The Use of Day 14 Bone Marrow Characteristics in the Decision to Pursue a Second Course of Induction Chemotherapy in the Treatment of Acute Myeloid Leukemia.

Patients with acute myeloid leukemia (AML) receiving induction or salvage chemotherapy often undergo bone marrow evaluation around day 14 to determine early treatment response and guide decision-making regarding need for immediate re-induction therapy. However, the parameters that define whether or not to initiate such therapy are seemingly not always objective. Thus we sought to define what factors influence physicians' decisions to pursue a second course of induction chemotherapy in AML.

We retrospectively reviewed 211 patients with newly diagnosed, relapsed, or refractory AML who received induction chemotherapy at the University of Washington /Fred Hutchinson Cancer Research Center from January 2008 to May 2012. 178 had a marrow aspirate and/or biopsy between days 13 to 17 after start of chemotherapy and were included in our “day 14” analysis. Bone marrows were assessed both morphologically and by multi-parameter flow cytometry (MFC). We used the Wilcoxon signed rank test and Fisher exact test to compare patient age, pre-treatment cytogenetics (SWOG criteria), newly diagnosed vs. relapsed or refractory AML, day 14 marrow cellularity, day 14 blast count by morphology or MFC, and induction regimen given on the first course (ara-C-containing vs. not) among patients who did and did not receive a second course of induction chemotherapy within 1 week of the “day 14” marrow. The second course could either be the same or different than the first course.

81% of the 178 patients had fewer than 10% blasts by morphology and MFC on day 14. None of these patients received course 2 within 1 week. 34 patients (19%) had greater than 10% blasts on the day 14 marrow by morphology or flow cytometry, of whom 18 (53%; 95% confidence interval 36–70%) received course 2 prior to day 21 and 16 (47%; 95% CI 30–64%) patients did not. As the decision of whether or not to begin course 2 within 1 week varied once blasts counts were >10%, we analyzed the 34 patients with >10% blasts between day 13–17 to assess what factors influenced the decision to begin a second course of treatment. The median age in those who received and did not receive course 2 within 1 week was 60 and 50.5, respectively (p=0.269). The fraction of newly diagnosed AML (as opposed to relapsed or refractory disease) in the two groups was 83% and 69%, respectively (p=0.429). The blast count by morphology was 69.5% in those who did vs. 24.4% in those who did not receive a second course within 1 week of the day 14 marrow (p=0.001) and by flow cytometry were 63% and 10%, respectively (p<0.001). 12 out of 27 (44%; 95% confidence interval 26–63%) patients who received ara-C on course 1 (“higher intensity”) received course 2 within 1 week of the first vs. 6 of 7 who received a lower intensity regimen on course 1 (p=0.09). 56% of the 9 patients with unfavorable cytogenetics received a second course within 1 week, as did the only patient with favorable cytogenetics, and 50% of those with intermediate risk cytogenetics. 1 of the 4 patients with hypocellular marrows on day 14 received a second course within 1 week as did 4 of the 5 with normocellular and 10 of the 21 with hypercellular marrows.

Although day 14 blast percentages were significantly higher among patients who received a second course of induction chemotherapy within 1 week of the day 14 marrow, 47% of patients with >10% blasts in a day 14 marrow and 56% who received a higher intensity regimen on course 1 did not begin a second course during the next week despite National Comprehensive Cancer Network (NCCN) guidelines that they do. We are studying whether this reflects objective patient and marrow characteristics, such as change in blast count from pre-treatment to day 14 marrow, or other more subjective factors in clinical practice.

Becker:Sanofi: Research Funding.