Sunday, May 18, 2014

Interactions between DNMT3A and other AML-related genes and treatment options

What other genes interact with DNMT3A that complicate the prognosis of AML?

Although in my previous posts, I have focused exclusively on the DNMT3A mutation and its mechanism in AML, clinical practitioners focus on multitude of other recurring genes and abnormalities that are overexpressed in AML patient genomes to aid in prognosis at initial diagnosis. Among such molecular markers are mutations in FLT3, NPM1 (which I have mentioned in my previous discussions of DNMT3A), and c-KIT. My goal is not to provide an exhaustive discussion about the involvement of these genes in the disease as they deserve entirely separate discussions of their own, but to provide a brief overview of them, as it turns out creating future therapies will need to target them as well. 

FLT3 is a receptor tyrosine kinase that spans the plasma membrane and has an important role in proliferation, survival and differentiation of hematopoietic progenitor cells. The incidence of this mutation in AML patients is about 25%, although this varies depending on age, clinical risk, and is more common in adult AML instead of pediatric AML and  myelodysplastic syndrome (MDS), which is a related malignancy that can give rise to AML. It was noted in this paper that several FLT3 inhibitors are currently in various stages of development.


It turns out that nucleophosmic (NPM1) is another gene most commonly mutated in AML, most frequently found near CpG islands that regulate gene promoters. You might recall from one of our earlier class discussions that the addition of methyl groups at a cytosine base in a 5’-CG-3’ dinucleotide pair (CpG) around promoter regions of our DNA can either decrease or eliminate transcription in particular genes and is a type of epigenetic genomic instability, an important hallmark of cancer. Indeed, it turns out that AML cancer cells exhibit areas of hypomethylation and hypermethylation of CpG islands near these promoters, that serves to inactivate transcription and has gene-silencing effects in tumor-suppressor genes.

DNMT is a whole family of methyl transferases that include DNMT1, DNMT3A, and DNMT3B, which catalyzes the addition of methyl groups to cytosine residues of CpG nucleotides mentioned earlier. An important piece of information that researchers have found with regards to DNMT3A specifically is that the majority of these mutations are of the missense type that occur at residue R882 at the carboxyl end of the DNMT3A protein. However there are also less common nonsense, frameshift and splice site mutations throughout the DNMT coding sequence (see Figure 1). Researchers currently do not understand why DNMT3A mutation at R882 is so prevalent. Yan et al. reports that the mutated DNMT3A enzymatic activity is substantially reduced at R882 in vitro as compared to wild-type DNMT3A. 

The involvement of DNMT3A though is nowhere even close to being deciphered as the paper presents conflicting pieces of evidence from various other sources that only fuel the controversy. For example, Ley et al. have reported that there are no differences in the total expression levels of DNMT3A protein of cells harboring mutant DNMT3A versus wild-type DNMT3A. This study also reported that although AML patients with mutated DNMT3A contained genomic regions with significantly different levels of methylation, there was no correlation between any of the differentially methylated regions and altered expression of nearby genes. Moreover, there were no clearly defined gene expression patterns that were associated with DNMT3A mutation status. Clearly, there is a still a lot of information that researchers don’t understand  regarding this important gene in AML.

One other gene I would like to address is the IDH1/2 genes that play an important role in metabolism. The mutation in this gene is heterozygous, resulting in proteins with newly acquired function (gain-of-function) that catalyzes nicotinamide adenine dinucleotide phosphate hydrogen-dependent reduction of α-KG to 2-hydroxyglutarate (2HG). This results in a decrease in α-KG and an increase in 2HG, with 2HG subsequently acting as a competitive inhibitor of α-KG-dependent reactions in the citric acid or Krebs cycle. In AML, increased cellular 2HG levels contribute to epigenetic mechanisms of pathogenesis by inhibiting α-KG-dependent enzymes that are important for normal DNA methylation. The frequency of IDH mutations in AML is 6–16% for IDH1 and 8–19% for IDH2, both which exhibit similar effects on DNA methylation (see also Figure 1).


