[959] PTEN Protein Loss Is More Frequent in ERG-Positive Prostate Tumors

Tamara L Lotan, Alcides Chaux, Sarah Peskoe, Jessica Hicks, Elizabeth A Platz, George J Netto, Angelo M De Marzo. Johns Hopkins Medical Institutions, Baltimore, MD; Johns Hopkins School of Public Health, Baltimore, MD

Background: Recent data from our group strongly suggests that ERG gene rearrangement occurs prior to PTEN loss in prostate cancer progression. Interestingly, PTEN gene deletion has been shown to occur more commonly in prostate tumors harboring the ERG gene rearrangement. In at least eight prior studies that have examined this association at the genetic level, typically by FISH, PTEN deletions occur 1.4- to 5.2-fold more frequently in ERG-rearranged tumors (average = 2.9-fold increase). We recently validated immunohistochemical (IHC) staining protocols for PTEN and ERG protein in prostate cancer and showed that each sensitively detects cases with genetic lesions by FISH. Here, we tested whether the strong association between PTEN gene deletion and ERG gene rearrangements can also be seen when PTEN and ERG status are ascertained at the protein level.
Design: Validated PTEN and ERG IHC assays were applied to a total of 1083 evaluable prostate tumors sampled in triplicate or quadruplicate on tissue microarrays (TMA) representing two previously published surgical cohorts. The first cohort (n=862) was a nested case-control study for tumor recurrence (biochemical or clinical), while the second cohort (n=221) was a consecutive series of patients from a single surgeon, all of whom experienced biochemical recurrence. As previously validated, a case was considered to have PTEN protein loss if any of the sampled tumor cores showed markedly decreased cytoplasmic PTEN protein when compared to surrounding benign glands and stroma. Similarly, a case was considered to express ERG protein if any sampled tumor cores showed nuclear ERG staining.
Results: In all, 49% of cases (534/1083) were ERG-positive and 52% of cases (564/1083) showed PTEN protein loss. Of ERG-negative cases, 44% (243/549) showed PTEN protein loss, compared to 60% (321/534) of ERG-positive cases (p<0.0001 by 2-sided Fisher's exact test). Thus, PTEN protein loss was 1.36-fold more frequent in ERG-positive cases by IHC.
Conclusions: PTEN protein loss is more frequently seen in ERG-protein positive prostate cancer cases, confirming prior studies that have examined this association at the genetic level. While some have proposed that PTEN loss and ERG rearrangement exhibit some cooperativity in prostatic oncogenesis in animal models, our data that ERG gene rearrangement occurs prior to PTEN loss could also be consistent with a causal role for ERG rearrangement in PTEN gene deletions, perhaps due to ERG-induced DNA damage.
Category: Genitourinary (including renal tumors)

Tuesday, March 5, 2013 9:30 AM

Poster Session III # 64, Tuesday Morning


Close Window