Does Early Myeloid Recovery Predict Superior Survival in Children and Young Adults Undergoing Hematopoietic Stem Cell Transplantation?.

Introduction: Slow rate of myeloid recovery at the end of induction chemotherapy reflected by a low absolute neutrophil count (ANC) is highly predictive of relapse in children with ALL (Laughton et al, 2005). Whether rate of myeloid recovery is predictive of outcome post-hematopoietic stem cell transplantation (HSCT) in children and young adults has not been studied.

Objective: To explore the association between early ANC recovery and survival in children and young adults undergoing allogeneic and autologous HSCT for leukemia or lymphoma.

Methods: Retrospective chart review was performed for all consecutive patients (age 6 months to 21 years) undergoing first HSCT for hematologic malignancies and lymphomas at Mayo Clinic, Division of Pediatric Hematology/Oncology from 1995 to 2004. Day 15 and day 28 ANC were calculated for all patients. The primary end point was 1-year relapse free (RFS) and overall survival (OS). To explore the association between ANC and primary outcome, patients were stratified in 2 groups: ≥ or < than median ANC and Kaplan Meier survival curves were generated for both the groups and compared with the log-rank test. Matched related donor (MRD), matched unrelated donor (MUD) and haploidentical (as one group) and autologous transplantations were analyzed separately.

Results: There were a total of 50 patients. 28 received MRD, 2 MUD, 10 haploidentical and 10 autologous. The underlying malignancies were (AML n =25, Biphenotypic Leukemia n=1, ALL n =7, NHL n=3, HD n=7, JMML n=5, CML n=2, CLL n=1). No patient was lost to follow up during that time and data was available for all except (day 28 ANC) for one patient. The 1-year RFS and OS rate was the same; 81% for the MRD, 50% for MUD and haploidentical and 70% for autologous transplant populations, respectively. In patients receiving MRD transplants, higher ANC at either day 15 or day 28 post transplantation was associated with better, albeit statistically non-significant, RFS/OS rates at 1 year (86% for ≥ median ^(0.68) vs. 76% for < median p=0.44 and 92% for ≥ median ^(2.2) vs. 72% for < median p= 0.18 respectively). Opposite results were observed for MUD and Haploidentical transplantation group (33% for ANC ≥ median ^(0.47) at day 15 vs. 66% for < median p=0.28, 43% for ANC ≥ median ^(1.9) at day 28 vs. 60% for < median p=0.48). For autologous transplant group, higher ANC at day 15 was associated with better, albeit statistically non-significant, RFS/OS rates at 1 year (80% for ANC ≥ median ^(0.42) vs. 60% for < median p=0.60) while opposite results were observed on day 28 (60% for ANC ≥ median ^(0.46) vs. 75% for < median p=0.54).

Conclusions: Our data generated conflicting findings of the association between ANC and RFS/OS in different patient populations undergoing HSCT. However, our results are limited by the small sample size. Further collaborative studies are needed to explore this association.