人类癌症中的磷脂酰肌醇3激酶（PI3K）通路突变
2016年7月7日，《美国医学会杂志肿瘤学分册》在线发表美国俄勒冈健康科学大学奈特癌症研究所空间系统生物医学中心生物医学工程学家的同期述评：人类癌症中的PI3激酶通路突变。
JAMA Oncol. 2016 Jul 7. [Epub ahead of print]
PI3 Kinase Pathway Mutations in Human Cancers.
Joe W. Gray.
Department of Biomedical Engineering, Center for Spatial Systems Biomedicine, Knight Cancer Institute, Oregon Health and Science University, Portland.
International genome analysis studies have now cataloged mutations in thousands of human cancers. An assessment of these studies using cBioPortal [1,2] and by Tamborero et al [3] shows that the phosphatidylinositol 3-kinase (PI3K) pathway genes, including PIK3CA, PTEN, and AKT1, are among the most commonly mutated in human cancers of all types. Only TP53 is mutated more frequently. PI3K signaling typically is activated by signaling from tyrosine kinases and other receptors and proceeds through recruitment of a regulatory subunit (eg, p85α encoded by PIK3R1) and a catalytic subunit (eg, p110α, encoded by PIK3CA) to produce phosphatidylinositol-3,4,5-trisphosphate (PIP3). PIP3 recruits protein kinase B (AKT) that, in turn, stimulates growth and cell survival. This process is negatively regulated by PIP3 dephosphorylation by the tumor suppressor, PTEN. PI3K pathway mutations enhance signaling along the pathway directly in the case of PIK3CA and AKT1 or by inactivating the negative regulator, PTEN. [4]
This is information has fueled the development of a growing number of therapeutic agents that target the PI3K pathway. [5] Notable examples include idelalisib, buparlisib, duvelisib, and alpelisib (ClinicalTrials.gov). These are being considered for assessment in a broad range of cancers carrying PI3K-activating aberrations. Importantly, the frequency of activating events varies considerably among cancer types. An analysis of 87 published studies of diverse cancer types reported in cBioPortal [1,2] shows that mutations in PIK3CA, PTEN, and AKT1 occur in more than 40% of cancers of the uterus, lung, breast, prostate, brain, and head and neck. However, few aberrations involving these genes have been found in pancreatic adenocarcinoma, central nervous system lymphoma, small cell carcinoma of the ovary, liver hepatocellular carcinoma, neuroblastoma, and thymic epithelial tumors. Notably, genome studies published to date were been intended as aberration discovery efforts and were not designed to assess the frequencies of aberrations in the overall clinical population. Moreover, the tumors analyzed often were selected to be relatively large and sufficiently necrosis free to enable high-quality genomic analysis. This may impose some selection bias. In addition, some studies were restricted in ethnic diversity.
Millis et al [6] now present a study of the frequency with which PIK3CA, PTEN, and AKT1 and other "actionable" genes are aberrant in a broad range of human cancers. They assessed aberrations involving these genes in 19,784 consecutive diverse solid tumors submitted by clinicians in more than 60 countries. Mutations in PIK3CA, PTEN, and AKT1 were assessed in a commercial Clinical Laboratory Improvement Amendment-certified laboratory (Caris Life Sciences) using massively parallel sequencing targeting mutational hot spots in 47 genes with an average coverage of ×1000. This sequencing depth may allow detection of mutations that might have been missed in published mutation discovery efforts in which sequencing depth was lower. Loss of PTEN was assessed using immunohistochemical analysis with loss defined as no protein expression in more than 50% of cells. The Millis et al [6] study supports previous reports of a high frequency of TP53 and PI3K pathway genes in a broad range of human tumors. It also supports previous reports that PI3K pathway aberrations are particularly prevalent in cancers of the uterus, breast, prostate, and lung and adds anal, liver, and colorectal cancers as frequent pathway aberration carriers. All of these cancers might be considered as candidates for treatment with PI3K-targeted drugs.
Millis et al [6] also assessed the frequency with which PI3K pathway aberrations interact with hormonal status or other actionable mutations. The associations of activating PIK3CA and AKT1 mutations and inactivating PTEN mutations with hormonal status are remarkable. The frequency of mutations that activate PI3K signaling is almost twice as high in cancers with high estrogen receptor (ER), progesterone receptor (PR), or androgen receptor (AR) expression as in cancers with low hormonal receptor expression. Interestingly, the association of PTEN loss as measured using immunohistochemical analysis was not so strongly associated with hormonal receptor expression. Thus, mutational status seems to be a more definitive indicator of PI3K activity than PTEN loss as measured by immunohistochemical analysis. This raises the possibility that the 50% positivity cutoff used to define PTEN loss may have been too generous. Requiring a more complete loss of staining might yield a more accurate but perhaps less sensitive indicator of PI3K activation. This may explain why PTEN loss assessed using immunohistochemical analysis was sometimes found in cancers with other PI3K pathway mutations.
Mutations in PIK3CA, AKT1, and PTEN also were less frequently found in cancers with mutations in the DNA guardian, TP53. This observation might seem counterintuitive because increased genome instability might facilitate development and selection of advantageous mutations. However, TP53 is mutated less frequently in tumors that express ER, PR, and/or AR, and PI3K genes are more frequently mutated in such tumors. This suggests that selection of PI3K mutations occurs more because they are strongly advantageous to tumors that depend on high hormonal signaling than because loss of genome stability enables their selection. Tumors with PI3K pathway mutations, high expression of ER, PR, or AR, and wild-type TP53 seem to be particularly interesting candidates for PI3K pathway targeted therapies.
Overall, Millis et al [6] have assessed the frequency of mutations in 47 genes, including PIK3CA, PTEN, and AKT1, in 19,784 consecutive diverse solid tumors from 60 countries using targeted massively parallel sequencing. They identify several cancer types in which PI3K pathway genes are present at high frequency that should be considered as candidates for PI3K pathway-targeted drugs. The use of immunohistochemical analysis to assess PTEN loss is novel, but additional work may be needed to define a scoring threshold that accurately identifies tumors in which PI3K signaling is aberrant. It is unfortunate that outcome information was not available for the Millis et al [6] study and that more PI3K pathway genes (eg, AKT2, PIK3R1) were not sequenced. Future studies might be designed to collect this information in order to further our understanding of the impact of PI3K pathway genes on cancer biology and treatment.
REFERENCES
Cerami E, Gao J, Dogrusoz U, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2(5):401-404.
Gao J, Aksoy BA, Dogrusoz U, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):pl1.
Tamborero D, Gonzalez-Perez A, Perez-Llamas C, et al. Comprehensive identification of mutational cancer driver genes across 12 tumor types. Sci Rep. 2013;3:2650.
Yuan TL, Cantley LC. PI3K pathway alterations in cancer: variations on a theme. Oncogene. 2008;27(41):5497-5510.
Wong KK, Engelman JA, Cantley LC. Targeting the PI3K signaling pathway in cancer. Curr Opin Genet Dev. 2010;20(1):87-90.
Millis SZ, Ikeda S, Reddy S, Gatalica Z, Kurzrock R. Landscape of phosphatidylinositol-3-kinase pathway alterations across 19 784 diverse solid tumors. JAMA Oncol. 2016 Jul 7. DOI:10.1001/jamaoncol.2016.0891 [Epub ahead of print]
DOI: 10.1001/jamaoncol.2016.0875