Patients with AML and an IDH2-R172 mutation exhibit a unique initial response to intensive chemotherapy induction

Abstract

The utility of a midcycle bone marrow biopsy (BMB) for early assessment of response in patients with acute myeloid leukemia (AML) after intensive chemotherapy (IC) induction is contested. Even when challenged, there is little consideration as to the possibility of different response dynamics among genetically defined subgroups. Clinical observations led to the hypothesis that patients with AML and mutations in IDH2-R172 (R172-m) exhibit particularly slow blast reduction after IC induction. The purpose of this study was to analyze response kinetics of patients with R172-m to IC and compare the dynamics to patients with AML and IDH2-R140 mutations (R140-m). A retrospective single-center analysis was conducted among patients with newly diagnosed IDH2-mutated AML who received IC induction. Dynamics of blast reduction were compared and correlated with outcomes. A total of 52 patients were identified; 33 with R140-m and 19 with R172-m. Patients with R172-m had significantly higher midcycle BMB median blast count (70% vs 5%; P < .001), and their BMBs were slightly more cellular (P = .045). Among the R140-m, 58% had ≤5% blasts vs 0 of the R172-m. Furthermore, it took significantly longer for patients with R172-m to achieve blast clearance (≤5% blasts in BMB) compared to those with R140-m (P = .017). However, there was no difference in overall survival between the 2 groups, and outcomes were similar and favorable. This type of slow blast reduction has only previously been described in patients with acute promyelocytic leukemia. These findings suggest judicial application of reinduction strategies in this subgroup and warrant further investigation.

Graphical abstract

Figure 1.

Patients with R172-m had significantly higher midcycle blast count and slightly more cellular midcycle marrows. (A) Blasts in midcycle marrow. Violin plot for comparison of blast percentage in BMB at diagnosis and at midcycle, n = 26 for R140-m, median midcycle blast percentage 5% (range, 0-81); n = 15 for R172-m, median midcycle blast percentage 70% (range, 25-87; P < .001). A Welch 2-sample t test was used for comparison. There was no significant effect of induction type or of addition of enasidenib to induction. (B) Cellularity of midcycle marrow. Among patients with R172-m, 40% had hypercellular or normocellular midcycle bone marrows (age-adjusted cellularity), compared with 12% of the patients with R140-m. n = 25 for R140-m; n = 15 for R172-m. There was a trend toward increased cellularity among the R172-m (P = .045). A Fisher exact test was used for comparison. Midcycle BMBs were done on days 13 to 22 from induction, most on days 14 to 15. Hyper, hypercellular BMB; Hypo, hypocellular BMB; Normo, normocellular BMB.

Figure 2.

Blast percentage in sequential bone marrow biopsies, among patients who achieved CR after single induction. n = 8 for patients with R172-m; n = 24 for patients with R140-m. Each circle represents a sequential BMB. One of the patients with R140-m had relapsed disease after day 50.

Figure 3.

Patients with R172-m had significantly slower blast clearance than patients with R140-m, despite no difference in outcomes. (A) Time to blast clearance. Kaplan-Meier curves for estimated time to blast clearance (defined as days from induction to first BMB with ≤5% blasts, censored at time of second treatment or death) demonstrate significantly slower blast clearance among patients with R172-m. Median days to blast clearance was 44 among patients with R172-m vs 24 days among patients with R140-m (P = .017). There was no significant effect of induction type or of addition of enasidenib to induction. (B) Overall survival. Kaplan-Meier curves for estimated overall survival reveal no difference between patients with R140-m vs patients with R172-m. Estimated 24-month survival was 78% (67%, 91%); 74% (60%, 91%) among patients with R140-m and 89% (76%, 100%) among those with R172-m.