Background: Acute leukemia is the most common pediatric malignancy, and high risk or relapsed disease accounts for the most frequent need for pediatric hematopoietic stem cell transplant (HSCT). The 5-year Event Free Survival for acute leukemia following HSCT approaches 55%, and we currently cannot predict which children will relapse following transplant. Prognostic determinants for allogeneic HSCT recipients are limited to risk status of underlying disease, type of transplant received (matched sibling donor vs. unrelated donor), and severity of graft vs. host disease (GVHD). Although some clinicians believe that an increase in RRT may actually reflect a better outcome, the prognostic significance of engraftment and regimen related toxicity (RRT) remains to be investigated. Patients receiving similar preparative regimens respond with varying degrees of toxicity which may reflect individual genetic variations and ultimately may determine their treatment outcome.

Objective: To determine whether engraftment or RRT correlates with outcome (sustained remission vs. relapse) in pediatric HSCT patients with acute leukemia.

Methods: A chart review was conducted on 96 consecutive pediatric allogeneic HSCT recipients with acute leukemia at Stanford University from 1996–2006 treated with TBI/ VP or TBI/CY. The previously validated Common Terminology Criteria for Adverse Events v3.0 (CTCAE) was used to assess stomatotoxicity, hepatoxicity, and nephrotoxicity as indicators of RRT based on a 5-point Likert scale (0=no toxicity, 5=death). Toxicity scores were recorded within a 15 and 30 day period following HSCT to minimize toxicity attributed to GVHD. Time to hematological recovery also was assessed. An independent sample, two-tailed t-test was used to compare children in different outcome groups (relapse vs. sustained complete remission [CR]) for both combined total acute leukemia (TOTAL) and separate subtypes (ALL and AML). Finally, the percentage of patients with CR was calculated above and below the mean for any significant variables. Patients who died of treatment-related mortality (TRM) were removed from the analyses regardless of their remission status to avoid confounding.

Results: Data were collected on 96 patients (ALL=53, AML=43). Twenty-one patients with TRM were removed, and the following number of HSCT patients analyzed: 75 TOTAL (CR=53, Relapse=22), 44 ALL (CR=29, Relapse=15), and 31 AML (CR=24, Relapse=7). CR patients took longer time to reach platelet recovery (P100) compared to relapsed pt’s for TOTAL (111 vs. 45 days, p=0.007), ALL (110 vs. 47 days, p=0.055), and AML (112 vs. 40 days, p=0.041). The TOTAL CR group also had higher creatinine-15 day scores (0.70 vs. 0.31, p=0.033) as did the AML CR group (0.83 vs. 0.28, p=0.05). This creatinine difference was no longer significant by Day-30. The percentage of CR patients above the mean P100 time to platelet recovery for each category was: TOTAL 89.5% (n=17/19, mean 97 days), ALL 82% (n=9/11, mean 92 days), and AML 100% (n=8/8, mean 104 days). The percentage of CR patients with creatinine-15 day scores above the mean was: TOTAL 79% (n=23/29, mean score 0.59) and AML 86% (n=12/14, mean score 0.7). The difference in denominators for each group is due to removal of patients without available data for specific variables or death due to TRM.

Conclusion: This study represents one of the first attempts to methodically investigate whether engraftment and RTT are related to outcome in pediatric HSCT patients. Our retrospective analysis found delayed platelet engraftment (defined by platelet level of 100K/uL x 3 consecutive days) and higher creatinine-15 day values (less than 1.5 x normal) in leukemia patients who achieved sustained remission vs. those who ultimately relapsed. The difference in RRT scores between children who relapsed and those who did not may reflect differences in bone marrow and kidney exposure to previous chemotherapeutic agents; however, another explanation may be individual genetic variation in drug metabolism which leads to rapid drug clearance with lower RRT and subsequent relapse. Prospective studies are needed to validate our findings, in addition to high-throughput genomic screening to identify biological differences between children with different engraftment and RRT scores.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author