Prognostic Impact of Pre-Transplant Serum C-Reactive Protein in Patients Receiving a Single-Unit Unrelated Cord Blood Transplantation

Background: C-reactive protein (CRP) is an acute-phase protein and its serum level rises in response to inflammation. There have been some reports about the relationship between the serum CRP level before allogeneic hematopoietic stem cell transplantation (HSCT) and transplant outcomes. This study particularly evaluated the impact of pre-transplant CRP on the outcome in patients who received a single-unit unrelated cord blood transplantation (UCBT) for hematologic malignancies.

Patients and methods: Adult patients receiving UCBT as first HSCT for hematologic malignancies between April 2009 and April 2016 at Kanagawa Cancer Center were retrospectively analyzed in this study. Serum CRP was measured by a latex agglutination assay (normal range 0.0-0.3 mg /dl). The time of measurement was within a few days before the conditioning procedure. Patients were divided into 2 groups according to the serum CRP levels (normal CRP group [≤0.03 mg/dl] and high CRP group [> 0.03 mg/dl)]). A total of 80 patients (pts) included 36 pts with acute myeloid leukemia (AML), 27 pts with acute lymphoid leukemia (ALL), 13 pts with myelodysplastic syndrome (MDS), and 4 pts with others. The median age was 59 years (range, 20-68 years). There were 45 males and 35 females. A disease risk at transplantation indicated a standard risk in 47 pts and a high risk in 33 pts. Myeloablative conditioning (MAC) was employed for 25 pts and reduced intensity conditioning (RIC) was for 55 pts. Data were analyzed with EZR statistical software.

Results: The median level of serum CRP at pre-transplant was 0.17 mg/dl (range: 0.01-12.93). No. of patients in normal and high CRP levels was 56 pts and 24 pts, respectively. Disease high risk was significantly associated with a high level of serum CRP. Other pre-transplant factors such as age, gender, diagnosis, comorbidity index, and serum ferritin levels were not related with serum CRP levels. The high incidence of gradeⅡ-Ⅳacute GVHD was significantly associated with high CRP group compared to normal CRP group (55% vs. 19%, p = 0.001). After a median follow-up of 16 months (range: 1-86 months), 3-year overall survival (OS) was 62%. The cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) at 3 years was 26% and 14%, respectively. According to univariate analysis, factors associated with worse 3-year OS included disease high risk at transplantation (vs. low risk; 49% vs. 70%, p = 0.025) and high CRP levels (vs. normal CRP levels; 28% vs. 77%, p < 0.001). Higher 3-year CIR was significantly correlated with high CRP levels (vs. normal CRP levels; 46% vs. 18%, P = 0.009). Adverse factors for NRM was high scores (≥3) of HCT-CI (vs. low scores (≤2) of HCT-CI; 25% vs. 5%, p=0.029). Multivariate analysis showed that high CRP levels (hazard ratio [HR], 4.13; 95% confidence interval [CI], 1.76-9.63; p = 0.001) was an independent determinant of 3-year OS. Prognostic factors related with CIR and NRM was high CRP levels (HR, 2.63; 95%CI, 0.96-7.21; p = 0.059) and high scores of HCT-CI (HR, 4.25; 95%CI, 0.89-20.29; p = 0.069) respectively, but those difference did not reach statistical significance.

Conclusions: Pre-transplant serum CRP might be a useful biomarker for predicting the transplant outcome and the development of acute GVHD following UCBT.