Abstract

Background: Next-Generation Sequencing (NGS) has been introduced to efficiently and simultaneously detect genetic variations in acute myeloid leukemia (AML). However, conventional PCR testing is still needed for rapid screening in AML diagnosis, targeted therapy implementation, and prognostic stratification. The PETHEMA group established a nationwide network of 7 reference laboratories in order to provide a rapid and standardized molecular diagnosis for adult patients with AML (Sargas et al., Haematologica 2021). The PCR-LMA study aimed to evaluate whether mutations detected by NGS correlate with those detected using conventional PCR: we present a preliminary report comparing results of FLT3-ITD, FLT3-TKD, NPM1, IDH1 and IDH2 gene mutations at AML diagnosis using PCR and NGS. Methods: Bone marrow samples were collected at diagnosis or at relapse/ refractory AML in any of the PETHEMA associated hospitals and centralized in the 7 laboratories. The PCR-LMA non-interventional study (NCT04446741) planned inclusion of 1400 patients in 2 years and 2500 samples. NGS analysis of 30 relevant genes was performed according to implemented protocols, only variants accomplishing the established quality control parameters were considered. Detection of FLT3, NPM1, IDH1 and IDH2 mutations was also performed using conventional methods based on PCR; cut-off for positivity was 3%. Technical cross-validation round (CVR) were performed to establish consensus parameters among 7 central laboratories for NGS and PCR analysis and variant reporting. The CVR for NGS showed error rate of 10.9% for variants with VAF>5% achieving high concordance. Variants with VAF<5% showed error rate of 40%. Then, NGS cut-off positivity was established as VAF≥5%, excepting variants with strong clinical evidence that were individually reported until VAF 1%. Results: From October 2019 to April 2021, 800 newly diagnosed AML patients with available essential data from 71 institutions were included in the PCR-LMA protocol to evaluate mutations detected by NGS and PCR. Paired NGS and PCR diagnostic samples were analyzed in central laboratories. Median age at diagnosis was 64 (range 19-98) years, 58% were male. Median leucocyte count was 9.2 ×109/L, and 81% had ECOG <2. All 800 patients had available results with NGS and conventional PCR was reported in 778 (97.3%) for NPM1, 791 (98.9%) for FLT3-ITD, 735 (91.9%) for FLT3-TKD, 570 (71.3%) for IDH1 and 573 (71.6%) for IDH2 patients. NGS positive results (i.e VAF>5%) were as follows: NPM1 in 185 (23.8%) with median VAF 35.7%, FLT3-ITD in 95 (12%) with median VAF 31.6%, FLT3-TKD in 41 (5.2%) with median VAF 24%, IDH1 in 43 (7.9%) with median VAF 42% and IDH2 in 82 (14.3%) patients with median VAF 41.8%. PCR positive results were: NPM1 in 184 (23.6%), FLT3-ITD in 127 (16.1%), FLT3-TKD in 44 (6%), IDH1 in 45 (7.5%) and IDH2 in 79 (13.8%) patients (Figure 1). Overall concordance between NGS and PCR was 98.9% (183/185) for NPM1, 73.4% (94/128) for FLT3-ITD, 89.1% (41/46) for FLT3-TKD, 95.6% (43/45) for IDH1 and 96.3% (79/82) for IDH2 (Figure 2). The NGS positive results were concordant by PCR as follows: 183 (98.9 %) for NPM1, 94 (98.9%) for FLT3-ITD, 35 (94.6%) for FLT3-TKD, 43 (95.6%) for IDH1, and 79 (96.3%) for IDH2. Regarding PCR, all the patients with positive PCR for NPM1 (n=184), IDH1 (n=43) and IDH2 (n=79) had concordant positive results by NGS. The PCR positive results were concordant by NGS: 94/127 (74%) for FLT3-ITD and 41/44 (93.2%) for FLT3-TKD. Discrepancies regarding FLT3-ITD were frequent (26.6%) (Table 1): 33 patients had positive PCR with negative NGS using VAF>5% as cut-off. Among 33 FLT3 negative NGS results, 23 patients had low VAF (<5%) variants and 10 patients remained undetected. NPM1, IDH1 and IDH2 showed a high concordance between both methods. An NGS with VAF >1% but <5% was found in 2 (0.3%) patients for NPM1, 20 (2.7%) for FLT3-TKD, 5 (0.9%) for IDH1 and none for IDH2. Regarding timing for analysis reports, median days until PCR results were 7 (interquartile range 4-13), and 28 (interquartile range 20-38) for NGS. Conclusion: We show high concordance between NPM1 and IDH results using PCR and NGS for AML patients. However sensible discrepancies are observed for FLT3 mutation that could be mitigated lowering the VAF cut-off to 1%. In our context, rapid screening for these mutations can be provided only by conventional PCR. This work was supported by Novartis grant, PI18-01340, PI19/00730, FI19/00059. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call