Correlation between NK function and response to trastuzumab in metastatic breast cancer patients

Abstract

Background: Trastuzumab is a monoclonal antibody selectively directed against Her2 and approved for the treatment of Her2 overexpressing breast cancer patients. Its proposed mechanisms of action include mediation of antibody-dependent cellular cytotoxicity (ADCC) by triggering FcgammaRIII on natural killer (NK) cells. This study addresses the correlation between overall NK function and trastuzumab's clinical activity.

Subjects and methods: Clinical and immunological responses were assessed in 26 patients receiving trastuzumab monotherapy as maintenance management after chemotherapy (8 mg/kg load and then standard doses of 6 mg/kg every 3 weeks). Cytotoxic activity against the MHC class I-negative standard NK target K562 cell line and HER2-specific ADCC against a trastuzumab-coated Her2-positive SKBR3 cell line were assessed in peripheral blood mononuclear cells (PBMC) harvested after the first standard dose. After six months, seventeen patients were scored as responders and nine as non-responders according to the RECIST criteria, while Progression-Free Survival (PFS) was calculated during a 12 months follow-up.

Results: The responders had significantly higher levels of both NK and ADCC activities (p < 0.05) that were not different from those of eleven normal controls. The NK activity of the non-responders was significantly (p < 0.05) lower than that of the normal controls. At twelve months, there was a marked correlation between PFS and NK activity only. PFS was significantly longer in patients with high levels of NK activity, whereas its pattern was unrelated to high or low ADCC activity.

Conclusion: One of the mechanisms of action of trastuzumab is NK cell-mediated ADCC lysis of the Her2-positve target cell. We show here that its potency is correlated with the short-term response to treatment, whereas longer protection against tumor expansion seems to be mediated by pure NK activity.

Figures

Figure 1

Different Her2 expression of the ADCC target cell lines. The human breast carcinoma cell lines MCF7 and SKBR3 were stained with the humanized anti-Her2 MoAb trastuzumab (Genentech Inc. San Francisco, California, USA) and analysed by flow cytometry.

Figure 2

NK and ADCC profile of patients classified as responders and non-responders after a six-month trastuzumab regimen. NK and ADCC activities of PBMC were tested at the start of trastuzumab therapy against the target K562 cell line (A) and the trastuzumab-coated Her2-positive cell line SKBR3 (B) in a 51Cr release assay. The trastuzumab-coated Her2-negative cell line MCF7 (C) was used as control for ADCC. Values of NK (A) and ADCC (B) cytotoxicity were always significantly higher in responders at all three E:T ratios.

Figure 3

Responders have normal NK and ADCC activities. NK and ADCC activities were assessed as described in legend to Fig. 2. Values (A and B) were not significantly different in responders compared to 11 normal donors at all three E:T ratios. As expected, negligible ADCC activity was found both in patients and normal donors against the Her2-negative cell line MCF7 (C).

Figure 4

Non-responder patients have defective NK activity. NK and ADCC activities were assessed as described in legend to Fig. 2. Values of NK activity (A) were significantly lower in patients compared to normal donors at three E:T ratios. ADCC activity (B) was significantly lower in patients only at the highest E:T ratio. As expected, negligible ADCC activity was found both in patients and normal donors against the Her2-negative cell line MCF7 (C).

Figure 5

Global time to tumor progression at twelve months. The Kaplan-Meier PFS curve shows that the percentage of progression-free patients is 75% at six months from start of Trastuzumab therapy and declines to 50% at twelve months.

Figure 6

Time to tumor progression is correlated with the NK activity. A cut-off of 5%. 3% and 2% cytotoxicity was chosen to discriminate between high and low NK levels at the three E:T ratios of 50:1, 25:1 and 12:1 respectively and a stratified PFS curve was created from these values. The figure shows that higher NK values are significantly (Log-Rank test p values) correlated with longer time to tumor progression.

Figure 7

Time to tumor progression is not correlated with the ADCC activity. A cut-off of 10%, 6% and 4% cytotoxicity was arbitrarly chosen to discriminate between high and low levels ADCC at the three E:T ratios of 50:1, 25:1 and 12:1 respectively and a stratified PFS curve was created from these values. The pattern of the two curves and the Log-Rank test p values demonstrated that time to tumor progression was not correlated with the ADCC.