Page 60 - Read Online

Saliba et al. Cancer Drug Resist 2021;4:125-42 I http://dx.doi.org/10.20517/cdr.2020.95 Page 135

Another approach to understanding resistance has involved genomic characterization of AMLs that exhibit
sensitivity vs. resistance. Through genomic analysis of AMLs treated on studies with venetoclax-based
combinations, DiNardo et al. [168] found that higher response rates and durable remissions were associated
with NPM1 or IDH2 mutations. AMLs that did not respond to venetoclax-based combination therapy
(primary resistance) or initially responded and then relapsed (adaptive or secondary resistance) had new
or expanded clones with activating mutations of FLT3 or RAS [168] . The variety of mutations occurring in
different kinase pathways (e.g., FLT3-TKD, FLT3 N676, RAS, or CBL among others) and the increase in
tumor heterogeneity following treatment indicate that resistance is likely driven by an evolving genetic
diversity in the cell pool rather than a single dominant gatekeeper mutation. Genomic changes between
diagnosis and relapse involved mutations in a variety of pathways, including kinase signaling such as FLT3-
ITD, NRAS, and JAK1 mutations; alternative RNA splicing due to mutations in U2AF1, U2AF2, SRSF2,
and ZRSR2; cancer-related transcription factors, including IKZF1, SETBP1, RUNX1, and STAT5A; tumor
suppressors such as TP53 and WT1; and epigenetic modifiers such as BCOR and CREBBP [169] . The observed
increase in AML cells harboring a FLT3-ITD mutation following treatment may indicate a clonal selection
in which a small pool of FLT3-ITD cells are able to selectively withstand treatment and repopulate the
marrow. The addition of inhibitors able to target mutated FLT3 such as sorafenib or midostaurin to the
HMA/venetoclax combination treatment may help to combat this particular form of resistance.

While some of these mechanisms of resistance can potentially be targeted therapeutically (e.g., FLT3
mutation or even possibly TP53 mutation), others such as monocytic differentiation are more difficult
to directly address pharmacologically. Accordingly, targeting a more general mechanism of resistance
that is not usually dependent on specific genetic mutations may be of clinical utility. AML cells take
advantage of stromal-dependent pro-survival signals to create a permissive niche within the bone marrow
microenvironment via interactions between upregulated cell surface receptors, including very late antigen-4
(VLA-4), CD44, E-selectin ligand-1, and CD98, with adhesion molecules, including vascular cell adhesion
molecule-1, fibronectin, hyaluronan, osteopontin, selectins, and integrins. This niche is favorable for the
maintenance and progression of chemotherapy-resistant disease [170,171] . This mechanism of resistance,
although less specific to the topic of this review, may be potentially modified or targeted; treatment with
anti-VLA-4 antibodies and cytarabine improves survival in AML mouse models relative to cytarabine
alone [172] . Additionally, patients with VLA-4-negative AML have a more favorable prognosis [172] .

Refined understanding of molecular determinants of sustained response, disease relapse, and secondary
resistance to venetoclax-based combination therapy will undoubtedly help to guide future strategies
for: (1) monitoring disease status; and (2) devising effective maintenance or salvage therapies. Multiple
venetoclax-based combinations are currently being tested with the goal of circumventing drug resistance.
These agents added to the HMA/venetoclax doublet include: a combination of cladribine and low-dose
cytarabine, the NEDD8-actviating enzyme inhibitor pevonedistat, the anti-CD33 antibody/calicheamicin
conjugate gemtuzumab ozogamicin, MDM2 inhibitors, and the anti-programmed death ligand-1 (PDL-1)
antibody avelumab [173] . How safe and effective any of these strategies are remains to be determined. Despite
the relatively high response rates with HMA/venetoclax combinations in patients newly diagnosed with
AML, the fact that most responders relapse despite ongoing therapy further underlines the importance of
understanding the mechanisms of acquired resistance [19,28] . Other outstanding questions include the optimal
duration of therapy with the HMA/venetoclax doublet in responders, the effects of treatment interruption
or discontinuation on long-term disease control, and the efficacy of the HMA/venetoclax doublet when
reintroduced at the time of relapse.

CONCLUSION
The HMA/venetoclax combination represents are major advance for AML patients who previously fared
poorly with conventional induction chemotherapy. As this combination is increasingly used, studies