Assessment of a Disease Risk Index in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation After Fludarabine and Pharmacokinetically Dose-Targeted Intravenous Busulfan: Effect On Overall and Progression-Free Survival

Background: Several methods are employed to predict the outcome of patients undergoing allogeneic hematopoietic cell transplantation (HCT), including patient characteristics (e.g., age, performance status, concurrent comorbidities), diagnosis, and disease status. Armand, et al. (Blood 2012;120:905-913) validated a disease risk index using diagnosis, cytogenetics, and disease status, which predicted for overall and progression-free survival. To test this disease risk index in our patient population, we applied it to patients receiving fludarabine and pharmacokinetically dose-targeted intravenous busulfan myeloablative conditioning followed by allogeneic peripheral blood HCT.

Methods: We analyzed a cohort of 322 patients transplanted between 2004 and 2009 and assigned them to disease (low, intermediate, and high) and stage (low and high) risk groups per the Armand classification. Overall and progression-free survival were compared between the overall risk groups (low, intermediate, high and very high).

Results: Median age (range) was 49.4 (19-73.8) years. 139 (43%) patients received matched sibling grafts, 120 (37%) from matched unrelated donors and 63 (20%) from unrelated one HLA-antigen/allele mismatched donors. AML was the most common diagnosis (40%); 31% of all patients were in CR1. HCT-CI scores were 0 (13%), 1-2 (32%) and 3 or greater (55%). 42% of patients had a Karnofsky performance status of 100%. 89% of patients were targeted to a busulfan AUC of 5300 mM*min. GVHD prophylaxis consisted of tacrolimus and methotrexate in 77% of patients. Disease risk by CIBMTR classification was Early 42%, Intermediate 34%, and Advanced 24%. Disease, stage, and overall risk groups, according to the criteria set forth by Armand, et al. as well as the corresponding overall (OS) and progression free survivals (PFS) are shown in the table::

Disease risk	Stage risk	% of patients	Overall risk	OS @ 3 yrs	PFS @ 3 yrs
Low	Low	11%	Low	46%	47%
Low	High	4%	Intermediate	45%	38%
Intermediate	Low	45%
Intermediate	High	15%	High	39%	34%
High	Low	18%
High	High	7%	Very High	33%	29%
p value				0.34	0.08

 Conclusion: Our outcomes were different in the Low and Very High risk groups reported by Armand, et al. This may be accounted for by a different distribution of diseases or stages within each overall risk category. Variations in other confounding factors also likely contribute to the disparate results. Further analyses of these differences will need to be done to evaluate the validity of this disease risk index in our patients.