ALBERT

Description: Key Points KT64/86 artificial antigen–presenting cells culture stimulation provides marked expansion of Tregs. In the context of sirolimus, mycophenolate mofetil immunosuppression, adoptive transfer of Tregs resulted in low risk of acute GVHD.

Description: Key Points Depletion of host regulatory T cells with IL2DT improves efficacy of haploidentical NK cell therapy for refractory acute myeloid leukemia. Depletion of Treg and persistence of NK cells for ≥7 days after NK cell adoptive transfer predicts beneficial clinical responses.

Description: Abstract 3731 Both T cell and natural killer (NK) cell reconstitution have been shown to affect clinical outcomes after hematopoietic stem cell transplantation (HSCT). Killer immunoglobulin-like receptor (KIR) interactions between alloreactive NK cells and their targets can prevent relapse, but may be dysregulated, especially after T cell replete HSCT. T cell recovery is also affected by the stem cell source and T cell content of the graft. To better understand the effects of various NK and T cell subsets we evaluated lymphocyte recovery in 304 adult patients who received either UCB (n=116), Sib (n=84) or Auto (n=94) HSCT for hematologic malignancies between 2003 and 2010 at the University of Minnesota. Peripheral blood mononuclear cells obtained at 3 months after HSCT were stained with CD56, CD3, CD4, CD8, and a cocktail of anti-NK cell KIR antibodies to determine the relative percentage of lymphocyte subsets by flow cytometry. The absolute lymphocyte count (ALC) was measured and used to calculate the absolute (Abs) number of T and NK cells and their subsets. ALC recovery at 3 months was similar among groups (UCB: 901.9 ± 74.5, Sib 890.2 ± 73.0 and Auto 1076.7± 69.4 cells/ul). Abs NK cells were highest in the UCB cohort (375.4 ± 24.9) vs. Sib (183.8 ± 15.4; p

Description: Allo-HCT is curative for hematological malignancies that are refractory to conventional chemotherapy. However, GVHD, CMV reactivation and relapse are common complications that either impair immune recovery (IR) or are secondary to the profound immune deficiency associated with these complications. We reasoned that patients who lacked these and survived beyond 1 year 'behaved' as if they had reconstituted functionally. To address this hypothesis we analyzed T, B and NK IR in 423 allo-HCT patients with hematologic malignancy without TRM in the first year. Patients who lacked GVHD, CMV reactivation and relapse were assigned to the "successful" IR group (n=115), while those who had any of these were assigned to the "suboptimal" IR group (n=308). Because some "successful" patients were CMV seronegative (CMV-) and thus at low risk for CMV viremia, we further divided this group into "successful-CMV+" (n=50) and "successful-CMV-" (n=65). In terms of 5-year DFS, "successful-CMV+" patients had the best outcome followed by "successful-CMV-" and "suboptimal" patients (90% [95% CI: 76-96%] vs 57% [36-72%] vs 43% [37-49%], p1480/ul) by D+60 as compared to that in "successful-CMV-" and "suboptimal" patients (57% vs. 13% vs. 24%, p=0.0005 and p=0.0025, respectively). This was primarily driven by faster CD3+CD8+ recovery to normal levels (64% vs. 15% vs. 26%, p=0.0004 and p=0.0016, respectively). These findings were durable over one year, with the "successful-CMV+" patients having significantly higher absolute numbers of CD3+CD8+ T cells at nearly every