Purpose: Adoptive cell transfer (ACT) using autologous tumor-infiltrating lymphocytes (TIL) was reported to yield objective responses in about 50% of metastatic patients with melanoma. Here, we present the intent-to-treat analysis of TIL ACT and analyze parameters predictive to response as well as the impact of other immunotherapies.Experimental Design: Eighty patients with stage IV melanoma were enrolled, of which 57 were treated with unselected/young TIL and high-dose interleukin-2 (IL-2) following nonmyeloablative lymphodepleting conditioning.Results: TIL cultures were established from 72 of 80 enrolled patients. Altogether 23 patients were withdrawn from the study mainly due to clinical deterioration during TIL preparation. The overall response rate and median survival was 29% and 9.8 months for enrolled patients and 40% and 15.2 months for treated patients. Five patients achieved complete and 18 partial remission. All complete responders are on unmaintained remission after a median follow-up of 28 months and the 3-year survival of responding patients was 78%. Multivariate analysis revealed blood lactate-dehydrogenase levels, gender, days of TIL in culture, and the total number of infused CD8 þ cells as independent predictive markers for clinical outcome.Thirty-two patients received the CTLA-4-blocking antibody ipilimumab prior or post TIL infusion. Retrospective analysis revealed that nonresponders to ipilimumab or IL-2 based therapy had the same overall response rate to ACT as other patients receiving TIL. No additional toxicities to TIL therapy occurred following ipilimumab treatment. Conclusion: Adoptive transfer of TIL can yield durable and complete responses in patients with refractory melanoma, even when other immunotherapies have failed.

Background Immunotherapy has changed the therapeutic landscape in oncology. Advanced uterine leiomyosarcoma (ULMS) remains an incurable disease in most cases, and despite new drug approvals, improvements in overall survival have been modest at best. The goal of this study was to evaluate programmed-death 1 (PD-1) inhibition with nivolumab in this patient population Methods This single-center phase 2 trial completed enrollment between May and October 2015. Patients received 3 mg/kg of intravenous nivolumab on day 1 of each 2-week cycle until disease progression or unacceptable toxicity. The primary endpoint was objective response rate. We assessed PD-1, PD-ligand 1 (PD-L1), and PD-L2 expression in archival tumor samples and variations in immune-phenotyping of circulating immune cells during treatment Results Twelve patients were enrolled in the first stage of the 2-stage design. A median of 5 (range, 2-6) 2-week cycles of nivolumab were administered. Of the 12 patients, none responded to treatment. The overall median progression-free survival was 1.8 months (95% confidence interval, 0.8-unknown). The study did not open the second stage due to lack of benefit as defined by the statistical plan. Archival samples were available for 83% of patients. PD-1 (>3% of cells), PD-L1, and PD-L2 (>5% and >10% of tumor cells, respectively) expression were observed in 20%, 20%, and 90% of samples, respectively. No significant differences were observed between pre- and posttreatment cell phenotypes. Conclusion Single-agent nivolumab did not demonstrate a benefit in this cohort of previously treated advanced ULMS patients. Further biomarker-driven approaches and studies evaluating combined immune checkpoint-modulators should be considered.