Characterization of Survival Outcomes and Clinical and Molecular Modulators in Adult Patients with Core-Binding Factor Acute Myeloid Leukemia (CBF-AML) Treated with Hidac Consolidation: An Alliance Legacy Study

Introduction

CBF-AML is a specific disease entity defined by the presence of either t(8;21)(q22q22)RUNX1-RUNX1T1, or inv(16)(p13.1q22) or t(16;16)(p13.1;q22) creating the CBFB-MYH11 fusion gene, and is associated with a relatively favorable prognosis. Yet, outcomes are variable and relapse occurs in ~50% of patients (pts). Whereas several clinical and molecular features have been suggested to modulate survival, consensus data are sparse. Moreover, there is persistent controversy about the optimal dose intensity of cytarabine consolidation, which complicates interpretation of correlative studies with respect to outcome associations. Thus, we analyzed outcomes of a large cohort of CBF-AML pts treated with high-dose cytarabine (HiDAC) consolidation with respect to cytogenetic subtypes and co-existing molecular features.

We analyzed clinical and molecular characteristics and outcomes of 304 pts with CBF-AML aged 18-59 years [y; inv(16)/t(16;16), n=186; t(8;21), n=118] who were similarly treated and achieved a complete remission upon standard cytarabine/anthracycline induction followed by HiDAC on frontline Cancer and Leukemia Group B/Alliance for Clinical Trials in Oncology protocols between 1986-2016. All pts had centrally reviewed cytogenetics and samples from 136 pts underwent comprehensive molecular profiling of 80 AML-associated genes for association studies. Kaplan-Meier method was used to estimate probabilities of disease-free survival (DFS) and overall survival (OS), and the log-rank test evaluated differences between survival distributions. Cox proportional hazard regression analyses was used for DFS and OS univariable and multivariable modeling.

Results

Thirty-eight percent of pts relapsed and 69% were alive 5y after diagnosis (median OS): 12.5y with median follow-up of 8.6y. Outcomes of t(8;21) and inv(16)/t(16;16) AML were similar; relapse rates: 35% vs 40%, 5y OS rates, 64% vs 72%, respectively. Although outcome of pts with inv(16) was not significantly different from outcome of t(16;16) pts, the latter had, by trend, a superior DFS (5y rates, 75% vs 47%, P=0.07); their OS was not significantly different (5y rates, 81% vs 64%, P=0.24). As secondary cytogenetic abnormalities are common and impact on survival has yielded controversial results, we next analyzed the impact on outcomes of our pt cohort. None of the frequently observed secondary abnormalities [i.e., loss of X or Y chromosomes, +8, or del(9q) in t(8;21) pts and +22 or +8 in inv(16)/t(16;16) pts] affected the outcomes of CBF-AML pts we analyzed, which contrasts with some previous reports. The presence of a complex karyotype also did not negatively affect the favorable survival of CBF-AML pts.

Next, we assessed the impact of recurrent gene mutations on pt outcomes. While KIT mutations have previously been shown to confer inferior survival, our data suggest that presence of a KIT mutation negatively affects DFS of both t(8;21) and inv(16)/t(16;16) AML pts [5y rates, t(8;21) 36% vs 67%, P= 0.03; inv(16)/t(16;16), 21% vs 61%, P<0.001], and also OS in inv(16)/t(16;16) pts (5y rates, 52% vs 74%, P=0.04). Co-existing FLT3-TKD mutations were almost exclusively found in inv(16)/t(16;16) in 16/18 cases. Though a negative prognostic impact of FLT3-TKD has been suggested, this was not found in our cohort (5y rates, DFS, 56% vs 55%, P=0.93; OS, 78% vs 68%, P=0.69). Furthermore, neither the presence of NRAS or KRAS mutations affected survival outcomes. In multivariable analyses for OS, a higher percentage of peripheral blood blasts was the only feature associated with inferior survival [HR, 1.14 (95% CI, 1.04-1.25), P=0.004].

Conclusions

Our analysis of a large cohort of younger CBF-AML pts treated on frontline protocols with 7+3-based induction therapy and HiDAC consolidation, showed though these pts are considered favorable risk, they demonstrate varying long-term survival. Clinically, the presence of high blast count at time of diagnosis, was a strong adverse prognosticator for inferior survival. With respect to recurrent co-occurring cytogenetic aberrations or gene mutations, only KIT mutations associated with inferior survival in pts with t(8;21) and in those with inv(16)/t(16;16).

Support: U10CA180821, U10CA180882, U24CA196171; https://acknowledgments.alliancefound.org.

Clinicaltrials.gov: NCT00048958(CALGB 8461), NCT00899223(CALGB 9665), NCT00900224(CALGB 20202)

*co-senior authors