Red cell distribution width (RDW) is an indicator of the variability in the size of circulating erythrocytes (anisocytosis); different conditions can increase the RDW levels; such as hemolysis, ineffective erythropoiesis and blood transfusions. Recently, different studies have shown an association between increased levels of RDW and inflammation in different diseases, being proposed as a surrogate marker of inflammation and a strong predictor of adverse outcome. The proposed mechanism of this association departs from the finding that Inflammatory cytokines like TNFand IL-6 (part of the classic inflammatory cascade), have been found to inhibit erythropoietin-induced erythrocyte maturation, which is reflected in the RDW increase.

Some reports have found a relationship between RDW and mortality related to age or several malignant or non-malignant conditions. However, there is no information about the role of RDW in overall survival (OS) of patients with DLBCL. We aim to evaluate the prognostic role of RDW levels in DLBCL patients at diagnosis.

METHODS

We retrospectively evaluated 83 patients with DLBCL homogenously treated in frontline with R-CHOP from 2002 to 2013 in the Son Espases University Hospital. To avoid selection bias patients were obtained from Pharmacy and Pathology Departments registries. Main clinical and prognostic factors at diagnosis were obtained from medical records. Cheson criteria were used for response assessment. The RDW was collected from the hemogram at diagnosis. The IBM SPSS STADISTICS program was used for all statistical analyses. PFS (time to progression/relapse) and overall survival (OS) (time to death) were measured from the date of ABVD onset, and were estimated according to the Kaplan-Meier method. We performed the comparisons between those interest variables with the log-rank test. A comparison between categorical variables was made with the chi-square of Fisher's exact test, as appropriate. All reported P-values were two-sided, and statistical significance was defined at P<0.05. For selecting cutoff values in RDW we used ROC curves.

RESULTS:

Main characteristics of patients were as follows: median age was 62 (20-86) years, 24% had ECOG PS>1, 64% advanced III-IV Ann Arbor (AA) stage, 39% B-symptoms, 51% adjusted-International Prognostic Index (a-IPI) and 39% belong to the high risk (3-5) subgroups of R-IPI Median RDW was 14.6 (11.1-21.1). Using ROC curves we selected the cutoff 14.05 for the death event. We evaluated the association of increased RDW with main prognostic factors at diagnosis. RDW >14.05 at diagnosis was associated with a more advanced age, worse ECOG PS, a more advanced AA stage, higher incidence of B symptoms and IPI>2. However, RDW was not related to disease control in terms of response to therapy (p=0.39) or relapse/progression (p=0.21) rates. Inversely, RDW>14.05 was in fact associated to a higher mortality (47%) compared to only 17% in patients with RDW≤14.05 (p=0.008). Median follow-up was 77 (20-137) months. Univariate survival analysis showed age>60 years (p=0.001), ECOG PS>1 (p=0.036), high risk R-IPI (p=0.005), a higher than 15% reduction in relative dose-intensity (RDI) (p=0.026) and RDW>14.05 (p=0.008) were significantly related to worse OS. By contrast, RDW did not significantly influence progression-free survival (p=0.19).

CONCLUSIONS:

Higher RDW at diagnosis in this series of DLBCL patients was related with older age, worse ECOG PS and more advanced disease but this was not translated into a worse control of disease in terms of only a small non statistically significant impact in response or PFS. By contrast higher RDW was linked to a significantly higher mortality and worse OS possibly related to a higher proinflammatory basal status and comorbidities. Patients with higher RDW may be at risk of reduction in RDI. These findings could justify including RDW in scores of comorbidities in DLBCL as well as in other malignant and non-malignant conditions.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution