Sensitive detection of BRAF V600E mutation by Amplification Refractory Mutation System (ARMS)-PCR
© Huang et al.; licensee BioMed Central Ltd. 2013
Received: 26 September 2012
Accepted: 9 October 2012
Published: 16 January 2013
BRAF mutations occur in approximately 8% of all human cancers and approach 50% in melanoma and papillary carcinoma of thyroid. These mutations provide potentially valuable diagnostic, prognostic and treatment response prediction markers. A sensitive, specific, low-cost assay to detect these mutations is needed.
To detect BRAF V600E mutation in formalin-fixed, paraffin-embedded (FFPE) tissue, we developed a method using Amplification Refractory Mutation System (ARMS)-PCR. This method was designed to amplify three products in a single reaction tube: a 200 bp common product serving as an amplification control, a 144 bp BRAF V600E specific product, and a 97 bp wild-type (wt) specific product. The sensitivity of this method was determined to be as low as 0.5% for the BRAF V600E allele in a wild-type background. This method was successfully validated in 72 thyroid tumors. It detected V600E mutation in 22 out of 33 (67%) of the conventional papillary thyroid carcinoma (PTC), 8 out of 12 (75%) of the tall-cell variant of PTC, whereas none of the 10 follicular variant of PTC showed BRAF V600E mutation. In addition, none of the 14 follicular adenomas and 3 follicular carcinomas had BRAF V600E mutation. As a comparison method, direct dideoxy sequencing found only 27 out of 30 (90%) mutations detected by ARMS-PCR method, suggesting that this ARMS-PCR method has higher sensitivity.
Our ARMS-PCR method provides a new tool for rapid detection of BRAF V600E mutation. Our results indicate that ARMS-PCR is more sensitive than automated dideoxy sequencing in detecting low BRAF V600E allele burdens in FFPE tumor specimen. The strategy of this ARMS-PCR design may be adapted for early detection of point mutations of a variety of biomarker genes.
BRAF gene encodes a serine/threonine protein kinase that functions downstream of RAS in the RAS-RAF-MEK-ERK signaling pathway, also known as mitogen activated protein kinase (MAPK) pathway, which is important in regulating cellular responses to extracellular signals including epidermal growth factor (EGF) . Upon activation by receptor binding, RAS proteins recruit RAF to the cellular membrane where RAF is phosphorylated . Activated RAF proteins further phosphorylate MEK1 and MEK2, which then phosphorylate ERK1 and ERK2. Phosphorylated ERK proteins regulate cellular functions through activation of transcription factors including p53, SMAD4, ELK1, c-Myc and c-Fos [2, 3]. Activating mutations in RAS and BRAF permit constitutive MAPK pathway activation independent of growth factor stimulation, thereby causing deregulation in cellular growth and survival. BRAF mutations occur in approximately 8% of all human cancers, with high mutation frequency in malignant melanoma (50-70%), classic papillary carcinoma of the thyroid (40-70%), colorectal cancer (CRC, 5-15%), ovarian cancer and hairy cell leukemia (5-100%) [4, 5]. A T1799A transversion resulting in valine-to-glutamate substitution at codon 600 (V600E) accounts for over 80% of all BRAF mutations [6, 7], and almost all mutations in thyroid tumors .
Molecular testing for BRAF mutation can provide valuable diagnostic information, treatment selection, and may help predict prognosis. Like KRAS mutations, CRC patients with activating BRAF mutations do not respond to anti-EGFR monoclonal antibody (MoAb) treatment [9, 10]. Because BRAF mutations occur in 5-15% of CRC and are usually mutually exclusive with KRAS mutations , a substantial number of these CRC patients may benefit from BRAF mutation testing before initiating MoAb treatment . In thyroid, BRAF V600E mutation is restricted to malignant tumors, and is associated with papillary thyroid cancer (PTC), and PTC-derived poorly-differentiated and anaplastic carcinoma [7, 8]. Therefore, BRAF mutation testing can be helpful as an adjunct method to improve diagnostic accuracy in difficult cases, such as thyroid FNA samples with indeterminate and atypical cytology [12, 13].
