My special interest is the myelodysplastic syndromes (MDS), and so I focus here primarily on new data presented at ASH 2011 related to that diverse and troublesome group of myeloid neoplasms.
All in all, 2011 was an exciting year for molecular biology discoveries related to MDS, but a slower “news cycle” clinically. There were few data pertaining to novel drugs for MDS at this ASH Annual Meeting, but investigators presented a lot of information on new molecular findings in myeloid neoplasms, including how the presence of these novel mutations in patients' marrow cells might influence the risk of progression to acute myeloid leukemia (AML) or death.
These new pathobiological data should allow us to finally move beyond the venerable 1997 International Prognostic Scoring System (IPSS), in a more substantive way than the minor, albeit helpful, tweaks that were incorporated into the recently-proposed-but-not-yet-published Revised IPSS (IPSS-R), which was first presented by Stanford's Peter Greenberg at the 13th MDS Foundation International Symposium in Edinburgh, Scotland in May 2011.
Recurrent Somatic Mutations in RNA Splicing-Associating Factors
Perhaps the most exciting MDS-related finding reported in the last several months is the discovery of recurrent somatic mutations in RNA splicing-associating factors in patients with MDS, especially in those patients with ring sideroblasts. These mutations — the fruit of ongoing whole exome sequencing projects in MDS — are the most common MDS-associated mutations yet to be described.
Kenichi Yoshida from Tokyo, Japan and his colleagues published a paper in Nature online last September (and subsequently published in the October 6, 2011 print issue (Nature 2011;478:64-69) describing mutations in eight different splicing factors in MDS, while Elli Papaemmanuil from the Institute for Cancer Research in England, Luca Malcovati from Pavia, Italy, and a group of European colleagues reported the frequency of SF3B1 mutations in October in The New England Journal of Medicine (2011;365:1384-1395); Jaroslaw Maciejewski's group at the Cleveland Clinic published a paper on the same topic in Leukemia (2 September 2011; doi:10.1038/leu.2011.232).
SF3B1 mutations appear to be the most common splicing mutations in MDS, present in approximately 30% of patients with MDS overall. Remarkably, an SF3B1 mutation is detectable in as many as 85% of patients with MDS associated with ring sideroblasts — almost as common as the frequency of the JAK2 V617F mutation in polycythemia vera. Splicing mutations are also found in about 55-65% of patients with chronic myelomonocytic leukemia (CMML), an MDS/myeloproliferative neoplasm overlap disorder.
Access the hyperlinks (shown in blue) in this article and throughout the issue by reading it on OT's free iPad app.
More News from the ASH Meeting on Oncology-Times.com
Check the Online-First articles for the following:
- Fractionated Dosing May Put Gemtuzumab Back in Play
- Stem Cell Transplants: Overall Survival Same with Unrelated-Donor PBSCs and BMSCs
- New Data Emphasize Importance of Alerting Patients Undergoing HCT of Long-Term Risks
- CML: Stopping Tyrosine Kinase Therapy May Be Option for Some Patients
- MDS: Iron Chelation Therapy Induces Erythroid Responses
These mutations are not limited to MDS: SF3B1 mutations can be found in up to 10% of patients with myelofibrosis or AML, 9-15% of patients with chronic lymphocytic leukemia (CLL), and rarely in other tumor types.
“We are definitely achieving a better understanding of the mutations associated with MDS and refining risk models, but it is difficult to know just how any of these might translate into new therapies.”
Several groups reported at ASH 2011 that mutations in SF3B1 or other splicing factors may be associated with a better prognosis in MDS, but Mrinal Patnaik and his colleagues in Ayalew Tefferi's research group at Mayo Clinic in Minnesota found that these mutations did not hold independent prognostic value, once existing prognostic data (i.e., IPSS-R score and World Health Organization MDS classification) are taken into account (Abstract 460).
Not all patients with lower-risk MDS as assessed by the IPSS actually have lower-risk disease. In fact, while 25% of these patients will live more than 10 years, at least 25% of patients will die within two years of diagnosis despite a reassuring IPSS score.
Rafael Bejar and colleagues from Dana-Farber Cancer Institute focused on this heterogeneous group by validating a model for lower-risk model for MDS (Abstract 969) that was published by Guillermo Garcia-Manero of MD Anderson Cancer Center in Leukemia in 2008, and then Bejar extended those results by analyzing 21 genes known to be mutated in MDS (including splicing genes) in 288 lower-risk patients.
The Garcia-Manero model takes five variables into account—cytogenetics, age of the patient, hemoglobin, platelet count, and bone marrow blasts—and divides patients with IPSS Low or Intermediate-1 risk disease into three categories. Bejar and colleagues found that most MDS-associated mutations cluster much more tightly with the risk category as assigned by the Garcia-Manero lower-risk model than with IPSS score, with higher-risk mutations such as RUNX1 or NRAS found only in the third (highest-risk) category. Only EZH2 mutations provided independent prognostic value beyond the MDACC lower-risk MDS model's five clinical parameters. The Cleveland Clinic group also validated the MD Anderson lower-risk model.
Furthermore, MDACC investigators presented a new seven-component risk model for therapy-related (secondary) MDS, a group of patients for which the IPSS was never validated and a clinical setting where the IPSS does a particularly poor job of predicting outcomes. So we are definitely achieving a better understanding of the mutations associated with MDS and refining risk models, but it is difficult to know just how any of these might translate into new therapies.
