Stumbling through the streets of New Orleans at American Society of Hematology Annual Meeting, like Ulysses on an unmoored ship, you are tempted by cloying beignets and a liberal open container law on the one hand, and all of the outstanding scientific research presented at the Ernest N. Morial Convention Center on the other. Even if you choose the latter, it's impossible to attend all of the sessions relevant to your practice. So, here's my synthesis of research that I found exciting in Myelodysplastic Syndromes (MDS) and Acute Myeloid Leukemia (AML). I hope you're able to read this while sitting at a small table at Café du Monde—at least in your imagination.
eHarmony Couldn't Have Brought Together a Nicer Pair of Drugs
The standard drug therapy for higher-risk MDS is azacitidine or decitabine—both often referred to as hypomethylating agents, as one purported mechanism of their activity is to reverse the methylation-induced silencing of tumor-suppressor genes. Histone deacetylase inhibitors such as vorinostat work synergistically with hypomethylating agents in vitro. Silverman and colleagues from the New York Cancer Consortium reported results from a Phase 2 trial of mostly higher-risk MDS patients treated with azacitidine and vorinostat (Abstract #386).
Of 33 patients evaluable for response, 23 (70%) achieved either a complete response or hematologic improvement. Median response duration was 16 months, and median overall survival was 21 months. This combination and the combination of azacitidine and lenalidomide are being compared with azacitidine monotherapy in the North American Intergroup MDS study (S1117, NCT01522976), which at the time of press had enrolled over 200 subjects.
I wrote last year (OT 1/25/13 issue) about the maddening story of FLT3 in AML. This lesion is detected in up to one-third of AML patients and confers a terrible prognosis (particularly in the absence of an NPM1 mutation). Attempts to treat relapsed/refractory AML patients who have the FLT3 internal tandem duplication abnormality with quizartinib (AC220) resulted (in studies presented at ASH 2012) in response rates of 32 to 53 percent as monotherapy, with the vast majority of responses comprised of complete response with incomplete platelet recovery (Cri). Responses were short-lived, though, with a median of 12 weeks or fewer across studies.
At ASH 2013, quizartinib was added to standard, cytarabine-based remission induction therapy in older (Abstract #622) and younger (Abstract #623) adults with untreated AML. In 48 evaluable older adults, presented by the United Kingdom group, quizartinib at 40 mg daily for 14 days given sequentially after cytotoxic chemotherapy was found to be the maximum tolerated dose. Complete response was achieved in 79 percent of all patients, including all four patients who were FLT3 internal tandem duplication positive.
In 18 younger adults, presented by the Northwestern group in collaboration with other U.S. sites, the maximum tolerated dose was 40 mg daily for 14 days or 60 mg daily for seven days given sequentially after cytotoxic chemotherapy. Complete response occurred in 61 percent, including six of eight patients who had the FLT3 ITD lesions. Multiple Phase 3 studies have been planned to explore these combinations further.
Dreaming of Genie
Next-generation sequencing has revealed a world of genetic abnormalities in cancer. Some of these have been suspected, some are surprises, and all are now contributing to our understanding of how cancer arises and evolves over time.
My patients always joke, as they complain about their age-related illnesses, that it stinks to get old but that it's better than the alternative. The Washington University group presented a Plenary Session talk (Abstract #5) in which they explored the role played by TP53 mutations in the evolution of therapy-related AML—AML that arises as a consequence of chemotherapy and/or radiation therapy administered for another cancer, years later.
As it happens, hematopoietic stem cells accumulate genetic mutations as a function of age—the longer you live, the more mutations you will develop. One of these mutations is TP53. The group identified patients who developed therapy-related AML, for whom they could obtain bone marrow samples that had been banked three to six years prior to the AML. They were able to identify the TP53 mutation at very low levels in these banked specimens (as low as 0.1%), and then ran experiments in which they showed that, while TP53 cells on their own do not have a growth advantage compared with other bone marrow cells, they do when exposed to chemotherapy.
In other words, we all may have some low level of TP53 mutations as a consequence of aging. Receiving chemotherapy for another cancer, such as lymphoma, may provide a growth advantage to those TP53 cells specifically (and not cause genome-wide damage, as previously thought), promoting the development of AML years later.
Exploring genetic mutations and comparing them with mutations seen in de novo AML, the Cleveland group, in collaboration with scientists from Japan, presented data on 707 patients with MDS or MDS-related conditions and 201 AML patients, using next-generation sequencing techniques (Abstract #518). They found that mutations including TET2, DNMT3A, ASXL1, and U2AF1 represent ancestral/founder events, while those of the IDH, receptor tyrosine kinase, and cohesion families are typically secondary abnormalities. Receptor tyrosine kinase, IDH family and NPM1 mutations were more frequently observed in the primary AML cohort, while mutations of SF3B1 and SRSF2 were more common in AML arising from MDS.
The take-home points of this research include:
* The sheer size of these cohorts analyzed;
* The demonstration of overlap of abnormalities across diseases; and
* The ability to identify the order in which genetic lesions occur in the development of these multi-step cancers.
One final abstract exploring genetic mutations and their clinical associations deserves mention: Another collaboration between the Cleveland and Japanese groups identified the mutation DDX41, which is a helicase—an enzyme involved in RNA metabolism, spliceosomal function, ribosome biogenesis, pre-mRNA splicing, and translation initiation (Abstract #655).
The gene was discovered in a pair of identical twins, both of whom had a biologically similar MDS and similar response to the drug lenalidomide, despite neither twin having the del(5q) cytogenetic abnormality that usually is associated with lenalidomide responses. DDX41, which interestingly does locate to chromosome 5, was then identified in 16 percent of other MDS/AML patients, and found to be associated with response to lenalidomide in those patients, as well.
In addition to showcasing a nice demonstration of bedside-to-bench-to-bedside research, this study found a novel area, helicases, where mutations can be found, and possibly identified the key abnormality responsible for the non-deletion-5q responses in MDS patients.