J Cancer Res Clin Oncol DOI 10.1007/s00432-014-1663-x
Original Article – Clinical Oncology
A GG allele of 3′‑side AKT1 SNP is associated with decreased AKT1 activation and better prognosis of gastric cancer Xiaoting Wang · Youdong Lin · Fenghua Lan · Yinghao Yu · Xuenong Ouyang · Xuzhou Wang · Qiaojia Huang · Lie Wang · Jianming Tan · Feng Zheng
Received: 21 January 2014 / Accepted: 25 March 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose v-akt Murine thymoma viral oncogene homolog (AKT) pathway is critically involved in cancer cell growth, invasion, and survival. We examined the correlation between the genetic variations in molecules of AKT pathway and clinical outcomes of gastric cancer. Patients and methods Six single nucleotide polymorphisms (SNPs) located in the four core genes of AKT pathway, namely the PIK3CA, PTEN, AKT1, and mTOR, were determined in 221 patients with stage T2 and T3 gastric cancer. Additionally, the activation of AKT1 in gastric cancer tissues was examined by immunostaining. The correlation between SNPs, AKT activation, and the progress of
gastric cancer was analyzed after an average of 51-month follow-up. Results The overall recurrence and survival rate in this study group were 54.8 and 46.6 %, respectively. The recurrence rate was reduced 30.4 %, and the survival rate was increased 33.7 % in patients with GG allele of a 3′-side AKT1 SNP (rs2498804). Significantly, GG allele was associated with lower AKT1 activation in gastric cancer tissues. On the contrary, CC allele of PTEN (rs701848) was associated with the increased risk of recurrence (hazard ratio [HR] 2.06, 95 % CI 1.19–3.58) and patient death (HR 2.01, 95 % CI 1.15–3.53). Conclusions The genetic variants in the PI3K/PTEN/AKT especially the GG allele in 3′ side of AKT1 are closely related to clinical outcomes of gastric cancer.
Xiaoting Wang and Youdong Lin are contributed equally to this paper. X. Wang · Y. Lin · F. Lan · Q. Huang (*) · J. Tan Department of Experimental Medicine, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
L. Wang (*) Department of General Surgery, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
Present Address: Y. Lin Department of Clinical Laboratory Medicine, Fujian Provincial Hospital, 134 Dongjie Road, Fuzhou 350001, Fujian, China
J. Tan (*) Organ Transplant Institute, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
Y. Yu · X. Wang Department of Pathology, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China X. Ouyang Department of Oncology, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China
F. Zheng (*) Department of Nephrology and Central Laboratory, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
13
Keywords SNPs · PI3K/PTEN/AKT/mTOR · AKT1 SNP (rs2498804) · Gastric cancer · Clinical outcome
Introduction Gastric cancer (GC) is the second most common malignant tumor in the world (Jemal et al. 2011). Surgery is the most effective treatment for GC patients with localized disease. Adjuvant radiotherapy and/or chemotherapy may reduce the recurrence (Cunningham et al. 2006; Sakuramoto et al. 2007; Paoletti et al. 2010). However, the prognosis of gastric cancer varies considerably even among patients with the same stage of disease and receiving the similar treatment. Previous studies have demonstrated that genetic and epigenetic factors are involved in the pathogenesis of gastric cancer (Figueiredo et al. 2013). For instance, exomes sequencing has revealed mutations in FAT4, ARID1A, and PIK3CA in gastric cancer tissues (Zang et al. 2012), and epigenetic alternations in DNA methylation were found in wide range of genes in gastric cancer (Park et al. 2011). Moreover, genetic variants in individuals have been found to affect the effectiveness of chemotherapy (Oh et al. 2013). Therefore, it is reasonable to postulate that genetic factors may play a role in determining the clinical outcomes of gastric cancer. The phosphoinositide 3-kinase (PI3K), phosphatase and tensin homolog (PTEN), v-akt murine thymoma viral oncogene homolog (AKT), and mammalian target of rapamycin (mTOR) are essential components of AKT signal pathway that is critically involved in multiple cellular processes (Nicholson and Anderson 2002; Bellacosa et al. 2005). Various growth