time point compared to the other two groups. Similarly, a significantly higher proportion of patients reached the normal range for B cells in the "successful-CMV+" and "successful-CMV-" groups relative to "suboptimal" IR group. After antigen encounter, naïve T cells differentiate into effector memory (EM) and central memory (CM) cells with some of the EM cells reacquiring CD45RA (EMRA); therefore, we evaluated the three groups for recovery of these T cell subpopulations. "Successful-CMV+" patients had significantly higher numbers of CD4+ and CD8+ EM and EMRA cells at nearly all time points compared to the other groups. Using a multivariate mixed model and taking time since transplant, donor stem cell source and recipient age into consideration, patients in "successful-CMV+" IR group showed higher ALC as well as higher absolute numbers of CD3, CD4EM, CD4EMRA, CD8EM, CD8EMRA, and NKG2C+CD57+ relative to the other group, Table 1). Together, these data define the phenotypic profile of the successful IR with the unexpected findings of better survival in CMV+ seropositive patients who are able to control CMV reactivation (and without GVHD or relapse). These results highlight the intriguing relationship between CMV and the human immune system and further suggest that strategies to control CMV reactivation (drugs, cellular therapy or vaccination) may have an unexpected impact on transplant outcomes. Table 1. Multivariable Regression Analysis on Lymphocyte Recovery Between Patient Groups. Linear mixed model was used on log transformed absolute counts at multiple time points with covariates of group, time since transplant, stem cell source, and recipient age. Regression coefficients reflect the "average" effect over time adjusted for stem cell source and recipient age. Lymphocyte Subsets Successful CMV+ vs Successful CMV- Non-Successful vs Successful CMV + Suboptimal vs Successful CMV- Coefficient P value Coefficient P value Coefficient P value ALC 0.29 0.014 -0.32 0.001 -0.04 0.664 CD19 0.09 0.77 -0.96 0.0001 -0.88 〈 0.0001 CD3 0.47 0.004 -0.38 0.006 0.09 0.454 CD3CD4 0.17 0.269 -0.33 0.014 -0.16 0.169 CD4 Naive -0.01 0.96 -0.25 0.22 -0.26 0.131 CD4CM 0.09 0.59 -0.25 0.099 -0.15 0.228 CD4EMRA 0.84

Description: In rodent graft-versus-host disease (GVHD) models, anti–IL-21 neutralizing mAb treatment ameliorates lethality and is associated with decreases in Th1 cytokine production and gastrointestinal tract injury. GVHD prevention was dependent on the in vivo generation of donor-inducible regulatory T cells (Tregs). To determine whether the IL-21 pathway might be targeted for GVHD prevention, skin and colon samples obtained from patients with no GVHD or grade 2 to 4 GVHD were analyzed for IL-21 protein expression. By immunohistochemistry staining, IL-21 protein-producing cells were present in all gastrointestinal tract samples and 54% of skin samples obtained from GVHD patients but not GVHD-free controls. In a human xenogeneic GVHD model, human IL-21–secreting cells were present in the colon of GVHD recipients and were associated with elevated serum IL-21 levels. A neutralizing anti–human IL-21 mAb given prophylactically significantly reduced GVHD-associated weight loss and mortality, resulting in a concomitant increase in Tregs and a decrease in T cells secreting IFN-γ or granzyme B. Based on these findings, anti–IL-21 mAb could be considered for GVHD prevention in the clinic.