There are several methods for BRAF V600E mutation testing, including dideoxy sequencing, colorimetric Mutector assay, allele-specific real-time PCR, pyrosequencing, high resolution melting (HRM) analysis and COLD-PCR [14–17]. These methods vary in their sensitivity, assay complexity and costs. Automated dideoxy sequencing is considered the “gold standard” method for mutation testing. However, it is limited by relatively complex procedure, overall high costs and low analytical sensitivity of detecting approximately 15-20% mutant allele in a wild-type background . ARMS-PCR is a highly sensitive, specific and low-cost mutation detection method with reported analytical sensitivity ranging from 0.1% to 2% for JAK2 V617F mutation detection [19, 20]. Here we report the development of a highly sensitive and specific ARMS-PCR assay to detect BRAF V600E mutation in formalin-fixed, paraffin-embedded (FFPE) tissue with sensitivity as low as 0.5%. We successfully used this assay to characterize the frequency of BRAF V600E in thyroid tumors.
Cell lines HT-29 (heterozygous for BRAF V600E) and K-562 (wt BRAF without V600E mutation) were cultured in RPMI1640 supplemented with 10% fetal calf serum. Anonymized FFPE thyroid tumor tissue blocks were obtained from the Department of Pathology for clinical protocol development.
3–5 sections of FFPE tissues (5 micron/section) were deparaffinized by xylene and ethanol treatment, and digested with proteinase K at 56°C overnight. Genomic DNA was extracted using ZR genomic DNA I kit (ZYMO Research Corp., Orange, CA) as recommended. DNA extraction from culture cells was otherwise the same as above except no deparaffinization was performed. DNA was then qualified/quantified with ND-1000 spectrophotometer (NanoDrop Products, Wilmington, DE).
ARMS-PCR and sequencing
The ARMS-PCR primer sequences were forward (Fo): 5′-CTCTTCATAATGCTTGCTCTGATAG-3′; reverse (Ro): 5′-GCCTCAATTCTTACCATCCAC-3′; forward wild-type identifying (Fiwt): 5′-GTGATTTTGGTCTAGCTACAGT-3′ and reverse mutation identifying (Rimut):5′-CCCACTCCATCGAGATTTCT-3′. PCR was performed in a 25 μl final volume containing 1 × Buffer, 2 mM MgCl2, 1 unit of Hotstar Taq DNA polymerase (Qiagen Science, Valencia, CA), 200 μM each dNTP, 400 nM primer Fo, 200 nM primer Ro and Fiwt, 800 nM primer Rimut and 30 ng genomic DNA template. PCR amplification was carried out by denaturation at 95°C for 5 min, followed by 40 cycles of 94°C for 20 sec, 68°C for 20 sec and 72°C for 20 sec with a final extension at 72°C for 5 min. PCR products were analyzed by 2% agarose gel electrophoresis. Automated dideoxy sequencing was performed by Macrogen USA (Rockville, MD) after PCR amplification using the Fo-Ro primer pair and the ampliocons purification using QIAquick PCR purification kit (Qiagen).
Assay sensitivity study
HT-29 cell DNA with heterozygous BRAF V600E mutation was serial diluted with DNA from a wt BRAF cell line K-562, to generate a sensitivity panel consisting of 0, 0.1%, 0.25%, 0.5%, 1%, 2.5%, 5%, 10%, 20% and 50% of BRAF V600E allele in a wild-type background. The PCR was performed under the optimized condition as above.
To assess the analytical sensitivity of the assay, BRAF V600E mutation containing DNA was spiked into wild-type BRAF DNA, the PCR was performed and showed that the assay we used could detect as low as, if not better than, 0.5% BRAF V600E allele in the background of wild-type DNA (Figure 1B). Therefore, the analytical sensitivity of the assay is 0.5%.
Assay validation and comparison with direct dideoxy sequencing using thyroid tumor specimens
BRAF V600E mutation rate in thyroid tumors
Positive rate (%)
Papillary Thyroid Carcinoma
Overall, our results indicate that ARMS-PCR is more sensitive than automated dideoxy sequencing in detecting low BRAF V600E allele burdens in FFPE tumor specimen.
