Search

For patients lacking a matched donor, transplant (HCT) from umbilical cord or T cell depleted (TCD) grafts, is potentially curative. However, these transplants are associated with high morbidity and mortality from reactivation of latent viruses. Adoptive therapy with CMV and EBV specific T cells expanded in vitro (CMV-CTLs and EBV-CTLs) is increasingly available to clear drug resistant CMV infections and induce durable remissions of refractory EBV+ lymphomas. Here we present studies that help elucidate mechanisms of T cell control of CMV and EBV in both the primary HCT and the adoptive T cell settings and provide rationale for pre-defined characterization of T cells generated for adoptive therapy. In a cohort (N=39) of CMV seropositive recipients of mismatched TCD-HCT, we established the HLA-restriction of the donor's endogenous CMV response. To evaluate instances where viral specific T cells failed to induce responses, we compared in vitro cytotoxicity of CMV-CTLs and EBV-CTLs against targets presenting CMVpp65 15-mers or lab strain EBV B95.8 to patient derived CMV or EBV targets.. In 15 of 39 recipients at risk for reactivation of CMV, the donor CMV response was restricted by an un-shared HLA allele.. Furthermore, 53% of these recipients developed persistent CMV viremia or invasive disease compared to none of those whosedonors CMV response was restricted by a shared HLA allele Our analyses show several mechanisms of immune escape leading to for the failure of CMV-CTLs or EBV-CTLs to recognize patient derived targets 1) deficient production of the viral protein (eg EBNA3B) 2) mutations of the viral epitopes (eg EBNA3B presented by HLA A1101) or 3) impaired processing or presentation of these epitopes (eg IPSI peptide of CMVpp65 presented by HLA B3501). In such cases, selection of 3rd party CTLs restricted by a different HLA allele (i.e. restriction switch) has induced remissions of EBV+ lymphomas in 60% of patients. In HLA disparate HCT recipients, donor-derived virus-specific T-cells may be ineffective if the HLA restriction is not shared between the T-cells and the recipient. Even in the 3rd-party setting, where there is appropriate selection of shared HLA restrictions, clinical responses may be impacted by immune escape. Our banks of 3rd party virus-specific T-cells characterized by HLA restrictions permit accurate selection of T-cells. Importantly, restriction switch can overcome viral mechanisms of immune evasion. [ABSTRACT FROM AUTHOR]

EBV PTLD is often a fatal disease. Tabelecleucel is an investigational off-theshelf, allogeneic T-cell immunotherapy specific for EBV antigens. Functional in vivo expansion approximated by a limiting dilution analysis of EBV-CTLp has previously been correlated to clinical response. We conducted analyses of reported peak circulating EBV-CTLp levels following tabelecleucel administration in correlation to response and safety across two studies (N = 42) in rituximab refractory or relapsing EBV PTLD following HCT or SOT. Two cohorts were defined by the lowest EBV-CTLp quartile (n = 11) and the combined upper 3 quartiles (n = 31). Twenty-five of 31 subjects (80.6%) in upper quartiles exhibited a clinical response compared to 3 of 11 subjects (27.3%) in the lowest quartile. Eighteen of 20 complete responses and 7 of 8 partial responses demonstrated peak CTLp values in the upper quartiles of the population. Overall survival (OS) rate in subjects exhibiting CTLp values in the upper quartiles was 83.9% (95% CI: 65.5-92.9%) at 1 year and 66.1% (95% CI: 45.9-80.2%) at 2 years, compared to 18.2% (95% CI: 2.9-44.2%) in the lowest quartile over the same periods. OS is commensurate with a hazard ratio of 0.168 in favor of subjects with higher EBV-CTLp (95% CI: 0.067-0.425; Log-Rank Test p-value < 0.001). Subjects demonstrating clinical responses experienced median CTLp values of 36.2 CTLp/ml (15.6-155 interquartile range), whereas non-responders exhibited significantly lower (p = 0.001, based on Wilcoxon Rank Sum test) median CTLp of 0.07/ml (0.01-5.5 interquartile range). Tabelecleucell was well tolerated across both studies, with a safety profile in each group consistent with the underlying clinical condition of patients These analyses illustrate the correlation of peak EBV-CTLp to clinical benefit in subjects receiving tabelecleucell. Significant correlation of EBV-CTLp with ORR and OS are consistent with potential use as a functional biomarker for the expansion of EBV specific T-cells in the treatment of EBV PTLD in the rituximab refractory/relapsing setting. [ABSTRACT FROM AUTHOR]