factors or stress signal(s) act via receptors and/or kinase(s) to activate PI3K and trigger a kinase cascade through AKT, leading to cell growth, proliferation, and survival. Increased PI3K, AKT, and mTOR and decreased negative regulator PTEN are observed in various cancers (Martelli et al. 2011; Hildebrandt et al. 2009; Pu et al. 2011; Hildebrandt et al. 2012; Chen et al. 2010; Oki et al. 2005). The abnormal activation of PI3K/AKT/mTOR may contribute to recurrence, metastasis, chemoresistant, and poor survival. In patients with gastric cancer, increased PI3K and phosphorylated AKT stainings in cancer tissues were observed in more than 50 % of patients and these increases were associated with poor clinical outcome. The underlying molecular mechanism(s) for abnormal PI3K/AKT activation in cancers are not clear. Functional germline mutation(s) in genes involved in this pathway may play an important role, but those kinds of mutation(s) are not regular. There are genetic polymorphisms in core components of PI3K/PTEN/AKT/mTOR, which may affect the levels of genes expression under normal and disease condition(s). Michelle et al. showed that polymorphisms in AKT1, AKT2, PTEN, and mTOR
13
J Cancer Res Clin Oncol
were associated with recurrence, survival, and responsiveness to chemotherapy in esophageal cancer (Hildebrandt et al. 2009). Lately, different AKT alleles were reported to be correlated with brain metastasis in lung cancer (Li et al. 2013). Moreover, distant progression and toxicity of platinum were related to genetic polymorphisms of PI3K/PTEN/AKT/mTOR in patients with lung cancer (Pu et al. 2011). Here, we examined for the first time the polymorphisms in PI3K, AKT1, PTEN, and mTOR in patients with gastric cancer. The association between genetic variants in PI3K, AKT1, PTEN, and mTOR and the levels of AKT1 activation and the prognosis in gastric cancer was assessed.
Materials and methods Patients Two hundred and twenty-one patients with resectable gastric cancer who underwent surgery and regular postoperative chemotherapy were recruited between 2003 and 2008 at Fuzhou General Hospital. Patients had no family history of gastric cancer and were all with Han ethnicity. Additionally, all cases selected met the following criteria: (1) gastric adenocarcinoma of T2 and T3 stages (NCNN guideline 2011); (2) underwent curative gastrectomy with lymph node dissection without surgery-related major or serious complications; and (3) received regular postoperative chemotherapy (fluoropyrimidines or fluoropyrimidines plus platinum). We selected patients with T2 and T3 stages because patients in T1 stage normally have a relative good prognosis, while the prognosis is generally poor in patients with T4 stage. Clinical data collection Gross examination during operation and postoperative histopathologic examinations included tumor size and location, gross and microscopic pathology, lymph node status, the width and depth of stomach wall invasion, TNM stages, and additional clinicopathological factors. Survival data were collected via regular follow-up, telephone interviews, community data system, and the social security data system. Patients were followed in every 6 months, and the checkup included physical examination, hematological analysis, serum tumor markers (CEA, CA125, CA199), abdominal ultrasonography, and chest X-ray or chest computed tomography. Gastric endoscopy was performed once a year. Computed tomography or MRI was ordered in suspicious of metastasis. The diagnosis of the recurrence was based on imaging results, endoscopy biopsy, or re-operation. The type of recurrence was classified as
J Cancer Res Clin Oncol
