Clinical Comorbidity Measures and Predictive Scores in Ex Vivo T Cell Depleted Allogeneic Hematopoietic Stem Cell Transplantation

Ex vivo T cell depletion (TCD) of the graft by CD34+ selection improves the tolerability of allogeneic hematopoietic stem cell transplantation (HCT). Clinical comorbidity measures enhance the estimates of HCT tolerance and outcomes. However, clinical prediction tools such as the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) have not been validated in TCD transplants. To improve outcome prediction we therefore evaluated published comorbidity measures and other standard clinical biomarkers of outcome in a series of patients with hematological malignancies receiving myeloablative TCD transplants. Pre-transplant factors compared were: HCT-CI, European Group for Blood and Marrow Transplantation (EBMT) risk score, Eastern Cooperative Oncology Group (ECOG) Performance Status, serum C-reactive protein (CRP), albumin (ALB), pre-albumin (PAB), ferritin, absolute lymphocyte counts (ALC), and absolute lymphocyte / monocyte ratio (LMR). All patients received a myeloablative preparative regimen (Flu 125 mg/m2, Cytoxan 120mg/kg, TBI 400 cGy if older than 55 or 1200 cGy if younger than 55) followed by the infusion of a CD34+ selected graft from a HLA-identical sibling containing 1 to 5x10e4 CD3+/kg. Of the 79 recipients, 34 (43%) had standard risk disease, and 45 (57%) recipients had high-risk disease. At a median follow up of ~ 5 years, overall survival (OS) was 42.9% and nonrelapse mortality (NRM) was 32.9%. In the initial analyses we eliminated pre-transplant ECOG, ferritin, ALB, PAB, ALC and EBMT. The independent pre-transplant comorbidity variables identified for further testing were HCT-CI, CRP and LMR. We demonstrated that LMR and HCT-CI score are important independent clinical predictors of OS and NRM in TCD HCT. In univariate analysis of OS, significant factors were HCT-CI scores ≥ 5 (Hazard ratio (HR) 2.125, p= 0.018), CRP (HR=1.016, p=0.058, a trend) and LMR ≤ 1.3 (HR=2.054, p= 0.036). In multivariate models of OS, CRP (HR 1.02, p= 0.029), HCT-CI≥5 (HR 1.966, p= 0.04) and LMR ≤1.3 (HR 2.206, p= 0.029) retained significance. Further refinement of the multivariable model was made by classifying the LMR into a low-risk group (LMR 3-25) versus high-risk groups (LMR< 3 or > 25), then the significant factors were HCT-CI ≥5 (HR 3.008, p= 0.002), LMR < 3 (HR=3.292, p= 0.0014) and LMR > 25 (HR=5.648, p= 0.0008) compared to the LMR 3-25 group. In multivariable analysis of NRM, significant factors were HCT-CI ≥5 (HR 3.283, p= 0.015), LMR < 3 (HR=5.891, p= 0.0008), and LMR > 25 (HR=9.008, p= 0.0011) compared to the LMR 3-25 group. To evaluate the relevance of these factors in predicting the 5-year mortality and NRM beyond traditional risk indicators, we used the continuous net reclassification improvement (NRI) and the C-statistic (area under the receiver operating characteristic (ROC) curve). LMR was found to be the most important risk predictor for both OS and NRM (NRI was 84% and the C-statistic was improved from 0.59 to 0.68 with the addition of LMR to the multivariate model for OS; NRI was 108% and the C-statistic was improved from 0.57 to 0.72 with the addition of LMR to the multivariate model for NRM). Based on our results, we propose a simple novel predictive score for CD34+ selected myeloablative transplantation calculated by summation of scores derived from the HCT-CI [≤4 =Score 0, ≥5=Score 1] and LMR [3-25= Score 0, <3=Score 1, >25=Score 2]. The scores [range, 0 to 3] differentiate three risk groups for hematopoietic cell transplantation OS (p=0.001) and NRM (p=0.004). In conclusion, this is the first study to explore pre-transplant comorbidity measures after TCD HCT, identifying HCT-CI and LMR as critical predictors of OS and NRM.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal