Leukemic Stem Cells of Monocytic AMLs Are Not-Resistant to BCL-2 Inhibition

Treatment with Hypomethylating agents (HMA) such as 5-Azazytidine (AZA) in combination with the BCL-2 inhibitor Venetoclax (VEN) has recently become the standard of care for AML patients unsuitable for intensive induction chemotherapy and shows results superior to treatment with AZA alone (DiNardo et al., 2020, NEJM). However upfront resistance and relapse following initial response remain major obstacles.

It has recently been proposed that monocytic differentiation predicts resistance to AZA/VEN treatment in AML (Pei et al., 2020 Cancer Discovery). This appears to be due to increased expression of other anti-apoptotic proteins such as MCL-1 in monocytic AMLs, which conveys resistance to AZA/VEN therapy, as survival of leukemic cells in these patients is no longer dependent on BCL-2. However, an independent study found no impaired outcome in patients with monocytic AMLs treated with HMA/VEN (Maiti et al., 2020, Blood, ASH abstract).

Here, we show that monocytic AML cell lines and bulk cells of monocytic primary AML cells are indeed intrinsically resistant to AZA/VEN treatment. However, in a collective of 30 patients treated with HMA/VEN at Heidelberg University Medical Center between 2018 and 2020, monocytic differentiation assessed by flow cytometry was not an independent risk factor for refractory disease. We hypothesized that the conflicting data may be caused by intra-patient heterogeneity of AZA/VEN sensivitity and assessed killing efficiency in various immunophenotypic subpopulations of 12 primary AML patient samples in vitro.

The CD64 ⁺CD11b ⁺, differentiated blast population made up >50% of leukemic cells in monocytic and <20% in primitive samples and showed high levels of resistance to AZA/VEN therapy in both primitive and monocytic leukemias but did not engraft when transplanted into NSG mice, arguing they do not contain leukemic stem cells (LSC).

In contrast, we found immature CD64 ^-CD11b ^- GPR56 ⁺ LSC to be sensitive to AZA/VEN treatment irrespective whether they were derived from monocytic or primitive types of primary AMLs. As expected, LSCs from either monocytic or primitive AMLs initiated disease in NSG mice, highlighting that targeting LSCs is essential for the effect of AML therapy.

Next, we investigated expression of BCL-2, MCL-1 and BCL-xL in the same primary patient samples and observed high MCL-1 expression in monocytic AML samples. However, MCL-1 expression was restricted to the CD64 ⁺CD11b ⁺ population whereas in the LSC sub-populations robust expression of BCL-2 but low levels of MCL-1 and BCL-xL were detected, independent of whether monocytic or primitive AMLs were analyzed.

To further validate the sensitivity of LSCs of monocytic AML to BCL-2-I, we established a platform combining BH-3 profiling with multi-color flow cytometry, allowing for single cell assessment of cellular dependencies on independent apoptotic pathways. We found that LSCs of both AML types show high VEN/BAD but low MS-1 induced apoptosis, functionally confirming the expression patterns of BCL-2 and MCL-1. As LSCs are rare in monocytic samples, investigation of samples in bulk are dominated by MCL-1 expressing and resistant non-LSCs, explaining the overall higher MCL-1 expression/survival of monocytic compared to immature AML cells. However, our data uncovers sensitivity of LSCs to AZA/VEN independent of overall monocytic or primitive sample classification and provide a mechanistic explanation for the clinical data of Maiti et al. and our Heidelberg AML collective, which found no increased resistance of monocytic AMLs to AZA/VEN treatment.