hematogenous, peritoneal, locoregional, or the combination. The follow-up was closed in March 2013. This study was approved by the Institutional Review Board and Ethics Committee of Fuzhou General Hospital and had obtained written informed consent from subjects. Investigators were blinded to the patient’s genotype status. SNP selection and genotyping SNP alleles on each gene locus were chosen based on the descriptions by Hildebrandt et al., who had demonstrated that these SNP were closely associated with the clinical outcomes of esophageal cancer (Hildebrandt et al. 2009) and were further confirmed via the NIEHS SNP database (available at [http://snpinfo.niehs.nih.gov/]). A total of six tagging SNPs in PIK3CA, PTEN, AKT1, and FRAP1 (mTOR) genes were selected for genotyping. SNP function prediction is obtained from [snpinfo/snpfunc.htm]. Genomic DNA from paraffin-embedded gastric cancer tissues was extracted using the QIAamp DNA kit (Qiagen GmbH, Hilden, Germany). SNPs were genotyped by first amplification using polymerase chain reaction, continued extension by SPA-iPlex systems, and finally genotyped by MassARRAY SpectroCHIP (SEQUENOM Inc, San Diego, CA). Data were analyzed by TYPER software (SEQUENOM Inc, San Diego, CA). Quality control measures were conducted throughout the study. Phospho‑AKT1 immunohistochemistry assay The activation of p-AKT1 in gastric cancer tissues was detected by immunohistochemistry (IHC) with a specific activated AKT1 antibody (anti-p-ser473, 1:100 dilution, Abcam Company, Cambridge, MA, USA) using a method as described by us previously (Huang et al. 2011). The intensities of p-AKT1 levels were determined by a score system introduced by Kawata. A pathologist blinded to genotype gave each slide a grade of 0, 1, 2, and 3. A grade of 0 or 1 was classified as weak staining, and a grade of 2 or 3 was classified as strong staining (Kawata et al. 2011). Statistical analysis The differences in patients’ characteristics and in survival and recurrence between two SNP alleles were evaluated by the χ2 test (or Fisher’s exact test when required). The association of the SNPs with clinical outcomes was assessed at study end points, mainly including survival and recurrence. Hazard ratios (HRs) and 95 % confidence intervals (95 % CI) in recurrence and survival among SNPs were calculated via Cox model after adjustment for gender, age, tumor site, histologic grade, gross findings, lymph node metastasis, and postoperative chemotherapy. The log-rank tests and
Kaplan–Meier survival function were applied to evaluate differences in recurrence-free and overall survival times. In addition, the cumulative effects of unfavorable genotypes were confirmed from the main effects of analysis of single SNPs. The levels of AKT1 activation and the rates of survival and recurrence and the difference in AKT1 activation between two SNP alleles were evaluated using the χ2 test. All statistical analyses were carried out using the SPSS software version 17.0 (SPSS Inc., Chicago, IL). Total p values were two-sided, and p
Original Article – Clinical Oncology
A GG allele of 3′‑side AKT1 SNP is associated with decreased AKT1 activation and better prognosis of gastric cancer Xiaoting Wang · Youdong Lin · Fenghua Lan · Yinghao Yu · Xuenong Ouyang · Xuzhou Wang · Qiaojia Huang · Lie Wang · Jianming Tan · Feng Zheng
Received: 21 January 2014 / Accepted: 25 March 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose v-akt Murine thymoma viral oncogene homolog (AKT) pathway is critically involved in cancer cell growth, invasion, and survival. We examined the correlation between the genetic variations in molecules of AKT pathway and clinical outcomes of gastric cancer. Patients and methods Six single nucleotide polymorphisms (SNPs) located in the four core genes of AKT pathway, namely the PIK3CA, PTEN, AKT1, and mTOR, were determined in 221 patients with stage T2 and T3 gastric cancer. Additionally, the activation of AKT1 in gastric cancer tissues was examined by immunostaining. The correlation between SNPs, AKT activation, and the progress of
gastric cancer was analyzed after an average of 51-month follow-up. Results The overall recurrence and survival rate in this study group were 54.8 and 46.6 %, respectively. The recurrence rate was reduced 30.4 %, and the survival rate was increased 33.7 % in patients with GG allele of a 3′-side AKT1 SNP (rs2498804). Significantly, GG allele was associated with lower AKT1 activation in gastric cancer tissues. On the contrary, CC allele of PTEN (rs701848) was associated with the increased risk of recurrence (hazard ratio [HR] 2.06, 95 % CI 1.19–3.58) and patient death (HR 2.01, 95 % CI 1.15–3.53). Conclusions The genetic variants in the PI3K/PTEN/AKT especially the GG allele in 3′ side of AKT1 are closely related to clinical outcomes of gastric cancer.
