[1657] Development of a Scoring System for PTEN Immunohistochemistry in Breast Cancer.

Rania Bakkar, Diana Urbauer, Russell Broaddus. MD Anderson Cancer Center, Houston, TX

Background: Loss of expression of PTEN, which inhibits the PI3K pathway, occurs in many types of cancer, including breast cancer. Patients with PTEN-deficient cancers may potentially benefit from PI3K pathway inhibitors, which are rapidly being developed and tested clinically. In breast cancer, loss of PTEN expression is associated with resistance to anti-hormonal therapy and Herceptin. Therefore, accurate identification of the subsets of breast cancer patients with PTEN loss is clinically important. Currently, there is no standardized protocol for pathological reporting of PTEN immunohistochemical results in breast cancer. Therefore, we wanted to design a practical scoring system with reliable clinical implications on patient outcome.
Design: Forty formalin fixed paraffin embedded breast carcinomas were immunohistochemically stained for PTEN using the Dako 6H2.1 antibody. Immunohistochemistry for anti-phosphorylated (serine 235/236) S6 ribosomal protein (pS6), a downstream member of the PI3K pathway, was scored as % positive. Cochran-Armitage trend and Fischer's exact tests were used for statistical analyses. The breast cancer results were compared to those from our previous study of 154 endometrial carcinomas.
Results: A 3-tiered scoring system (negative, reduced, positive) was devised. In all cases, the stroma stained strongly positive for PTEN. Positive was defined as tumor PTEN expression comparable to stroma PTEN expression. Negative was defined as a tumor with complete lack of PTEN. Reduced was defined as a tumor with less PTEN expression compared to internal control stroma. Significantly elevated pS6 expression in the negative tumors helped to validate this scoring system. Positive PTEN was present in 27.5% of cases; this correlated with ER positivity (p=0.009) and the absence of the triple negative subtype (p=0.04). The expression pattern of PTEN in the in situ component tended to reflect that of the invasive component. Interestingly, the reduced PTEN category is not identified in endometrial carcinoma.
Conclusions: Positive PTEN expression in breast cancer is associated with ER expression and the lack of the triple negative subtype, factors indicative of a better clinical outcome. The accuracy of our 3-tiered scoring system relies, in part, on careful comparison of tumor PTEN expression to stromal PTEN expression. The pattern of PTEN expression in breast carcinoma is distinct from that we have previously reported in a large study of endometrial carcinoma. This suggests that it may be necessary to devise unique PTEN scoring systems for cancers derived from different organ systems.
Category: Pathobiology

Wednesday, March 2, 2011 9:30 AM

Poster Session V # 211, Wednesday Morning


Close Window