CCR2 Expression Signature Can Classify and Predict Outcome in a Subpopulation of Chronic Lymphocytic Leukemia (CLL) Patients

Despite the recent advances in chronic lymphocytic leukemia (CLL) treatment with the development of novel targeted agents such as Bruton's tyrosine kinase (BTK) inhibitor ibrutinib and the BCL2 antagonist Venetoclax, the disease remains incurable for most patients, and resistance mechanisms to these novel agents have already been identified. Therefore, identifying new therapeutic vulnerabilities to treat and prevent resistant tumors is of great importance.

Stromal cells provide CLL cells a protective niche which increases resistance to drug-induced apoptosis. Therefore, a current challenge is to develop new strategies by targeting not only CLL cells, but also the microenvironment. We found that soluble factors can recapitulate much of the protective function of stromal cells. To understand the role of microenvironmental factors on CLL cells, we took advantage of the conditioned media (CM) collected from the stromal cell line NK-tert, which protects CLL cells from spontaneous cell death. First, we identified three major cytokines (IL6, IL8, CCL2) from the conditioned media using Proteome Profiler Human XL Cytokine Array. Next, we evaluated the effect of each cytokine on peripheral blood (PB) CLL survival (n=5, treatment naïve, unselected regarding genetic background). CLL cells were cultured ex-vivo in CM and treated with neutralizing antibody against either Il6, IL8 or CCL2 for 48hrs and found a significant decrease in cell viability in presence of anti-CCL2 antibody (median % decrease in cell viability over isotype control is 50%, p<0.0001). Similarly, we also found an increase in cell viability when CLL cells were cultured in RPMI in presence of recombinant-CCL2 protein than those are cultured in RPMI only, suggesting that CCL2 has an important role in CLL survival.

BH3 profiling quantifies apoptotic signaling based on cytochrome c loss as a measure of mitochondrial outer membrane permeabilization (MOMP) by flow cytometry in response to BH3-only synthetic peptides that mimic pro-apoptotic BCL-2 family proteins, applied to digitonin permeabilized cells. We performed BH3 profile on PB derived -CLL samples (n=5, treatment naive) cultured in CM and exposed to either isotype control or anti-CCL2 antibody. Anti-CCL2 treated CLL samples showed an increase in MOMP in response to BIM, PUMA, MS1 and HRK peptides compared to CLL samples exposed to isotype control. This suggests that CCL2 suppresses apoptotic signaling in CLL cells. Furthermore, Dynamic BH3 profiling (DBP), a measure of drug-induced apoptotic signaling as previously described (Montero et al., Cell, 2015) showed that a combination of anti-CCL2 and venetoclax increased dependence on BCL-2 (BAD peptide)MCL-1(MS1 peptide) as well as BCL-xL (HRK peptide) in CLL cells.

To our surprise, we also found a subgroup of treatment naive PB-CLL samples (CM independent) that undergo minimal spontaneous cell death in the absence of stroma support. In addition to this, IL-6, IL8, CCL2 neutralizing antibodies showed no effect on their survival. These CM independent CLL samples(n=5) are less sensitive to venetoclax treatment compared to the CLL samples that need stroma support (CM dependent). Moreover, baseline BH3 profile showed that CM independent CLL cells are less primed for apoptosis compared to CM dependent CLL cells, and exposure to venetoclax and anti-CCL2 antibody did not increase any drug-mediated apoptotic priming. Next, we checked mRNA expression level of the CCL2 receptor, CCR2 from CM dependent and CM independent CLL samples and found that CM dependent CLL cells have higher expression of CCR2, suggesting a distinct signature discriminating CM dependent and independent patients.

In summary, our study shows that CCL2 may serve as a novel biomarker to target in CLL, especially in relapsed/refractory patients.