Adoptive therapy with 3rd party virus-specific T cells has demonstrated efficacy in the treatment of refractory infection complicating hematopoietic stem cell (HCT) and solid organ (SOT) transplant. A striking feature of responses observed at our center has been their durability. We have previously demonstrated that expansion of the infused viral specific T cells lasting for 3-6 wks is observed in patients (pts) responding to transplant donor-derived or 3rd party T cells. The durability of response to HCT donor-derived T cells is expected; however, the durability of response to 3rd party T cells, which are expected to be rejected, is unexplained. We have evaluated pts responding to 3rd party EBV-specific T cells (tabelecleucel) and CMV CTLs for the durability of response through 12-mo and whether the durability of response correlates with overall survival (OS) and depends on persistence of 3rd party viral specific T cells. We evaluated HCT pts treated for CMV viremia and/or disease who responded to 3rd party CMV-CTLs (N=32) and HCT and SOT pts with rituximab refractory EBV+ PTLD who responded to tabelecleucel (N=29). To detect in vivo persistence of 3rd party T cells we sorted viral specific T cells and performed analysis by short tandem repeats (STRs). This method is especially well suited to detection of even rare CMV specific T cells where the epitope recognized by the infused population has been defined. Clinical responses were durable at 1 year in 57/61 pts. This durability correlated to improved OS. Pts responding to CMV-CTL therapy had 12-mo OS of 80.3%. In contrast 12-mo OS for pts not responding to CMV CTLs was 33.3%. Pts responding to tabelecleucel had 12-mo OS of 93.1%. with 12-mo OS of 12.5% in non-responders. We demonstrated three patterns of persistence of CMV-CTLs in HCT pts ranging from very transient detection up to detection for 9 mos after infusion. Our studies provide evidence that, while initial response can be ascribed to the function and expansion of the T cells administered, the durability of response in HCT pts does not depend on detectable 3rd party cells. [ABSTRACT FROM AUTHOR]

Our program has used 2 T-cell depleted HSCT approaches for patients with FA for transplants from donors other than HLA matched siblings: 1) the Isolex 300i Magnetic Cell Selection System (Baxter) + sheep RBC depletion (E-) (TCD1) was used to achieve an approximate 5 log 10 depletion of CD3+ cells, and (2) the CliniMACS device (Miltenyi; TCD2) was used as a sole TCD method with an approximate 4.5 log 10 depletion of CD3+cells. We retrospectively reviewed charts of consecutive patients with FA who received a first HCT from alternative donors. We identified 42 patients with aplastic anemia, MDS or AML, 24 who received TCD1 (1999-2009) and 18 who received TCD2 (2010-present). Most patients in TCD1 received a TBI-based conditioning, while most in TCD2 received busulfan-based conditioning. Donors were matched unrelated, mismatched unrelated, or mismatched related. Because of the differences in the treatment periods, conditioning regimens, and disease breakdown, we analyzed the data specifically related to the T-cell depletion. Distributions of continuous variables were compared using a Mann-Whitney-Wilcoxon test, while binary and categorical variables were compared using a Fisher's exact test. Grafts were G-CSF mobilized peripheral blood progenitor cells. T-cell depletions were performed as described. Table 1 provides demographics, cell doses and hematologic recovery data for TCD1 and TCD2. Median CD34 cell doses of the grafts did not differ significantly while median CD3 dose did (N=0.048). All but one patient engrafted. Two patients receiving grafts from mismatched unrelated donors developed grade 2-4 acute GvHD in TCD1, none in TCD2. Neutrophil engraftment did not significantly differ while platelet engraftment was more rapid with TCD1 (N=0.0045). CD4 recovery (Abs CD4 >200) and PHA recovery (50th%ile) differences were comparable (NS). In conclusion, no difference were noted between the Isolex E- T-cell depletion and the CliniMACS CD34 selection system, except for a more rapid platelet reconstitution for the CliniMACS column. There was no difference in risks of graft rejection and GvHD, and immune reconstitution was comparable in both cohorts. Future plans include the comparison of these methods to the TCR alpha beta depletion in this patient population. [ABSTRACT FROM AUTHOR]

