Conflict of interest br Authors contributions br Acknowledgm

Conflict of interest
Authors׳ contributions
Acknowledgments This work was supported by grants from the Associazione Italiana Lotta alle Leucemie, Linfoma e Mieloma (AIL) – Sezione ‘Luciano Pavarotti’ Modena-ONLUS (to FF and LP) and from the Associazione Italiana per la Ricerca sul Cancro (AIRC, IG 14797-2013 to ML).
Introduction Mutations in the FLT3 gene have been described in about 25% of acute myeloid leukemia (AML). They are somewhat more common in acute promyelocytic leukemia (APL), and have been associated with an increased risk of relapse, decreased disease-free survival, decreased event-free survival, and decreased overall survival [1]. These mutations result in constitutive activation of the FLT3 protein and are of two types: internal tandem duplication (ITD) mutations in exon 14 resulting from the duplication and tandem insertion of a portion of the juxtamembrane (JM) domain of the FLT3 gene and missense mutations in exon 20 which alter the aspartic nitric oxide booster residue at position 835 (D835) within the kinase domain of the FLT3 protein. In the case of ITD mutations, the duplicated segment length ranges in size from 3 to several hundred base pairs and is always in-frame and therefore expected to produce a functional protein [2]. Rare deletion and deletion/insertion mutations affecting the FLT3 juxtamembrane region have been described in childhood acute lymphoblastic leukemia [3,4]. Here, we report two cases of deletion and deletion/insertion mutations in the juxtamembrane domain of FLT3 in adult AML. Proper identification of these mutations may have prognostic and therapeutic significance for AML patients.
Methods
Results and discussion Upon FLT3 ITD fragment analysis during routine molecular diagnostics work-up at presentation, both patient samples showed an unusual peak in the electropherogram (Fig. 1A). Besides the wild-type allele of 330bp, a shorter PCR product in the same reaction pointed to the presence of a mutated allele showing a deletion in the PCR-amplified juxtamembrane domain region. Fragments shorter by 3-bp (327-bp) in patient #1 with a mutant allele/wild-type FLT3 ratio of 0.29, and by 12-bp (318-bp) in patient #2 with a ratio of 0.49 were detected. These fragments were further analyzed by Sanger sequencing to elucidate the nature of the deletions. Compared to the wild-type FLT3 sequence, patient #1 had a 5-bp deletion (CTACG) mutation combined with a 2-bp (GT) insertion: c.1770_1774delCTACGinsGT mutation (Fig. 1B), giving an overall 3bp deletion as detected by FLT3 fragment analysis. The deletion resulted in a p.F590_V592delinsLF amino acid change in the juxtamembrane domain. Patient #2 had a c.1780_1791delTTCAGAGAATAT (12-bp deletion) mutation (Fig. 1B) resulting in p.F594_Y597del amino acid deletion in the juxtamembrane domain. Notably, these deletion and deletion/insertion mutations were in-frame and the reading frame of the FLT3 gene was preserved in both cases. Subsequently, the samples were tested for mutant versus wild type allele expression using cDNA fragment analysis. Both wild-type and mutant alleles were expressed at ratios comparable to the results of the FLT3 ITD assay (Fig. 2). FLT3 deletion and deletion/insertion mutations were previously reported in cases of pediatric acute lymphoblastic leukemia [3,4], but seldom described in adult acute leukemia. While the biological significance of this type of FLT3 mutations is unknown in human disease, a small 10-amino acid (Tyr589 to Tyr599) deletion in the juxtamembrane domain of FLT3 has been previously shown to lead to constitutive activation of the FLT3 protein in transformed murine IL3-dependent myeloid progenitor 32D cell line [7]. Similar deletion mutations are found in receptor tyrosine kinase KIT in gastrointestinal stromal tumors (GIST) [8,9]. An in-frame deletion of 7-amino acids in the juxtamembrane domain of the KIT gene resulted in receptor autophosphorylation and malignant transformation of mast cells [10]. These studies and our findings that both patients showed in-frame deletions with mRNA expressed (unfortunately, the samples did not yield enough material for a Western blot) suggest that deletion and deletion/insertion mutations in FLT3 juxtamembrane domain may lead to receptor activation. Animal models would be a way to prove this hypothesis and show, if inhibition of FLT3 can be therapeutically exploited in such cases. Whether the presence of these mutations in adult acute leukemia has prognostic significance warrants further investigation of a larger patient cohort.