A Phase II Trial Evaluating the Efficacy and Safety of Sargramostim Post Infusion of T -Replete HLA Mismatched Peripheral Blood Haploidentical Hematopoietic Stem Cells with Post Transplant Cyclophosphamide

Phase II trial evaluating the Efficacy and Safety of Sargramostim Post Infusion of T -Replete HLA Mismatched Peripheral Blood Haploidentical Hematopoietic Stem Cells with Post-Transplant Cyclophosphamide

Background: T-replete HLA mismatched haploidentical transplantation (Haplo) with post-transplant cyclophosphamide (PTCY) is widely used for patients lacking a suitably matched donor. Patients who receive haplo with PTCY usually start growth factor support (G-CSF) on day +5 for neutropenia prophylaxis. G-CSF is a specific stimulator of the growth and differentiation of hematopoietic progenitor cells committed to the neutrophil lineage. Granulocyte-macrophage colony-stimulating factor (GM-CSF) on the other hand, is not restricted to neutrophils because it also affects monocytes and eosinophils and plays a central role in antifungal host response. In a prior randomized study among 206 matched allogeneic HCT patients (wang et al, J Clin Oncol. 2015;33(34):3999-4006), neutropenia prophylaxis with GM-CSF +/- G-CSF had significantly lower 100-day TRM, lower 100-day cumulative mortality, lower infection related mortality and lower 600-day invasive fungal disease related mortality compared to G-CSF. Given the increase number of HLA mismatched haplos with PTCY performed each year and the high risk of infectious complications associated with this type of transplant, we suggest that GM-CSF administration post haplo and PTCY can yield similar count recovery rate to GCSF with a potential of lowering risk of infectious complications.

Endpoints: The primary endpoint is to establish equivalent effectiveness of sargramostim to a matched control cohort of G-CSF treated patients in time to neutrophil engraftment (ANC>500 x 3 days) post haplo with PTCY.

Secondary endpoints: estimates of OS, DFS, relapse rate, GVHD and TRM. Donor chimerism at days 30, 60, 100 and 6 months post-transplant. Incidence and grade of cytokine release syndrome between day 0 and 30. Incidence of infectious complications and infection related mortality at day 100. Immune profiling assay "CYTOF" will be measured at days 30, 60, 90 and 180 for functional and phenotypic immune analysis. Other secondary endpoints include length of hospital stay and charges-based cost comparison during the first 100 days between sargramostim and G-CSF matched controls. All primary and secondary endpoints will be reported for the study group and compared to a matched control cohort from our database containing patients who received haplo with PTCY using G-CSF for neutropenia prophylaxis. A set of 20 patients, who are not on the Sargramostim study and receiving haplo transplant with G-CSF at the same time of the study accrual period, will be consented for immunoprofiling assay at days 30, 60, 100 and 180.

Methods: Patients aged 18-78 years with an available 5/10-8/10 HLA- matched related donor with high risk hematological malignancy with adequate organ function will be enrolled for the study. Patients will receive Sargramostim 250mg/m² /day subcutaneous starting day +5 post haplo/PTCY and continue till ANC>1000x 3 days or >1500x 1 day. If grade 3 or 4 adverse reactions occur, reduce dose by 50% or temporarily discontinue the dose until the reaction abates. Conditioning regimens to be used include previously published and currently used at our center (Flu/Cy/TBI, Flu/high dose TBI or Flu/Mel). Treatment will be in the outpatient setting and admissions if clinically indicated.

Sample Size Determination: Time to ANC>500 will be evaluated between the prospectively enrolled patients and matched historical controls. Based on the data of haplo transplants done at Northside in 2010-2016, the standard deviations (SD) was 3 days for time to ANC500. The correlation in measurements between prospective patient and matched control is conservatively assumed to be 0.1. SD for differences between matched pairs would be 3.55 days for time to ANC500. To control the type I error to be 5%, the 90% paired t confidence interval for difference in time to ANC500 should be constructed and its upper boundary will be compared to the non-inferiority margin 2 days. The size of 35 prospective patients provides the power 90% for concluding that time to ANC500 for Sargramostim is no more than 2 days longer than historical controls. The actual enrollment size is decided to be 38 to accommodate 10% of drop off rate.

Enrollment: The study is expect to start enrolling patients October 2019.