Evidence for serial antibody-dependent cell-mediated cytotoxicity (ADCC) by primary unstimulated human natural killer (NK) lymphocytes

Abstract

Serial killers are more desirable than single-hit attackers when it comes to treating cancer. Antibody-dependent cell-mediated cytotoxicity (ADCC) mediated by NK lymphocytes is a significant component of many monoclonal antibody anti-cancer immunotherapies. However, there is a significant problem associated with ADCC: the NK cells lose their CD16A Fc-receptors and can no longer recognize target cells coated with antibodies, due to the action of NK cell metalloproteases. In this report, we searched for evidence for serial ADCC before the effector cells lose their receptors. Human ADCC measured by 51Cr-release at multiple NK effector to tumor cell ratios (E:Ts) was largely finished after 4 hours when the targets were Daudi B leukemia cells coated with non-fucosylated obinutuzumab anti-CD20. We reasoned that serial killing is evident when the number of dead targets exceeds the number of initial CD16A NK effectors. We assessed the dead Daudi per NK effector (D/E) at an E:T of 1:4 to detect serial killers given an excess of targets. For 28 healthy human subjects, the total ratio of dead targets per CD16Apos NK effector cell ranged from 0.6 to 2.2, with a median of 1.5 and average of 1.5 +/− 0.4. Four donors (14%) had D/Es >2. The D/E values were similar for CD16A AA158 F/F donors compared to F/V & V/V donors, even though the V-positive donors had greater initial NK cellular levels of CD16A. We conclude that this gross measurement (D/E) can detect serial ADCC even without the information provided by time-lapse cinematography to indicate the actual fraction of effectors that are serial killers. This simple measurement may be useful to screen for promotion of serial killing by reagents that reduce proteolytic loss of NK cell CD16A.