ALBERT

Description: Abstract 4521 Continuity of care (COC) is acknowledged as a core quality measure in HIV, heart failure and family medicine. Allogeneic hematopoietic stem cell transplantation (Allo-HCT) is complex therapeutic option where is the selection of patient, donor, conditioning and immune-suppression plays a pivotal role in overall survival (OS) outcome. Although the team concept is an integral part of care in Allo-HCT, there is little literature known about the impact of personnel COC (care from the same provider) on OS. Method: Between July 2009 and May 2012, 74 consecutive Allo-HCT were performed at our center. The patient's clinical care for the first consecutive 41 patients was shared between the physicians independent of primary transplant physician (Non- COC group). To assess the impact of COC on OS after Allo-HCT, the subsequent 33 patients (COC group) were followed by their transplant physician both as in-patient and outpatient. Physician's contribution into the care of each individual patient was calculated from physicians billing visits. Patient characteristics of COC & Non-COC groups are shown in table I. Graft vs. host disease (GVHD) prophylaxis was Tacrolimus/MTX or Cyclosporine/Mycophenolate with the addition of Thymoglobulin for MUD and mismatched RD. Results: The average contribution of the primary transplant physician into their patients care during year one post-transplant was 49% vs. 80% for Non-COC and COC groups respectively (P=0.01). There was no difference in patient characteristics between COC and Non-COC groups except for older patients in Non-COC. With median duration of follow up of 815 days for Non-COC and 320 days for COC groups, the 1- year OS was 56% vs. 75% respectively (P=0.07). Similarly, there was a trend toward improved DFS for COC (1-year DFS of 68% vs. 48%, P=0.11). On Univariate analysis, Age (≤ 55, P=0.26), Donor source (MUD vs. RD, p=0.65), diagnosis (acute leukemia vs. other, p=0.18), status at transplant (P= 0.23), cytogenetic risk (p=0.79) and conditioning (FIC vs. RIC, p=0.62) were not predictive of improved OS. Both cumulative incidence of relapse and treatment related mortality (TRM) at 1-year were lower in COC compared to Non-COC groups; 9.5% vs. 25% and 17% vs. 25% respectively. The cumulative incidence of grade II –IV acute GVHD (aGVHD) at day 100 and day 180 was 64% & 64% for Non-COC vs. 46% & 72% for COC respectively. There was more patients with grade III/IV aGVHD; 13/41 (32%) in Non-COC compared to 6/33(18%) in COC, however this difference was not statistically significant (p=0.27). Additionally, there was no difference in OS in patients with grade III/IV aGVHD in Non-COC (13 patients) vs. COC (6 patients), P=0.85. In contrast, Patients without grade III/IV aGVHD had a statistical OS advantage in favor of COC (27 patients) vs. Non-COC (28 patients) with one year OS of 90% vs. 68% respectively, P=0.05. Cumulative incidence of chronic GVHD at one year was 77% for COC and 48% for Non-COC patients, P=0.02. Conclusion: Physician-Patient continuity of care may favorably impact OS in Allo-HCT for hematological malignancy. The reason for lower relapse and TRM in COC group is unclear but could be attributed to older patients and differences in aGVHD management in Non-COC group. In this small study, COC did not impact OS in patients with severe aGVHD but may result in OS advantage in Allo-HCT patients without grade III/IV aGVHD. Larger studies are needed to address the impact of COC on outcomes after Allo-HCT. Disclosures: Khaled: Celgene and Takeda Pharmacutical: Honoraria, Speakers Bureau. Solh:Celgene: Speakers Bureau.

