A Precision Medicine Heirarchical Classification Developed Using Variant Allele Frequency (VAF) for Treatment of Older Patients (Pts) with Acute Myeloid Leukemia (AML): Alliance Clinical Trials in Oncology (Alliance) Historical Patient Control

Introduction: In contrast to younger pts with AML, those ages 60 years (y) and above often have multiple, sequentially acquired, somatic mutations that drive leukemogenesis. Long-term 10-y event-free survival (EFS) of non-transplanted older AML pts receiving intensive cytarabine/daunorubicin induction (7 and 3) is 2% and the median overall survival (OS) is less than 1 y (Vasu et al., Blood Adv 2018;2:1645-50). Divergent pathogenesis based upon sequentially acquired mutations with targetable founder alterations and recent introduction of new targeted therapies have prompted interest in precision medicine-based treatment of AML. In order to determine if new therapeutic treatment strategies perform better than current standards of care, it is important to understand the historical outcomes for defined genomic groups in older AML pts.

Methods: In anticipation of a precision medicine-based trial initiated by the Leukemia and Lymphoma Society (LLS AML Master Study), we designed a precision medicine-based stratification considering "assignment to curative therapy with 7 and 3" for known responsive groups [i.e., core-binding factor (CBF) AML and NPM1-mutated(+)/FLT3-ITD- pts] followed by pts with defined driving cytogenetic aberrations [11q23/MLL-rearranged and TP53 wild-type/complex karyotype (≥3 chromosome abnormalities)] and pts with mutation clone with VAF ≥0.3 by next generation sequencing. For pts not assigned to any genomic group during the initial stratification, a second run-through of the algorithm was performed assessing for a mutation clone with VAF ≥0.2 (Figure 1). The following priority (in order of highest to lowest) was used: CBF-AML, NPM1+/FLT3-ITD-, 11q23/MLL-rearranged, IDH2+, IDH1+, TP53+, TP53 wild-type/complex karyotype, FLT3-ITD+ (both high and low allelic ratios included) or FLT3-TKD+, WT1+ or TET2+, and marker negative (all other karyotypes and mutations not defined by previous grouping). A well-characterized data set of 589 pts aged ≥60 y (range, 60-92) from the Alliance was analyzed to show outcomes of pts within specific genomic groups (Table 1). Identification of a second multi-institutional pt data set is currently on-going to validate these findings.

Results: In the Alliance cohort, 516 of 589 (88%) pts were assigned solely based upon cytogenetic group or dominant mutational clone with VAF ≥0.3 and 11 (1%) patients were assigned based off of VAF ≥0.2, thus providing support for this algorithm. The remaining 62 pts were assigned to the marker-negative group. As this algorithm is designed to assign therapy and assess outcome on an intent-to-treat basis, pts who suffered early death (ED), defined as death within 30 days of starting therapy irrespective of cause, were included. Baseline characteristics among groups were similar with the following exceptions: 1) CBF-AML and NPM1+/FLT3-ITD- pts having equal male and female pts with other groups having male predominance; 2) white blood cell counts being highest in NPM1+/FLT3-ITD-, 11q23/MLL-rearranged and FLT3-ITD+ or TKD+ groups; 3) percent of bone marrow blasts were lowest in the TP53+ group. Outcome endpoints are summarized in Table 1. Early death (ED) occurred in 20% of pts, most commonly in TP53+ (41%), IDH1+ (24%) and FLT3-ITD+ or TKD+ (23%) groups. Considering ED, the complete remission (CR) rates were above 50% in the CBF-AML (72%), and NPM1+/FLT3-ITD- (67%) groups, whereas in other groups they ranged between 30% and 50%, except for TP53+ at 17%. The 3-y disease-free survival (DFS) ranged from 0 to 13% for most groups, whereas only NPM1+/FLT3-ITD- (27%) and CBF-AML (30%) pts appeared to gain long-term benefit from 7 and 3. Three-y OS was 0-16% for most groups, except NPM1+/FLT3-ITD- pts at 27% and CBF-AML pts at 33%.

Conclusion: Our analysis of a large older AML pt cohort provides evidence of the feasibility of classifying this disease based upon long-term benefit from 7 and 3. We also use cytogenetic/dominant mutation clone identification for precision medicine-based assignment to novel targeted therapies where 7 and 3 lacks impact on long-term outcome. These data can also serve as a historical control population for pts treated as part of the LLS AML Master Study and for other efforts brought forth to advance novel therapies in select genomic groups of older AML pts.

Support: U10CA180821, U10CA180882, U10CA180861, U24CA196171

ClinicalTrials.gov Identifiers: NCT00048958 (8461) and NCT00900224 (20202)

View largeDownload slide

View largeDownload slide

Close modal