| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cancer Diagnostics |
1 Departments of Pathology and 2
Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia.
3 Departments of Medicine and 4
Pathology, University of Melbourne, VIC, Australia.
5 Medical Department, Novartis Pharmaceuticals, North Ryde, NSW, Australia.
aAddress correspondence to this author at: Department of Pathology, Peter MacCallum Cancer Centre, Locked Bag No. 1 A’Beckett Street, Melbourne, Victoria, 8006, Australia. Fax 61-3-9656-1460; email angela.tan{at}petermac.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We developed 2 methods for detection of KIT D816V in SM patients. The first uses enriched sequencing of mutant alleles (ESMA) after BsmAI restriction enzyme digestion, and the second uses an allele-specific competitive blocker PCR (ACB-PCR) assay. We used these methods to assess 26 patients undergoing evaluation for SM, 13 of whom had SM meeting WHO classification criteria (before variation testing), and we compared the results with those obtained by direct sequencing.
Results: The sensitivities of the ESMA and the ACB-PCR assays were 1% and 0.1%, respectively. According to the ACB-PCR assay results, 65% (17/26) of patients were positive for D816V. Of the 17 positive cases, only 23.5% (4/17) were detected by direct sequencing. ESMA detected 2 additional exon 17 pathogenic variations, D816Y and D816N, but detected only 12 (70.5%) of the 17 D816V-positive cases. Overall, 100% (15/15) of the WHO-classified SM cases were codon 816 pathogenic variation positive.
Conclusion: These findings demonstrate that the ACB-PCR assay combined with ESMA is a rapid and highly sensitive approach for detection of KIT D816V in SM patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
KIT encodes the protein KIT, a member of the type III transmembrane receptor protein–tyrosine kinase family. On binding of the ligand stem cell factor, dimerization of 2 KIT proteins occurs, leading to phosphorylation and a signaling cascade promoting cell growth and proliferation (1). Activating pathogenic variations of KIT occur in the juxtamembrane domain and the kinase domain, such as D816V, which stabilizes the activated conformation of the protein (2). The D816V pathogenic variation causes ligand–independent constitutive phosphorylation and activation of KIT, leading to uncontrolled growth (3)(4)(5)(6). Because of its transforming ability, D816V may play a major role in SM and is included in the consensus WHO SM classification criteria (7).
To fulfill SM WHO criteria, either the major and at least 1 minor criterion or 3 of the 4 minor criteria must be satisfied. Briefly, the major criterion is the visualization of multifocal dense aggregates (>15 mast cells) in a trephine section (confirmed using mast cell tryptase). The minor criteria are (a) >25% of the mast cells are immature or spindle shaped in the trephine section, (b) a KIT point mutation at codon 816, (c) presence of extra–cutaneous infiltrate of mast cells that coexpress CD117 with CD2 and/or CD25, and (d) total serum tryptase >20 µg/L.
Juxtamembrane pathogenic variations and SM patients without the D816V pathogenic variation are sensitive to the tyrosine kinase inhibitor imatinib mesylate (imatinib), but the D816V pathogenic variation confers resistance to imatinib both in vitro and in vivo (8)(9)(10). Thus detection of D816V in SM patients would indicate that an alternative treatment to imatinib should be sought.
Crystal structure analysis of KIT has elucidated a mechanism for imatinib resistance (11). Imatinib can bind to the kinase domain only in the inactive state. The D816V variation stabilizes the active kinase state, so imatinib is unable to bind and inhibit KIT.
Although D816V is common in SM patients, the reported incidence has been highly inconsistent (5)(12)(13)(14)(15)(16), possibly due to patient heterogeneity (8)(14) or, more importantly, due to the tissues tested and the detection methods used.
Peripheral blood has been used in some SM studies because D816V is thought to originate in a pluripotent hematopoietic progenitor cell and may be present in cell lineages other than mast cells (12)(17). The number of mast cells present in peripheral blood and the presence of the pathogenic variation in other mature myeloid cell lineages, however, appear to depend on the severity of the disease (17). Thus bone marrow samples, where the affected progenitor cells and mast cells are most likely to reside, may yield more useful information than peripheral blood, but the mast cell population of bone marrow aggregates can also be low (18).