Description: Immune mediated demyelinating disease after allogeneic stem cell transplantation is a rare entity with unclear etiology. Acute inflammatory demyelinating polyneuropathy (AIDP) has been reported post related and unrelated allogeneic stem cell transplantation but no such case has been reported post unrelated cord blood transplantation. We hereby present the first case of GBS post double umbilical cord blood transplantation (DUCBT). A 55 year old male with relapsed refractory CLL received DUCBT with two 5/6 matched cord units (antigen level HLA-A, HLA-B and allele level HLA-DRB1) with fludarabine, cytoxan and total body irradiation based reduced intensity conditioning regimen. Graft versus host disease (GVHD) prophylaxis was with cyclosporine and mycophenolate. Early post transplant course was complicated by grade 4 acute GVHD of the gut with a complete resolution with steroid therapy and successful taper of all immunosuppression by day 180. 7 months post transplantation, patient presented with skin rash and tingling in both feet that progressed rapidly to lower extremity paralysis over the course of 2 days. Physical exam showed macula-papular rash affecting his upper extremities, upper chest and back area. Neurologic exam was significant for motor weakness in lower extremities 2/5, plantar flexion and knee flexion 3/5. He had loss of deep tendon reflexes in both lower extremities (Achilles and Patellar) and upper extremities (biceps and triceps).Workup revealed normal blood counts, organ function, vitamin B12, folate, TSH level, free cortisol. Serum electrophoresis and immunfixation was also normal. Magnetic resonance imaging of the central nervous system showed mild neural foramina narrowing at the L4-L5 level. Serology for Lyme disease, Epstein Bar virus (EBV), syphilis, cytomegalo virus (CMV), Hepatitis, HIV, toxoplasma, enterovirus and human herpes virus 6 was negative. Blood tests for autoimmune markers including (anti-nuclear antibody)ANA, acetylcholine esterase and volted calcium channel antibodies were normal. A lumbar puncture was performed and showed a high protein level of 67mg/dl, 1 nucleated cell/mm3 and normal glucose level. Cerebrospinal fluid was negative for oligoclonal bands, West Nile virus, cryptosporidium, HHV6, herpes virus 1 and 2, gram stain and cultures. Nerve conduction studies and needle electormyegraphy was suggestive of acute demyelinating polyneuropathy. Based on the above workup, he was diagnosed with GBS and started on therapy with intravenous immunoglobulin at 0.5gm/kg for 4 days and prednisone 1mg/kg daily for the treatment of GVHD. Etiology of GBS was presumed to be related to GVHD as his workup was negative for campylobacter, HIV and CMV. He improved significantly over the next 4 weeks and became ambulatory without assistance but his weakness symptoms relapsed as his prednisone was tapered. Prednisone was increased again to 1mg/kg and sirolimus was started. Patient was successfully tapered of prednisone and remains fully ambulatory without assistance or evidence of GVHD on single agent sirolimus 13 months post DUCBT. This is the first case of autoimmune demyelinating polyneuropathy post DUCBT with association of GVHD that was managed successfully with a combination of intravenous immunoglobulins, steroids and sirolimus. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Reduced and full intensity conditioning (RIC, FIC) regimens employing intravenous (i.v.) busulfan plus fludarabine (Bu-Flu) have resulted in improved treatment related mortality and comparable overall survival in patients undergoing allogeneic HCT for AML/MDS who are not candidates for more intense regimens. However, relapse after HCT remains a leading cause of treatment failure after such conditioning regimens. Methods In Order to assess relapse following allogeneic HCT for AML/MDS, a retrospective analysis was performed to evaluate the outcomes of 55 consecutive patients with AML/MDS (49/6) who received i.v. Bu-Flu based conditioning. Blood samples were collected post HCT in a subset of those patients (30 patients). Serum values of 42 biological markers were measured at day 30 post HCT (2/30 patients were day 60 samples) using multiplex Luminex assay. Patients characteristic are shown in Table-1. Patients received single daily dose of iv Bu 3.2 mg/kg for 2 days (RIC, Bu2-Flu) or 4 days (FIC, Bu4-Flu) based on age, older or younger than 65 