Search

We retrospectively analysed overall survival (OS) and potential predictive biomarkers of OS in patients with metastatic melanoma treated with ipilimumab plus nivolumab in a single institution. Electronic medical records of patients with advanced melanoma receiving ≥ 1 dose of a combined ipilimumab plus nivolumab regimen between March 3, 2016 and March 7, 2020 in a single institution, were reviewed. OS was analysed using the Kaplan–Meier method. Sub-group analyses were conducted to examine several endpoints according to relevant clinical, molecular and pathological variables using logistic and Cox models. Forty-four cases were reviewed, 38 (86.4%), of whom had cutaneous melanoma, 21 (47.7%) were BRAF mutant, 21 (47.7%) presented high lactate dehydrogenase (LDH) values, 23 (52.3%) had ≥ 3 disease sites, and 10 (22.7%) patients had brain metastases. The median follow-up was 37.7 months, and the median OS was 21.1 months (95% CI 8.2–NR). In the multivariate analysis, the OS was significantly longer in patients with an Eastern Cooperative Oncology Group (ECOG) score of 0, LDH ≤ upper limit of normal, absence of liver metastases and neutrophil-to-lymphocyte ratio (NLR)

Background Immune checkpoint combination therapy with nivolumab and Ipilimumab for untreated metastatic melanoma has shown higher rates of response rates, progression free survival and overall survival compared to monotherapy (Long et al. ESMO 2019), including patients with brain metastasis, at a cost of higher toxicity. Several clinical factors had been identified as prognostic of response to immunotherapy, such as LDH, dLNR and toxicity. Methods We analysed all patients with melanoma treated with ipilimumab and nivolumab between March 2017 and September 2019 institution. Statistical analysis was preformed with IBM SPSS Statistics 24.0. Results 42 patients (p) treated with ipilimumab and nivolumab were included. 22 (52.4%) men, 38 (90%) ECOG 0-1 (%). Stages at treatment: IVA (n = 4), IVB (n = 11), IVC (n = 17), IVD (n = 9). 20p (48%) presented a positive BRAF mutation. 10p (24%) had a previous adjuvant treatment. LDH: N (n = 25, 60%), 1.2xULN (n = 10, 24%). Cycles completed: 1 (n = 4), 2 (n = 16), 3 (n = 7), 4 (n = 12). Best Overall Response (BOR): Progressive disease (n = 22, 52%), stable disease (n = 4, 10%), partial response (n = 10, 24%), complete response (n = 6, 14%). Immunorelated adverse events (irAEs): Yes (n = 27, 64,3%), no (n = 15, 35,7%). Hepatitis 15, colitis 5, hypophysitis 5, thyroiditis 1, Nephritis 1. Grades: 1 (n = 2), 2 (n = 14), 3 (n = 8), 4 (n = 3), Number of CNS lesions: 1 (n = 0), 2 (n = 2), 3 (n = 0) and >3 (n = 8). 9 patients presented CNS disease, 4 of them had symptoms requiring steroid treatment. OS from the whole population was 34.7 months (m) (22.6-46.8, IC 95%). Independent predictors of OS where ECOG (p = 0.1, IC 95%), no previous treatment (p = 0.07, IC 95%), LDH (p = 0.02, IC 95%), irAES (p = 0.05), No 25.6m (8,9-42,4), Yes 39.1m (26.1-52.2), IC 95%). BRAF status, cycles, number of extracranial lesions, number of intracranial lesions and dNLR were not found to be predictors of OS. A significant association was found between BOR and irAEs (p = 0.009), LDH (p = 0.018), ECOG (p = 0.027). Conclusion In our cohort, LDH, ECOG and irAEs and no previous treatment were found predictors of response and overall survival. Our median OS and ORR were slighly slower than reported in trials. Legal entity responsible for the study The authors. Funding Has not received any funding. Disclosure A.M. Arance Fernandez: Honoraria (self): Bristol-Myers Squibb. All other authors have declared no conflicts of interest.