Abstract PR11: Enhanced in vivo persistence and therapeutic activity of combined transferred polyclonal melanoma-reactive CTL generated in presence of IL-21 and anti-CTLA4

Monotherapy with a blocking human anti-CTLA4 mAb achieves disease control in 20-28% of patients (pts), but failure to benefit in the remaining patients may reflect inadequate induction and/or expansion of melanoma reactive T cells following CTLA4 blockade. We have been exploring a combined biologic strategy involving the adoptive transfer of melanoma-reactive CD8+ cytotoxic T lymphocytes (CTL) with concurrent administration of anti-CTLA4 as a means to enhance and broaden the anti-tumor effect of the transferred CTL. Autologous melanoma-specific CTL were generated by stimulating CTL lines in the presence of the γc-chain cytokine Interleukin-21 (IL-21), which programs responding antigen-specific cells to retain characteristics of central memory T cells (Tcm). Responding cells were sorted for binding of a peptide-MHC (pMHC) multimer to obtain a purified tumor-reactive polyclonal product. We report on 7 pts with bulky, therapy-resistant melanoma who received 1010/m2 polyclonal IL-21-derived CTL lines targeting MART1. The infusions were preceded by intravenous cyclophosphamide 300 mg/m2 and followed by low-dose subcutaneous IL-2 (250,000 IU/m2 BID) x 14 days to promote in vivo expansion and survival. Anti-CTLA4 (ipilimumab) was administered at 3mg/kg every three weeks for four doses, starting on the day after the CTL infusion. The treatment was well tolerated, with no unexpected or serious (grade ≥3) related toxicities. 5/7 evaluable pts exhibited stable disease (SD) or partial responses (PR) by RECIST 1.1 criteria at 6 weeks, and 4/6 exhibited SD at 12 weeks. One pt who had failed previous ipilimumab monotherapy and prior CTL therapy targeting MART1 without CTLA4 blockade had a noteworthy response to the current protocol. He had SD at 12 weeks, then experienced a delayed objective response with 66% reduction in tumor mass at 28+ weeks. Persistence of the infused CTL at frequencies >2x baseline levels was documented at 6 weeks (6/6 evaluable pts) and 12 weeks (4/4 evaluable pts), whereas such levels of persistence beyond 6 weeks was previously typically observed in only 15-20% of pts in previous studies. The cells infused into each pt expressed moderate/detectable levels of CD28 prior to infusions, but analysis of persisting CTL detected by specific multimer binding in vivo in responding pts revealed progressively higher expression of CD28, consistent with enhanced CD28:CD80/86 interactions in the presence of CTLA4 blockade providing survival advantages. Additi