respectively. Fludarabine was given as a single daily dose of 40 mg/Kg for 4 days. Graft versus host disease prophylaxis was Tacrolimus/Methotrexate in FIC recipients and Tacrolimus/Mycophenolate in RIC recipients. Low dose thymoglobulin of 4.5 mg/kg was used in unrelated donor HCT recipients. Results With a median follow up of 18 month, the overall survival (OS) at 1 & 2 years was 73 ± 6% and 67 ± 7%, respectively, (Fig 1). Similarly, disease free survival at 1 and 2 years was 64 ± 7%. As expected, there was low cumulative incidence of treatment related mortality of 8 ± 3% at 1 and 2 years while the cumulative incidence of relapse was 28.0 ± 3% and 31± 2% at 1 and 2 years respectively, (Fig 2). Cumulative incidence of grade II-IV acute GVHD was 54% with grade III-IV of 25% at day 100. Cumulative incidence of chronic GVHD was 49, 54% at 1 and 2 years respectively. In a subset of patients where chemokine analysis was performed (30 patients), only MCP-1 levels at day 30 post HCT were predictive of relapse out of the 42 biological markers tested. The 7 out of 30 patients who relapsed in this subset (23%) had higher mean level of MCP-1 at day 30 of 537, SD ±213 versus 324, SD ± 160, P=0.007, (Fig 3). MCP-1 was predictive of leukemic relapse 82 days in advance on average prior to overt hematological relapse. Full chimerism (〉95%) was detected at Day 30 in 5/7 patients who relapsed in the biological marker group. Conclusion Bu-Flu based conditioning regimens result in improved OS in patients with AML/MDS but do not impact relapse rate after allogeneic HCT. Serum MCP-1 levels in the early post-transplant period were predictive of relapse in subset of patients where post HCT biomarkers were available. Future larger studies may find potential role of MCP-1 in predicting relapse in patients at risk after HCT with Bu-Flu for AML/MDS. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4536 Elevated white blood cell count is considered an independent risk factor for thrombotic events among patients with hematological malignancies undergoing intensive chemotherapy, as well as autologous or allogeneic hematopoieitc cell transplant. Engraftment among allogeneic HCT recipients is associated with endothelial damage, cytokine release and elevated white count. The impact of persistent leukocytosis after engraftment on survival and transplant related complications such as graft versus host disease is not well established. In this study, we retrospectively reviewed the charts of 74 consecutive patients who underwent allogeneic HCT in a single center between 2009 and 2011. Neutrophil engraftment was defined as an absolute neutrophil count of 0.5 ×103/μl for 3 consecutive days. Patients blood counts were assessed for 60 days post engraftment (three times per week). Persistent Leukocytosis (group A) was defined as a white blood cell count of 〉10 ×103/μl for more than four days. Patients who had leukocytosis for three or less days were called group B. Conditioning regimens included myeloablative MA (Fludarabine/Busulfan or cytoxan/total body irradiation) and reduced intensity RIC (fludarabine/busulfan or fludarabine/cytoxan/low dose TBI). Graft versus host disease prophylaxis included methotrexate/tacrolimus for myeloablative and tacrolimus/Mycophenolate for reduced intensity conditioning. The median age was 51 years (range: 18–74 years). Graft source included matched related (n=24), Unrelated donor (n=48) and cord blood (n=2). 36 patients received MA conditioning and 37 had RIC conditioning. Median follow up for the cohort was 556 days. Median time to neutrophil engraftment was 13 days. 22 patients had persistent leukocytosis (group A) during the first 60 days post engraftment. The two groups did not differ in terms of age, gender, diagnosis (acute leukemia versus other), disease risk and conditioning regimen (table 1). Group A had a higher proportion of related donor source (50% versus 31%; p=0.05). One year overall survival was significantly worse in group A (40.6% versus 74.7%; p=0.009). Grade II-IV acute graft versus host disease was higher among group A patients (68% versus 48%; p=0.18) although not statistically significant. The cause of death among patients with persistent leukocytosis included acute GVHD (n=6), relapse (n=5), infection (n=2) and thrombosis (n=1). In conclusion, Persistent leukocytosis post engraftment in allo HCT recipients is associated with worse overall