Taxanes enhance trastuzumab-mediated ADCC on tumor cells through NKG2D-mediated NK cell recognition

Abstract

Recent clinical data indicate a synergistic therapeutic effect between trastuzumab and taxanes in neoadjuvantly treated HER2-positive breast cancer (BC) patients. In HER2+ BC experimental models and patients, we investigated whether this synergy depends on the ability of drug-induced stress to improve NK cell effectiveness and thus trastuzumab-mediated ADCC. HER2+ BC cell lines BT474 and MDAMB361 treated with docetaxel showed up-modulation of NK activator ligands both in vitro and in vivo, accompanied by a 15-40% increase in in vitro trastuzumab-mediated ADCC; antibodies blocking the NKG2D receptor significantly reduced this enhancement. NKG2D receptor expression was increased by docetaxel treatment in circulating and splenic NK cells from mice xenografted with tumor cells, an increase related to expansion of the CD11b+Ly6G+ cell population. Accordingly, NK cells derived from HER2+ BC patients after treatment with taxane-containing therapy expressed higher levels of NKG2D receptor than before treatment. Moreover, plasma obtained from these patients recapitulated the modulation of NKG2D on healthy donors' NK cells, improving their trastuzumab-mediated activity in vitro. This enhancement occurred mainly using plasma from patients with low NKG2D basal expression. Our results indicate that taxanes increase tumor susceptibility to ADCC by acting on tumor and NK cells, and suggest that taxanes concomitantly administered with trastuzumab could maximize the antibody effect, especially in patients with low basal immune effector cytotoxic activity.

Keywords: ADCC; NK cell; NKG2D; breast cancer; docetaxel.

Figure 1. Modulation of NKG2D ligands on breast carcinoma cells in response to docetaxel treatment

A, B. BT474 (A) and MDAMB361 (B) cells were treated with 100 nM docetaxel for the indicated times and analyzed by flow cytometry. Shown are fold-increases of ligand expression in treated versus untreated cells at the same time points. Data are mean ± SEM (n = 3). C. Fold-increase in MICA and ULBP2 protein expression levels, as assessed by Western blot and quantified by densitometric analysis using Quantity One software, in MDAMB361 breast carcinoma cells grown in SCID mice and treated with 20 mg/Kg docetaxel versus untreated tumors. Data are mean ± SEM (n = 5). *p < 0.05 by paired Student's t-test.

Figure 2. Docetaxel treatment increases trastuzumab-mediated cell cytotoxicity on tumor cells, as assessed by ⁵¹Cr release

A, B. To evaluate ADCC, ⁵¹Cr-labeled breast cancer cell lines BT474 (A) and MDAMB361 (B) were treated for 6 hours with 100 nM docetaxel (DTX) or not treated (NT) and cultured for 4 hours with PBMCs from healthy donor blood samples in medium containing trastuzumab (4 μg/ml). Shown are percentages of lysis of target cells at 50:1 effector:target cell ratio of PBMCs from independent healthy donors (A: n = 19, p = 0.0004; B: n = 13, p = 0.0006). C, D. BT474 and MDAMB361 cells, respectively, treated with DTX or not treated were cultured as above with PBMCs pre-incubated for 30 minutes with blocking NKG2D blocking antibodies (1 μg/ml). Values are median, interquartile range (box), minimum and maximum. (C: n = 6; D: n = 6). E, F. PBMCs from independent healthy donors (n = 4) were treated with 100 nM PGE2 for 24 hours, analyzed by flow cytometry for NKG2D expression (MFI on NK cells, E) and used in ADCC assay (F) against BT474 cells as described above. *p < 0.05, **p < 0.01, ***p < 0.001 by paired Student's t-test.

Figure 3. Chemotherapy alters NKG2D expression in circulating NK cells in mice injected with tumor cells

A, B. SCID mice were xenotransplanted with MDAMB361 breast carcinoma cells and, when tumors reached a volume of 150 mm³, treated with 20 mg/Kg docetaxel (DTX) or not treated. Six days after treatment, NKG2D expression (MFI of NK cells, A) and NK cell proportions (B) were analyzed in mouse blood. The same analysis was performed in tumor-free SCID mice of the same age (A, B) *p < 0.05 by unpaired Student's t-test. C, D. PBMCs from independent healthy donors were cultured for 24 hours in medium conditioned or not by BT474 breast carcinoma cells (C) or with BT474 cells treated or not with docetaxel (D) and analyzed for NKG2D expression (MFI of NK cells) by flow cytometry. *p < 0.05 by paired Student's t-test.

Figure 4. Chemotherapy alters NKG2D expression of NK cells through myeloid cells

A, B, C, D. Mice bearing MDAMB361 tumors were treated with docetaxel (DTX) and 6 days later, spleens were analyzed for NK and myeloid cell expansion. CD11b+Ly6G+ (A) CD11b+Ly6C+ (B) NK (CD49b+) cells (C) and NKG2D expression (MFI of NK cells, D) were analyzed by flow cytometry. E. Splenocytes from tumor-free SCID mice were co-cultured or not with purified GR1+ myeloid cells for 24 hours and analyzed for NKG2D expression by flow cytometry (n = 6). *p < 0.05, **p < 0.01 by unpaired Student's t-test.

Figure 5. Chemotherapy alters the NK cell phenotype in human patients

A, B, C. PBMCs isolated from patients at different time points during neoadjuvant treatment (pre: before any treatment, post: after chemotherapy) were analyzed by flow cytometry. The percentage of NK cells, identified as CD3– CD56+ (A) and their proportion of CD16+ (B) and the NKG2D expression (C) are shown. D. NKG2D expression on NK cells (CD3− CD56+ CD16+) of PBMCs from independent healthy donors treated in vitro with plasma derived from patients pre and post treatment. **p < 0.01, ***p < 0.001 by paired Student's t-test.

Figure 6. Expression of NKG2D in patients is associated with trastuzumab-mediated ADCC

A. Trastuzumab-mediated lysis of ⁵¹Cr-labeled BT474 cells induced by healthy donor PBMCs (n = 4) treated with plasma obtained from 5 patients (P1-P5) before docetaxel administration. Data are mean and range. ADCC post > pre refers to the percentage of times that PBMCs treated with post-treatment plasma induced greater trastuzumab-mediated ADCC of BT474 cells than when treated with the pre-treatment plasma. B. ADCC of BT474 cells induced by PBMCs obtained from healthy donors treated with pre- and post-treatment plasma of patients P1 and P5 (n = 7). *p < 0.05, **p < 0.01 by paired Student's t-test. C, D. NKG2D expression as evaluated by qPCR using RNA obtained from the blood buffy-coat of 18 HER2-positive breast cancer patients before any treatment according to response to one cycle of trastuzumab alone (C) and according to pCR (D) RD: residual disease. *p < 0.05 by unpaired Student's t-test.