Description: Abstract 513 UCB transplantation (UCBT) is associated with greater risk of graft rejection and, in particular in double UCBT, with graft vs. host disease (GVHD). Our group has shown in mouse models that ex vivo activated and expanded Tregs improve engraftment and reduce and treat GVHD. We here report on the first clinical trial to study the safety of the infusion of UCB Tregs in patients receiving a nonmyeloablative double UCB transplant. We enrolled 19 patients (pts) (age 25-65 yrs; weight 52-133; male 9; CMV seropositive 6) with high risk or advanced leukemia (n=12) and lymphoma/CLL (n=7). All patients received the same nonmyeloablative conditioning consisting of cyclophosphamide 50mg/kg/×1day, fludarabine 40 mg/m2/×5d, and total body irradiation 200 cGy/×1d; immunosuppression was cyclosporine A/mycophenolate mofetil (MMF) in the first 17 and sirolimus (Siro)/MMF in the last 2 pts. Tregs were obtained from a 3rd UCB unit and after CD25+ selection (Miltenyi CliniMACS) were activated and expanded in the presence of anti-CD3/CD28 mAb coated beads and IL-2 for 18±1 days. Treg dose escalation levels were 1, 3, 10, 30 × 105/kg on day +1, and 30 × 105/kg on days +1 and +15 after UCBT. After CD25+ selection the median % CD4+/CD25+ was 62% (range , 11-87%) and following culture was 86% (r, 62-97%). Median fold expansion was 198 (r, 13-1169); median mixed lymphocyte reaction suppression at a 1:4 ratio post-culture was 86% (39-95%). All products achieved lot release. Pts were monitored for 48h post-infusion and no dose limiting or serious adverse event was observed. After infusion an increase in the proportion of peripheral blood comprised of CD4+FoxP3+CD127- cells was observed. Donor Tregs were clearly detected in all pts receiving Tregs that were HLA disparate with the pt and graft donor units (Figure 1). Mixed day 21 donor graft UCB chimerism was observed in 59% (10/19) of pts vs. 35% (38/107) in historical controls. The cumulative incidence of grades II-IV acute GVHD was 47% (95%CI, 24-71) and III-IV 16% (95%CI, 0-33). In pts receiving ≥ 30 × 105/kg (n=14), grades II-IV acute GVHD was 43% (95%CI, 16-70) and III-IV 7% (95%CI, 0-21). Compared to historical controls receiving the same conditioning regimen, the median time to the development of acute GVHD was longer (Figures 2), although not statistically significant. Neutrophil recovery was 95% (95%CI, 79-100) at median of 8.5 days (r, 3-36), platelet recovery 〉50,000/mL was achieved in 74% (95%CI, 53-95) at a median 43 days (r, 27-57), cumulative incidence of opportunistic infections (fungal + viral) at 100 days was 32% (95%CI, 11-53), treatment related mortality at 100 days was 11% (95%CI, 0-25), and disease free-survival was 40% (95%CI, 15-65%). None differed from historical controls. Based on our results, the co-infusion of ex vivo expanded and activated UCB-derived Treg to recipients of nonmyeloablative UCBT 1) is safe at the tested dose levels, 2) leads to detectable increased in Treg-donor unit derived circulating CD4+FoxP3+CD127- cells, and 3) results in an increased the proportion of mixed chimerism at day +21. Identification of the Treg MTD in the context of Siro/MMF immunosuppression will set the stage for the planned definitive studies that will establish the true impact of Tregs on engraftment and GVHD and for future studies in the treatment of autoimmune disease. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 346 Expansion of NK cells after adoptive transfer is a major determinant of their anti-tumor efficacy but the mechanisms of their expansion are not completely understood. To study in vivo NK cell expansion and the relative merits of fresh versus frozen NK cell products, we used xenotransplantation of human GMP NK cell products (provided by two PACT centers funded by the NHLBI) into immune deficient mice. NOD/IL-2Rγc/Rag (NOG) that lack T-, B-, and NK cells and that have a macrophage defect that renders recipients highly amenable to human cell engraftment, were given 250 cGy total body irradiation, and infused with fresh or frozen NK cells. Two products were tested: 1) an enriched fresh NK cell product obtained by CD3 and CD19 depletion followed by overnight activation with IL-2 or IL-15 (U. of Minnesota) or 2) ex vivo expanded NK cells using membrane bound IL-15 and 41BB-L transduced K562 cells with 10 U/ml IL-2 in G-Rex devices for 10 days (Baylor). NK cells (1–2 million NK cells for each individual experiment) were given IV and mice were treated with no cytokines, 5 or 10 mg of IL-2 (Novartis), or 5 mg IL-15 (produced by the NCI for clinical use) as 3 injections per week for 2 weeks. Peripheral blood (Table), spleen and marrow were evaluated on days 7, 14, 21 and 28 after adoptive transfer. Counts were converted to absolute cell numbers per 100 μL of blood or the number of cells recovered per spleen or after flushing