survival and may be an independent risk factor for acute GVHD. Despite a larger proportion of related donors among the leukocytosis group, our data showed a trend towards higher incidence of acute GVHD. The exact mechanism leading to prolonged leukocytosis is not clear and further studies to confirm the association between leukocytosis and clinical outcomes are warranted. Table 1. Clinical and Transplant Related Characteristics of Group A and Group B Patients Group A Group B P value Leukocyte count 〉10 × 103/μl for ≥4 days Leukocyte count 〉10 × 103/μl for ≤ 3 days Number of Patients 22 51 Diagnosis Aplastic anemia 0 2 0.191 Acute Leukemia 19 35 MDS/MPD 1 3 Lymphoma/CLL 1 9 MM 1 2 Graft Source Related 11 12 0.05 Unrelated 10 38 Cord Blood 1 1 Conditioning regimen RIC Conditioning 13 23 0.39 MA Conditioning 9 28 Time to Neutrophil engraftment (Days) 13 12 GVHD Prophylaxis Tacro/MMF 13 23 Tacro/Methotrexate 9 28 0.46 Median follow up (Days) 533 567 Cause of Death GVHD 6 3 0.03 Relapse 5 9 0.86 Infection 2 2 Thrombosis 1 0 MDS: myelodysplastic syndrome; MPD: myeloproliferative disorder; CLL: chronic lymphocytic leukemia; RIC: reduced intensity conditioning; MA: myeloablative; GVHD: graft versus host disease; CSA: cyclosporine; MMF: mycophenolate; 1 P value for acute leukemia versus all others. Disclosures: Solh: Celgene: Speakers Bureau. Khaled:Celgene and Takeda Pharmacutical: Honoraria, Speakers Bureau.

Description: Abstract 4715 Allogeneic hematopoietic stem cell transplantation (HCT) represents potential curative option for patients with high risk hematological malignancy. However, HCT is still associated with high treatment related mortality (TRM). Extensive research in graft vs. host disease (GVHD) prevention and treatment of relapse post-transplant is being done to improve HCT outcomes. Although the transplant center experience and transplant volume has been correlated with improved outcomes after allogeneic HCT, the effect of physician's experience and continuity of care on outcomes after HCT in hematological malignancies is not known. We hypothesized that continuity of care along with physician experience would improve outcomes after HCT. Retrospectively, we reviewed the survival outcomes of 28 consecutive patients with high risk malignancy that underwent HCT from July 2010 to May 2012. A single 9 years experienced physician followed up those patients both as inpatient and outpatient from the time of initial consult up to last follow. The average percentage of care provided by the transplant physician to the 28 patients was 84% of the total billing visits. Patients characteristics including age, diagnosis, disease risk, status at HCT, donor source, conditioning regimen, GVHD prophylaxis and survival status are shown in table 1. All the patients received an established conditioning regimens and standard GVHD prophylaxis according to the donor source (table 1). Median age at transplant was 53 y/o. Donor source was 9 related donors (RD), 18 MUD donors and 1 double cord blood. All patients engrafted except two patients. One graft failure was in a patient who received a marrow product with low CD34 count. The other graft failure was in a myelofibrosis patient who failed to engraft after 3 attempts. With median duration of follow up of 422 days; at one year the overall survival (OS) was 92.8% (C.I. 83%–100%) and disease free survival (DFS) was 80% (C.I. 64%–96%). The cumulative incidence of acute GVHD was 52.5%, 55% for RD and 51% for MUD. Only 2/28 patients developed grade III and IV acute graft vs. host disease (Both MUD). Surprisingly, the cumulative incidence of chronic graft vs. host disease was higher for RD compared to MUD, 100% vs. 64%, respectively (P=0.1). Overall cumulative incidence of chronic GVHD was 74%. Both cumulative incidence of relapse and TRM were low of 7% at 1 year post transplant. Conclusion: Continuity of care when combined with the physician experience may result in superior survival outcomes in patients with high risk malignancy undergoing allogeneic HCT. Despite the small sample size, this intervention resulted in unexpected improved survival rate with lower TRM and relapse rate. Design of future prospective studies to address the impact of continuity of care and physician experience on outcomes after HCT is needed. Disclosures: Khaled: Celgene and Takeda Pharmacutical: Honoraria, Speakers Bureau. Solh:Celgene: Speakers Bureau.

