Allogeneic Virus-specific T Cell Lines (VSTs)

Overview

The primary purpose of the study is to evaluate whether most closely HLA-matched multivirus-specific T cell lines obtained from a bank of allogeneic virus-specific T cell lines (VSTs) have antiviral activity against three viruses: EBV, CMV and adenovirus. Reconstitution of anti-viral immunity by donor-derived VSTs has shown promise in preventing and treating infections associated with CMV, EBV and adenovirus post-transplant. However, the time taken to prepare patient-specific products and lack of virus-specific memory T cells in cord blood and seronegative donors, limits their value. An alternative is to use banked partially HLA-matched allogeneic VSTs. A prior phase II study at our institution using trivirus-specific VSTs generated using monocytes and EBV-transformed B cells gene-modified with a clinical grade adenoviral vector expressing CMV-pp65 to activate and expand specific T cells showed the feasibility, safety and activity of this approach for the treatment of refractory CMV, EBV and Adenovirus infections. However, the production process was lengthy, requiring 8-12 weeks, with exposure to biohazards (B95.8 EBV viral strain and adenovector), while antigenic competition between different viral components precluded increasing the spectrum of specificity beyond these three viruses. Investigator have overcome these limitations and in the current trial, they will evaluate whether rapidly generated, allogeneic most closely HLA-matched multivirus-specific VSTs, activated using overlapping peptide libraries spanning immunogenic antigens from CMV, adenovirus and EBV will be safe and produce anti-viral effects in allogeneic HSCT recipients infected with one of more of the targeted viruses that are persistent despite conventional anti-viral therapy. The study agent will be assessed for safety (stopping rules defined) and antiviral activity.

Investigators evaluated the clinical utility of Multivirus VSTs in recipients of matched related, matched unrelated, or haploidentical donor transplants. To date, 10 clinical-grade multivirus-directed VSTs have been generated from donor PBMCs. These lines were polyclonal, comprising both CD4+ (57±5%) and CD8+ (35±5%) cells and retained expression of the memory markers CD45RO+CD62L+ (58±8%). Their specificity was dependent on the prior viral exposure of the cell donor; 7/8 tested lines had activity against Adv,8/8 against CMV, 6/8 against EBV. None of the lines reacted against recipient PHA blasts - indicating lack of alloreactive potential in these rapidly generated lines. Investigators administered these multivirus-specific donor-derived VSTs to 3 allogeneic HSCT recipients in a dose escalation study all on DL1 (5x106/m2). There were no immediate infusion toxicities, and no de novo acute GvHD, demonstrating the in vivo safety of these mVST. Further, antiviral efficacy has been observed in 1 patient with refractory CMV. In addition, in a recently published paper from Baylor College of Medicien (Anapoulous et al, STM 2014) they have treated VSTs manufactured with a similar methodology, but targeting 5 viruses instead of 3 with specificity to BK virus and HHV6. In that study 10 patients were treated with 4 on DL1 (5x10e6/m2), 4 on DL2 (1x10e7/m2) and 2 on DL3 (2x10e7/m2) and again saw no immediate infusion toxicities, and no de novo acute GvHD, demonstrating the in vivo safety of these mVST. Three patients received the cells as viral prophylaxis (days 38-43 post-HSCT) and none developed viral infections at up to 3 months post-treatment. The other 7 patients received the cells as treatment for one or more active infections between days 59-139 post-HSCT. Based on viral load measurements by day 42 post-infusion, the VSTs were successful in controlling active infections with CMV (1 complete (CR) and 1 partial response (PR)), EBV (2 CRs, including a case of frank PTLD); Adv (1 CR); HHV6 (1 CR); and BK (3 CR, 1 PR, 1NR). Of note, 3 BK virus responders had tissue disease with severe hemorrhagic cystitis and all had marked improvement or disappearance of hematuria following infusion. One patient subsequently had an episode of transient but severe bladder pain in association with inflammation seen on cytoscopy coincident with a 6 log fall in urine BK viral load. Only non-responder was a patient with BK infection whose line lacked activity for this virus, likely reflecting the serostatus of the donor. In addition, 3 patients subsequently reactivated other viruses than those for which they were initially treated, but all cleared these infections by week 12, without the requirement for additional cell infusions (CMV: 1CR; EBV: 1CR; BK: 1CR; HHV6: 1CR). Finally, 1 patient received multivirus specific VSTs under a single patient protocol as an emergency treatment for widespread and bulky rituximab-resistant EBV-PTLD. Post VST treatment there was an immediate decline in the patient's EBV viral load with complete and sustained resolution of PTLD, coincident with an increase in circulating EBV-specific T cells. However, the profound anti-tumor activity mediated by the rapidly-expanding EBV-directed T cells also produced a transient systemic inflammatory response syndrome, which was controlled with steroids and anti-TNFR antibody, with no long term adverse effects. Thus, infusion of donor-derived, multivirus specific VSTs generated with clinical grade pepmixes and infused either prophylactically or as treatment for one or more viral infections has been safe and is associated with the appearance of virus-reactive T cells in peripheral blood that have been able to control infection with above mentioned viruses.