of a single femur. On day 7, the number of huNK cells in blood in the absence of cytokine administration was low with both cell products and numbers continued to diminish by day 14. In marked contrast, every other day IL-2 or IL-15 significantly increased huNK cells in murine blood. On day 7 and 14, ex vivo expanded NK cell products resulted in significantly higher numbers of huNK in mice receiving 5μg IL-2 compared to mice who received fresh NK cells activated overnight with IL-2. With 5μg IL-2, huNK number decreased between day 7 and 14 irrespective of cell product. In contrast, only IL-15 lead to increased numbers of huNK cells between days 7 and 14. Previously cryopreserved NK cells showed significantly worse survival for both cell products, with a larger fold decrease seen with ex vivo expanded cells. Blood collected from cytokine treated mice at day 14 after infusion of fresh products contained huNK cells that were fully functional as assessed by potent CD107a degranulation, TNF and IFN production after exposure to K562 target cells as well as augmented IFN production induced by IL-12 and IL-18. On day 8, Ki67+ (proliferating) huNK cells were significantly higher with IL-15 compared to IL-2 (both at 5μg) in marrow for fresh (76% vs 35%, p=

Description: Introduction: Monocyte recovery following allogeneic hematopoietic cell transplantation (allo-HCT) has previously been correlated with improved overall survival (OS). Three distinct monocyte subpopulations have been identified based on CD14 and CD16 surface expression, including: CD14+CD16++ (non-classical), CD14+CD16+ (intermediate), and CD14++CD16- (classical). Among these subpopulations, non-classical monocytes have greater capacity to produce inflammatory cytokines, whereas classical monocytes have phagocytic activity, are less inflammatory and produce counter regulatory cytokines (i.e., IL-10), and intermediate monocytes are a transitionary population sharing features of both non-classical and classical monocytes. To date, there has not been a direct comparison between stem cell source and monocyte recovery, nor has there been an analysis of monocyte subpopulation recovery and HCT outcomes. We hypothesized that recovery of monocytes and their subpopulations would vary based on stem cell source and that recovery of the monocyte subpopulations would be associated with clinical outcomes. Methods: This analysis included individuals who underwent a first allo-HCT at the University of Minnesota between 2010-2014 and who enrolled onto an institutional immune reconstitution protocol. Absolute monocyte count (AMC), as well as the absolute counts of non-classical, intermediate and classical monocyte subpopulations were assessed at days 28, 60, 100 and 365 post-HCT. Optimal cut points of each monocyte subpopulation at day 28 were calculated using the Contal and OQuigley method and were used for analysis. This time point was selected because the outcomes of interest were unlikely to have occurred prior to this time. Individuals who experienced acute graft versus host disease (aGVHD) prior to day 28 were excluded from the aGVHD analysis (n=39). Patient demographics, disease and treatment characteristics were included in univariable analyses to determine associations between monocyte subpopulations and transplant outcomes, including 2-year OS, 1-year transplant related mortality (TRM), 2-year disease free survival (DFS), 2-year relapse, 180-day aGVHD and chronic GVHD. Factors with P-values /= 0.25x 109/L at day 28 was associated with improved OS (HR=0.33, 95% CI 0.12-0.86, P=0.02) (Figure) and DFS (HR=0.36, 95% CI 0.15-0.85, P=0.02). Absolute numbers of non-classical monocytes 〉/=0.02 x 109/L at day 28 was associated with decreased rates of grade II-IV aGVHD (HR=0.58, 95% CI 0.34-0.99, P=0.04) and decreased cGVHD (HR=0.48, 95% CI 0.26-0.90, P=0.02). None of the day 28 monocyte subpopulation counts were associated with TRM or relapse. Conclusions: Monocyte recovery differs among stem cell sources. UCB recipients experience more robust monocyte recovery, specifically within the intermediate and classical monocyte subpopulations, at day 60 and at all measured time points beyond, compared to other stem cell sources. Early monocyte recovery is associated with HCT outcomes. Increased classical monocyte count at day 28 is independently associated with improved OS and DFS. Paradoxically, despite their inflammatory properties, increased non-classical monocyte count predicts decreased acute and chronic GVHD within this large series of patients. If reproduced, these findings could allow for enhanced outcome prediction models and could lead to novel interventions to improve transplant outcomes. Disclosures Cooley: Fate Therapeutics: Research Funding. Miller:Oxis Biotech Scientific Advisory Board: Membership on an entity's Board of Directors or advisory committees.