Description: Abstract 4538 CD20 is a trans-membrane protein expressed on mature B cells through all stages of their differentiation. However, its expression is down regulated at the point of differentiation into plasma cells and expressed only in 16–22% of mature plasma cells. CD20 expression on plasma cells has been described with both favorable prognosis in association with translocation t(11;14) and unfavorable prognosis in association with plasma cell leukemia. The incidence of CD20 expression on plasma cells from patients with relapsed/refractory multiple myeloma is not well addressed in literature and CD20 testing is not routinely done in this patient population. Additionally, it is not known if CD20 represents a primary aberrant expression in newly diagnosed cases of multiple myeloma or represents a secondary genetic change at the time of relapse related to hypothesized myeloma stem cells. In order to study the above questions, we retrospectively reviewed the medical records of 92 patients with symptomatic MM who underwent ASCT between January 2008 and December 2011. As of July 2012, 38 patients have relapsed. Bone marrow biopsy and flow cytometry results were available for 33 patients at time of diagnosis and relapse.CD20 expression was positive on plasma cells by flow cytometry in 11 out of 33 patients (33%) at time of relapse. Interestingly, CD20 expression at diagnosis was negative in 4 out of those 11 patients. This up-regulation of CD 20 expression at time of relapse was associated with clonal evolution in two patients (deletion-17 and complex hypodiploid cytogenetics, respectively). Confirmatory immune-histochemical staining for CD20 was positive only in 2 of those 4 patients. Rituximab Outcomes: Three of the 11 patients with relapsed disease after autologous transplant were treated with Rituximab/Bortezomib. Rituximab was given weekly for 4 weeks at dose of 375 mg/m2 and Bortezomib was given at dose of 1.3 mg/m2on day 1, 8 &15 for 2–4 cycles. Time to progression for these three patients was 12, 5 and 3 months. These patients were CD20 positive both at time of diagnosis and relapse. An additional patient with primary refractory myeloma and negative CD20 expression at diagnosis was found to be CD20 positive after 5 unsuccessful lines of therapy suggesting CD20 up regulation. This patient achieved complete remission after treatment with weekly Rituximab/Bortezomib × 4 weeks followed by an autologous stem cell transplantation. Conclusion: CD20 expression on plasma cells at time of relapse/progression can occur in one third of patients with multiple myeloma and may provide an additional therapeutic target. The cause of discrepancy between CD20 expression by flow cytometry and immune-histochemical staining is unclear and suggests that the two methods may be complementary for a comprehensive evaluation of CD20 expression in multiple myeloma.CD20 up-regulation in patients with relapsed/refractory myeloma who were previously CD20 negative at diagnosis may represent a secondary genetic event heralding a more aggressive disease. Future prospective studies evaluating CD20 expression on plasma cells at different stages of disease progression may optimize the timing for anti-CD20 therapy while harnessing the concept of myeloma stem cells. Disclosures: Khaled: Celgene and Takeda Pharmacutical: Honoraria, Speakers Bureau. Solh:Celgene: Speakers Bureau.