Accurate detection of variations present in a small fraction of cells can be difficult. A method with high sensitivity is required in which the rare variant allele will not be overwhelmed by the large proportion of wild-type alleles. Numerous techniques have been developed to enhance the detection of the rare allele signal (19).
In this study, we developed 2 methods to rapidly and sensitively detect codon 816 pathogenic variations in SM patients. The first method, enriched sequencing of mutant alleles (ESMA), is an adaptation of quantitative enriched PCR (22). The second method, allele-specific competitive blocker PCR assay (ACB-PCR), is a modified method of allele-specific amplification (25).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
dna extraction
For the dilution experiments, HMC-1 cells and normal peripheral blood mononuclear cells were combined in the appropriate proportions before DNA extraction. The samples for the dilution experiments were extracted with standard salting-out techniques (27). DNA from peripheral blood and bone marrow aspirate samples was extracted with the Wizard® Genomic DNA Purification Kit (Promega) according to the manufacturer’s instructions. Paraffin-embedded samples (2–5 microns) of bone marrow trephine, skin, spleen, or liver were sectioned and DNA was extracted with a 4-day, 56 °C proteinase K digest, then isolated by the Magnesil® Genomic Fixed Tissue System (Promega) or the DNeasy Tissue Kit (Qiagen) according to the manufacturers’ instructions.
primers
All primers are listed in Table 1
. The primer binding sites and BsmAI digestion restriction sites of KIT exon 17 are shown in Fig. 1
.
|
|
direct sequencing of exon 17
Before performing the sequencing reactions, we PCR–amplified 10–100 ng DNA in a total volume of 25 µL containing 200 nmol/L of each primer Ex17_f and Ex17_r, 2.0 mmol/L MgCl2, 0.2 mmol/L each dNTPs, 0.5 units HotStar Taq (Qiagen), and 1x Buffer. After an initial DNA polymerase activation step (95 °C for 15 min) 45 cycles of 94 °C, 30 s/60 °C, 30 s/72 °C, 30 s were performed, with a final 72 °C extension for 10 min. Because the majority of samples were obtained from paraffin-embedded tissue, in which DNA may be highly degraded, this PCR reaction was also used to assess the amplification ability of the extracts before the allele-specific competitive blocker assay (see Fig. 1
in the online Data Supplement. The 204-bp products were visualized on a 2% ethidium bromide-stained agarose gel.
We prepared the 204-bp products for sequencing with EXO-SAP-IT (USB) according to the manufacturer’s instructions. Then, 3.5 µL of the enzymatically cleaned PCR products were used as a template for the cycle sequencing reaction. Both forward and reverse sequencing were performed with BDT version 3.1 Chemistry (Applied Biosystems) according to the manufacturer’s instructions. Briefly, 2.0 µL BDT premix, 2.0 µL of 5x sequencing buffer, 1.0 µL of either the Ex17_f or Ex17_r primer (10 µmol/L) and 6.5 µL dH2O were combined with 3.5 µL of template. The cycling conditions were 96 °C for 1 min, followed by 25 cycles of 96 °C, 10 s/50 °C, 5 s/60 °C, 4 min. After cycle sequencing was completed, products were ethanol precipitated and analyzed on an ABI 3100.
