Bevacizumab specifically decreases elevated levels of circulating KIT+CD11b+ cells and IL-10 in metastatic breast cancer patients

Abstract

Background: Whether bevacizumab exerts its anti-tumor properties through systemic effects beyond local inhibition of angiogenesis and how these effects can be monitored in patients, remain largely elusive. To address these questions, we investigated bone marrow-derived cells and cytokines in the peripheral blood of metastatic breast cancer patients undergoing therapy with bevacizumab.

Methods: Circulating endothelial cells (CEC), circulating endothelial progenitor (CEP) and circulating CD11b+ cells in metastatic breast cancer patients before and during therapy with paclitaxel alone (n = 11) or in combination with bevacizumab (n = 10) were characterized using flow cytometry, real time PCR and RNASeq. Circulating factors were measured by ELISA. Aged-matched healthy donors were used as baseline controls (n = 12).

Results: Breast cancer patients had elevated frequencies of CEC, CEP, TIE2+CD11b+ and KIT+CD11b+ cell subsets. CEC decreased during therapy, irrespective of bevacizumab, while TIE2+CD11b+ remained unchanged. KIT+CD11b+ cells decreased in response to paclitaxel with bevacizumab, but not paclitaxel alone. Cancer patients expressed higher mRNA levels of the M2 polarization markers CD163, ARG1 and IL-10 in CD11b+ cells and increased levels of the M2 cytokines IL-10 and CCL20 in plasma. M1 activation markers and cytokines were low or equally expressed in cancer patients compared to healthy donors. Chemotherapy with paclitaxel and bevacizumab, but not with paclitaxel alone, significantly decreased IL-10 mRNA in CD11b+ cells and IL-10 protein in plasma.

Conclusions: This pilot study provides evidence of systemic immunomodulatory effects of bevacizumab and identified circulating KIT+CD11b+ cells and IL-10 as candidate biomarkers of bevacizumab activity in metastatic breast cancer patients.

Keywords: IL-10; KIT; angiogenesis; breast cancer; monocytes.

Figure 1. Increased frequency of CEC and CEP in the blood of metastatic breast cancer patients

Quantification of (A) CEP, (B) CEC, (C) CEP expressing VEGFR2, (D) dying CEP using 7AAD staining, and (E) dying CEC using 7AAD staining in the blood of healthy donors and metastatic breast cancer patient before therapy start. (F) Quantification of CEC and CEP under chemotherapy ± bevacizumab treatments in the blood of metastatic breast cancer patients. Cell quantifications were performed by flow cytometry and data are represented as mean +/− SD.

Figure 2. Increased frequency of TIE2⁺CD11b⁺ and KIT⁺CD11b⁺ cells in the blood of metastatic breast cancer patients and decreased frequency of KIT⁺CD11b⁺ cells by bevacizumab therapy

Quantification of CD11b⁺ cells expressing (A) VEGFR1, (B) JAM1, (C) TIE2 and (D) KIT in the blood of healthy donors and breast cancer patients before therapy start. Quantification under both chemotherapy treatments of CD11b+ cells expressing (E) VEGFR1, (F) JAM1, (G) TIE2 and (H) KIT in the blood of metastatic breast cancer patients. Cell quantifications were performed by flow cytometry and data are represented as mean +/− SD.

Figure 3. CD11b⁺ myelomonocytic cells from metastatic breast cancer patients and healthy donors show different expression profiles

(A) PCA plot representing differential clustering of cancer patients (C1-4) and healthy donors (H1-4) based on mRNA expression profiles of CD11b⁺ cells. (B) Self-organizing heat-map of the top 100 genes with greatest variability across all samples, showing different expression profiles in CD11b⁺ cells from cancer patients compared to healthy donors. (C) Gene ontology analysis showing the most up-regulated biological processes comparing cancer patients and healthy donors. (D) Gene ontology analysis showing the most down-regulated biological processes comparing cancer patients and healthy donors.

Figure 4. Evidence for a M2 activation phenotype of CD11b⁺ cells in the blood of metastatic breast cancer patients

(A) Quantification of the mRNA expression levels of M2 markers CD163, (B) ARG1 and (C) IL-10 in CD11b+ cells derived from the blood of healthy donors compare to cancer patients before therapy start. (D) Quantification of mRNA expression levels of M1 markers CCR7, (E) CD86 and (F) IL-12α at mRNA level in CD11b⁺ cells of the same healthy donors and cancer patients. Analysis was performed by real time qPCR. All data are represented as mean +/− SD.

Figure 5. Increased expression of the M2 cytokines IL-10 and CCL20 in the plasma of metastatic breast cancer patients and decrease of IL-10 levels by bevacizumab

(A) Quantification of M1 cytokines TNFα, IL-12p70 and CCL2 in plasma of healthy donors and cancer patients before therapy start. (B) Quantification of M2 cytokines IL-10 and CCL20 in the plasma of the same healthy donors and cancer patients before therapy start. (C) Quantification of CD163, (D) ARG1 and (E) IL-10 mRNA expression variation in CD11b⁺ cells of cancer patient during therapy with paclitaxel ± bevacizumab using real time qPCR. (F) Quantification of IL-10 in the plasma of cancer patients during therapy with paclitaxel ± bevacizumab. Cytokines were measured by ELISA. All data are represented as mean +/− SD.