Description: Abstract 3036 Poster Board II-1012 The potential role of allogeneic natural killer (NK) cells for therapy of refractory lymphoma is supported by the curative potential of allogeneic transplantation for lymphoid malignancies. Haploidentical donor derived NK cells may overcome Class I MHC Ag mediated inhibition and deliver an NK versus lymphoma effect. In a Phase II study we evaluated allogeneic NK cell infusions with Rituximab and IL-2 in a non-transplant setting to determine the expansion of NK cells in vivo and the clinical response in patients with refractory B-cell non-Hodgkin lymphoma (NHL). Six patients with advanced NHL received conditioning with Rituximab 375mg/m2 days -8,-1,+6,+15; Cyclophosphamide 60 mg/kg IV day -5; Fludarabine 25 mg/m2 IV days -6 through -2 as immunosupression to permit homeostatic expansion of allogeneic donor NK cells. Peripheral blood cells were obtained by lymphapheresis from unmobilized, HLA-haploidentical donors and selected for “killer immunoglobulin receptor” (KIR) ligand mismatch when available (3 out of 6 patients). Donor peripheral blood cells were enriched for NK cells with the Miltenyi CliniMACS device by depletion of T (CD3+) cells. The donor NK cells were then activated by overnight incubation with IL-2 (1,000 U/mL) and infused at a median nucleated cell dose of 2.27 ±0.4 × 107/kg. Subcutaneous IL-2 10×106 units (qod x 6 doses) was given to facilitate NK cell survival and expansion. All patients were evaluable for toxicity and efficacy. Patients tolerated the NK infusion well with only transient grade 1-2 toxicity and 5 received all 6 scheduled doses of IL-2. IL-2 activated donor NK cell products showed 〉 55% cytotoxicity against K562 targets. After IL-2 therapy, we observed a median absolute lymphocyte count of 980 ±440/μL. All cells were of recipient origin with no detectable donor NK cells. Importantly, in all patients the median number of host regulatory T cells (T regs phenotype CD4+Foxp3+CD127−) post treatment was significantly increased compared to pre-treatment (day 14 T regs: 134 ±141 cells/μL versus pre-treatment T regs: 24 ±12 cells/μL; P=0.06). To investigate the possibility of NK trafficking to affected lymph nodes, we performed fine needle aspiration of palpable tumor in 1 patient and demonstrated a low level of donor DNA by RFLP testing (2.5% donor chimerism). Simultaneous absence of NK cells in peripheral blood in the same patient suggested NK cell tissue homing to lymphoma-bearing nodes. Three patients achieved a partial remission (PR), one of whom proceeded to non-myeloablative cord blood allograft 2 month after NK cell infusion; two remain in partial remission after 1 and 4 months of follow-up. The trial failed to achieve prospective statistical parameters established to detect circulating NK cell expansion rate and will be modified. Conclusions This “proof of principle” study demonstrated lack of in vivo expansion of haploidentical NK cells in peripheral blood of patients with lymphoma. However, we identified host factors that interfered with NK cell expansion, including T reg proliferation and possibly inadequate immunosupression, and additionally, the finding of donor DNA in sites of tumor suggested donor NK cell localization to extravascular or tumor sites. Novel approaches to adoptive NK cell therapy trials should incorporate strategies to eliminate or prevent T reg expansion using alternate lymphodepleting regimens. Disclosures No relevant conflicts of interest to declare.

