Targeting of CLEC12A In Acute Myeloid Leukemia by Antibody-Drug-Conjugates and Bispecific CLL-1×CD3 BiTE Antibody

Abstract 2890

Response rates of ±80% in Acute Myeloid Leukemia (AML) are observed after conventional therapy but ±30% of patients experience a relapse. In the elderly the outcome is even worse. A small population of therapy resistant leukemia cells, minimal residual disease (MRD), are thought to be responsible for relapse of AML. The leukemic stem cells (LSC) herein have self renewal potential and reside in the CD34+CD38- stem cell compartment and side population (SP) compartment and can be identified via aberrant marker expression and scatter properties. Several markers are identified that show differential expression on AML (stem) cells versus normal hematopoietic stem cells (HSC). Previously we showed that CLEC12A (CLL-1, MICL, KLRL1, DCAL-2) is expressed on blasts of 90% of AML patients with varying expression. Importantly, CLEC12A is expressed on LSC but not on normal HSC (van Rhenen, Blood 110(7), 2007). This unique expression pattern paves the way to develop therapies that potentially eliminate CLEC12A-positive LSC and preserves CLEC12A-negative HSC. Drug-conjugated antibodies (ADCs) targeting CLEC12A and Bispecific T cell Engager (BiTE) scFv-antibodies targeting T-cells to CLEC12A positive cells could be instrumental to achieve this goal.

We evaluated the response of AML cells to ADCs conjugated via cleavable and non-cleavable linkers to the maytansine derivates DM1 and DM4 and to the BiTE antibody CLL-1×CD3. ADC activity was assessed by colony formation capacity after 24 hours exposure to 0.1–100 nM ADC in 29 freshly obtained AML samples. The response to the BiTE antibody was tested by flow cytometry in 9 AML samples via induction of apoptosis (Annexin V/7AAD) after 24 hours exposure. To determine the effect of ADC on self-renewal in normal bone marrow (NBM), colony formation capacity was asseses during long term liquid culture after 24 hours exposure to 1–100 nM ADC. Furthermore internalisation of CLEC12A in AML progenitor and stem cells was tested. Several splice variants of CLEC12A are reported (CLL-1, MICLα, MICLβ, MICLγ) that have different intra-cellular signalling motifs or lack the transmembrane motif or the extra-cellular c-type lectin-like domain. Since these variants could not all be distinguished or detected by extra-cellular antibody binding, we evaluated these splice variants by Q-RT-PCR.

After 24 hours exposure, a median IC₅₀ value of >100 nM was observed for the unconjugated antibody CR2357. The median IC₅₀ values for ADCs with non-cleavable linkers were 10 nM for CR2357-SMCC-DM1 (4,3 DM1/mAb), 2 nM for CR2357-PEG₄-MAL-DM1 (5.9 DM1/mAb) and 0.8 nM for CR2357-PEG₄-MAL-DM1 (10 DM1/mAb). For CR2357-SPDB-DM4 (4 DM4/mAb) which has a cleavable linker the median IC₅₀ was 4 nM. The median IC₅₀ of ADCs with non-cleavable linkers were significantly correlated to each other (r=0.730-0.784, p<0.01). CR2357-PEG₄-MAL-DM1 (10 DM1/mAb) was significantly correlated to CLEC12A membrane expression (r=0.649, p<0.05). Prelimanary data of colony formation capacity during long term liquid culture of NBM showed that at >5 weeks after exposure, this was reduced to 15–50% for CR2357 and CR2357-PEG₄-MAL-DM1 (10 DM1/mAb) relative to the untreated control. Exposure of AML cells to the CLL-1×CD3 BiTE antibody with donor T-cells (E:T=10:1 and 1:1) showed a dose dependent activation of T-cells as measured by increased CD25 and CD69 expression on CD4+ and CD8+ T-cells. Importantly, besides T-cell activation, Annexin V/7AAD staining of AML cells showed a specific decrease of CLEC12A-positive viable cells while in CLEC12A-negative cells viability remained constant. Internalisation of CR2357 antibody in CD34+/CD38+ progenitor cells and in CD34+/CD38- LSC was clearly demonstrated. Q-RT-PCR of CLEC12A splice variant expression showed highest expression for MICLα > MICLβ ∼F MICLγ > CLL-1 indicating that MICLα is the main variant expressed on the cellular membrane. Downstream signalling will therefore mainly be mediated by SHP-1/2 phosphatases. Although expression levels in AML, NBM, and sorted sub-populations varied, the ratio between the splice variants remained almost similar suggesting that the individual splice variants play a similar role in the different cell populations.

In conclusion: these result show that targeting of CLEC12A-positive AML cells by ADCs and BiTE antibodies results in specific cell kill and might be a promising approach for the eradication of LSC that survive conventional therapy.