High interobserver variability of PTEN immunohistochemistry defining PTEN status in low- to intermediate-risk prostate cancer: results of the first German ring trial

Abstract

The prognostication of individual disease trajectory and selection of optimal therapy in patients with localized, low-grade prostate cancer often presents significant difficulty. The phosphatase and tensin homolog on chromosome 10 (PTEN) has emerged as a potential novel biomarker in this clinical context, based on its demonstrated prognostic significance in multiple retrospective studies. Incorporation into standard clinical practice necessitates exceptional diagnostic accuracy, and PTEN's binary readout-retention or loss-suggests its suitability as a biomarker. This multi-institutional ring trial aimed to validate the diagnostic precision of PTEN immunohistochemistry in localized, low- to intermediate-risk prostate cancer, across ten university pathology institutes in Germany. The trial incorporated 90 cases of patients diagnosed with acinar adenocarcinoma of the prostate of grade groups 1 (n = 8, 8.9%) and 2 (n = 82, 91.1%) post-radical prostatectomy. Remarkably, the interpretation of PTEN immunohistochemistry displayed substantial variation (12.5-51.2% PTEN loss rates) within an identical cohort of prostate cancer. Fluorescence in situ hybridization analysis demonstrated PTEN hemizygous deletions in 5.5% (5/90) of cases. All cases with hemizygous deletions presented a distinct loss of PTEN expression by immunohistochemistry and were unanimously identified as PTEN loss by all participants (sensitivity 100%). However, negative (loss) immunohistochemistry was relatively non-specific for an underlying genomic deletion. Improved inter-observer agreement was observed in a subsequent ring trial. Finally, we identify S473-pAKT immunohistochemistry as a useful marker in equivocal cases. In summary, this multi-institutional ring trial illustrates surprisingly heterogeneous outcomes in defining PTEN status by immunohistochemistry.

Keywords: AKT; Active surveillance; PTEN; Prognostic biomarker; Prostate cancer.

Fig. 1

Results of the first ring trial. a Graphical and b tabular overview of PTEN immunohistochemistry results among the institutes in the first ring trial

Fig. 2

Representative PTEN immunohistochemistry. a Homogenous loss of PTEN expression in tumor glands with intact staining in peritumoral stroma. These cases was rated ‘PTEN loss’ by all institutes. b Homogenous and strong PTEN expression in tumor glands and stroma. This case was rated ‘retained PTEN expression’ by all institutes. c Equivocal case with weak PTEN expression. This case was rated ‘PTEN loss’ by 5 institutes and ‘retained PTEN expression’ by 5 institutes

Fig. 3

PTEN FISH analysis. The LSI PTEN probe hybridizes to the 10q23 region on chromosome 10 (red signal). The CEP 10 SpectrumGreen probe hybridizes to chromosome 10 (green signal). a Tumor sample with retained PTEN expression and two intact PTEN alleles. b Tumor sample with PTEN loss and hemizygous PTEN deletion (red)

Fig. 4

Results of the second ring trial in comparison to the first trial. Results of a all cases, b equivocal cases only and c and unequivocal cases only

Fig. 5

Expression of PTEN, S473-pAKT, and CD24 in prostate needle core biopses. a–c Prostate cancer with retained PTEN expression, absent S473-pAKT, and no expression of CD24. d–f Prostate cancer with loss of PTEN expression, positivity for S473-pAKT, and expression of CD24. g–i Prostate cancer with retained PTEN expression, weak equivocal staining for S473-pAKT and diffuse overexpression of CD24