Immunohistochemistry with Anti-BRAF V600E (VE1) Mouse Monoclonal Antibody is a Sensitive Method for Detection of the BRAF V600E Mutation in Colon Cancer: Evaluation of 120 Cases with and without KRAS Mutation and Literature Review

The major aim of this study was to evaluate the performance of anti-BRAF V600E (VE1) antibody in colorectal tumors with and without KRAS mutation. KRAS and BRAF are two major oncogenic drivers of colorectal cancer (CRC) that have been frequently described as mutually exclusive, thus the BRAF V600E mutation is not expected to be present in the cases with KRAS mutation. In addition, a review of 25 studies comparing immunohistochemistry (IHC) using the anti-BRAF V600E (VE1) antibody with BRAF V600E molecular testing in 4041 patient samples was included. One-hundred and twenty cases with/without KRAS or BRAF mutations were acquired. The tissue were immunostained with anti-BRAF V600E (VE1) antibody with OptiView DAB IHC detection kit. The KRAS mutated cases with equivocal immunostaining were further evaluated by Sanger sequencing for BRAF V600E mutation. Thirty cases with BRAF V600E mutation showed unequivocal, diffuse, uniform, positive cytoplasmic staining and 30 cases with wild-type KRAS and BRAF showed negative staining with anti-BRAF V600E (VE1) antibody. Out of 60 cases with KRAS mutation, 56 cases (93.3%) were negative for BRAF V600E mutation by IHC. Four cases showed weak, equivocal, heterogeneous, cytoplasmic staining along with nuclear staining in 25–90% of tumor cells. These cases were confirmed to be negative for BRAF V600E mutation by Sanger sequencing. Overall, IHC with anti-BRAF V600E (VE1) antibody using recommended protocol with OptiView detection is optimal for detection of BRAF V600E mutation in CRC. Our data are consistent with previous reports indicating that KRAS and BRAF V600E mutation are mutually exclusive.


Introduction
Colorectal cancer is the third most common cancer and the fourth most prevalent cause of death in the world [46]. Approximately 35-45% of patients with colorectal tumors have mutation in KRAS gene, while BRAF V600E mutation is found in about 5-15% of colorectal adenocarcinomas [8,9,26,31]. Both these mutations are considered to be oncogenic driver mutations, since they are both responsible for the initiation and maintenance of the tumor [10]. Importantly, many studies have indicated that BRAF V600E mutation occurs only in tumors that do not carry mutations in KRAS gene and it is widely accepted that these two mutations are mutually exclusive [10,23,25,30].
The BRAF gene encodes a cytoplasmic serine-threonine kinase that is frequently mutated in various cancers, including melanoma, papillary thyroid carcinoma, and colorectal carcinoma, among others. The oncogenic mutations in BRAF gene result in constitutive activation of the MAPK signaling pathway, leading to increased cell proliferation, resistance to apoptosis and tumor progression. The most common of these mutations, the V600E mutation, occurs in exon 15 and results in a substitution from valine to glutamic acid at position 600 within the BRAF kinase domain.
BRAF V600E mutation occurs in about 5% of microsatellite stable (MSS) CRC tumors. These tumors are associated with a distinct molecular and clinical phenotype with a poor prognosis [40]. The presence of BRAF V600E mutation in CRC is associated with poor survival [44]. BRAF V600E mutation is also detected in sporadic CRC tumors with microsatellite instability (MSI) [27]. Particularly, BRAF V600E mutation is observed in about two thirds of MSI tumors with the loss of MLH1 expression due to MLH1 promoter methylation [18]. In contrast, BRAF V600E mutation is very rare in CRC patients with Lynch syndrome [27]. In clinical practice it is much easier to detect BRAF V600E mutation than methylation status of MLH1 promoter [26]. Therefore, it was suggested that assessment of BRAF V600E mutation can be used to triage patients for mismatch repair (MMR) genetic testing to differentiate MLH1-deficient sporadic CRC from Lynch syndrome caused by germ-line MLH1 mutations [7,14,17,26,41,43]. Currently, the American National Comprehensive Cancer Network (NCCN) guidelines recommend that BRAF V600E mutational status should be evaluated in all colorectal carcinomas to identify 1) the patients with Lynch syndrome in MMR deficient group and 2) to identify the MMR proficient/BRAF V600E group with poor prognosis [15,38,43].
The most common approach for the detection of BRAF mutation is sequencing of tumor DNA, for example Sanger sequencing, pyrosequencing and high resolution melting. All of these methods are able to detect a mutant allele in a background of 5-20 fold excess of wild-type alleles. In contrast, immunohistochemistry allows direct visualization of the mutant protein in the tumor cells in tissue context. The anti-BRAF V600E (VE1) antibody is currently used to evaluate the BRAF V600E mutation status in various cancers including CRC [32]. This antibody is a mutation-specific mouse monoclonal antibody that was raised against a synthetic peptide representing the BRAF V600E mutated amino acid sequence from amino acids 596 to 606 (GLATEKSRWSG) [5,6].
The primary goal of this study was to compare the performance of the anti-BRAF V600E (VE1) antibody to detect BRAF V600E mutation by IHC in colon cancer cases with/ without KRAS mutation. Since concomitant KRAS and BRAF tumor mutations are considered mutually exclusive we wanted to confirm that the CRC cases carrying KRAS mutation show negative BRAF V600E staining by IHC with anti-BRAF V600E (VE1) antibody. In addition, we performed a review of 25 studies that compared IHC using anti-BRAF V600E (VE1) antibody with molecular testing for BRAF V600E mutation.

