Lilliam Ribeiro

3040 Background: Cytokine release syndrome (CRS) is observed with adoptive T-cell therapies (ACT) with varying frequency. CRS can be life threatening and may require steroids, which hinder the effectiveness of ACT. CRS prevalence and severity within the context NY-ESO-1C259T across 5 studies in 4 tumor types were examined: synovial sarcoma (SS), ovarian cancer (OC), myeloma and melanoma. Methods: A review of CRS including evaluation of concurrent AEs and reported symptoms, cytokine levels and CRP was performed in pts treated with NY-ESO-1C259T. AEs that are known manifestations of CRS were summarized for each pt. diagnosed with CRS. CRP was measured by the hospital laboratory. Serum cytokines were measured by Luminex. Results: Of 53 pts treated with NY-ESO-1C259T, 7 were diagnosed with CRS (Table). 2/7 pts received immunomodulatory treatment for mitigation of CRS: methylprednisolone 2 mg/kg x 2 doses with prednisone taper (n=1, G3) and Tocilizumab 4 mg/kg x 1 dose (n=1 G4). Diagnosis of CRS was made within the first 2 wks in all 7 pts and resolved within 32 days after the T-cell infusions. Onset of CRS coincided with T-cell expansion. Elevated IL-6 and IL-8 were observed during CRS. Timing of the increase of IL-6 and IL-8 suggests involvement in the effector phase of CRS instead of its induction. IL-5, IL-15, INF-γ and GM-CSF showed no discernable pattern associated with CRS. Conclusions: CRS was observed in 13% of pts treated with NY-ESO-1C259T. Symptoms manifest in the first 2 wks and have been effectively managed with supportive care measures. Two pts required immunomodulatory therapy. While there are differences in the pt. populations, incidence of CRS with NY-ESO-1C259T appears to be of lower frequency and severity than reported with CD19 CAR-T therapy. Data summary. Trial/Patient SS 205 SS 207 SS 208 SS 261 SS 264 OC 101 OC 203 Grade CRS/duration (days) 2/3 1/8 3/24 3/11 4/11 3/10 2/7 Responder Y N Y N Y N × Fever/Rash/Diarrhea Y/Y/Y Y/NR/Y Y/Y/Y Y/NR/Y Y/Y/NR Y/Y/Y Y/Y/NR CNS symptoms NR Y Y NR NR Y Y Dyspnea/hypoxia NR NR NR Y Y Y NR Hypotension Y NR NR NR Y Y Y Fatigue N N Y Y NR Y NR CRP (mg/L) 261 157 198 175 182 ND 143 IL-6 peak/IL-8 (pg/mL) 385/129 114/671 72/226 ND 987/394 80/150 ND × too early to assess response; NR – not reported; ND – not done

Abstract Background Despite recent therapeutic advances, multiple myeloma (MM) remains primarily an incurable cancer. Patients experiencing rapid recovery of T cells post autologous stem cell transplant (auto-SCT) may have improved outcomes, and spontaneous cellular responses to tumor can occur, suggesting immune mediated control of tumor is possible. We and others have investigated therapeutic cancer vaccines that have shown promise in pilot studies, in particular following post-transplant infusion of activated autologous T cells. However, efficacy of these approaches may be limited by thymic selection which restricts the repertoire of T cell receptors (TCRs) to low affinity TCRs that cannot recognize the low level of antigen present on most tumor cells. We hypothesized that incorporation of affinity-enhanced tumor antigen-specific TCRs into autologous T cells infused post-transplant would overcome this limitation and improve response rates in the post auto-SCT setting. Methods We report interim results of a Phase II clinical trial (NCT01352286) to evaluate the safety and activity of autologous T cells genetically engineered to express an affinity-enhanced TCR that recognizes the NY-ESO-1/LAGE-1 peptide complex HLA‐A*0201‐SLLMWITQC; these cells are infused in the setting of profound lymphodepletion that accompanies high dose chemotherapy administered during auto-SCT. Patients with high risk or relapsed MM, who are HLA‐A*0201 positive, and whose tumor is positive for NY-ESO-1 and/or LAGE-1 by RT-PCR are eligible. CD25 depleted CD4 and CD8 T cells are activated and expanded using anti-CD3/CD28 antibody conjugated microbeads, and genetically modified with a lentiviral vector containing the TCR construct at a multiplicity of infection of 1. Engineered T cells are administered four days after high dose melphalan and two days following auto-SCT, at a dose range of 1-10 billion total cells with a minimum gene modification requirement of 10%. Patients are evaluated for MM responses in accordance with the IMWG criteria at 6 weeks, and 3 and 6 months post infusion. At 3 months, patients start lenalidomide maintenance. The initial 6 patient phase is complete and a 20 patient extension phase is ongoing. Results Prior to enrollment on study, patients had received a median of 3 prior therapies including 6 with prior transplant. 50% of tumors contained high risk chromosomal abnormalities, and NY-ESO-1 expression is correlated with adverse prognosis. 20 patients (average age of 57) have been infused with an average of 2.3 X 109 engineered T cells (range 4.5 X 108-3.9 X 109); this reflects an average clinical scale transduction efficiency of 34% (range 18% – 49%). Infusions have been well tolerated, and the majority of adverse events were related to the high dose melphalan. Possibly related SAEs were neutropenia and thrombocytopenia, and GI and metabolism disorders including diarrhea, colitis, hyponatremia and hypomagnesemia. 10, 4, 2, and 2 patients have reached the 1 year, 9 month, 6 month and 3 month assessment timepoints, respectively, and 17/20 patients are alive. Best response by day 100 is sCR/CR in 2/15 (13%), nCR in 10/15 (67%), and PR in 3/15% (20%), which compares favorably to historic responses in patients undergoing first or second transplant. Engineered T cells expanded and persisted in blood and marrow at 180 days by Q-PCR and flow-cytometry in all but one case (Figure). 7 patients progressed after day 100, which was accompanied either by loss of engineered T cells or loss of tumor antigen. Detailed phenotyping and functional analysis of engineered T cells, and correlates with clinical responses, is underway. Summary This is the first clinical evaluation of engineered T cells in the MM setting. Infusions are safe, well tolerated, and are associated with encouraging responses in a high risk myeloma population. A study evaluating the engineered T cells in a non-transplant study is underway. Disclosures: Stadtmauer: Celgene: Consultancy. Binder-Scholl:Adaptimmune: Employment. Smethurst:Adaptimmune: Employment. Brewer:Adaptimmune: Employment. Bennett:Adaptimmune: Employment. Gerry:Adaptimmune: Employment. Pumphrey:Adaptimmune: Employment. Tayton-Martin:Adaptimmune: Employment. Ribeiro:Adaptimmune: Employment. Levine:Novartis: cell and gene therapy IP, cell and gene therapy IP Patents & Royalties. Jakobsen:Adaptimmune: Employment. Kalos:Novartis corporation: CART19 technology, CART19 technology Patents & Royalties; Adaptive biotechnologies: Member scientific advisory board , Member scientific advisory board Other. June:Novartis: Patents & Royalties, Research Funding.

