Population Based Study on the Utilization and Cost of Azacitidine In a Canadian Province.

Azacitidine (AZA) is an efficacious therapy for high risk Myelodysplastic Syndrome (MDS). In a randomized trial by Fenaux et al. patients were treated with a median of 9 cycles of AZA at 75mg/m²/day × 7 days every 28 days. Based on the published treatment regimen (Lancet Oncol 2009; 10: 223-32), the average cost of AZA is estimated to be $10,000 Canadian dollars per month. With a median of nine full dose cycles of AZA cost per patient is anticipated to be $90,000. In many health care systems, AZA has not been approved, due to financial constraints. Population based studies may provide a more realistic estimate of AZA drug utilization and cost.

The province of Manitoba (population 1.2 million) offers a unique opportunity to perform population based studies, as provincial health and pharmacy services capture information regarding administration of parenteral drugs like AZA. Although AZA is not funded by the province, it was provided by the manufacturer on compassionate basis for any patient who met the approved criteria from April 2009 to 31 March 2010. Enrolled patients continued to receive the drug even after March 2010. The drug was administered at a single site.

All patients enrolled during the period of compassionate use (April 2009-31 March 2010) who received one or more doses of AZA were prospectively followed for drug administration, response, toxicity and outcome. For this analysis, the last follow up was 31 July 2010.

Eleven eligible patients received at least one cycle of AZA. The diagnosis was MDS in 8, AML in remission with underlying MDS in 2 and CMML in 1. Four of eleven patients had therapy-related MDS and two patients were ECOG >2 at onset of therapy. A total of 42 cycles were given with a median of 3.0 (range 1–9) cycles per patient. There was a 25–50% dose reduction in 9 of 42 cycles (21%) due to cytopenias. Eleven of 42 cycles (26%) were complicated by febrile neutropenia, with 7 of these episodes requiring hospital admission. Reasons for discontinuing AZA were: (a) progressive cytopenia or transformation to AML in four (b) acceptance for allogeneic hematopoietic stem cell transplantation (HSCT) in three (c) severe sepsis during therapy in two (d) lack of objective response in one. Only one patient continues on AZA. Hematologic improvement and transfusion independence was seen in 5/11 and 3/11 patients respectively. There were two patients with bone marrow complete response (CR) in addition to hematologic improvement, one of whom underwent HSCT. The approximate cost of AZA in this cohort is $34,090 per patient.

In this population based study, we demonstrate that many MDS patients are withdrawn from further treatment due to adverse effects or lack of early response, in contrast to patients participating in clinical trials. Despite receiving dose reductions and fewer total cycles of AZA, a comparable number of patients (45.5%) had objective responses. Our study suggests that the actual drug acquisition cost would be just over one third (38%) of the projected cost based on clinical trial data. Decisions to fund or deny expensive drugs should include data from population based studies.

Kumar:Celgene: Membership on an entity's Board of Directors or advisory committees.