Tumor Specimens
A total of 120 formalin-fixed paraffin embedded (FFPE) tissues from patients with colorectal cancer were ordered from Avaden Biosciences and GLAS/Consultants in Human Biologics. The requested cases included 60 CRC cases with confirmed KRAS mutation, 30 CRC cases with confirmed BRAF V600E mutation and 30 CRC cases confirmed to be wild-type BRAF and wild-type KRAS. The presence/absence of these mutations was confirmed by molecular testing by the vendor. ]. This antibody is not directed against any known epitope present in human tissue. In addition, slides containing 2 cases positive for BRAF V600E mutation (CRC, thyroid papillary carcinoma) and one case negative for BRAF V600E mutation (CRC) were used as run control slides. The absence/presence of the BRAF V600E mutation in the tissues was confirmed by Sanger sequencing. These slides were included with each individual run to assess the expected quality of the antibody and all components of the assay. The overall run was accepted if: 1) the BRAF V600E positive tissue control stained with anti-BRAF V600E (VE1) antibody showed specific cytoplasmic staining pattern and had acceptable background; 2) the positive tissue control stained with Negative Control Monoclonal shows no specific staining and had acceptable background; 3) the BRAF V600E negative tissue control stained with anti-BRAF V600E (VE1) antibody showed no specific staining pattern and acceptable background; and 4) the BRAF V600E negative tissue control stained with Negative Control (Monoclonal) shows no specific staining and has acceptable background.

BRAF V600E Immunohistochemistry
The stain intensity of anti-BRAF V600E (VE1) antibody in tumor cells was recorded on a 0-3 scale. Strong cytoplasmic staining was scored as 3, medium cytoplasmic staining as 2, weak cytoplasmic staining as 1 and the absence of staining was scored as 0. In addition, any nuclear staining and the percentage of tumor cells stained positive with anti-BRAF V600E (VE1) antibody was recorded. The criteria for positive BRAF V600E staining included unequivocal, diffuse, uniform, cytoplasmic staining at intensity ≥1 in majority of malignant cells. The cases were scored as negative for BRAF V600E mutation if they showed no staining or weak, cytoplasmic, non-granular, uniform staining (stain intensity <1). The cases with staining of isolated tumor cells in a tumor that otherwise showed no staining were also scored as negative. The cases were scored as equivocal if they displayed ambiguous, heterogeneous, non-uniform cytoplasmic staining in tumor cells with or without nuclear staining. The equivocal cases were sequenced by Sanger sequencing to confirm the presence/absence of BRAF V600E mutation.

BRAF V600E Sanger Sequencing
Genomic DNA was extracted from 20 μm thick sections from FFPE samples using the QIAamp FFPE Tissue Kit (Qiagen, Redwood, CA) according to manufacturer's instructions. The primers for Sanger sequencing were designed to amplify region of the exon 15 of the BRAF gene coding sequences at mutation site and a few nucleotides in the intron on both ends. Two primers were used including 1) BRAF-ex15F-TGCTTGCTCTGATAGGAAAATG and 2) BRAF-ex15R-AGCATCTCAGGGCCAAAAAT. Both forward and reverse strands were sequenced on an Applied Biosystem's 3730xl DNA Analyzer and analyzed using DNASTAR Lasergene 12 software (DNASTAR, Madison, WI).