Description: Introduction Plerixafor is a reversible CXCR4 antagonist that has been approved by the food and drug administration for autologous hematopoietic stem cell mobilization in patients with multiple myeloma and non-Hodgkin’s lymphoma. Patients mobilized with Plerixafor were shown to have a higher proportion of primitive stem cells (CD34+/CD133+/CD38-), CD4+ T cells and Natural killer cells (CD3-/CD16+/CD56+) in the graft composition when compared to patients mobilized with chemotherapy plus G-CSF alone .We investigated the effect of Plerixafor on immune reconstitution at thirty and sixty days post autologous stem cell transplantation. Methods Patients eligible for autologous stem cell transplantation were enrolled on a single arm prospective immune reconstitution trial. A complete blood count, differential and lymphocyte flow cytometry panel (T cell, NK cell and B cell markers) was checked on Days 30 and 60 post autologous transplantation. Stem cell mobilization was carried per our institutional standards. All patients received subcutaneous G-CSF at a dose of 10 µg per kilogram body weight for four consecutive days. On Day4, patient with a peripheral CD34 count of ≤20/µl received plerixafor 0.24mg/kg. Collection was started on day 5 and continued till collection goal was reached or patient failed to get the minimal cell dose after 4 consecutive days of aphaeresis. Results 49 patients were enrolled during the period from September 2010 till May 2012. Median age at time of transplantation was 54 years (range 21; 72 years). 35 patients had multiple myeloma and 14 had non-Hodgkin’s lymphoma. 16 patients received GCSF alone (group A) and 33 had plerixafor plus GCSF (group B) for mobilization. All patients achieved the minimum target of CD34 collection. The mean number of collection days was 1.9 and 1.4 days (p=0.05) with a total collection dose of 7.76 and 7.61 CD34 x106/Kg for groups A and B respectively. The percentage proportion of CD34 in the aphaeresis product was 0.73% and 0.75% (p=0.9) for group A and B. Total infused CD34 dose was similar in both groups (4.88 and 4.56 CD34x106/kg) with time to engraftment of 11.68 vs 11.69 days for neutrophils and 20.62 vs 21.39 for platelets in groups A and B respectively. There was no difference between day 30 absolute lymphocyte count (1.09 vs 1.44 x103/mm3 p=0.18); Absolute NK cell (0.31 vs 0.35 x103/µl; p=0.51); absolute T cell count (0.71 vs 0.96 x103/µl p=0.33) and absolute neutrophil count (2.98 vs 2.63 x103/mm3 p=0.37). The cell count recovery was also not significantly different when analyzed per disease (myeloma or non-Hodgkin’s lymphoma) and at day 60. Conclusion Our study shows that patient mobilized with plerixafor and G-CSF have similar immune reconstitution at 30 and 60 days post autologous transplantation compared to patients mobilized with G-CSF alone. Disclosures: No relevant conflicts of interest to declare.

Description: Low Dose Thymoglobulin Result in Improved Outcomes after Allogeneic Unrelated Hematopoietic Stem Cell Transplantation (HCT) for Patients with Acute Myeloid Leukemia/ Myelodysplastic Syndrome Conditioned with Intravenous Busulfan and Fludarabine. Intravenous Busulfan/Fludarabine (Bu/Flu) based conditioning regimens have resulted in lower treatment related mortality (TRM) after allogeneic HCT in both reduced and full intensity dosing in AML/MDS. The addition of thymoglobulin to such regimens to minimize the risk of graft versus host disease after unrelated donor (MUD) HCT has been linked to the increased concerns of relapse risk. Method In order to study the impact of addition of low dose thymoglobulin to Bu/Flu regimens in MUD recipients, the medical records of 38 consecutive AML/MDS patients who underwent MUD HCT with BU/Flu and low dose thymoglobulin were retrospectively reviewed. All the patients received single daily dose of iv Bu 3.2 mg/kg for 2 days (RIC, Bu2-Flu) or 4 days (FIC, Bu4-Flu) based on age,  older or younger than 65 respectively. Fludarabine was given as a single daily dose of 40 mg/Kg for 4 days. Graft versus host disease prophylaxis was Tacrolimus/Methotrexate in FIC recipients and Tacrolimus/Mycophenolate in RIC recipients. Additionally, all the patients received thymoglobulin 1.5 mg/Kg on day -3, -2 and -1. All patients received peripheral stem cells except one patient with mismatch MD who received bone marrow product. Results Patient Characteristics are shown in Table 1. All the patient engrafted except one who received marrow product.  All the patients but 3 (8%) achieved 90% or more donor chimerism by day 100. With Mean follow up of 500 days (range, 100-1242) the overall survival (OS) was 77% ± 7 (CI 63-91%) at 1 year and 67% ± 9 (CI 49-85%) at 2 years. Similarly, disease free survival was 66% ±8 (CI 50-82%) at 1 and 2 years. Cumulative incidence of acute GVHD grade II-IV was 55% with grade III-IV 12%. Cumulative incidence of chronic GVHD at 1 year was 43% with extensive chronic GVHD in 17%. The regimen was associated with low treatment related mortality (TRM) with cumulative incidence of only 5% at one year, CI 14-21%. The cumulative incidence of relapse at one year was 29%, CI 17-49%. On univariate analysis only high risk CIBMTR status was predictive of poor OS (p=0.05). Conclusion The addition of low dose thymoglobulin to RIC and FIC regimens with iv Busulfan/Fludarabine prior to MUD HCT results in low TRM and improved OS for patients with AML/MDS. Relapse rate does not seem to be increased in this cohort by the addition of low dose thymoglobulin in comparison to historical control. Disclosures: No relevant conflicts of interest to declare.

