Quantitative Profiling of Codon 816 KIT Mutations Can Aid in The
Total Page:16
File Type:pdf, Size:1020Kb
Letters to the Editor 1574 Acknowledgements 3 Hurt EM, Wiestner A, Rosenwald A, Shaffer AL, Campo E, Grogan T et al. Over expression of c-maf is a frequent oncogenic event in This study was supported by a grant from Cancer Research Society multiple myeloma that promotes proliferation and pathological Inc. and a grant from Leukemia and Lymphoma Research Society interactions with bone marrow stroma. Cancer Cell 2004; 5: 191–199. of Canada to HC. 4 Suzuki A, Iida S, Kato-Uranishi M, Tajima E, Zhan F, Hanamura I Hong Chang1,2,QQi1,2,WXu3 and B Patterson1,2 et al. ARK5 is transcriptionally regulated by the Large-MAF family 1Department of Laboratory Hematology, Toronto General and mediates IGF-1-induced cell invasion in multiple myeloma: ARK5 as a new molecular determinant of malignant multiple Hospital, University Health Network, Toronto, Ontario, myeloma. Oncogene 2005; 24: 6936–6944. Canada; 2 5 Avet-Loiseau H, Facon T, Grobois B, Magrangeas F, Rapp MJ, Department of Laboratory Medicine and Pathobiology, Harousseau JL et al. Intergroupe Francophone du Myelome. Toronto General Hospital, University Health Network, Oncogenesis of multiple myeloma: 14q32 and 13q chromosomal Toronto, Ontario, Canada and 3 abnormalities are not randomly distributed, but correlate with Department of Biostatistics, University Health Network, natural history, immunological features, and clinical presentation. University of Toronto, Toronto, Ontario, Canada Blood 2002; 99: 2185–2191. E-mail: [email protected] 6 Fonseca R, Blood E, Rue M, Harrington D, Oken MM, Kyle RA et al. Clinical biologic implications of recurrent genomic aberrations in References myeloma. Blood 2003; 101: 4569–4575. 7 Chang H, Samiee S, Qi WY, Yi QL, Mikhael J, Chen C et al. Genetic 1 Nishizawa M, Kataoka K, Goto N, Fujiwara KT, Kawai S. v-maf,a risk identifies myeloma patients who do not benefit from autologous viral oncogene that encodes a ‘leucine zipper’ motif. Proc Natl stem cell transplantation. Bone Marrow Transplant 2005; 36: Acad Sci USA 1989; 86: 7711–7715. 793–796. 2 Chesi M, Bergsagel PL, Shonukan OO, Martelli ML, Brents LA, Chen 8 Chang H, Stewart AK, Qi XY, Li ZH, Yi QL, Trudel S. T et al. Frequent dysregulation of c-maf proto-oncogene at 16q23 by Immunohistochemistry accurately predicts FGFR3 aberrant expres- translocation to an Ig locus in multiple myeloma. Blood 1998; 91: sion and t(4;14) in multiple myeloma. Blood 2005; 106: 4457–4463. 353–355. Quantitative profiling of codon 816 KIT mutations can aid in the classification of systemic mast cell disease Leukemia (2007) 21, 1574–1576; doi:10.1038/sj.leu.2404680; were also analyzed by direct sequencing using the standard published online 5 April 2007 dideoxy-chain termination (Sanger) method. Cases that were negative for codon 816 KIT mutation were also assessed using a D816V mutation specific-probe TaqMan real-time quantitative The KIT receptor tyrosine kinase is normally highly expressed in PCR method. To increase the sensitivity of analysis in BM biopsy myeloid cells and mast cells and promotes growth through material, mast cell clusters, identified on parallel hematoxylin interaction with its ligand, stem cell factor.1 Activating kinase and eosin-stained section, were manually dissected from domain point mutations in KIT at codon 816 in exon 17, formalin-fixed paraffin-embedded sections before DNA extrac- predominantly Asp816Val and less commonly Asp816Phe, are tion and PCR amplification in a subset of cases. highly associated with mast cell disorders (MCD) and result in The pyrosequencing genomic PCR assay assessed exon 17 of ligand-independent kinase signaling.2,3 A number of studies KIT between codons 815 and 822, with a sequencing dispensa- have shown that KIT mutations can be present in a variety of cell tion order maximized for interrogation of codon 816. PCR lineages4 and this results in a spectrum of systemic MCD with (forward 50-TCATGGTCGGATCACAAAGAT, reverse 50-Biotin- associated hematological clonal non-mast cell lineage disease CAGGACTGTCAAGCAGAGAATGG) was done for 35 cycles at (AHNMCD) that have differing clinical behavior.5,6 Mast cells in 941C for 30 s, 561C for 90 s, and 721C for 45 s, followed by bone marrow (BM) in MCD can be easily enumerated by direct purification of product over streptavidin sepharose beads and microscopic examination, with the aid of mast cell tryptase or sequencing using a third primer (50-TGATTTTGGTCTAGCCAG) CD117 immunostaining, or by flow cytometric analysis (e.g. performed on a PSQ HS 96 Pyrosequencer (Biotage, Uppsala, gating on CD117 þ CD25 þ CD2À mast cells). However, MCD Sweden). PCR were run in duplicate and quantification of the with AHNMCD is not as easily recognized. We show here that