esma
To reduce the number of normal alleles relative to variant alleles, 8 µL of the 204-bp exon 17 PCR product was digested with 5 units of BsmAI overnight at 55 °C and visualized on an agarose gel. In normal DNA, after digestion with BsmAI, which recognizes GANAC, 3 products are generated, 98, 72, and 34 bp (Fig. 1
). In DNA carrying the D816V pathogenic variation, 1 BsmAI recognition site is lost (GANTC), leaving 1 cut site as a result; a 34-bp fragment and a larger 170-bp variant fragment are generated. The larger variant fragment is also generated if any other alteration affects the GANAC recognition sequence. Therefore, variations affecting the 2nd and 3rd base of codon 816 as well as the 2nd and 3rd base of codon 815 will be detected. The sequence of the 170-bp digested product was assessed by stabbing the band with a sterile tip and placing it into a 2nd-round seminested PCR reaction. The PCR conditions were the same as described for the exon 17 presequencing PCR reaction, except the forward primer BsmAI_f was used. After checking that the expected 160-bp fragment was of acceptable quality on an agarose gel, we analyzed the product (enriched for the variant allele) by sequencing. The 160-bp PCR products were prepared with EXO-SAP-IT, cycle sequenced using BDT version 3.1, and ethanol precipitated as described above for direct sequencing of the exon 17 PCR products (see Fig. 1
in the online Data Supplement).
acb-pcr
Approximately 10–100 ng of DNA was amplified in a total volume of 25 µL containing 200 nmol/L of primer Normal_f, 400 nmol/L each of Variant_f and ACB_r, 2.0 mmol/L MgCl2, 50 µmol/L each dNTPs, 0.5 units HotStar Taq and 1x Buffer. After an initial DNA polymerase activation step (95 °C for 15 min), 45 cycles of 94 °C, 30 s/64 °C, 30 s/72 °C, 30 s were performed with a final 72 °C extension for 10 min. The 99-bp products were visualized on an agarose gel.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Additional mismatched bases are introduced near the 3' end, which aids in differentiation (20). Specificity of the reaction is further enhanced by use of a 3' end label or "blocker" introduced into the 3' end of the normal primer, which physically prevents amplification of the normal allele. Under stringent conditions, only the variant allele is amplified. The optimal conditions for the ACB-PCR assay were determined by adjusting the annealing temperature, concentration of dNTPs, and primer concentration, all of which affect the specificity of allele-specific PCR reactions (28). The HMC-1 cell line and normal peripheral blood mononuclear cells were used to determine the optimal conditions for the ACB-PCR assay. If nonoptimal conditions are used in the ACB-PCR assay, allele-differentiating primers do not bind specifically, and mispriming will occur from the normal template generating a false-positive product. None of the 30 negative control samples were amplified in repeated ACB-PCR assays under the optimized conditions used in this study (results available on request).
detection sensitivity
The detection sensitivity of both methods was assessed using dilutions of the HMC-1 cell line with normal peripheral blood mononuclear cells. Dilutions of 1/10, 1/30, 1/100, 1/300, and 1/1000 were tested.
With ESMA, at 100% of HMC-1 cells, the variant T allele peak was substantially larger than the wild-type A peak (Fig. 2A
). At 10% HMC-1 cells, the variant T peak was equivalent in size to the wild-type A peak. The variant T peak was still observed easily at 3% HMC-1 cells, but at 1% HMC-1 cells the variant peak was less readily detected. It appears that ESMA is reproducibly sensitive to the level of at least 1/30 or possibly 1/100, or 1 cell in 100. ESMA is more sensitive than direct sequencing (for which the variant cell population must be at least 20% to be detectable).
|
The detection sensitivity of HinfI enzymatic digestion of exon 17 PCR products was also assessed as a method for detecting the D816V pathogenic variation, because HinfI recognizes GANTC, which is created in samples with the D816V variation (5). However, the variation could not be detected at 10% HMC-1 cells, so this method would not be sufficiently sensitive for many cases of SM (results available on request).