Description: Invariant natural killer cells are immunomodulatory T cells with a proven effect in Graft versus Host disease. Recent data have shown that the infused iNKT cell dose is associated with decreased risk of aGVHD in matched sibling graft recipients. This study was conducted to assess the effect of the infused iNKT cell dose and early INKT recover at day 30 and 60 post HCT on relapse and GVHD. We also assessed the effect of INKT cells among patients receiving thymoglobulin as part of their HCT protocol. Methods: 48 adult allogeneic HCT recipients were enrolled on a single arm prospective trial between 2012 and 2014. All donors were mobilized with G-CSF per institutional and NMDP guidelines. Conditioning regimen consisted of Flu/Bu4 in the myeloablative setting and flu/Bu2 in RIC setting. Thymoglobulin at 1.5mg/kg/day x3 days was used for unrelated donor recipients. A small portion (2.5mL) of the donor peripheral blood stem cell product was collected prior to transplant and recipient peripheral blood (~10 mL) was collected on days +30, and +60 post HCT. Flow cytometric analysis was performed on the samples with an antibody cocktail that contained the following pre-conjugated monoclonal antibodies: CD56-PE (Miltenyi Biotech, Auburn, CA), CD3-APC, CD16-FITC, (Beckman Coulter, Brea, CA), CD19-PE-CY7 (BD Biosciences, San Jose, CA). For iNKT analysis and for CD4/CD8 analysis, 2 x 106 total cells were stained with CD3-APC, TCR Vα24-PE, TCR Vβ11-FITC (Beckman Coulter), CD4-APC-H7 and CD8-PE-CY7 (BD Biosciences). Data were acquired using BD FACSCanto II and analyzed with the FACSDiva software (BD Biosciences) to quantify CD3+ T cells, CD3+ CD56+ NK-like T cells, CD56+ CD16+ and CD56+ CD16- NK cells as well as CD19+ B cells. INKTs were quantified as CD3+, TCR Vα+, TCR Vβ+ lymphocytes. A total of 30,000 lymphocytes were collected for NK, T, and Bcell analysis. For iNKT, an end point of 200 iNKT or 500,000 total events was set. The graft and early cell subsets were assessed for their impact on acute GVHD and relapse. Results: 48 patients with a median age of 55 years received Flu/Bu4 (n= 29) and Flu/Bu2 (n=19). Donor source was a matched sibling (n=13, 27%) or a matched unrelated donor (n=35, 73%). The collected and infused graft cell doses are shown in table 1. Median time to Neutrophil and platelet engraftment was 11.5 and 18.8 days respectively. The propability of one year survival was 76% with a 1 year cumulative incidence of relapse of 34%. The pre-transplant parameters that predicted a more robust iNKT cell recovery by multivariate analysis were young recipient age, young donor age, dose of CD34 infused and RIC regimen (p10-3developed CMV reactivation. Conclusion: The infused iNKT cell dose in the peripheral graft inversely affects relapse rates post allogeneic HCT. This effect is independent of thymoglobulin use or donor source. Trials aiming at expanding the INT cell population in the infused grafts may help decrease risk of relapse post HCT. Table 1: Cellular Content of the Infused Graft Cell Subset Mean ± sd Median (min-max) CD3/kg infused 250 ± 111 251 (23-428) CD 34 infused 6.2 ± 2.1 6.0 (1.9-13.5) NK cell x106/kg infused 9.5 ± 11.0 6.6 (0-67) T cell x106/kg infused 192 ± 102 187 (0-516) B cell x106/kg infused 39.8 ± 33 35 (0-151) NKT x106/kg infused 9.5 ± 11.0 6.6 (0 -67) iNKT x106/kg infused 0.43 ± 0.7 0.23 (0-4.6) %CD34 product 0.99 ± 0.5 0.89 (0.15-2.7) %NK product 2.4 ± 1.5 1.9 (0.19-6.5) %T cell product 27.0 ± 11.0 27 (1.21-51.8) %NKT product 1.3 ± 1.2 1.0 (0.5- 5.9) %iNKT product 0.26 ± 0.7 0.12 (0 -4.5) %B cell product 5.4 ± 3.8 4.5 (0 -19.7) %WBC product 236 ± 100 209 (115-508) % Lymph product 32.4 ± 10.3 32 (13-51.8) NK cells, T cell, NKT and B cells are percentage of total nucleated cells iNKT cells are percentage of CD3+ cells Disclosures No relevant conflicts of interest to declare.

