. Author manuscript; available in PMC: 2025 Jan 7.

Published in final edited form as: J Oral Pathol Med. 2024 Jan 7;53(1):79–87. doi: 10.1111/jop.13506

Table 3. Comparison of BRAFV600E detection methods.

The table summarizes reliability of the three BRAFV600E testing methods. Differences between the three BRAFV600E detection methods was statistically significant [χ²(2) = 31.34, p < 0.0001]. Direct comparison of IHC with ARMS-PCR (p < 0.0001) or IHC with DSS (p < 0.0001) was statistically significant.

Comparison of BRAFV600E Detection Methods (n = 30 samples)
	BRAFV600^−ve	BRAFV600E^+ve	% BRAFV600E^+ve ****
Immunohistochemistry (IHC)	2	28	93.33
ARMS-PCR	17	13	43.33
Direct Sanger sequencing	23	7	23.33
**** χ² test (2, n=90 BRAFV600E analyses) = 31.34, p < 0.0001
	IHC **** (Reference: Direct Sanger sequencing)	IHC**** (Reference: ARMS-PCR)	ARMS-PCR (Reference: Direct Sanger sequencing)
Sensitivity (95% CI)	0.8 (0.64 – 0.90)	0.7 (0.53 – 0.80)	0.7 (0.43 – 0.82)
Specificity (95% CI)	0.9 (0.75 – 0.99)	0.9 (0.67 – 0.98)	0.6 (0.42 – 0.72)
Positive Predictive Value (95% CI)	0.9 (0.79 – 0.99)	0.9 (0.79 – 0.99)	0.4 (0.27 – 0.61)
Negative Predictive Value (95% CI)	0.8 (0.59 – 0.88)	0.6 (0.39 – 0.73)	0.8 (0.59 – 0.88)
**** Fisher’s exact test (two-sided)	p <0.0001	p <0.0001	ns