Virus-specific T-cells generated from transplant or 3rd party donors can induce durable remissions of post-transplant severe infections or EBV lymphomas. T-cells simultaneously sensitized with antigens from multiple viruses have also shown promise. We hypothesized that in any individual donor, immunogenic peptides from different viruses might elicit T-cell responses restricted by different HLA alleles. In HLA non-identical patients, the T-cells, specific for one virus or tumor antigen will be effective only if they are restricted by an HLA allele shared by the patient. We analyzed HLA restrictions (HLA-R) of 42 CTL lines generated by dual sensitization with either autologous EBV-transformed B-cells (BLCL) loaded with a pool of overlapping 15-mer peptides spanning the sequence of CMVpp65 (BLCL/CMV CTL) (n=20) or autologous BLCL loaded with a pool of 15-mers spanning the oncofetal protein WT-1 (BLCL/WT1 CTL) (n=22). EBV CTL lines were generated by sensitization with BLCL alone. The HLA-R of the CMVpp65-specific T-cells(CMV-T), WT1 specific T cells (WT1-T) and EBV specific T cells (EBV-T) in each T cell product were assessed by their cytotoxic activity against a panel of Cr51 labeled peptide-loaded dendritic cells and BLCL sharing a single HLA allele with the T cell donor. In 13/20 BLCL/CMV CTL(65%) and 17/22 BLCL/WT1 CTL(77%) the CMV-T or WT1-T were restricted by single HLA alleles. In 10/20(50%) BLCL/CMV CTL, CMV-T and EBV-T in the same line were restricted by the same HLA allele. However, in the other 10(50%) the CMV-T were restricted by an HLA allele different from that of the EBV-T. Similarly, in 13/22(59%) BLCL/WT1 CTL, WT1-T and EBV-T in the same line were restricted by different HLA alleles. In 2/4 BLCL/CMV CTL and 2/5 BLCL/WT1 CTL, in which the HLA-R of CMV-T or WT1-T differed from that of EBV-T in the same culture, the HLA allele differentially presenting CMV or WT1 antigen but not an EBV antigen was a prominent restricting allele of EBV-T in the EBV CTL of the same origin. This suggests that while immunodominant EBV and CMV epitopes presented by certain prevalent HLA alleles such as HLA B0702 or HLA A0201 often induce responses to both viruses, in certain dual-sensitized cultures, immunogenic epitopes from different antigens may compete for presentation by a specific HLA allele or differ in their potential to induce a dominant T-cell response. [ABSTRACT FROM AUTHOR]

Abstract: Manifestations of and risk factors for graft-versus-host disease (GVHD) after double-unit cord blood transplantation (DCBT) are not firmly established. We evaluated 115 DCBT recipients (median age, 37 years) who underwent transplantation for hematologic malignancies with myeloablative or nonmyeloablative conditioning and calcineurin inhibitor/mycophenolate mofetil immunosuppression. Incidence of day 180 grades II to IV and III to IV acute GVHD (aGVHD) were 53% (95% confidence interval, 44 to 62) and 23% (95% confidence interval, 15 to 31), respectively, with a median onset of 40 days (range, 14 to 169). Eighty percent of patients with grades II to IV aGVHD had gut involvement, and 79% and 85% had day 28 treatment responses to systemic corticosteroids or budesonide, respectively. Of 89 engrafted patients cancer-free at day 100, 54% subsequently had active GVHD, with 79% of those affected having persistent or recurrent aGVHD or overlap syndrome. Late GVHD in the form of classic chronic GVHD was uncommon. Notably, grades III to IV aGVHD incidence was lower if the engrafting unit human leukocyte antigen (HLA)-A, -B, -DRB1 allele match was >4/6 to the recipient (hazard ratio, 0.385; P = .031), whereas engrafting unit infused nucleated cell dose and unit-to-unit HLA match were not significant. GVHD after DCBT was common in our study, predominantly affected the gut, and had a high therapy response, and late GVHD frequently had acute features. Our findings support the consideration of HLA- A,-B,-DRB1 allele donor–recipient (but not unit–unit) HLA match in unit selection, a practice change in the field. Moreover, new prophylaxis strategies that target the gastrointestinal tract are needed. [Copyright &y& Elsevier]