Comparison of PTEN protein loss and PI3K gene mutation status in breast cancer

Abstract

Abstract Background: The phosphatidylinositol 3-kinase (PI3K) family of enzymes plays key roles in the transduction of cell signals, and several inhibitors of this pathway are currently in clinical development for the treatment of breast and other cancers. PI3K signaling is activated by growth factor receptors and regulated and terminated by multiple factors including dephosphorylation of the 3′ phosphate of PIP3 by the phosphatase PTEN. Reduced PTEN expression in breast cancer confers susceptibility to inhibitors of the PI3 kinase/Akt pathway. In a subset of breast cancer cases, PTEN protein has been found to be expressed at lower levels (e.g., as compared with normal breast epithelium), however, human breast cancer cells rarely have PTEN mutations. On the other hand PI3K gene mutations have been found in a majority of breast tumors. Methods: To examine the relationship between PTEN protein loss and PI3K gene mutation, PTEN immunoreactivity was assessed semi-quantitatively using histoscores in freshly-cut sections of paraffin blocks from 26 anonymized breast cancer specimens. Sections from the same tumor blocks were used for PI3K mutation detection. One H&E-stained section was used to mark the area of tumor and 2-4 additional unstained sections were macrodissected to enrich for tumor cells, which then underwent DNA extraction. The DNA was analyzed with a commercial PI3K mutation test kit (DxS PI3K Mutation Test Kit, Manchester, UK) that detects 4 somatic mutations in exons 9 and 20 of PIK3CA in three discrete mutation groups or “buckets.” Results: Of the 26 breast cancer specimens examined, 7 (27%) had absence of PTEN protein expression (histoscore=0) and an additional 4 (15%) had a reduction in PTEN expression (histoscore>0 to <50), with positive internal controls. PI3K mutations were detected in 8 (31%) of the 26 breast cancer specimens examined. Of the mutations detected, there were 4 (15%) mutations of H1047R (3140 A>G), 1 (4%) of E542K (1624 G>A), and 3 (12%) of E545D (1635 G>T) or E545K (1633 G>A). Conclusions: There was a statistically significant correlation between PTEN loss (histoscore<50) and the detection of PI3K mutations (Fisher's exact test, 2-tail P=0.003). The percentage of cases with overall PTEN loss in this specimen set was similar to previous reports of PTEN loss in breast tumors. In contrast, another report did not find concomitant PTEN loss and PI3K mutations. However, the investigators evaluated fewer samples, studied HER2 amplified breast cancer specifically, and used a different method (sequencing) to detect mutations. A three-way study that includes HER2 amplification or overexpression may be necessary to explain the discrepancy.