Xiaoting Wang and Youdong Lin are contributed equally to this paper. X. Wang · Y. Lin · F. Lan · Q. Huang (*) · J. Tan Department of Experimental Medicine, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
L. Wang (*) Department of General Surgery, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
Present Address: Y. Lin Department of Clinical Laboratory Medicine, Fujian Provincial Hospital, 134 Dongjie Road, Fuzhou 350001, Fujian, China
J. Tan (*) Organ Transplant Institute, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
Y. Yu · X. Wang Department of Pathology, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China X. Ouyang Department of Oncology, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China
F. Zheng (*) Department of Nephrology and Central Laboratory, Fuzhou General Hospital (Dongfang Hospital), Clinical College of Fujian Medical University, 156 North Xi‑er Huan Road, Fuzhou 350025, Fujian, China e-mail: [email protected]
13
Keywords SNPs · PI3K/PTEN/AKT/mTOR · AKT1 SNP (rs2498804) · Gastric cancer · Clinical outcome
Introduction Gastric cancer (GC) is the second most common malignant tumor in the world (Jemal et al. 2011). Surgery is the most effective treatment for GC patients with localized disease. Adjuvant radiotherapy and/or chemotherapy may reduce the recurrence (Cunningham et al. 2006; Sakuramoto et al. 2007; Paoletti et al. 2010). However, the prognosis of gastric cancer varies considerably even among patients with the same stage of disease and receiving the similar treatment. Previous studies have demonstrated that genetic and epigenetic factors are involved in the pathogenesis of gastric cancer (Figueiredo et al. 2013). For instance, exomes sequencing has revealed mutations in FAT4, ARID1A, and PIK3CA in gastric cancer tissues (Zang et al. 2012), and epigenetic alternations in DNA methylation were found in wide range of genes in gastric cancer (Park et al. 2011). Moreover, genetic variants in individuals have been found to affect the effectiveness of chemotherapy (Oh et al. 2013). Therefore, it is reasonable to postulate that genetic factors may play a role in determining the clinical outcomes of gastric cancer. The phosphoinositide 3-kinase (PI3K), phosphatase and tensin homolog (PTEN), v-akt murine thymoma viral oncogene homolog (AKT), and mammalian target of rapamycin (mTOR) are essential components of AKT signal pathway that is critically involved in multiple cellular processes (Nicholson and Anderson 2002; Bellacosa et al. 2005). Various growth factors or stress signal(s) act via receptors and/or kinase(s) to activate PI3K and trigger a kinase cascade through AKT, leading to cell growth, proliferation, and survival. Increased PI3K, AKT, and mTOR and decreased negative regulator PTEN are observed in various cancers (Martelli et al. 2011; Hildebrandt et al. 2009; Pu et al. 2011; Hildebrandt et al. 2012; Chen et al. 2010; Oki et al. 2005). The abnormal activation of PI3K/AKT/mTOR may contribute to recurrence, metastasis, chemoresistant, and poor survival. In patients with gastric cancer, increased PI3K and phosphorylated AKT stainings in cancer tissues were observed in more than 50 % of patients and these increases were associated with poor clinical outcome. The underlying molecular mechanism(s) for abnormal PI3K/AKT activation in cancers are not clear. Functional germline mutation(s) in genes involved in this pathway may play an important role, but those kinds of mutation(s) are not regular. There are genetic polymorphisms in core components of PI3K/PTEN/AKT/mTOR, which may affect the levels of genes expression under normal and disease condition(s). Michelle et al. showed that polymorphisms in AKT1, AKT2, PTEN, and mTOR
13
J Cancer Res Clin Oncol
were associated with recurrence, survival, and responsiveness to chemotherapy in esophageal cancer (Hildebrandt et al. 2009). Lately, different AKT alleles were reported to be correlated with brain metastasis in lung cancer (Li et al. 2013). Moreover, distant progression and toxicity of platinum were related to genetic polymorphisms of PI3K/PTEN/AKT/mTOR in patients with lung cancer (Pu et al. 2011). Here, we examined for the first time the polymorphisms in PI3K, AKT1, PTEN, and mTOR in patients with gastric cancer. The association between genetic variants in PI3K, AKT1, PTEN, and mTOR and the levels of AKT1 activation and the prognosis in gastric cancer was assessed.
