SOX-2 (SP76)

It has been reported that anti-SOX-2 antibody recognizes lung squamous cell carcinoma (LSCC). A recent study1 has demonstrated that extensive anti-SOX-2 staining was seen in over 90% of LSCC and largely paralleled p63 expression. Extensive anti-SOX-2 staining was seen in 21% of lung adenocarcinomas (LACA), including cases that were anti-p63-negative or only anti-p63 focally-positive. However, another recent study showed only 4.5% of LACA is positive for anti-SOX-2 expression.2 In a study by Sholl et al, 29% of LACA cases exhibited at least focal p63 expression.1 Combined p63 and SOX-2 expression was seen in 94% of LSCC and 12% of LACA with a statistically significant difference (P<0.0001) versus p63 alone. This study also showed that anti-CK5/6 had a good sensitivity but poor specificity for LSCC. Combined anti-CK5/6 and anti-p63 positivity was seen in 93% of LSCC and 24% of LACA. Anti-CK5/6+/anti-p63+/anti- SOX-2+ was detected in 93% of LSCC and only 9% of LACA. These results indicate that the sensitivity of anti-p63 is equally high but its specificity is similarly variable; it was seen at least focally in close to 30% of LACA. When used together, anti-p63+/anti-SOX-2+ applied to the same tumor cell population is >90% specific for LSCC. Anti-SOX-2 produced moderate-to-intense staining in all 50 cases of embryonal carcinoma components. The only other component that showed reactivity was the primitive neuroectodermal component in 11 of 14 (79%) of immature teratomas. In each of these positive staining foci, the staining varied from moderate-to-strong. Yolk sac tumor, seminoma, mature teratoma, choriocarcinoma, and intratubular germ cell neoplasia, unclassified (IGCNU) were uniformly negative, as were all the non-neoplastic parenchymal and stromal structures.3