PDL1 Positivity By FISH in Patients Not in Complete Remission at Transplantation

Introduction Failure to achieve morphological complete remission occurs in 30% AML patients resulting in dismal outcomes. Data addressing influence of pre-transplant PDL-1 status on post-transplant outcomes is limited. We hypothesize that PDL-1 expression by immunohistochemistry (IHC) or Fluorescent in situ hybridization (FISH) prior to transplantation may predict a poor outcome. Methods This was a retrospective, observational single center pilot study involving refractory AML patients who underwent allogeneic stem cell transplantation (ASCT) at Karmanos Cancer Institute from 01/01/2012 to 01/01/2018. A planned sample size of 24 patients were selected and assigned to two cohorts based on their respective responses to ASCT: a) poor responders: persistent leukemia by bone marrow biopsy within 30 days of ASCT, and b) good responders: no evidence of leukemia at 1-year post-transplant. Descriptive analyses evaluated differences in demographic, disease and treatment variables. FISH assay was used to evaluate PDL-1 alterations in the pre-transplant archival, formalin-fixed, paraffin-embedded bone marrow biopsy specimens including frequency and magnitude of 9p24.1 alterations— low-level disomy, polysomy, copy gain, and amplification. The expression of PD-L1 was evaluated by IHC in a semi-quantitative (0 through +3) manner. Results Of the total 24 patients, 12 patients were poor responders and 12 were good responders; among poor responders, median age was 58 years (range, 31- 72) and gender distribution was 1:1. Seven patients (58%) had prior MDS, Myelofibrosis or CMML, while five patients had de-novo AML. Treatment information is provided in table 1. At pre-transplant bone marrow biopsy assessment, median number of bone marrow and peripheral blood blast counts were 8.1% (1.7-37) and 2% (0-37), respectively. Bone marrow cytogenetics assessed prior to transplant included intermediate risk in 2 (17%) and high risk disease in 9 (75%) patients and one patient without cytogenetic evaluation. PD-L1 expression was negative by IHC in all samples, but FISH testing was positive in 3 of 12 (25%) poor responders, characterized by copy gain. In good responders, median age was 59 years (range, 43-70) with a 3:2 male predominance. Four patients (25%) had prior MDS or Myelofibrosis and 8 had de-novo AML. Five patients (42%) had intermediate risk and 7 (58%) had high-risk cytogenetics. At pre-transplant bone marrow biopsy assessment, median number of bone marrow and peripheral blast counts were 6% (1-40) and 1% (0-7), respectively. PD-L1 expression was negative by both IHC and FISH examination in this cohort. Conclusion Our study demonstrates that PDL-1 expression by FISH was specifically detected in 25% of patients with poor responders following ASCT, which could possibly provide an insight into the poor outcome. A larger sample will be required to validate this observation in this population.