evaluated PTEN IHC using 5 potential PTEN antibodies on standardized cell lines. improved and standardized methodologies of PTEN assessment are required to clarify the role of PTEN as a biomarker in colorectal cancer. copy number variation, immunohistochemistry, no tumour identified, PF 477736 phosphatase and tensin homologue. Taqman? copy number PCR Using a PTEN Taqman? copy number assay, 25/51 specimens (49%) had 1.5 copy number and were thus classified as PTEN loss. Concordance between IHC and Taqman?PCR The 37 specimens with concordant IHC assessment were included in the IHC versus TaqmanPCR concordance analysis. Fifteen specimens had PTEN loss on IHC of which 10 (67%) also had PTEN allelic loss on FLT3 Taqman? PCR. Seventeen specimens had PTEN allelic loss on Taqman? PCR of which 10 (58%) had PTEN loss on IHC. Fifteen specimens had preserved PTEN on both IHC and Taqman? PCR analysis. Overall concordance between IHC and Taqman? copy number in PTEN loss assessment was 25/37 (68%) (Table ?(Table22). Table 2 Concordance PF 477736 of PTEN loss between IHC and Taqman copy number Immunohistochemistry, phosphatase and tensin homologue. Shaded squares?=?discordant results. Discussion In this validation study of PTEN assessment in CRC we evaluated inter-observer variability in PTEN assessment with IHC and subsequently the discordance of PTEN assessment between IHC and PCR based methodologies. IHC assessment yielded rates of PTEN loss of 33% and 57% between two pathologists, while Taqman? PCR demonstrated 49% of specimens contained PTEN allelic loss. Our analysis provides particular insight into the relationship between PTEN protein expression and allelic loss. Specifically how is protein expression maintained in the setting of allelic loss, and why do samples show absence of PTEN expression despite allelic loss? In samples with PTEN allelic loss 41% maintained protein expression. Of these specimens all PF 477736 had IHC staining intensity of 1+ suggesting possibly a reduced level of PTEN protein. The maintenance of protein expression in these cases is likely due to the remaining functional PTEN allele, which allows transcription of a normal PTEN protein. In cases of PTEN haploinsufficiency (monoallelic loss) whether protein expression is reduced and whether such reduction confers a growth advantage is unknown. Sood et al. also demonstrated monoallelic PTEN dysfunction (by mutation or promoter methylation) resulted in loss of protein expression in only 38% of samples, while biallelic inactivation resulted in loss of PTEN expression in PF 477736 80% of cases [16]. Ali et al. reported a higher PTEN expression loss of 71% in samples with a single PTEN gene mutation, though allelic loss and methylation were not assessed [19]. In our cohort 25% of cases without PTEN allelic loss demonstrated complete absence of PTEN expression on IHC. These findings confirm alternative genetic mechanisms, beyond allelic loss, are responsible for loss of PTEN protein expression. Several authors have undertaken more comprehensive analysis of PTEN status on CRC specimens and provide an important insight into the often coexisting genetic mechanisms of PTEN dysfunction. Goal et al. demonstrated hypermethylation of the PTEN promoter region occurred in 10/132 (7.6%) sporadic CRC specimens, with a higher rate (19.1%) in microsatellite unstable CRCs. PTEN mutations coexisted in 4/10 (40%) of hypermethylated PTEN specimens. Eighty percent of patients with promoter hypermethylation had reduced (+1) or loss of PTEN protein expression and in the 3 cases of complete loss of PTEN staining, promoter hypermethylation coexisted with PTEN mutation or allelic loss [13]. Nassif et al. assessed allelic loss and PTEN mutation in 41 primary CRC specimens, finding 15 (37%) contained one or both aberrations. Nine of these cases contained biallelic.