EHA Library - The official digital education library of European Hematology Association (EHA)

Daratumumab (DARA), an anti-CD38 monoclonal antibody, induces lysis of multiple myeloma (MM) cells through complement dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC). Immunotherapy with DARA is clinically effective, but there is a significant variability in quality of response among patients.

We examined potential determinants of DARA sensitivity before start of treatment, and mechanisms of resistance during therapy, with special emphasis on the target protein CD38. Since several studies have stressed the importance of CDC for therapeutic efficacy of antibodies, we also focused on the complement inhibitory proteins (CIPs), including CD46, CD55, and CD59. 

We investigated a panel of cell lines, as well as patient samples by flow cytometry for CD38, CD46, CD55, and CD59 and assessed their sensitivity to DARA-mediated CDC in vitro. We also evaluated samples from relapsed/refractory MM patients treated with DARA as a single agent (GEN501 study).

MM and lymphoma cell lines (n=33), as well as primary MM cells (n=31) expressing high levels of CD38 and low levels of CD55 and CD59 were significantly more sensitive to DARA-mediated CDC, compared to cells with low CD38 expression and high levels of CD55 or CD59. However, CD46 expression was not associated with CDC sensitivity.

We also analyzed the expression of CD38 and CIPs on MM cells from 21 relapsed/refractory MM patients directly before start and during DARA treatment. The CD38/CD59 and CD38/CD55 ratios prior to treatment were predictive of clinical response. Also the sensitivity of pretreatment MM cells towards DARA-mediated CDC ex vivo, correlated positively with clinical response to DARA, expressed as maximum reduction in tumor load (R=0.70, P=0.005).

At the time of progression during DARA therapy, there was a significant reduction of CD38 expression and significant increase in CD59 and CD55 expression on both bone marrow-localized and circulating MM cells, whereas CD46 levels did not change. As expected, at the time of relapse, a significant decrease in DARA-mediated CDC was found when compared to baseline values. 

Interestingly, before start of DARA therapy, two out of 21 patients had two coexisting populations of MM cells based on differential expression of CD55 and/or CD59. In both cases, the dominant relapsing clone was strongly positive for both CD55 and CD59, whereas these clones were present as minor populations before initiation of DARA treatment. Altogether, these data suggest that elevated CIP expression results in reduced sensitivity to DARA-mediated CDC, and that upregulation of CD55 and CD59 during DARA treatment may contribute to the development of resistance. 

With the intention to improve CDC sensitivity, we evaluated phospholipase C, which enzymatically removes the GPI-anchored CD55 and CD59 molecules from the cell surface. Indeed, loss of CD55 and CD59 improved DARA-mediated CDC sensitivity in vitro. Similarly, all-trans retinoic acid (ATRA) enhanced CD38 expression levels and reduced CD55 and CD59 expression resulting in a significant enhancement of CDC in MM cell lines and primary MM cells. Also in a humanized mouse model, ATRA significantly enhanced the anti-MM effect of DARA.

These results show that baseline CD38/CD55/CD59 expression may be useful to predict response to DARA, and that targeting CIPs could be a potential adjuvant therapy for DARA or other monoclonal antibodies with CDC activity.