Description: NK cell based immunotherapy can be used to treat advanced acute myelogenous leukemia and data shows that success depends on the function of NK cells and how long the cells are maintained in the recipient. An important determinant of NK cell function is expression of Killer-cell immunoglobulin-like receptors (KIRs), which are involved in the maintenance of self tolerance and acquisition of NK cell functional competence, a process termed education. Little is known about how NK cell education influences NK cell survival. Utilizing a serum starvation assay we found that KIR+ and educated NK cells survived better when compared to those that did not have a KIR or KIR matching cognate HLA-ligand respectively (Figure 1A and 1B). Under basal conditions both KIR+ and educated NK cells had more anti-apoptotic proteins (Bcl-2 and Bcl-xL) and expressed less Fas. When NK cells were incubated with serum and IL-15 (10 ng/ml) followed by IL-15 withdrawal, a substantial increase of pro-apoptotic Bim was found on uneducated NK cells (1.4 fold increase; P = 0.01). Under this same condition educated NK cells expressed more FasL (1.5 fold increase; P = 0.0003), indicating that they could be driving cell death on neighboring NK cells when cytokine is limiting. Since both NK cell survival and homeostasis are mediated by cytokine signaling, we studied expression of IL-15 and IL-2 signaling components. The most significant change was IL-2Ra, which was expressed at higher levels on uneducated NK cells (3 fold increase; P = 0.0004) when the cells were stimulated with serum and IL-15 (1 ng/ml) for 72 hrs. Higher IL-2Ra expression correlated well with cell death and Bim expression so we decided to test if modulating its expression would alter NK cell survival in an IL-15 withdrawal setting. Compared to the control, transient overexpression of IL-2Ra lead to a 1.55 fold decrease in NK cell numbers (P = 0.01) while siRNA knockdown of IL-2Ra lead to a 1.5 fold increase in NK cell numbers (P = 0.07). Importantly, at the time of harvest there was a 1.4 fold decrease in cell death in the IL-2Ra knockdown condition when compared to the control (P = 0.0007). Given that no IL-2 or crosslinking signals were present, the mechanism must differ from the well-described role of IL-2 on activation induced cell death. Since FasL is upregulated on educated NK cells when IL-15 is limiting and IL-2Ra renders NK cells more sensitive to cell death, we tested if educated cells could drive cell death of IL-2Rahi NK cells when survival signals are scarce using a co-culturing assay and FasL blocking antibodies. IL-2Rahi NK cells were more sensitive to apoptosis when co-cultured with KIR+ NK cells (presumably enriched for educated subsets) than with KIR- NK cells from the same donor that were subjected to IL-15 stimulation and withdrawal (34±4.8% vs. 20.6±4.7%; P = 0.002). The effect was reduced when FasL was blocked (P = 0.003), and no differences in killing were seen on the IL-2Ralo NK cells regardless of the treatment. Taken together these findings indicate that educated NK cells outlive their counterparts through two mechanisms: decreased expression of proteins involved in cell death and by killing competing NK cells when IL-15 is limiting. Finally, since we have previously reported on expansion of educated NK cells on transplant patients post CMV reactivation and since CMV reactivation can be associated with decreased relapse, we wanted to investigate if CMV reactivation could alter NK cell survival. There was increased survival on the NK cells from adult donor HCT (2 fold increase at 6 months; P = 0.03) and umbilical cord blood (3 fold increase at 100 days; P = 0.01) allogeneic transplant patients that had undergone CMV reactivation supporting the physiologic role of NK cell survival in vivo from a pathologic challenge. Taken together, these findings show that NK cell functional repertoires are determined by class I interactions, infection, and NK-NK interactions through IL-2Ra, Bim, and FasL to mediate clonal dominance that might be exploited in order to enhance NK cell survival and function after adoptive transfer of allogeneic NK cells for therapeutic use in cancer. Disclosures: No relevant conflicts of interest to declare.