Meeting abstracts We report on a 26 patient phase I/II clinical trial ([NCT01352286][1]) to investigate the safety, feasibility and anti-tumor activity of T cells engineered to express an affinity-enhanced TCR that recognizes the NY-ESO-1/LAGE-1 peptide complex HLA-A*0201-SLLMWITQC. Patients with

Abstract Adoptive immunotherapy for cancer has been limited by a lack of antigen specificity, low levels of target expression, and failure to break self-tolerance. We hypothesized that infusion of genetically modified tumor-specific T cells following autologous stem cell transplant (ASCT) may overcome these barriers for multiple myeloma (MM). To test this, we conducted a phase I/II clinical trial (NCT01352286) in which T cells engineered with an HLA-A*0201 restricted, affinity-enhanced TCR recognizing NY-ESO-1 / LAGE-1 peptides (NY-ESOc259-T), were infused in the setting of profound lymphodepletion that accompanies high-dose chemotherapy given with ASCT. HLA-A*0201 MM patients eligible for ASCT, with antigen positive tumor were enrolled. NY-ESOc259-T was manufactured in a 10 day process using anti-CD3/CD28 microbeads and lentiviral vector, and was administered two days following ASCT. IMWG criteria were used to assess response at day 100 with the addition of a near complete response category (nCR) due to the common occurrence of oligoclonal banding observed following rapid post-ASCT immune reconstitution. Blood and marrow samples were taken at multiple timepoints for serum cytokine analysis, NY-ESOc259-T persistence and trafficking, multiparameter flow analysis to examine the phenotype and function of NY-ESOc259-T, and tumor biomarker analysis. 25 of 29 enrolled patients were infused. A mean of 2.8 × 109 engineered cells were administered (range 8.3 × 108-4.2 × 109), and the average transduction efficiency was 33% (range 30%-45%). Patients tended to have advanced disease (64% chromosomal abnormalities, and 24% prior ASCT). At 3 months, 67% (16/24) and 58% (14/24) of patients were in VGPR and nCR or better, respectively. Infusions were well-tolerated and no cytokine release syndrome was reported. NY-ESOc259-T persisted at 6 months in all but one patient, and in a subset of patients at 2 years; marrow infiltration was consistently observed from day 7 through day 180. NY-ESOc259-T initially displayed a dominant activated effector phenotype which converted towards a dominant effector memory phenotype by 1 year post infusion, in a pattern that mirrored clinical responses. Persisting cells demonstrated a polyfunctional response (IFN-γ and TNF-α) with a cytotoxic (CD107a and granzyme B) signature without overexpression of exhaustion markers (PD-1, LAG-3, and TIM-3). Tumor biomarker analysis is ongoing. MM relapse occurred in 13/25 patients. This data show that NY-ESOc259-T cells exhibit robust trafficking and expansion, durable persistence without exhaustion, and follow a natural immune expansion and contraction pattern consistent with an antigen-driven mechanism of action. Relapse correlated with a loss of persistence or tumor antigen escape, suggesting that targeting multiple antigens and maintenance infusions may increase durable remissions. Citation Format: Aaron Rapoport, Edward Stadtmauer, Luca Melchiori, Ryan Wong, Eduardo Davila, Gwendolyn Binder-Scholl, Tom Holdich, Dan Vogl, Brendan Weiss, Jeffrey Finkelstein, Simon Lacey, Sarah Bond, Marylene Fortin, Yoav Peretz, Joanna Brewer, Alan Bennett, Andrew Gerry, Nick Pumphrey, Helen Tayton-Martin, Lilliam Ribeiro, Ashraf Badros, Saul Yanovich, Nancy Hardy, Jean Yared, Naseem Kerr, Sunita Philip, Sandra Wesphal, Bruce L. Levine, Carl June, Michael Kalos, Bent Jakobsen. NY-ESO T cells administered post ASCT for MM exhibit extended functionality without exhaustion in a natural pattern of effector and memory programming. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4701. doi:10.1158/1538-7445.AM2015-4701