Comorbidity Measures In Ex Vivo T Cell Depleted Allogeneic Hematopoietic Stem Cell Transplantation (HCT)

Clinical comorbidity measures enhance the estimates of HCT tolerance and outcomes, thereby aiding therapeutic decisions. Ex vivo T cell depletion (TCD) of the graft reduces the risk of graft-versus-host disease and improves the tolerability of allogeneic transplantation in patients with impaired pretransplant performance. However, prediction tools such as the hematopoietic cell transplantation-specific comorbidity index (HCT-CI), have only been validated in conventional T-replete HCT. To improve outcome prediction in TCD transplants we therefore evaluated published comorbidity measures and other potential biomarkers of outcome in a series of myeloablative TCD transplant recipients. Pre transplant (week -2) factors measured were: HCT-CI, ECOG performance status, serum C-reactive protein (CRP), albumin (ALB), pre-albumin (PAB), ferritin, absolute lymphocyte counts (ALC), and absolute lymphocyte / monocyte ratio (LMR). CRP was also studied serially post transplant. CRP increased after conditioning, peaked (p =0.0001) in the first week of HCT, with recovery to baseline at 5 weeks post-HCT. We evaluated outcomes in a cohort of 79 patients in our institute with hematological malignancies receiving myeloablative total-body irradiation, and an HLA-identical sibling peripheral blood HCT, T cell depleted using Miltenyi CD34+ selection. The median age of recipients was 43 years (range 13-68 years). Of the 79, 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 33.9%. Comorbidity measures were first screened to eliminate highly-correlated covariates. Univariate Cox regression models were used to identify significant factors (p<0.05) associated with OS and NRM (with relapse as a competing risk), which were then further evaluated by multivariable Cox regression models. In the initial analysis we eliminated pre-HCT ECOG (94% were ECOG 0) and ferritin (median 1408 mcg/L, range 11-1444), which were highly correlated (p =0.003) to the HCT-CI score (median 3, range 0-9). Similarly, ALB (median 3.9 g/L, range 2.1-5.1, p = 0.002), PAB (median 25.6 mg/dL, range 6.9-56, p =0.001) and ferritin levels (p =0.005) were excluded as highly correlated to the pre-HCT CRP (preCRP) (median 5.8 mg/L, range 0.92-96.8). CRP levels measured at 2 weeks post transplant (postCRP) (median 10.4 mg/L, range 1.9-180) and ALC (median 0.96 K/uL, range 0.02-104.6) were not correlated to the pretransplant CRP. The independent pretransplant comorbidity variables identified for further testing were HCT-CI, preCRP, post CRP, ALC and LMR. In univariable analysis of OS, significant co-variates were HCT-CI scores ≥ 5 (HR 2.09 p= 0.02), preCRP (HR=1.016, p=0.07, a trend), postCRP (HR=1.014, p<0.001) and LMR ≤ 1.3 (HR=2.04, p= 0.04). In multivariable models of OS, postCRP ≥ 15 (HR 2.39, p =0.009) and LMR ≤ 1.3 (HR 2.25, p =0.04) retained significance. Confining the model to pretransplant data, then significant factors were pre CRP (HR 1.024 p= 0.02), HCT-CI≥5 (HR 2.08 p= 0.03) and LMR ≤1.3 (HR 2.43 p= 0.02). In univariable analysis of NRM, only continuous postCRP and postCRP ≥10 were significantly associated with NRM (HR=2.5, p =0.03) and in multivariable modeling of NRM, only postCRP ≥10 (HR=2.57, p=0.04) or continuous postCRP (HR=1.018, p=0.004) was found to be significant. LMR ≤ 1.3 and ECOG > 0 were found to be significantly associated with the cumulative incidence of relapse (both p=0.01). In conclusion, this is the first study to explore comorbidity scores and biomarkers to predict outcome after ex vivo TCD HCT. Our results suggest that HCT-CI score, preCRP, postCRP and the LMR are important independent clinical predictors of OS and NRM in TCD HCT.