Patterns of Secondary Resistance Differ in Patients (pts) with Acute Myeloid Leukemia (AML) Treated with Type I Versus Type II FLT3-Inhibitors (FLT3i’s)

Background

Despite promising responses achieved with FLT3i’s in AML, response durations remain short (4 – 14 months) (ADMIRAL, QUANTUM-R), frequently driven by the emergence (acquisition or clonal expansion) of mutations that drive secondary resistance. These include mutations involving the activating loop or gatekeeper residues of FLT3 (on target), or in parallel pro-survival signaling pathway genes such as RAS/MAPK, BCR-ABL (off target). Understanding the profile of secondary mutations in pts treated with type I versus type II FLT3i-based therapies may help design concomitant or sequential strategies to abrogate resistance.

Methods

We used a next generation sequencing (NGS)-based myeloid panel to compare bone marrow mutational profiles pre- and post-FLT3i based therapy, to identify emergent mutations at relapse, in FLT3-mutated AML (ITD and/or D835) pts treated with FLT3i-based therapies (single agent or combinations) at our institution between Jan 2012 and Dec 2019. Pts with an initial CRc response (CR + CRp + CRi), as defined in the ADMIRAL trial, and subsequent relapse with available FLT3 and NGS-based myeloid profiles pre-therapy and at relapse were included. A mutation was deemed emergent if the mutation was detected at relapse but was not detected pre-therapy.

Results

Among 810 consecutive pts with FLT3-mutated AML who received FLT3i-based therapy, 546 had initial response and subsequent relapse. Of these 67 pts had available FLT3 and NGS panel pre-therapy and at relapse. Pt characteristics are in Table 1.

Among 67 pts, 31% received type I (gilteritinib 12 pts, midostaurin 7 pts or crenolanib 2 pts), and 69% type II FLT3i-based therapies (sorafenib 39 pts, quizartinib 7 pts). Sixty-five (97%) pts received FLT3i’s in combination with either low intensity [43 (64%)] or cytotoxic chemotherapy [22 (33%)], and only 2 (3%) pts received single agent gilteritinib in salvage. Twenty-four (36%), 17 (25%) and 26 (39%) patients achieved a CR, CRp and CRi, respectively (only responders included for this analysis). 21 pts (31%) underwent allogeneic transplant in CRc.

Thirty-seven (55%) pts had emergent mutations at relapse (Table 2). The most frequent emergent mutations across all pts were FLT3-D835 in 21%, non-canonical FLT3 mutations (N676K, N841K) in 5%, RAS/MAPK pathway mutations (including N/K-RAS, PTPN11, CBL) in 13%, IDH1/IDH2 in 9%, WT1 in 7%, and TP53 in 7% .

The most common emergent mutations at relapse post type II FLT3i-based therapies (N=46) were FLT3-D835 in 14 (30%), IDH1/IDH2 in 5 (10%), TP53 in 5 (10%) and WT1 in 5 (10%). In addition, two pts had non-canonical emergent FLT3 mutations at relapse: FLT3-N676K and FLT3-N841K in one pt each. Importantly, mutations in the RAS/MAPK were noted in only 3 (6%) pts, treated with type II FLT3i-based therapies. Pts with emergent FLT3-D835 post type II inhibitors had inferior overall survival (OS) at relapse (2.6 months versus 6.7 months, P=0.002) (Fig 1A).

The most common emergent mutations in pts at relapse post type I FLT3i-based therapies (n=21), were in RAS/MAPK genes in 6 (29%): NRAS in 4, PTPN11 and CBL in 1 each. RAS/MAPK emergent mutations were noted in 4 of 14 (29%) pts treated with LIT + Type I FLT3 inhibitor, 2 of 2 pts treated with single agent gilteritinib, and no pts treated with CCT + type I FLT3i (n=5), suggesting potential differential secondary resistance pathways to specific combinations. Pts with emergent RAS/MAPK mutations had inferior OS at relapse (2.4 months versus 6.8 months, P=0.009) (Fig 1B). No pts had emergent FLT3-D835 or FLT3-F691L mutations post type I FLT3i-based therapies, however one pt had an emergent non-canonical FLT3-N676K, and another with FLT3-D835 alone pretherapy had an emergent FLT3-ITD at relapse.

Eighteen of 67 (26%) pts lost the FLT3 mutation at relapse with improved OS in pts with undetectable compared with persistent FLT3 mutations at relapse (9.9 months versus 4.6 months, P =0.029) (Fig 1C).

Conclusion

Specific, and in some cases targetable, pathways of secondary resistance were noted at relapse post type I and II FLT3 inhibitors, and possibly post specific combination partners. Understanding these differences and selecting type I or type II inhibitors with the optimal combination partner to target specific scenarios, may improve response durations. FLT3 mutations may no longer be detectable at relapse post FLT3i-based therapies highlighting the importance of repeated FLT3 testing at relapse.