For the development and implementation of personalized medicine, a sensitive, specific, low-cost and easy-to-implement assay to identify various genetic alterations such as BRAF V600E mutation is needed. By selecting primer pair and optimizing PCR cycling conditions, we achieved this goal by using ARMS based PCR. One of the advantages of ARMS-PCR is that the assay is designed to amplify a relative larger common fragment of DNA that flanks the mutation site in all samples regardless of their mutation status. This common fragment conveniently serves as an internal control for template DNA quality as well as potential PCR inhibition. The mutant or wild-type specific PCR amplifications take place in the same reaction tube, thereby allowing the mutant or wild-type specific PCR primers to compete for binding to very limited templates. To increase the sensitivity for detecting this mutation, we tried to change the ratio of four primers and found that the combination of 400 nM primer Fo, 200 nM primer Ro and Fiwt, 800 nM primer Rimut resulted the strongest mutant band. The primer binding to template varies based on their annealing conditions, but relative high Fo and the highest Rimut primer concentrations might favor the production of mutant product in our design. In addition, relative low concentration for Fiwt and Ro primers might reduce full length and wild type product formation by the competition effect of each primer. By such a design, the three products were produced at reasonably comparable levels. Otherwise, the wild type or whole length bands might overshadow the weak mutant band if the mutant BRAF V600E abundance is low, which is frequently the case for clinical specimens. With this study, we improved assay sensitivity to as low as 0.5% mutant allele level in the background of wild-type DNA, which is quite sensitive among ARMS-PCR methods . This sensitivity is very difficult to reach with direct dideoxy sequencing which has the sensitivity of 15-20%, and with pyrosequencing which has greatly increased sensitivity to about 2% [22, 23]. Considering DNA samples obtained from FFPE tissue is often fragmented due to damage by formalin fixation, we designed our PCR product size to be small (no more than 200 bp) to maximize the chance of successful amplification. This was proven to be helpful since all 72 FFPE samples had successful amplification.
The reported prevalence of BRAF V600E mutation for conventional PTC varied from 36% to 67% in the US and European studies [7, 8, 24–27], and was 83% in a Korean study . The difference in prevalence could be due to different study population or different methodologies used. The prevalence of BRAF V600E mutation in the tall-cell variant PTC is in general higher than that of the conventional type, with a mean prevalence of 77% in several studies . On the contrary, the follicular variant of PTC is rarely involved by BRAF V600E mutation, with a mean prevalence of 12% . Using ARMS-PCR, we tested 72 thyroid tumors and determined that the frequencies of BRAF V600E mutation in the conventional, tall-cell and follicular variants of PTC were 66%, 75% and 0%, respectively. Our frequency of BRAF V600E mutation in conventional PTC was at the high end of the reported results, possibly reflecting our improved assay sensitivity compared with automated DNA sequencing, which was the method used in many of the reported studies. Our study showed that DNA sequencing method missed 3 out of 30 PTC samples with BRAF V600E detected by ARMS-PCR, resulting in a false negative rate of 10%. In our validation study, the assay specificity is 100% (14/14) based on negative mutation in follicular adenoma. This is consistent with previous reports since benign follicular adenoma has not been found to harbor the BRAF mutation .
BRAF V600E mutation testing has demonstrated utility in helping select CRC patients who are considering monoclonal antibody therapy as wild-type BRAF is required for response to anti-EGFR antibodies  and improve diagnostic accuracy in thyroid FNA samples [12, 13]. In addition, BRAF V600E mutation is associated with sporadic microsatellite instable CRC, but not hereditary non-polyposis colorectal cancer (HNPCC) syndrome [29, 30]. Therefore, the presence of BRAF V600E mutation is an exclusion criteria for HNPCC genetic testing [29, 30]. BRAF V600E mutation testing can also help facilitate clinical studies of BRAF-targeted therapies . The improved understanding of the role BRAF mutations in cancer diagnosis, prognosis and treatment has increased the need for BRAF mutation testing [4, 32].
New methods are constantly being developed for BRAF V600E mutation detection. Shifted termination assay (STA) developed by TrimGen Corporation declared the sensitivity of 1-5% . Lang et al. developed an allele-specific, also known as ARMS, real-time PCR using Taqman probe to increase the sensitivity to 1% and using internal control to exclude false negative results . Morandi et al. developed an allele specific locked nucleic acid (LNA) quantitative PCR assay using LNA-modified allele specific primers and LNA-modified beacon probes to achieve sensitivity of 0.1% . Dual-priming oligonucleotide (DPO)-based multiplex PCR was commercially available from Seegene (Seoul, Korea) and has a sensitivity of 2% . Qiagen developed a BRAF mutation detection PCR kit with a sensitivity of 1.27%. Our method using ARMS-PCR therefore appears to have high sensitivity (0.5%) for BRAF V600E mutation detection.
We have developed a sensitive, specific and low-cost ARMS-PCR assay to detect BRAF V600E mutation. Comparative study showed that this assay is superior to automated dideoxy sequencing in terms of assay sensitivity, turn-around time and costs. The ARMS-PCR assay can be easily implemented by many molecular laboratories for BRAF V600E mutation testing. Our method provides one of the most sensitive methods for BRAF V600E gene mutation detection. The principal of our study design can be potentially adapted to detect other low abundance point mutations such as tumors with rich background stroma and post-treatment tumor samples.
We thank Dr. Minghao Zhong for critical reading of the manuscript.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.