD
Assistant Attending
Memorial Sloan Kettering Cancer Center, United States
To compare the safety and efficacy of hepatic arterial embolization (HAE) and Yttrium-90 (Y90) radioembolization in the treatment of hepatic metastatic melanoma using propensity score matching (PSM).
Materials and Methods:
This retrospective study included 75 patients with hepatic metastatic melanoma treated with either HAE (63 patients) or Y90 radioembolization (12 patients) between 04/2009 and 04/2024 at a tertiary cancer center. All patients received immune checkpoint inhibitors in the study period. Demographic data, tumor characteristics, treatment details, laboratory values, and dates of death or last clinical encounter were collected. Propensity scores were generated based on age, gender, tumor number, size, and pathology, and were used to match HAE and Y90 patients at a 3:1 ratio using a nearest neighbor algorithm. Treatment response was assessed using contrast enhanced CT or MRI according to RECIST criteria. Overall survival (OS), progression-free survival (PFS), and time to local progression (TTLP) were compared using Kaplan-Meier analysis. Hepatobiliary and hematologic toxicities were classified per NCI CTCAE v5.0.
Results:
Before PSM, Y90 patients were older than HAE patients (mean age: 76 vs 62 years; p = 0.00001). After matching (n = 36 in HAE group, n = 12 in Y90 group), no significant differences in baseline characteristics remained (mean age: 76 vs 72 years; p = 0.07). The Y90 group had significantly longer median TTLP (not reached vs 5 months; p = 0.0055) and a trend toward higher disease control rates (DCR) at 3 months (100% vs 68%, p = 0.06) and 6 months (78% vs 37%, p = 0.07). No significant differences were observed in OS (7 vs 11 months; p = 0.99) or PFS (2 vs 1 month; p = 0.89). No grade 3 or higher hepatobiliary toxicity was noted 1-month post-procedure. However, grade 3-4 toxicity of lymphocytopenia were observed in both Y90 and HAE groups at 1-month post-procedure (3/12 vs 3/36, p = 0.15) and at 3 months post-procedure (2/12 vs 3/36, p = 0.58). The decrease in lymphocyte counts was significantly greater in the Y90 group compared to the HAE group at both 1 month (mean absolute lymphocyte count, -650 vs +150/ µL, p < 0.0001) and 3 months (-400 vs +200/ µL, p = 0.0005) post-treatment.
Conclusion:
Y90 offers superior local control compared to HAE in patients with hepatic metastatic melanoma, though it is associated with a more significant decrease in lymphocyte counts, an important factor when combining Y90 with immunotherapy.