Phosphatase and Tensin Homolog Deficiency and Resistance to Trastuzumab and Chemotherapy

Abstract

The phosphatase and tensin homolog (PTEN) gene is a tumor suppressor that is mutated in a large number of cancers at high frequency. Oncogenic receptor tyrosine kinases such as human epidermal growth factor receptor 2 (HER2 [ERBB2]) potently activate phosphatidylinositol-3 kinase (PI3K) by autophosphorylating their C-termini and/or phosphorylating associated adaptor proteins such as GAB1/2, insulin receptor substrates 1/2, and HER3. In turn, these phosphorylated tyrosines bind the p85 regulatory subunit of PI3K, thus relieving the p110 catalytic subunit of PI3K from p85-mediated inhibition. Active p110 phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) to produce the second messenger phosphatidylinositol-3,4,5-trisphosphate (PIP3), which then recruits pleckstrin homology domain– containing proteins like AKT, PDK1, steroid glucocorticoid kinase, and so on to the plasma membrane. These kinases become activated at the plasma membrane and mediate enhanced cell proliferation and survival. The protein encoded by the PTEN gene is the 3 lipid phosphatase of PIP3. Therefore, PTEN-deficient tumors lack negative regulation of PI3K signaling and, as a result, exhibit aberrant activation of this transforming pathway. This regulation is particularly important in breast cancers with HER2 gene amplification, which rely on HER2-activated PI3K for their progression and survival. Indeed, sustained inhibition of PI3K is required for the antitumor action of the HER2 inhibitors trastuzumab and lapatinib. In HER2-positive/PTEN-deficient tumors, the hyperactivity of PI3K, as a result of both HER2 overexpression and loss of PTEN function, is predicted to attenuate the action of HER2 inhibitors and, therefore, potentially lead to drug resistance, as shown in several mechanistic and clinical correlative studies. The report by Perez et al that accompanies this editorial is the largest study to date that evaluates the correlation between PTEN status and response to trastuzumab in patients with HER2overexpressing breast cancer. In this study, intensity of PTEN expression was scored by immunohistochemistry (IHC) in 1,802 HER2-positive tumors from patients who were enrolled onto the North Central Cancer Treatment Group N9831 adjuvant trial of trastuzumab. N9831 compared adjuvant chemotherapy (arm A) versus adjuvant chemotherapy followed by trastuzumab (arm B) versus adjuvant chemotherapy with concomitant trastuzumab (arm C) in patients with resected HER2-positive breast cancer. PTEN-positive tumors were those with any staining of more than 0%. Appropriate for the scoring of a tumor suppressor, tumors with 0% staining were compared with tumors that exhibited any detectable PTEN protein. A total of 1,286 tissue microarray sections with three cores per block and 516 whole-tumor sections were analyzed. The rate of PTEN loss was 26%. In comparing disease-free survival (DFS) within each treatment arm, there was no difference between patients with PTEN-positive and PTEN-negative tumors. The investigators also compared DFS between the treatment arms for both the PTEN-positive and PTENnegative groups. Hazard ratios (HRs) for concurrent trastuzumab (arm C) versus no trastuzumab (arm A) were 0.65 (P .003) for PTEN-positive and 0.47 (P .005) for PTEN-negative tumors. Comparing sequential trastuzumab (arm B) with no trastuzumab (arm A), HRs were 0.7 (P .009) for PTEN-positive cancers and 0.85 (P .44) for PTEN-negative tumors. From these data, the authors concluded that patients with HER2-positive early breast cancer with or without PTEN benefit from adjuvant trastuzumab. Inconsistent with this report, several smaller retrospective clinical studies have shown that loss or low levels of PTEN are associated with a reduced clinical benefit from trastuzumab. Furthermore, well-conducted mechanistic studies have shown that genetic inactivation of PTEN in HER2-overexpressing cells counteracts the effect of HER2 inhibitors like trastuzumab. How might the difference between this large, carefully conducted study and other clinical correlative and preclinical studies be reconciled? Several possibilities to explain these differences should be considered. First, for practical reasons, most studies in HER2-overexpressing breast cancer have assessed PTEN status (which we interpret as PTEN function) by IHC. However, we should note that not all molecular alterations that result in a loss in PTEN function are detected by IHC. Loss of PTEN function can result from mutation or deletion of the PTEN gene, gene inactivation by loss of heterozygosity at 10q23, reduced RNA/protein levels via transcriptional dysregulation, microRNA expression, and methylation, or reduced phosphatase activity by post-translational alterations in phosphorylation, acetylation, or ubiquitination of the PTEN protein. This suggests that a fraction of HER2 gene–amplified breast cancers that are PTEN positive by IHC may express otherwise nonfunctional PTEN. In the Catalogue of Somatic Mutations in Cancer database, PTEN mutations were detected in approximately 6% of all subtypes of breast cancer, but in the recent analysis of 453 breast tumors by The Cancer Genome Atlas, PTEN mutations were observed in 16 of 453 tumors (3.5%) and did not JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 31 NUMBER 17 JUNE 1