Heterogeneity of PTEN and ERG expression in prostate cancer on core needle biopsies: implications for cancer risk stratification and biomarker sampling.
Author(s): Shah RB, Bentley J, Jeffery Z, DeMarzo AM
Publication: Hum Pathol, 2015, Vol. 46, Page 698-706
PubMed ID: 25724568 PubMed Review Paper? No
Purpose of Paper
This paper investigated the effects of specimen heterogeneity in prostate cancer specimens on assigned Gleason score, and ETS-related gene (ERG) and phosphatase and tensin homolog deleted on chromosome 10 (PTEN) status.
Conclusion of Paper
Gleason score, ERG overexpression and PTEN loss were consistent across all cores in 51%, 58% and 32% of cases. When the Gleason score was inconsistent among cores, the core with the highest tumor content had the highest Gleason score in 76% of cases. Further, among cases that displayed an inconsistent Gleason score, the core with the highest Gleason score reflected the ERG and PTEN status of the cases as a whole in 92% and 98% of cases, respectively. When the Gleason score was consistent across cores, the core with the highest tumor content identified ERG overexpression in 90% of cases (90%), but only identified PTEN loss in 76% of cases. Generally, ERG staining was consistent across sites in unilateral prostate cancer, but this was not true for PTEN staining or for bilateral prostate cancer.
Studies
-
Study Purpose
This study investigated the effects of specimen heterogeneity in prostate cancer specimens on assigned Gleason score and ERG and PTEN status. Six to fourteen needle core biopsies were obtained from 194 patients with prostate cancer, fixed in formalin and H&E stained. A Gleason score was assigned for individual biopsy cores as well as for the case (which was the highest score of the cores). PTEN immunostaining was considered intact when both the nucleus and cytoplasm were stained, partial loss if staining was only observed in a portion of the tumor, but a complete loss if the tumor failed to exhibit cytoplasmic staining. Specimens were considered as positive for ERG if the tumor displayed nuclear staining, and either diffuse if staining occurred throughout the entire tumor or partial if staining was limited to select regions. Intra-tumor variation was defined as variation within a core, while inter-tumor variation occurred between cores in a single case.
Summary of Findings:
Fifty-one percent (91 of 194) of cases had a uniform Gleason score across cores. When the score was not uniform across cores, the highest Gleason score belonged to the core with the highest tumor content in 76% (73 of 194) of cases. ERG overexpression was observed in 57% (111 of 194) cases, 95% of which were classified as diffuse (105 of 111), while 5% were partial (6 of 111). PTEN loss was observed in 36% (69 of 194) cases, of these 68% (47 of 69) were considered a partial loss and 32% were considered a complete loss (22 of 69). ERG overexpression was strongly associated with PTEN loss (p<0.0001) but not Gleason score; however PTEN loss was strongly associated with increasing Gleason score (p<0.0001). Inter-tumoral variation in ERG staining was observed in 42% (48 of 194) of cases and intra-tumoral variation occurred in 5% (6) of cases. Intra-tumoral variation for ERG staining was attributed to separation of tumor foci by normal tissue in 4 cases, and differences in expression between continuous tumor foci in the remaining 2 cases. For PTEN, inter- and -intra-tumoral variation occurred in 68% of cases each. For cases in which the Gleason score was inconsistent across cores, 51% (32 cases) had ERG overexpression and 24% (11 cases) had PTEN loss across all cores. For cases with a consistent Gleason score across all cores, the core with the most tumor content was representative of the case for ERG overexpression and PTEN loss 90% (43 cases) and 76% (19 cases) of the time, respectively. When the Gleason score was inconsistent across cores, the core with the highest tumor content and Gleason score was representative of ERG overexpression or PTEN loss for the case in 92% (58 of 63 cases) and 98% (45 of 46 cases) of cases, respectively. For patients diagnosed with unilateral prostate cancer, ERG staining was concordant across sites in 85% (17 of 20 cases) of cases, but PTEN staining was only concordant across 38% (8 of 20 cases) of cases. For patients diagnoses with bilateral prostate cancer, ERG and PTEN staining concordance across sites only occurred in 31% (15of 46 cases) of cases each.
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Protein Immunohistochemistry Morphology H-and-E microscopy Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Preaquisition Prognostic factor Unilateral prostate cancer
Bilateral prostate cancer
Immunohistochemistry Specific Targeted peptide/protein ERG
PTEN
Immunohistochemistry Specific Technology platform H&E
Biospecimen Aliquots and Components Biospecimen heterogeneity Intratumoral sampling (exact positions not specified)