Abstract 2559

Introduction:

Activating mutations in FMS-like tyrosine kinase receptor-3 (FLT3) represent one of the most commonly detected mutations in acute myeloid leukemia (AML). Recently, a potent, investigational inhibitor of FLT3, AC220, has shown remarkable promise in clinical studies of relapsed/refractory AML patients with activating FLT3 mutations (Cortes et al, ASH 2011 abstract #2576). However, many patients ultimately relapse within weeks to several months. Next-generation deep sequencing using Pacific Biosciences RS Single Molecule Real Time (SMRT) technology revealed AC220-resistant FLT3 kinase domain (KD) mutations in 8/8 primary samples isolated at the time of loss of response to AC220 (Smith, et al, Nature 2012). Mutations isolated in patient samples included the F691L “gatekeeper” mutation and mutations at the activation loop residue D835. This observation and the fast rate of relapse seen in patients suggest blast cells conferring resistance against AC220 are likely present prior to initiation of AC220. While SMRT sequencing provides sensitive detection of mutations, it is not yet commonly available and requires significant bioinformatics analysis. Furthermore, the sensitivity of SMRT sequencing is estimated to be ∼1–2%, which is likely inadequate for the detection of resistant alleles pretreatment. Here, we sought to assess the sensitivity and reliability of a competitive allele-specific TaqMan PCR (castPCR) (Life Technologies) assay for the detection of AC220-resistant mutations in pre-treatment and relapsed primary AML samples, compared to that of SMRT sequencing.

Results:

The sensitivity of the FLT3 castPCR assay was determined for each codon substitution that can result in D835Y/V/F, Y842C/H and F691L point mutations utilizing FLT3 plasmids containing each mutation diluted from 100,000 to 10 copy numbers. The sensitivity of detection for D835Y/V and Y842H was 0.01%, Y842C was 0.1%, and 0.1–0.01% for the three different substitutions resulting in F691L. D835F and Y842H showed the greatest sensitivity of detection at 0.001%. The ability of castPCR to identify FLT3 mutations in cDNA was determined by utilizing Molm14-R cells which have an AC220 resistant D835Y mutation in FLT3-ITD. Molm14-R cells were serially diluted in parental Molm14 cells to test the sensitivity of the assay. The assay was capable of detecting D835Y with a sensitivity of 1:1000 cells. We then assessed the correlation of mutant allele burden determined by castPCR with previous results obtained from SMRT sequencing in pre-treatment and relapse samples from 8 patients who were treated with AC220. Briefly, RNA was isolated from primary patient samples, cDNA was generated and samples were assessed by castPCR for the specific D835 mutation that had previously been identified. Cast PCR detected all assayed D835Y, D835V, and D835F mutations in patient samples. The lowest allele expression detected for D835Y, D835V, and D835F was 1.4%, 15.6%, and 2.5% respectively. For the majority of samples (14/16), mutant allele burden determined by castPCR correlated well with mutation burden obtained by SMRT sequencing. Mutations in pretreatment samples were not detected by castPCR. Efforts are ongoing improve the sensitivity through further optimization, and to analyze primary samples for F691L mutations.

Conclusion:

In our current analysis, we determined that castPCR has a sensitivity of 0.1–0.001% for the detection of FLT3 KD mutations. In preliminary studies, we find good correlation between mutant allele burden previously obtained by SMRT sequencing and our current results with castPCR. As with SMRT sequencing, the sensitivity of the castPCR assay was not sufficient to detect mutations in pretreatment samples., CastPCR has the potential to rapidly identify the presence of AC220-resistant FLT3 KD mutations in primary samples obtained at the time of relapse, and to detect these mutations prior to overt relapse in patients treated with AC220 or other effective next-generation FLT3 inhibitors. With further optimization, this assay may enable detection of drug-resistant alleles prior institution of TKI therapy.

Disclosures:

Mokang:Life Technologies: Employment. Bao:Life Technologies: Employment. Shah:Ariad: Consultancy, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution