Expression of M2 macrophage markers YKL-39 and CCL18 in breast cancer is associated with the effect of neoadjuvant chemotherapy.

PURPOSE: High activity of enzyme TOP2a in tumor cells is known to be associated with sensitivity to anthracycline chemotherapy, but 20% of such patients do not show clinical response. Tumor microenvironment, including tumor-associated macrophages (TAM), is an essential factor defining the efficiency of chemotherapy. In the present study, we analyzed the expression of M2 macrophage markers, YKL-39 and CCL18, in tumors of breast cancer patients received anthracycline-based NAC. METHODS: Patients were divided into two groups according to the level of doxorubicin sensitivity marker TOP2a: DOX-Sense and DOX-Res groups. Expression levels of TOR2a, CD68, YKL-39 and CCL18 genes were analyzed by qPCR, the amplification of TOR2a gene locus was assessed by the microarray assay. Clinical and pathological responses to neoadjuvant chemotherapy were assessed. RESULTS: We found that the average level of TOP2a expression in patients of DOX-Sense group was almost 10 times higher than in patients of DOX-Res group, and the expression of CD68 was 3 times higher in the DOX-Sense group compared to DOX-Res group. We demonstrated that expression levels of M2-derived cytokines but not the amount of TAM is indicative for clinical and pathological chemotherapy efficacy in breast cancer patients. Out of 8 patients from DOX-Sense group who did not respond to neoadjuvant chemotherapy (NAC), 7 patients had M2+ macrophage phenotype (YKL-39+CCL18- or YKL-39-CCL18+) and only one patient had M2- macrophage phenotype (YKL-39-CCL18-). In DOX-Res group, out of 14 patients who clinically responded to NAC 9 patients had M2- phenotype and only 5 patients had M2+ macrophage phenotype. Among pathological non-responders in DOX-Sense group, 19 (82%) patients had M2+ tumor phenotype and only 4 (18%) patients had M2- phenotype. In DOX-Res group, all 5 patients who pathologically responded to NAC had M2 phenotype (YKL-39-CCL18-). Unlike the clinical response to NAC, the differences in the frequency of M2+ and M2- phenotypes between pathologically responding and non-responding patients within DOX-Sense and DOX-Res groups were statistically significant. CONCLUSIONS: Thus, we showed that in patients with breast cancer who received anthracycline-containing NAC the absence of clinical response is associated with the presence of M2+ macrophage phenotype (YKL-39-CCL18 + or YKL-39 + CCL18-) based on TOP2a overexpression data.

Deletions of multidrug resistance gene loci in breast cancer leads to the down-regulation of its expression and predict tumor response to neoadjuvant chemotherapy.

Neoadjuvant chemotherapy (NAC) is intensively used for the treatment of primary breast cancer. In our previous studies, we reported that clinical tumor response to NAC is associated with the change of multidrug resistance (MDR) gene expression in tumors after chemotherapy. In this study we performed a combined analysis of MDR gene locus deletions in tumor DNA, MDR gene expression and clinical response to NAC in 73 BC patients. Copy number variations (CNVs) in biopsy specimens were tested using high-density microarray platform CytoScanTM HD Array (Affymetrix, USA). 75%-100% persons having deletions of MDR gene loci demonstrated the down-regulation of MDR gene expression. Expression of MDR genes was 2-8 times lower in patients with deletion than in patients having no deletion only in post-NAC tumors samples but not in tumor tissue before chemotherapy. All patients with deletions of ABCB1 ABCB 3 ABCC5 gene loci--7q21.1, 6p21.32, 3q27 correspondingly, and most patients having deletions in ABCC1 (16p13.1), ABCC2 (10q24), ABCG1 (21q22.3), ABCG2 (4q22.1), responded favorably to NAC. The analysis of all CNVs, including both amplification and deletion showed that the frequency of 13q14.2 deletion was 85% among patients bearing tumor with the deletion at least in one MDR gene locus versus 9% in patients with no deletions. Differences in the frequency of 13q14.2 deletions between the two groups were statistically significant (p = 2.03 × 10(-11), Fisher test, Bonferroni-adjusted p = 1.73 × 10(-8)). In conclusion, our study for the first time demonstrates that deletion MDR gene loci can be used as predictive marker for tumor response to NAC.