The disparity in clinical outcomes for patients with acute myeloid leukemia (AML) illustrates the heterogeneity of this disease. Overall survival (OS) in AML is impacted by many clinical, cytogenetic, and molecular factors. Although multiple clinical trials in AML have been completed, the standard of care for frontline treatment has remained largely unchanged since the completion in 1973 of a pilot trial using infusional cytarabine with daunorubicin. Since then, new combinations and dosing schedules with these same drugs have resulted in only modest improvements in outcomes.
Despite the abundance of research, long-term survival in AML remains poor, and the search continues for more effective treatments with less toxicity. Genetic profiling is increasingly being used for prognostic purposes and treatment decisions. In the therapeutic arena, while targeting of genetic alterations with specific therapies could conceivably improve outcomes, the presence of multiple genetic “drivers” in AML makes it unlikely that targeting a single molecular anomaly will translate into clinical success. Trials investigating targeted treatments often involve large groups of patients who have been selected without regard to the presence or absence of a particular anomaly. The molecular heterogeneity of AML may limit the efficacy of these treatments, which are directed at one particular target, thereby resulting in low response rates and subsequent lack of new drug approval. Although a subset of patients may benefit from some of these new treatments, this beneficial effect will be obscured when the drug is tested in such a diverse group of subjects. By selectively choosing study patients who are known to have the targeted aberration, we may begin to see the desired response rates. This review will discuss some of the novel agents currently being studied in AML and will provide insight into what may become available in the near future.
Small Molecule Inhibitors
The FMS-like tyrosine kinase 3 (FLT3) gene found on chromosome 13 plays a role in early hematopoiesis and development of myeloid precursors. Upregulation of the FLT3 ligand and receptor occurs in most leukemia cell lines. In the presence of the FLT3 internal tandem duplication (ITD) mutation – which occurs in nearly 25 % of patients with AML – hyperactivity of the FLT3 tyrosine kinase occurs, resulting in constitutive activity of FLT3 and dysregulation of cellular proliferation. FLT3-ITD mutations commonly occur in conjunction with leukocytosis and diploid cytogenetics, resulting in higher relapse rates when compared with patients who have wild-type FLT3.
Given the frequency of this mutation as well as its adverse prognostic effects, targeting FLT3 is a reasonable therapeutic strategy. However, the heterogeneity of AML implies that the targeting of FLT3-ITD alone will not be sufficient to result in durable remissions. FLT3 mutations are not considered founding mutations, but rather driver mutations, bringing into question their role in the pathogenesis of AML. Furthermore, the appropriate timing of FLT3 inhibitors during treatment is not entirely clear. Nonetheless, a number of FLT3 inhibitors have been studied in AML, both as single agents and in combination. Many of the earlier-generation drugs were less selective and potent than the newergeneration drugs. Three of the most advanced FLT3 inhibitors include Sorafenib, Midostaurin, and Quizartinib (formerly known as AC220).
Cell Cycle Inhibitors
Flavopiridol is a synthetic flavone derivative that inhibits multiple serine-threonine cyclin-dependent kinases that result in apoptosis in hematopoietic cell lines, including those from AML. In vitro data indicate that flavopiridol has direct cytotoxic effects, but when it is administered sequentially with ara-C and mitoxantrone – both of which are most efficacious when cells are in S phase – it sensitizes the remaining leukemic cells to these drugs as they reenter the cell cycle.
Novel Chemotherapeutic Agents
Vosaroxin (formerly voreloxin, SNS-595) is a first-in-class quinolone derivative known to inhibit topoisomerase II by DNA intercalation, causing site-selective DNA doublestrand breaks, G2 arrest, and apoptosis. The mechanism of action bears similarities to anthracyclines. However, the volume of distribution of vosaroxin is markedly lower compared to anthracyclines, suggesting the likelihood that off-target organ toxicity is lower. Furthermore, unlike anthracycline drugs, vosaroxin is not a P-gp substrate, and it has shown activity in anthracycline-resistant models.
Elacytarabine is the lipophilic 5’-elaidic acid ester of cytarabine (ara-C). Ara-C is a nucleoside analog that incorporates into DNA, thereby inhibiting the activity of DNA polymerase and ultimately resulting in apoptosis. Ara-C serves as the backbone for nearly all treatment regimens in AML, utilizing the human equilibrative nucleoside transporter (hENT1) to enter cells. A large proportion of AML patients,
however, have hENT1 deficiency which results in ara-C resistance. Elacytarabine was designed to enter cells independent of hENT1. Preclinical data indicate activity of elacytarabine in AML cell lines with known resistance to cytarabine.
Sapacitabine is a rationally designed oral cytosine nucleoside analog that incorporates into DNA, leading to strand breaks and apoptosis after subsequent rounds of DNA replication. The recommended dose from a phase I trial is either 325 mg PO BID x 7 days or 425 mg PO BID x 3 days weekly for two weeks. Sapacitabine was well-tolerated, and 28 % of patients had objective responses. In light of the oral administration and relatively mild side effect profile, further trials have been carried out with sapacitabine in the elderly population, where palliation and quality of life become significant factors.
CPX-351 is a liposomal encapsulation of cytarabine and daunorubicin at a 5:1 molar concentration ratio. Building upon the concept that drug combinations may act synergistically, additively, or antagonistically, preclinical studies revealed that a liposomally encapsulated combination of cytarabine and daunorubicin produced the best synergy at the 5:1 molar ratio in AML cells, maintaining this ratio in vivo for over 24 hours. In the first-in-human phase I study with CPX-351 in patients with primarily advanced AML, the compound was well tolerated, with very infrequent grade 3/4 toxicities that included mucositis, maculopapular rash, decreased ejection fraction, and increased liver enzymes, with a recommended phase II dose of 101 units/m2. Amongst the 43 AML patients treated in this study, an overall response rate of 23 % (CR+ CRi) and median remission duration of 6.9 months were observed. Importantly, pharmacokinetic studies confirmed that the 5:1 molar ratio was maintained for up to 24 hours on both days 1 and 5, while the half-lives of the cytarabine and daunorubicin were observed to be much longer than with free drugs.
Other Novel Anti-CD33 Therapeutics
The attractiveness of CD33 as a biologically important and accessible therapeutic target has led to continued pursuit of novel anti-CD33 strategies in the clinic. One such example is lintuzumab (SGN-33; HuM195), a humanized monoclonal antibody directed against CD33 with effector function through multiple mechanisms. This agent recently underwent testing in a large randomized trial against low-dose cytarabine as initial therapy for elderly patients with CD33-positive AML, but conferred no overall survival advantage. Perhaps more promising is the compound SGN33a, which is currently in phase I testing. This agent is an ADC consisting of an anti-CD33 monoclonal antibody conjugated to a novel synthetic pyrrolobenzodiazepine dimer, leading to DNA crosslinking and cell death. In preclinical studies, this compound demonstrated greater cytotoxic potency against AML cell lines and primary cells than gemtuzumab ozogamicin, including in AML cells demonstrating multidrug resistance phenotype. In light of the above-stated recent evidence of the efficacy of gemtuzumab ozogamicin, the appeal of second-generation ADCs appears justified.
Kendra Sweet & Jeffrey E. Lancet. Curr Hematol Malig Rep (2014) 9:109–117