A rapid, one step assay for simultaneous detection of FLT3/ITD and NPM1 mutations in AML with normal cytogenetics

Acute myeloid leukemia (AML) is a heterogeneous disease with diverse genetic abnormalities and variable responsiveness to therapy (Brunning et al, 2001; Baldus et al, 2007). Cytogenetic karyotyping for the recurrent chromosomal abnormalities used to define disease entity and to risk-stratify AML patients has been adopted by the World Health Organization (WHO) classification of hematopoietic and lymphoid tissues (Brunning et al, 2001). However, accumulating evidence demonstrates that the prognostic stratifications of AML patients based on karyotyping alone are still far from satisfactory (Baldus et al, 2007). One of the major reasons is that AML with normal karyotypes, considered to have intermediate prognosis, represents approximately 40–50% of de novo AML (Brunning et al, 2001; Baldus et al, 2007). Stratified prognostic determinants are therefore required to predict which patients in this category have increased risk of relapse, resistance to therapy or long-term disease outcomes. Recently, mutations in two highly frequent genes, Fms-like tyrosine kinase-3 internal tandem duplication (FLT3/ITD) and nucleophosmin (NPM1), were identified in AML patients, particularly those with normal cytogenetics (Rombouts et al, 2000; Döhner et al, 2005; Falini et al, 2005, 2007). In this AML subgroup, FLT3/ITD and NPM-1 mutations represent unfavorable and favorable prognosis, respectively (Rombouts et al, 2000; Döhner et al, 2005; Falini et al, 2005, 2007), although they have been frequently found to co-exist in these patients (Döhner et al, 2005; Falini et al, 2005). Patients having only NPM1 mutations had a significantly better overall and disease-free survival and a lower cumulative incidence of relapse than AML patients carrying both NPM-1 and FLT3/ITD mutations (Fröhling et al, 2002; Döhner et al, 2005; Falini et al, 2005). Therefore, combined analysis of both FLT3/ITD and NPM1 mutations is usually required for their appropriate prognostic relevance in AML patients with normal cytogenetics and identification of those mutations would help to determine therapeutic options and predict patient outcomes. This study developed and validated a novel, one step, rapid multiplex polymerase chain reaction (PCR) method for simultaneous detection of FLT3/ITD and NPM1 mutations by using genomic (g)DNA in de novo AML patients with normal cytogenetics. gDNA extracted from fresh blood or bone marrow aspirates of AML patients was used for a multiplex PCR procedure, with primer sets designed to detect the presence or absence of FLT3/ITD and NPM1 4 bp insertion mutations (Kottaridis et al, 2001; Döhner et al, 2005). Primers for FLT3 spanned approximately 330 bp to include the common internal duplication site (Kottaridis et al, 2001). Primers for NPM1 spanned approximately 170 bp to cover the clustered multiple insertional or insertion/deletion sites including so-called mutation A-F (Table I) (Döhner et al, 2005; Falini et al, 2005). These primers should positively detect all or almost all of the currently known mutations in AML patients. In addition, a specific set of HBG primers with an anticipated fragment length of 469 bp was also used to ensure DNA integrity and PCR amplification (Table I). The 20 μl multiplex PCR reaction solution consisted of 200 ng gDNA, 20 pmol of each primer, 25 mmol/l MgCl2, 2·5 mmol/l dNTP, 2 μl 10 × PCR buffer, 0·2 μl Qiagen HotTaq DNA polymerase and water. The PCR conditions were: initial denaturing for 15 min at 95°C, followed by 35 cycles at 95°C for 20 s, 60°C for 40 s and 72°C for 40 s with a final cycle of 30 min at 72°C. The PCR products were then diluted 1:30 and analyzed by capillary electrophoresis on an ABI 310 or ABI 3100 Genetic Analyzer according to the manufacturer's protocol. Selected cases were further analyzed by PCR amplification, subcloning and direct DNA sequencing as described previously (Huang et al, 2006). Peripheral blood and bone marrow aspirates from 35 AML patients with normal cytogenetics determined by conventional karyotype analysis were included in this study (15 male, 20 female; median age, 54 [range, 19–78] years). The leukemic blasts in the specimens ranged from 20% to 85%. The expected amplified products for wild type FLT3 (FAM), wild type NPM1 (HEX) and HBG were at 328 bp, 167 bp and 469 bp, respectively. FLT3/ITD mutation produced larger peaks with a range from 349 to 409 bp, 21 to 81 bp larger than the wild type, while NPM1 mutation yielded a single peak at 171 bp, 4 bp larger than the wild type (Fig 1A). Of 35 cases analyzed, 10 cases of FLT3/ITD (29%) and 11 cases of NPM1 (31%) mutations were identified. Four of 35 (11%) cases were positive for both FLT3/ITD and NPM1 mutations, while 18 of 35 (51%) cases were negative for either FLT3/ITD or NPM1 mutation (Fig 1B). The mutation profile and distribution were similar to those studies reporting a large cohort of AML patients with or without normal cytogenetics (Falini et al, 2005, 2007). The lowest detection limits were approximately 5% of leukemic blasts for both mutations, as determined by serial dilution experiments. The sensitivity of multiplex PCR was further validated by comparing the results with that of each individual tests for FLT3/ITD or NPM1 mutations. The results (with or without multiplex) were very similar (data not shown) and suggested that multiplex PCR had no significant impact on the sensitivity for two mutations in this assay. No false positive peaks were identified in various negative controls including leukemic or non-leukemic specimens. The diagnosis of no detectable mutations was only given for FLT3/ITD or NPM1 when the wild type FLT3, NPM and HBG were successfully amplified. Selected cases were further validated by PCR amplification, subcloning and direct DNA sequencing. (A) Capillary electrophoresis of FLT3/ITD and NPM1 mutations in AML with normal cytogenetics. Arrows indicate wild type NPM (167 bp); NPM1 mutation (171 bp); wild type FLT3 (328 bp); FLT3/ITD mutation (371 bp) and internal control HBG (469 bp), respectively. (B) Mutational profile of FLT3/ITD and NPM1 in three representative cases of AML with normal cytogenetics showing FLT3/ITD positive/NPM1 mutation negative (upper panel); both FLT3/ITD positive and NPM1 mutation positive (middle panel) and FLT3/ITD negative and NPM1 mutation positive (low panel) results. Specific diagnosis, subclassification and risk-stratification are now increasingly required for the management of AML patients in many Western countries. Direct gDNA sequencing is usually preferred for determining mutation status although the requirements of this method often make it impractical for a clinical diagnostic laboratory. Using a multiple PCR approach, Noguera et al (2005) reported the simultaneous detection of FLT3/ITD and NPM1 mutation by RT-PCR assay. However, RNA instability, needed for cDNA synthesis, and lack of a reliable internal control precluded a clinically useful assay. Our current method provides a one-step, rapid assay for simultaneous detection of FLT3/ITD and NPM1 insertion mutations. This method demonstrates several merits: (1) relatively stable and reliable starting test materials by using gDNA, (2) a simple and rapid, one-step test for simultaneous detection of two high frequent mutations by multiplex PCR, (3) the ability to cover all or almost all currently known mutations of FLT3/ITD and NPM1 by using specifically designed premiers, and (4) inclusion of a reliable internal control to ensure both DNA integrity and PCR amplification. The test provides excellent sensitivity and specificity for these mutations and can be readily performed in a clinical diagnostic laboratory. Thus, the assay provides clinical applicability for rapid pathological diagnosis and prognostic stratification of AML patients with normal cytogenetics.