Immunohistochemical Staining for p-ERK: A Potential Screening Tool for BRAF Mutations in Melanoma
J S Ko, S D Billings, L Durkin, W L Wang, A J Lazar, E D Hsi. Cleveland Clinic, Cleveland, OH; MD Anderson Cancer Center, Houston, TX
Background: BRAF V600E mutations are frequent in malignant melanoma (MM), resulting in constitutive ERK signaling and tumor cell proliferation. Documentation of BRAF exon 15 mutation in MM is required for treatment with the BRAF-mutant kinase inhibitor, vemurafenib. We hypothesized that phosphorylated ERK (pERK) immunohistochemical staining might distinguish BRAF wild type (WT) and mutated tumors and serve as a screening tool.
Design: pERK1/2THR202/Tyr214 immunostains performed on 27 MM with known BRAF mutational status were scored as percentage of positive cells (0-100) and intensity of cytoplasmic and nuclear staining (1-3 for each). The summed products of percentage positive and cytoplasmic and nuclear staining intensity (total stain score) was tabulated. Mutations in BRAF exon 15 and NRAS exons 1 & 2 were determined by PCR and DNA sequencing from formalin-fixed paraffin embedded, microdissected tissue. Two tailed Mann-Whitney U test was employed.
Results: Of 27 MMs, 7 harbored a BRAF mutation: 6 (V600E) and 1 (K601N). Of 20 BRAF WT tumors, 5 were NRAS mutants (4xQ61K; 1xQ61R). 15/15 BRAF WT MM had a total stain score of ≤100 (mean+/-SEM=35+/-9.6, median= 40, range=0-100) and each showed pERK positivity in ≤ 25% of tumor cells. For BRAF mutated cases, 6/7 had a total stain score of ≥150 (mean+/-SEM=220+/-37, median=200, range=75-375). All had at least 25% of tumor cells staining, with 5/7 having ≥50% positivity (mean 52%). NRAS mutated cases were more heterogeneous but tended to have increased pERK over WT (mean+/-SEM=134+/-31, median=100, range=45-225). The pERK total stain scores between the BRAF mutated and WT groups was statistically significant (p<0.001). Using a cutoff of 25% cell positivity, the positive and negative predictive value of pERK immunostaining for BRAF mutation was 50% and 100%. NRAS mutated cases, which also activate the ERK pathway, lowered the positive predictive value.
Conclusions: In this cohort, use of the 25% cutoff would have identified all BRAF mutated cases, and reduced testing of ultimately negative cases from 20 tests to 7 tests (65% reduction). pERK immunostains may be useful as a screening tool for BRAF mutations and confirm this known signaling cascade in fixed specimens. Given the known time-dependent lability of phosphoproteins, we would recommend application only in rapidly fixed, small biopsy or cytologic samples.
Wednesday, March 21, 2012 1:00 PM
Poster Session VI # 106, Wednesday Afternoon