High Hematopoietic Transplantation Comorbidity Index Is Not Associated with Increased Transplant Related Mortality: Review of a Large Cohort of Pediatric Patients Undergoing Allogeneic Stem Cell Transplantation Following Busulfan-Based Regimens for Malign

Introduction: Identification of patients at high risk for transplant related mortality (TRM) prior to allogeneic hematopoietic cell transplantation (alloHCT) is crucial. However, the heterogeneity of alloHCT indications, conditioning regimens, patient characteristics and center-specific practices makes defining generalizable risk factors for TRM a perpetual challenge. In a retrospective study, Smith et al. (Blood, 2011) concluded that Sorror's 17-item pre-alloHCT adult comorbidity index was also an effective tool for risk-stratifying pediatric alloHCT patients. In this study, we attempt to re-validate the HCT-CI published by Smith et al. in a cohort of patients receiving well- defined Busulfan (Bu)-based conditioning regimens.

Methods: Retrospective cohort study of 165 patients (<22y) who underwent alloHCT. Patients conditioned with 1 of 3 Bu-based regimens were included: reduced intensity (RIC): Bu (6.4-8mg/kg)+ Fludarabine (Flu) [150mg/m2), reduced toxicity (RTC): Bu (12.8-16mg/kg)+ Flu (180mg/m2) and myeloablative (MAC): Bu (12.8-16mg/kg)+ cyclophosphamide (120-200mg/kg) or melphalan (135mg/m2). 

Results: Median age at alloHCT was 9.5y (0.1-22y), malignant group n=102, non-malignant group n=63. Pre-alloHCT comorbidities were scored using HCT-CI and risk group numbers were as follows:  32 (19.4%) group 0, 69 (41.8%) group 1-2, 64 (38.8%) group ³3. The difference in incidence of TRM at 1y in each HCT-CI group was not statistically significant for malignant vs non-malignant patients.  1yr TRM of patients receiving RIC, RTC and MAC regimens was also not significantly impacted by pre-transplant HCT-CI scores in each of these cohorts. Data outlined in supplemental tables.

Conclusion: Preliminary results suggest that the HCT-CI as it stands may not be a generalizable risk stratification tool for all pediatric alloHCT patients. Contrary to aforementioned studies, we were unable to demonstrate poorer outcomes in patients with HCT-CI score ³3. One potential reason could be a smaller sample size in our study (n=166 vs. 252). We excluded total body irradiation as it is only used in MAC regimens at our center which might have skewed the comparison between RIC/RTC vs. MAC thus making findings generalizable. Further analysis is currently being conducted to identify risk factors distinct from the HCT-CI that may have a higher impact on TRM in pediatric patients undergoing alloHCT.

HCT-CI Score	1y TRM All Patients			p-value
	Number	Deceased	Living
0	32	9 (28.1%)	23 (71.9%)	0.299
1-2	69	19 (27.5%)	50 (72.5%)
³3	64	11 (17.2%)	53 (82.8%)
All	165	39 (23.6%)	126 (76.4%)

HCT-CI Score	1y TRM: Conditioning Regimens			p-value
	RIC	RTC	MAC
0	16.7% (1/6)	0% (0/11)	20% (3/15)	0.376
1-2	5% (1/20)	13% (3/23)	15.4% (4/26)	0.576
³3	0% (0/18)	10.7% (3/28)	11.1% (2/18)	0.576

1yr TRM: HCT-CI Score Comparisons
	p-value
0 vs 1-2	0.180
1-2 vs³3	0.238
³3	0.706