Materials and methods Patients Two hundred and twenty-one patients with resectable gastric cancer who underwent surgery and regular postoperative chemotherapy were recruited between 2003 and 2008 at Fuzhou General Hospital. Patients had no family history of gastric cancer and were all with Han ethnicity. Additionally, all cases selected met the following criteria: (1) gastric adenocarcinoma of T2 and T3 stages (NCNN guideline 2011); (2) underwent curative gastrectomy with lymph node dissection without surgery-related major or serious complications; and (3) received regular postoperative chemotherapy (fluoropyrimidines or fluoropyrimidines plus platinum). We selected patients with T2 and T3 stages because patients in T1 stage normally have a relative good prognosis, while the prognosis is generally poor in patients with T4 stage. Clinical data collection Gross examination during operation and postoperative histopathologic examinations included tumor size and location, gross and microscopic pathology, lymph node status, the width and depth of stomach wall invasion, TNM stages, and additional clinicopathological factors. Survival data were collected via regular follow-up, telephone interviews, community data system, and the social security data system. Patients were followed in every 6 months, and the checkup included physical examination, hematological analysis, serum tumor markers (CEA, CA125, CA199), abdominal ultrasonography, and chest X-ray or chest computed tomography. Gastric endoscopy was performed once a year. Computed tomography or MRI was ordered in suspicious of metastasis. The diagnosis of the recurrence was based on imaging results, endoscopy biopsy, or re-operation. The type of recurrence was classified as
J Cancer Res Clin Oncol
hematogenous, peritoneal, locoregional, or the combination. The follow-up was closed in March 2013. This study was approved by the Institutional Review Board and Ethics Committee of Fuzhou General Hospital and had obtained written informed consent from subjects. Investigators were blinded to the patient’s genotype status. SNP selection and genotyping SNP alleles on each gene locus were chosen based on the descriptions by Hildebrandt et al., who had demonstrated that these SNP were closely associated with the clinical outcomes of esophageal cancer (Hildebrandt et al. 2009) and were further confirmed via the NIEHS SNP database (available at [http://snpinfo.niehs.nih.gov/]). A total of six tagging SNPs in PIK3CA, PTEN, AKT1, and FRAP1 (mTOR) genes were selected for genotyping. SNP function prediction is obtained from [snpinfo/snpfunc.htm]. Genomic DNA from paraffin-embedded gastric cancer tissues was extracted using the QIAamp DNA kit (Qiagen GmbH, Hilden, Germany). SNPs were genotyped by first amplification using polymerase chain reaction, continued extension by SPA-iPlex systems, and finally genotyped by MassARRAY SpectroCHIP (SEQUENOM Inc, San Diego, CA). Data were analyzed by TYPER software (SEQUENOM Inc, San Diego, CA). Quality control measures were conducted throughout the study. Phospho‑AKT1 immunohistochemistry assay The activation of p-AKT1 in gastric cancer tissues was detected by immunohistochemistry (IHC) with a specific activated AKT1 antibody (anti-p-ser473, 1:100 dilution, Abcam Company, Cambridge, MA, USA) using a method as described by us previously (Huang et al. 2011). The intensities of p-AKT1 levels were determined by a score system introduced by Kawata. A pathologist blinded to genotype gave each slide a grade of 0, 1, 2, and 3. A grade of 0 or 1 was classified as weak staining, and a grade of 2 or 3 was classified as strong staining (Kawata et al. 2011). Statistical analysis The differences in patients’ characteristics and in survival and recurrence between two SNP alleles were evaluated by the χ2 test (or Fisher’s exact test when required). The association of the SNPs with clinical outcomes was assessed at study end points, mainly including survival and recurrence. Hazard ratios (HRs) and 95 % confidence intervals (95 % CI) in recurrence and survival among SNPs were calculated via Cox model after adjustment for gender, age, tumor site, histologic grade, gross findings, lymph node metastasis, and postoperative chemotherapy. The log-rank tests and
Kaplan–Meier survival function were applied to evaluate differences in recurrence-free and overall survival times. In addition, the cumulative effects of unfavorable genotypes were confirmed from the main effects of analysis of single SNPs. The levels of AKT1 activation and the rates of survival and recurrence and the difference in AKT1 activation between two SNP alleles were evaluated using the χ2 test. All statistical analyses were carried out using the SPSS software version 17.0 (SPSS Inc., Chicago, IL). Total p values were two-sided, and p
