CHICAGO—Patients with myelodysplastic syndromes (MDS) treated with hypomethylating agents (HMAs), especially decitabine, are more likely to achieve hematologic responses when treated with a greater number of cycles, investigators from Georgia Cancer Specialists in Marietta reported at the 2010 annual meeting of the American Society of Clinical Oncology.
“The likelihood of improvement in anemia, thrombocytopenia, and neutropenia is associated with the number of cycles. For every extra cycle, the chance of improvement increases,” said Rodolfo Bordoni, MD. Responses increased by 15% to 19% with each additional cycle beyond the first.
Although clinical trials of HMAs have demonstrated a range of clinical benefits, concern exists about the applicability of trial results in the community practice setting where patients may differ in terms of demographics, risk level, comorbidities, and treatment regimen, Bordoni noted.
“We therefore examined hematologic outcomes in MDS among patients treated with the HMAs azacitidine and decitabine in a large community hematology/oncology practice,” he said.
This was a retrospective observational study of patients through electronic medical record data for 2006-2009. The population included 1070 patients with intermediate- or high-grade MDS, 137 (13%) of whom received azacitidine (n = 53) or decitabine (n = 84). The groups were well-balanced except that decitabine recipients were somewhat sicker, he noted.
The Charlson comorbidity index was ≥1 in the month preindex for 40/53 on azacitidine and 66/84 on decitabine. Mean Charlson index was 1.2 for the azacitidine group and 1.7 for the decitabine group. Mean number of cycles was five for azacitidine and four for decitabine.
Response increased per cycle
The total number of cycles significantly predicted hemoglobin level (odds ratio [OR], 1.19; P = .029), platelet count (OR, 1.15; P = .031), and absolute neutrophil count (ANC) (OR, 1.16; P = .047), Bordoni reported.
This improvement, however, was largely driven by decitabine, which was associated with a 58% improvement in hemoglobin response per additional cycle (P = .0035). There was also a trend toward a greater likelihood of hemoglobin ≥11 g/dL for decitabine over azacitidine (OR, 2.70; P = .085) after controlling for the number of cycles given. Although decitabine decitabinetreated patients were also more likely to have a platelet response, the difference was not statistically significant after controlling for the number of cycles.
The drugs were equally likely to produce an ANC response. The use of growth factors during treatment was negatively associated with ANC response (OR, 0.96; P = .046) as well as hemoglobin response (OR, 0.85; P = .007), he added.
The findings suggest that “we should not give up on slow responders,” Bordoni said. “Continuing on treatment is the key to hematologic response.”