Introduction: Chronic Lymphocytic Leukemia (CLL) is a malignancy characterized by B-lymphocytes with aberrant expression of CD5. In normal T-lymphocytes, CD5 is an important regulator of T-cell receptor signaling. In CLL, CD5 acts as a repressor of B-cell receptor (BCR) signaling, whereby phosphorylated CD5 anchors SHP-1 at the cell membrane, resulting in inhibition of BCR-mediated second messenger signaling. The significance of CD5 on outcomes of CLL patients has not been evaluated. Because BCR signaling is an oncogenic stimulus in CLL, we hypothesized that higher levels of CD5 may be associated with superior clinical outcomes.
Methods: Median fluorescence intensity (MFI) of CD5 on CLL cells was determined by flow cytometry. CLL samples and clinical data were obtained from patients enrolled in IRB-approved protocols at the Duke University and Durham VA Medical Centers. Molecular prognostic markers were measured as described previously. Descriptive and time to event (Cox proportional hazard and Kaplan-Meier) statistical analyses were performed in the statistical environment, R.
Results: Between March 2011 and April 2015, 961 CLL samples were evaluated from 352 unique CLL patients. One to 10 samples were evaluated per CLL patient. The mean CD5 MFI was calculated for each unique patient, with a median of 206.59 (range 0.50 - 768.31) and median standard deviation of 33.02 (range 0.09 - 315.46) for CLL patients with multiple samples collected. 150 patients received therapy (44%), and there was no significant difference in CD5 MFI mean or standard deviation between patients who received and did not receive therapy. CD5 MFI values did not significantly differ among CLL demographic or prognostic groups. Cox proportional hazard ratio analyses showed that higher CD5 MFI is associated with longer time to therapy (TTT, p = 0.037), but not overall survival. Based on the distribution of CD5 MFI, we divided the CLL cohort into thirds, with low CD5 defined as MFI < 137, high CD5 defined as MFI > 283.5, and medium CD5 with MFI between these values. Kaplan-Meier analyses demonstrated a significant difference in TTT between these three groups (p = 0.004), with higher CD5 MFI associated with longer TTT. Moreover, CD5 MFI added to established molecular prognostic markers to risk stratify CLL patients (p < 0.0001 for CD38 and IGHV, p = 0.0003 for FISH, and p = 0.004 for ZAP70). Importantly, for those with good prognostic markers such as IGHV mutation or low CD38 expression, lower CD5 identified patients with shorter TTT (p = 0.01 for IGHV mutated; p = 0.0009 for CD38 negative). However, TTT was not significantly different in unmutated IGHV CLL patients or CD38 positive CLL patients with varying degrees of CD5 expression.
Conclusions: Surface CD5 expression varies among CLL patients. For most individual patients, there appears to be a low level of variability in CD5 expression, regardless of CLL-directed therapy. Higher CD5 expression is associated with superior clinical outcomes in CLL, consistent with prior in vitro determination of CD5 as a negative regulator of BCR signaling. The relevance of CD5 in risk-stratifying IGHV mutated and CD38 negative CLL patients is of particular relevance, since these subgroups of CLL are particularly dependent on BCR-signaling as an oncogenic process. These findings are important both for prognostication and for development of therapies for this group of CLL patients.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.