Description: NK cell immunotherapy shows exciting promise, but inconsistency and variability remain as a significant challenge. Since NK cells comprise a small fraction (∼5%) of the peripheral blood mononuclear cell fraction, expansion of NK cells in vivo or ex vivo is a critical requirement to attain therapeutically effective dosages and to observe consistent positive clinical outcomes. Most of currently developed ex vivo expansion protocols depend on co-culture with various engineered and/or cancer derived stimulator/feeder cells to induce the proliferation of NK cells. The use of accessory cells poses significant challenges to clinical transfer. Our laboratory has developed a nanoparticle-based expansion technology that utilizes particles, few hundred nanometers in size, derived from the plasma membrane (PM) of K562 feeder cells expressing IL-15 and 41BBL on their surface (PM-mb15-41BBL). These particles in combination with low concentration of IL-2 induce selective and efficient expansion of NK cells within human peripheral blood mononuclear cells (PBMC). When PBMC are stimulated with PM-mb15-41BBL over 21 days the NK cell numbers increase exponentially between days 6 and 18 of culture. The numbers of NK cell increased on average 200 fold (range 104-557, n=11, 4 donors) after 12-13 days of culture in the presence of PM-mb15-41BBL particles (at 200 µg of membrane protein/mL). The expansions with the PM particles are comparable to those in the presence of live feeder cells that gave ∼200 fold (79-895, n=11, 4 donors). The PM-particle based NK expansion is far better in comparison to NK stimulation with soluble purified 41BBL, IL-15 and IL-2, at matching concentrations, that yielded only 3 fold (1-4, n=6, 3 donors) increase in NK cells. Furthermore, the NK cells expand selectively under these conditions where they initially consisted only about 10% of the population of PBMC isolated from fresh peripheral blood, but increased to more than 95% of the cell suspension after 14 days in culture. The extent of expansion and NK cell content on day 12 of culture was dependent on the concentration of PM particles used with 200 µg of PM protein/mL being the optimal dose. Thus, PM nanoparticles can expand NK cells as efficiently and selectively as feeder cells. Furthermore, the PM-particle based expansion is more reproducible between trials and with different donors as compared to NK cell expansion induced with feeder cells (coefficient of variation 63% vs. 88%, respectively). The NK cells expanded in presence of PM-particles were highly cytotoxic against several leukemia cell lines and also against patient derived AML blasts. Expanded NK cells were 4 to 9 times more potent against AML cell lines K562, KG1 and HL-60 as compared to freshly isolated NK cells that were pre-activated with a high dose of IL-2. The PM-particle expanded NK cells also were selectively cytotoxic where they efficiently killed patient derived CD34+ leukemia blasts while sparing healthy CD34- peripheral blood cells. The expanded NK cells were observed to have an increase in the expression of major activating receptors such as NKG2D, NKp44, NKp30 and of the death receptor ligand FasL. This expression difference corresponds well with the activated cytotoxic phenotype and is likely responsible for their increased cytotoxicity against AML cells. Pilot trials in NSG mice are currently ongoing. Disclosures: Solh: Celgene: Speakers Bureau.