Anti-BRAF V600E (VE1) Immunohistochemistry
All 120 cases were examined for presence of BRAF V600E mutation by IHC using the anti-BRAF V600E (VE1) antibody on the automated VENTANA BenchMark ULTRA platform.
All 30 cases with BRAF V600E mutation exhibited uniform, unequivocal, diffuse, cytoplasmic staining in majority of tumor cells with stain intensities of 1-2.75 and background ≤0.25. Out of 30 cases, 28 cases showed positive BRAF V600E signal in 100% of tumor cells and 2 cases showed positive staining in 90% and 85% of tumor cells, respectively. These data are consistent with previous reports indicating that majority of tumor cells express mutated BRAF V600E protein, since this mutation is driving tumor proliferation. All 30 cases with no BRAF V600E and no KRAS mutations showed intensities of ≤0.5 and background ≤0. 25. In 28 cases the stain intensities were 0-0.25, in remaining two cases 10% and 85% of malignant cells stained at the stain intensities 0.5. The sensitivity and specificity was 100% for cases with confirmed BRAF V600E mutational status. Representative images are shown in Fig. 1.
Out of 60 CRC cases with KRAS mutations 56 cases were scored as negative for BRAF V600E mutation (stain intensity <1). There were four cases where the stain intensities were scored as 1. However, these four cases exhibited ambiguous, heterogeneous, non-uniform cytoplasmic staining along with nuclear staining and thus they were scored as equivocal. In the first case, only 40-50% of tumor cells were positively stained with anti-BRAF V600E (VE1) antibody, the cells with positive staining showed the signal in cytoplasm and also strong signal in nuclei, the cytoplasmic staining was non-diffuse and non-uniform. Representative images are shown in Fig. 2A-C. In the second case only small portion of tumor showed uneven, cytoplasmic staining (25% tumor cells). In addition, tumor cells also exhibited nuclear staining. Representative images are shown in Fig. 2D-F. Third case showed staining in 70% of cells, however the staining was heterogeneous and clearly nuclear along with lighter non-uniform cytoplasmic staining. Representative images are shown in Fig. 2 G-I. Fourth case showed high degree of nuclear staining, overall 90% of tumor cells showed positive staining in cytoplasm, however the strong staining was observed in nuclei with some signal in cytoplasm. This cytoplasmic staining was scattered and uneven. Representative images are shown in Fig. 2J-L. Since these four case exhibited ambiguous, non-uniform, heterogeneous and nuclear staining pattern, they were assigned as equivocal for BRAF V600E mutation.
In addition, there were 4 cases that were scored with stain intensity of 0.75 in all three evaluated slides in 30-90% of tumor cells. All these cases exhibited nuclear staining and non-diffuse, weak, heterogeneous cytoplasmic staining. Since the stain intensity was <1 these cases were scored as negative for BRAF V600E mutation.

DNA Sequencing
Overall, out of the all 120 cases, there were 60 cases with KRAS mutation. Out of these 60 cases, 4 cases showed BRAF V600E stain intensity 1 in 25-90% of tumor cells. This was not expected since these KRAS and BRAF V600E are mutually exclusive mutations. These cases exhibited the staining pattern that was not consistent with the recommendations based from studies shown in Table 1 that include uniform, diffuse, cytoplasmic staining in majority of malignant cells. In addition, 4 KRAS mutated cases showed stain intensities 0.75 with similar nuclear/heterogeneous staining pattern. These cases were also sequenced for BRAF V600E mutation. All these cases were negative for BRAF V600E mutation by Sanger sequencing.