Field Moving Towards Combination Approaches in Higher-Risk Patients
With respect to treatment, the field is moving towards combination approaches in higher-risk patients with MDS, building on the azacitidine (Vidaza) platform, since azacitidine monotherapy is known to extend survival compared with conventional care in MDS (Fenaux et al: Lancet Oncology 2009; 10:223-232).
Mikkael Sekeres from Cleveland Clinic presented updated data at the ASH 2011 meeting from a multicenter trial of combination therapy with azacitidine plus lenalidomide (Revlimid) in 36 patients with MDS (Abstract 607). Using go-forward doses defined in a previous Bone Marrow Failure Consortium Phase 1 trial (i.e., 75 mg/m2 subcutaneously on days 1-5 of azacitidine, and 10 mg orally daily for 21 out of 28 days with the lenalidomide), the overall response rate in the Phase 2 study was 72%, including a 42% complete response rate, and the median duration of complete response was greater than 16 months.
“All in all, 2011 was an exciting year for molecular biology discoveries related to MDS, but a slower ‘news cycle’ clinically. There were few data pertaining to novel drugs for MDS at this ASH Annual Meeting, but investigators presented a lot of information on new molecular findings in myeloid neoplasms, including how the presence of these novel mutations in patients' marrow cells might influence the risk of progression to AML or death.”
“The upcoming SWOG S1117 cooperative group study will formally compare azacitidine monotherapy with azacitidine and lenalidomide combination and the azacitidine/vorinostat combination.”
Although this was an uncontrolled Phase 2 study, the results compare favorably with historical reports of azacitidine monotherapy.
Oliver Ottmann from Frankfurt, Germany reported results of another combination study on behalf of an international group of collaborators: a Phase 1 trial with panobinostat and azacitidine in higher-risk MDS, CMML, and oligblastic AML patients, using the seven-day standard azacitidine dosing as a backbone (Abstract 459).
Panbinostat is a deacetylase inhibitor, and there is in vitro synergy (in terms of modification of gene expression) between deacetylase inhibitors like panobinostat and DNA hypomethylating agents such as azacitidine. The maximum tolerated dose of panobinostat was 30 mg when administered orally on days 3, 5, 8, 10, 12, and 15 of each azacitidine cycle.
It's unknown if concurrent administration or sequential administration of these agents is best; the ECOG E1905 randomized cooperative group trial of azacitidine monotherapy versus azacitidine plus concurrently administered entinostat (MS-275) failed to show a benefit from the combination approach, and some investigators blamed the concurrent schedule for these results, since entinostat is a cell cycle inhibitor and azacitidine requires cells to cycle to be cytotoxic. Because the panobinostat + azacitidine study was a Phase 1 trial enrolling patients with three different diseases, it is difficult to get any insight into response rate; a randomized Phase 2 trial is planned.
Finally, Garcia-Manero presented final results of an interesting Phase 2 study of azacitidine combined with a deacetylase inhibitor, vorinostat (Zolinza), which is FDA-approved for treatment of cutaneous T-cell lymphoma (Abstract 608). This study enrolled patients with MDS or AML who were not eligible for another clinical trial because they had an elevated bilirubin, creatinine, or poor performance score.
The complete response rate for the 30 patients in this study was 26%, and the combination was surprisingly well tolerated for a group of patients that sick. The upcoming SWOG S1117 cooperative group study will formally compare azacitidine monotherapy with azacitidine and lenalidomide combination and the azacitidine/vorinostat combination.
Optimal Timing of Allogeneic Stem Cell Transplant
The only curative treatment for MDS remains allogeneic stem cell transplantation, but the optimal timing of transplant during the course of disease is unknown.
John Koreth and his colleagues from DFCI, working with the Center for International Blood and Marrow Transplant Research (CIBMTR), presented data comparing a cohort of 80 CIBMTR and DFCI patients with MDS who were between ages 60 and 70 and underwent reduced-intensity transplant conditioning from a fully HLA-matched donor, with cohorts in the same age group who received supportive care alone, erythropoiesis-stimulating agents, or hypomethylating agents, but did not undergo transplant (Abstract 115).
“Because the panobinostat + azacitidine study was a Phase 1 trial enrolling patients with three different diseases, it is difficult to get any insight into response rate; a randomized Phase 2 trial is planned.”
Using a Monte Carlo statistical model, Koreth and colleagues found that in patients with IPSS Intermediate-2 or High risk disease, there appeared to be net benefit from proceeding to transplant earlier rather than later, whereas in those with IPSS Low or Intermediate-1 risk disease, it was best to wait until the time of disease progression so as not to cost the patient net years of good quality life.
While Koreth's study was not a randomized trial, these data complement a classic analysis of younger patients undergoing transplant for MDS published by Corey Cutler in 2004 in Blood (104:579-585) and will inform our approach for patients with MDS who are older or otherwise ineligible for stem cell transplantation with full-intensity conditioning.
“It is somewhat troubling that although the three FDA-approved disease-modifying therapies for MDS (azacitidine, decitabine, and lenalidomide) have putative mechanisms of action that have been suggested based on in vitro studies, none were designed based on knowledge of MDS pathobiology, and the actual mechanism of action for each drug remains speculative. Hopefully the future promises better!”
We can certainly hope that the new molecular biology insights in MDS in 2011 will translate into more targeted, rationally designed therapies in the near future. It is somewhat troubling that although the three FDA-approved disease-modifying therapies for MDS (i.e., azacitidine, decitabine, and lenalidomide) have putative mechanisms of action that have been suggested based on in vitro studies, none of these agents were designed based on knowledge of MDS pathobiology, and the actual mechanism of action for each drug remains speculative. Hopefully the future promises better!