Figure 1. Location and frequency of DNMT3A, IDH1, and IDH2
mutations as determined by the Cancer Genome Atlas. Mutation
data was obtained from the cBio portal (Im A.P. 2014)



What does seem to be clear at this point in the majority of studies thus far is that older age, higher white blood cell and platelet counts, normal cytogenetics and the presence of NPM1, FLT3-ITD and IDH1 mutations have been found to be more common in patients with DNMT3A mutations versus wild-type DNMT3A. Therefore it seems that the complex interplay between these genes is what leads to a worse prognosis and on the whole, a negative clinical outcome.

So is there any hope for treatment??

The current goal of AML chemotherapy is complete remission (CR) with return to normal hematopoiesis, and this goal has been successful to some extent over the past four decades, with the combined use of anthracycline (that is, daunorubicin or idarubicin) with cytarabine (60-70% success rate in newly diagnosed AML patients). However, as with many anti-cancer drugs, the disease manages to relapse in these patients, and so new and novel therapies need to be created. 

Age seems to play a significant role in whether the disease can be cured or never return. It seems that for patients under the age of 60, the combination of cytarabine and allogeneic hematopoietic cell transplantation (HCT) improves cure rates to 50%. However, for individuals over the age of 60, the cure rate remains quite low.

The promise of hypomethylating agents in AML

Hypomethylating agents, such as decitabine and azacitidine, are DNA methyltransferase inhibitors, and FDA approved agents for treatment of high-risk MDS. Decitabine is a deoxycytidine analog that is incorporated into DNA during S-phase of the cell cycle and binds to DNA methyltransferase, rendering it inactive. Azacitidine is a cytidine analog that primarily is incorporated into RNA, inhibiting RNA processing and function. To a lesser extent, it is incorporated into DNA, similarly to decitabine.

In a recent Experimental Hematology and Oncology 2014 paper by Smith, a retrospective comparative analysis study was conducted to study outcomes for AML patients treated with either decitabine (DEC) or azacitidine (AZA) between January 2006 and June 2012. 487 patients were eligible for this study and over 70% of patients in each cohort were at least 65 years old (mean age was AZA 70.3 ± 11.8 years, DEC 69.4 ± 11.6 years).  According to the study, most patient characteristics were similar between the cohorts except that the decitabine cohort had significantly more hospitalizations than the azacitdine cohort (62% AZA, 71% DEC; p = 0.0323). 

Below is a Kaplan Meier graph, analyzing the overall survival (OS) of the two cohorts.
Figure 2. Overall Survival (OS) for AML patients treated with decitabine or azacitidine between January 2006 and June 2012 (B Douglas Smith 2014).
Overall survival was significantly better in the AZA-treated cohort compared with patients in the DEC-treated cohort (10.1 months vs. 6.9 months respectively; p = 0.007, Figure 2) and treatment with azacitidine resulted in a significantly longer time to death when compared with decitabine treatment.

The most crucial aspect of the entire study, however, was that patients were not randomly assigned to treatment and administration schedules (i.e. dosing and adherence) for each therapy were not controlled. Since this was a retrospective, non-randomized control study, there is a possibility that patients who were at the more advanced stage of the disease or who showed worse prognosis were prescribed decitabine by their doctors, and hence had to be hospitalized more often than the other cohort. The reasons for hospitalization were stated to be primarily infections, bleeding events, or both. Now whether these hospitalizations were largely due to the cancer itself or side-effects of the treatment is not supported by thorough discussion in the paper. Neither does the paper state what stage of disease these patients were in prior to receiving either treatment as that could have a major impact on the clinical outcome and results of the study. A study analyzing the effects of both azacitidine and decitabine on the patient should also be conducted by these researchers in the future.

References:
B Douglas Smith, Charles L Beach, Dalia Mahmoud, Laura Weber, Henry J Henk
Exp Hematol Oncol. 2014; 3: 10. Published online 2014 March 25. doi: 10.1186/2162-3619-3-10
PMCID: PMC3994315

Im, A. P., A. R. Seghal, M. P. Carroll, B. D. Smith, D. E. Johnson, and M. Boyiadzis. "DNMT3A and IDH Mutations in Acute Myeloid Leukemia and Other Myeloid Malignancies: Associations with Prognosis and Potential Treatment Strategies." Leukemia 28.4 (2014): 727-980. Nature. Web. 12 May 2014. <http://www.nature.com/leu/journal/vaop/ncurrent/full/leu2014124a.html>.