The ACB-PCR assay was reproducibly sensitive to the level of 0.1% or 1 cell in 1000 (Fig. 2B
). No amplification was observed in the normal DNA samples or in the no-template blank sample. Strong amplification products were observed at 100%, 10%, 3%, and 1% HMC-1 cells. Although amplification was weaker at the levels of 0.3% and 0.1%, any amplification in this assay indicates the presence of D816V variant alleles.
sm patient samples
Both the ACB-PCR assay and ESMA were used to analyze the 26 patient samples. Direct sequencing was also performed for comparison. Each sample was analyzed in duplicate, and a positive control (HMC-1 cell line) and a no-template control blank were included in each PCR assay. At least 3 different negative control DNA samples were run alongside test samples in the ACB-PCR assay. Potential false-negative samples in the ACB-PCR assay were controlled for by assessing the PCR amplification of KIT exon 17. The ACB-PCR assay detected 17 patients who were D816V positive (see Fig. 2
in the online Data Supplement); 12 were from the SM-WHO group and 5 were from the subdiagnostic group. In comparison, ESMA detected only 12 D816V-positive patients (9 SM-WHO, 3 subdiagnostic), all of whom were detected by the ACB-PCR assay (see Fig. 3A in the online Data Supplement). However, pathogenic variations were detected in 2 other patients, D816Y (SM-WHO) and D816N (subdiagnostic) [see Fig. 3, B and C, in the online Data Supplement].
After variation testing results were obtained, a total of 16 patients were found to fulfill SM WHO criteria. However, because variation testing was performed only on the skin lesion of patient 5, the incidence of KIT variations in SM patients in this study is based on 15 SM patients.
In patients in whom more than 1 type of tissue was tested, identical variation states were obtained with the exception of patient 10. In patient 10, although the bone marrow was involved morphologically, the variation D816Y was detected only in the liver sample. This presentation is unusual, but because only small amounts of tissue were available for diagnostic variation analysis, the variation may have gone undetected in the bone marrow trephine.
The number of D816V-positive patients identified was consistent with the detection sensitivity of the method used. The ACB-PCR assay (the most sensitive method) detected the highest number of D816V cases (17), whereas direct sequencing (the least sensitive method) detected only 4 D816V-positive cases. The combined results of all 3 methods are summarized in Table 2
, and a summary of the variation status of each patient is shown in Table 1
of the online Data Supplement.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We used 2 cost-effective approaches to enable detection of low levels of variant alleles. The first approach was modified from a previous sensitive method, in which 2 rounds of PCR amplification are performed and the variant allele is selectively amplified in the 2nd reaction after restriction enzyme digestion (22). ESMA uses 2 rounds of PCR amplification in a seminested PCR reaction after BsmAI digestion of the normal alleles. The second approach used primer 3' end-base discrimination. Blocking amplification of the normal allele with a 3'-end dideoxynucleotide or locked nucleic acid has also been used to detect variations at the level of 3 x 10–9 in mouse genomes and 100-fold excess of wild-type in BRAF cell line studies, respectively (35)(36). The cost of manufacturing primers with 3' modified locked nucleic acids is high, however, and primer with added 3' dideoxynucleotides is not widely available commercially, so this expertise-requiring modification must be performed in the laboratory.
Costly and time-consuming approaches such as cell sorting (30) and laser microdissection(31) have been used previously to increase the proportion of variant mast cells, but additional mast cell purification steps were not required for sensitive detection in either method described in this study. Furthermore, both methods were designed to enable detection in archival paraffin-embedded tissues. Similar to other groups, we observed that to maximize DNA recovery, bone marrow trephines should be fixed with 10% buffered formalin (32).
High detection sensitivity (1/1000) was achieved in an assay with peptide nucleic acid-mediated clamping and melt-point analysis of 2 fluorescent-labeled oligonucleotide probes, but this procedure is expensive because of the required probes and peptide nucleic acid clamps (33). Ligation-based methods also have high detection sensitivities but were not explored in this study because of the relative complexity of the method for the detection of low-abundance point variations (34).
Although the incidence of the D816V pathogenic variation in SM-WHO patients found in this study, 100% (15/15), was similar to previous studies that used sensitive methods of variation detection, e.g., 80% (16) or 100%(13), this figure was much higher than other reports of 25% when only peripheral blood was investigated (12) and 31% when peripheral blood was analyzed by direct sequencing (14). The low incidence reported is likely due to the methods used, i.e., direct sequencing (without any purification of the mast cell population) and sampling of peripheral blood where the level of variant allele is too low for detection.