Disscussion
This study evaluated 120 CRC cases with and without KRAS mutation to access the performance of IHC using anti-BRAF V600E (VE1) antibody for detection of BRAF V600E mutation. Overall, the results of these experiments demonstrates that IHC using the anti-BRAF V600E (VE1) antibody with the VENTANA OptiView DAB detection system and BenchMark ULTRA platform is a highly specific and sensitive method for the detection of BRAF V600E in colon cancer.
There is strong evidence from multiple studies that the IHC using anti-BRAF V600E (VE1) antibody is highly concordant with molecular tests for the BRAF V600E mutation. Table 2 shows a summary of 25 studies that evaluated sensitivity and specificity of IHC with anti-BRAF V600E (VE1) antibody in comparison with molecular tests using different methods (Sanger, pyroseqencing, SNapShot PCR, NGS, etc.). Altogether, 4041 patient samples were evaluated in these studies, the overall sensitivity and specificity of IHC assay using anti-BRAF V600E (VE1) antibody compared to molecular tests was 93% (934/1008) and 96% (2922/3033), respectively.
Out of these 25 studies, 4 publications reported lower sensitivity and/or specificity of anti-BRAF V600E antibody compared to sequencing [1, 12,20,21]. However, there were several problems with these studies. First, the study by Adackapara et al. analyzed 52 colorectal carcinomas with known BRAF mutation status determined by pyrosequencing and found that IHC had a low sensitivity (71%) and specificity (74%) for detecting BRAF V600E mutation compared to pyrosequencing (Table 2). They concluded that IHC using anti- BRAF V600E (VE1) antibody is not a useful surrogate for detecting BRAF mutation in colorectal carcinoma. However, in their experiment, manual staining with citrate buffer as antigen retrieval was employed. In our experience and the experience of others the use of acid for antigen retrieval step results in suboptimal staining that is difficult to interpret [19]. TRIS or EDTA buffers at pH = 8 proved to be retrieval agents that produced the most robust and homogenous cytoplasmic staining with anti-BRAF V600E (VE1) antibody. Similarly, Lasota et al. used in their studies Bond Epitope Retrieval Solution 1 (pH = 6) which is not an optimal solution for antigen retrieval for this assay [20].
Another important factor that may contribute to the different outcome of the studies is the interpretation of the IHC results. As multiple studies have highlighted, a proper scoring system is necessary to reduce false-positive and false-negative cases. Since BRAF V600E mutation is a driver mutation, a majority of tumor cells should express this mutated protein.
The scoring criteria shown in Table 1 were used in the individual studies presented in Table 2 that compare IHC using     (Table 2). All studies that used these interpretation criteria (and appropriate protocol using antigen retrieval at alkaline pH) reached close to 100% sensitivity and specificity compared to sequencing. Additional 7 studies did not include scoring criteria in the method section, however they reported that homogeneous, diffuse staining pattern was observed in cases with confirmed BRAF V600E mutation. Three studies provided no details. In one study the cases were scored as positive for BRAF V600E staining when only ≥20% tumor cells showed positive signal in one study [12]. The sensitivity and specificity reported in this study was only 89% and 57% (when IHC assay on BenchMark ULTRA platform was used) and 75% and 93% (when IHC assay on Leica Bond was used). Several studies suggested that additional analysis is required for minority of equivocal cases with ambiguous, focal, heterogeneous staining (Table 2). False positive staining was noted in signet ring tumor cells [33]. Importantly, the nuclear staining was described as the most common artifact (Table 2) [22]. For example, Bledsoe et al. noted that BRAF-mutant cases showed homogeneous, finely granular, cytoplasmic staining with varying intensities, however, non-specific nuclear and heterogeneous non-diffuse cytoplasmic staining of variable intensity was observed in a minority of non-BRAF mutant cases [3]. Therefore, it was recommended by Marin et al. that Bthe interpretation should be made with caution in the presence of nuclear staining^ [22].
Our study also suggests that the cases showing the presence of heterogeneous cytoplasmic staining with or without nuclear staining should be carefully interpreted.
Overall, the evidence from the publications presented in Table 1 and from the current study suggests that the cases should be scored as positive for BRAF V600E mutation if they display unequivocal, diffuse, uniform, granular, cytoplasmic staining in the majority of tumor cells at stain intensity ≥1. They should be scored as negative for BRAF V600E mutation if they exhibit no staining or weak, cytoplasmic, non-granular, non-uniform staining (stain intensity <1). The cases with staining of isolated tumor cells in a tumor that otherwise showed no staining should be considered negative. The cases should be considered as equivocal if they display ambiguous, heterogeneous, cytoplasmic staining with or without nuclear staining in tumor cells. If these interpretation criteria are followed the IHC with anti-BRAF V600E (VE1) antibody using recommended protocol with OptiView detection is optimal for detection of BRAF V600E mutation in CRC. In our study all 30 cases with BRAF V600E mutation showed unequivocal positive cytoplasmic staining in 85-100% tumor cells; all 30 cases with wild-type KRAS and BRAF were negative; 6.7% (4/60) cases with KRAS mutation showed heterogeneous, cytoplasmic/nuclear staining at stain intensity 1. However, the staining was heterogeneous and the presence of distinct nuclear staining was noted in these four cases along with cytoplasmic staining. Therefore, these cases were assigned as equivocal for BRAF V600E mutation. These cases were sequenced and confirmed to be negative for BRAF V600E mutation.
In summary, this study indicates that IHC with the anti-BRAF V600E (VE1) antibody performed on the Benchmark ULTRA automated stainer is a highly sensitive and specific detection method for determination of BRAF V600E mutation status in CRC. The results presented in this study are consistent with previous reports indicating that KRAS and BRAF V600E mutation are mutually exclusive. Based on our findings and consistent with other literature reports, the majority of BRAF V600E positive cases demonstrate a uniform or nearly uniform, diffuse staining pattern present in the majority of tumor cells. We propose that in the minority of cases with an equivocal staining pattern, additional molecular testing should be done to assess BRAF mutational status.