Outcomes of combined ipilimumab/nivolumab in metastatic uveal melanoma: A prevalence meta-analysis.

Abstract

9534 Background: Uveal melanoma is the most common primary intraocular malignant tumor in adults and, approximately 40-50% of the patients eventually develop metastatic disease. Metastatic uveal melanoma has a dismal prognosis with an overall survival of < 50% at 1-year. Single-agent check point inhibitors revealed minimal benefit and novel approaches are underway to improve the outcomes with the recent approval of Tebentafusp in uveal melanoma. The purpose of this metanalysis was to assess the safety and efficacy of combined immunotherapy with ipilimumab (3 mg/kg x q3w for 4 cycles) and nivolumab (1 mg/kg x q3w for 4 cycles) in metastatic uveal melanoma. Methods: A comprehensive literature search on PubMed, Embase, Cochrane and Web of Science was conducted. Two independent reviewers screened the literature and extracted data. Our search strategy included MeSH terms and key words for metastatic uveal melanoma; ipilimumab and nivolumab. OpenMeta[Analyst] software was used for the analysis. The endpoints included the prevalence of overall response rate (ORR), complete response (CR), ≥grade-III diarrhea/colitis, and ≥grade-III hepatic toxicity. Additional endpoint included the median overall survival (OS) was reported as a range. Random-effects model (DerSimonian-Laird method) was applied. Results: Overall, eight studies (n = 379) were included for the analysis. Five studies were phase II, three studies were retrospective. 52% of the patients were male with good performance status. Median follow-up ranged from 9.2 mo-28 mo. 40% of the patients (n = 142) had elevated LDH at the time of treatment. GNAQ and GNA11 mutations were reported in three studies, BRAF mutation was reported in two studies. The pooled prevalence for ORR was 13.7% (95% CI: 9.2–18.2%, N = 6 studies, n = 33/226) with CR of 2.1% (95% CI: 0.3–3.9%, N = 6 studies, n = 8/226). The median OS ranged from 12.7 to 19.1 months (N = 7 studies). Majority of the patients experienced treatment-related adverse events. Most common side effects included diarrhea, colitis, hepatic toxicities, skin disorders, hypothyroidism. The pooled prevalence for ≥grade III hepatic toxicity was 26.2% (95% CI: 13.9–0.385%, N = 5 studies, n = 60/219). Conclusions: Combined checkpoint blockade with ipilimumab and nivolumab showed an ORR of 13.7% which appears to show better clinical activity than single-agent checkpoint inhibitor. Most common treatment side effect was hepatic toxicity.