The pathogenic variation D816Y has been previously reported in a patient with acute myeloid leukemia with a background of SM and also in pediatric patients with urticaria pigmentosa (6)(33)(37). Our patient with the D816Y variation met the criteria for a diagnosis of SM but also had morphologic features of essential thrombocytosis. He subsequently developed a mild chronic thrombocytosis of >500 x 109/L. According to the WHO classification, this patient’s disease was formally classified as SM with associated clonal hematological non–mast-cell lineage disease.
D816N has been reported in a patient with de novo childhood acute myeloid leukemia and sinonasal lymphomas (38)(39). The patient with D816N in this report had a clinical diagnosis of SM, with splenomegaly, rash, an increased tryptase, andeosinophilia, and the bone marrow showed a substantial abnormal loose mast cell infiltrate. FIP1L1-PDGFRA was negative. These features are suspicious but do not formally reach WHO criteria for SM.
The D816H and D816F pathogenic variations reported to be present in SM patients were not observed in this cohort of patients, although ESMA can detect both variations. Also, because all of the SM patients were D816V positive, investigation of whether some patients were carriers of the CHIC2 deletion resulting in FIP1L1-PDGFRA fusion was not investigated. This type of variation has been shown to be present in a subset of SM patients who are D816V negative (40).
The D816V pathogenic variation confers resistance to the drug imatinib. However, new tyrosine kinase inhibitors with activity against the KIT tyrosine kinase such as PKC412 are being examined and may be a suitable treatment for SM patients (41).
After KIT variation testing was performed, the SM WHO criteria fulfilled in each case were reassessed. All of the patients in the SM-WHO group (n = 13) had a point variation in codon 816 (D816V, n = 12; D816Y, n = 1). Of the 13 patients in the subdiagnostic group, 4 had a point variation in codon 816 (D816V, n = 3; D816N, n = 1). This feature in combination with the other WHO criteria originally fulfilled confirmed the diagnosis of SM in 3 patients. Of the remaining 10 patients in the subdiagnostic group, SM could be excluded in 2 patients, diagnoses in 5 patients were inconclusive, and in 3 patients other diseases were diagnosed (patient 2 had acute basophilic leukemia, patient 6 had cutaneous mastocytosis, and patient 7 had reactive anemia). Overall, 15/15 (100%) of SM patients (excluding patient 5, because in this case only skin was tested) were positive for codon 816 pathogenic variation.
The ACB-PCR assay was the most sensitive method for detection of the D816V pathogenic variation, and led to identification of 12/13 (92%) of variations in the SM-WHO group and 5/13 (38%) of the subdiagnostic group (Table 2
). The ESMA results for these D816V negative patients showed that 2 were in fact positive for other 816 codon alterations (D816Y in the SM-WHO group and D816N in the subdiagnostic group). D816Y and D816N could be detected only by ESMA, a result that highlights the complementary nature of these 2 methods in their ability to effectively analyze the patient samples for both D816V and other exon 17 alterations. However, additional ACB-PCR assays could be designed to sensitively detect the other codon 816 pathogenic variations such as D816Y and D816N. Once samples were found to be negative by the ACB-PCR assays, in clinically relevant patients ESMA could then be used to scan for other rare variations.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: SM, systemic mastocytosis; Imatinib, imatinib mesylate; ESMA, enriched sequencing of mutant alleles; ACB-PCR, allele-specific competitive blocker PCR; SM-WHO, WHO classified SM.
3 Human genes: KIT, cellular homologue of v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homologue, CHIC2, cysteine-rich hydrophobic domain 2.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Val in microdissected pooled single mast cells and leukemic cells in a patient with systemic mastocytosis and concomitant chronic myelomonocytic leukemia. Leuk Res 2002;26:979-984.[CrossRef][ISI][Medline]
[Order article via Infotrieve]The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J A Schumacher, K S J Elenitoba-Johnson, and M S Lim Detection of the c-kit D816V mutation in systemic mastocytosis by allele-specific PCR J. Clin. Pathol., January 1, 2008; 61(1): 109 - 114. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