ratio of codon 816 mutated versus unmutated PCR product was quantitative pyrosequencing and mutation-specific quantitative determined by averaging the peak heights. To optimize the polymerase chain reaction (qPCR) assays for the detection of KIT relative quantitative features of the pyrosequencing assay, PCR codon 816 mutation can be used to highlight the mixed and conditions were optimized to allow as small a number of pure forms of MCD to better aid in disease classification. amplification cycles as possible to generate adequate product Pyrosequencing is a DNA sequencing-by-synthesis method for sequence determination. Assuming most tumor cells will developed for single-nucleotide polymorphism analysis7 that we have one normal and one mutant allele of KIT, the percentage of have adapted here to provide quantitation of the levels of mutated PCR product would be expected to be half as high as mutated and unmutated product at codon 816 of KIT. For this the number of tumor cells. In cases with low level of mutated study, genomic DNA was isolated from BM core biopsy and/or PCR product, qPCR was also performed using KIT-816V BM aspirate samples from 39 patients with morphological (mutated) and KIT-816D (unmutated) FAM dye-labeled MGB evidence of MCD using WHO criteria,6 and subjected to probes in separate PCRs using the TaqMan method on a 7900HT pyrosequencing of KIT exon 17. The majority of cases (n ¼ 34) detection system (Applied Biosystems, Foster City, CA, USA). Leukemia Letters to the Editor 1575 Figure 1 Correlation of levels of KIT D816V mutation with myeloid cell counts as an aid in classification of the type of mast cell disorder. Mast cell and combined non-mast cell myeloid cell counts were done on BM aspirates, and compared to percentage of KIT mutated PCR product detected in a parallel BM aspirate sample. Final diagnoses listed in groups 1 (cases 1–6), 2 (cases 7–10) and 3 (cases 11–16) were based on a review of all pathological material and clinical features. Table 1 Correlation of KIT mutation levels and clinical disease patterns in MCD with D816V No. Age/gender Mast Other myeloid PB Eos (K/UL) %KIT mutant Final diagnosis (pyrosequencing) 1 66/F 94 3 0.02 20 Mast cell leukemia 2 65/F 50 23 0.04 30 Pure MCD 3 47/M 20 43 0.17 17 Pure MCD 4 58/F 20 56 0.06 6 Pure MCD 5 60/F 10 56 0.05 6 Pure MCD 6 71/F 10 56 0.11 13 Pure MCD 7 42/F 25 36 0.44 32 MCD-EOS 8 56/F 30 39 1.09 19 MCD-EOS 9 59/F 2 71 0.82 5 MCD-EOS 10 66/F 15 69 0.57 47 MCD-EOS 11 70/M 60 15 0.04 33 MCD-CMML 12 57/M 30 30 0.8 21 MCD-CMML 13 48/M 12 50 0.04 31 MCD-MDS 14 57/M 1 71 0 45 MCD-MDS 15 67/M 1 82 0.48 45 MCD-MPD 16 67/M o1 71 0.12 40 MCD-MDS Abbreviations: ANMHD; associated clonal hematologic non-mast cell lineage disorder (unspecified type), CMML; chronic myelomonocytic leukemia, EOS; associated isolated eosinophilia (normal reference range 0.04–0.4 Â 109/L), MCD; mast cell disorder, MDS; myelodysplastic syndrome, MPD; myeloproliferative disorder (unspecified type). Thermocyling conditions were 951C for 10 min followed by 40 the Sanger method correlated with pyrosequencing in all cycles of 951C for 15 s and 611C for 1 min. The sensitivity of samples with greater than 15% mast cells but failed to detect each assay was determined by serially diluting DNA with KIT- unequivocally mutations below that level. KIT mutation was not 816V into DNA obtained from cells with umutated KIT. The unequivocally detected by pyrosequencing in any case where sensitivity established by this technique was 5% mutated cells the mast cell counts were below 5% establishing a lower limit of for pyrosequencing, 1% for qPCR and 20% for direct Sanger sensitivity of detection of this assay. Pyrogram peak heights sequencing. corresponding to the mutated product which were below 5%, Using the exon 17 pyrosequencing assay, we tested 74 were considered equivocal and those cases were subjected to samples (29 BM biopsies, 45 BM aspirates) from 39 patients with further analysis with mutation-discrimination qPCR and de- morphologic evidence of MCD. The levels of marrow involve- tected D816V KIT mutation in five additional patients for an ment determined by mast cell number in a 400-cell aspirate overall mutation incidence of 54% for all MCD cases. There differential or semi-quantification of the area occupied by mast were no discordances between qPCR and pyrosequencing when cell aggregates in a parallel section of the biopsy varied from mast cell counts were above 5%. 1 to 60% but was below 5% in 21 of the cases. Codon 816 KIT Figure 1 shows the correlation between the numbers of mast mutation was detected in 16 of 39 patients (41%) by cells and myeloid cells in BM aspirate samples and the levels of pyrosequencing and included D816V in 15 cases and D816F mutated PCR product determined in the 16 cases with KIT in one.