Abstract
Background Retrospective studies have revealed that overexpression of the epidermal growth factor receptor ErbB-2 (HER2) reduced the efficacy of tamoxifen therapy, which is associated with an increased risk for cardiovascular events. The aim of this study was to evaluate the effects of the HER2 I655V polymorphism (ATC/isoleucine to GTC/valine) on lipid profiles after tamoxifen treatment.
Methods Thirty-four women diagnosed with breast cancer were recruited in the study and followed up for 6 months. The presence of the HER2 I655V polymorphism and fasting plasma lipid profiles before and after tamoxifen treatment were determined for each subject for the duration of the study.
Results In response to tamoxifen therapy, we found that plasma total cholesterol (TC, P = 0.041), low-density lipoprotein-cholesterol (LDL-C, P < 0.001), and high-density lipoprotein-cholesterol (HDL-C, P = 0.012) levels decreased significantly in the A-allele (AA genotype) carriers compared with the baseline measurements. Plasma LDL-C (P < 0.001) and HDL-C (P < 0.001) levels decreased significantly, and the TC/HDL-C (P = 0.027) ratio increased significantly in the G-allele (AG/GG genotypes) carriers. According to the repeated-measures analysis, G-allele carriers had a lower ratio of LDL-C/HDL-C than A-allele carriers did (P = 0.016). Notably, G-allele carriers had a greater reduction in HDL-C concentration than A-allele carriers (P = 0.039; G-allele, −12.4 ± 6.8 mg/dL vs A-allele, −5.6 ± 9.5 mg/dL).
Conclusions This study shows that the HER2 codon 655 G-allele was associated with a larger reduction in plasma HDL-C levels in breast cancer patients under tamoxifen therapy. Therefore, we suggest that the polymorphism of HER2 655 codon should be considered for the prevention of cardiovascular events in breast cancer patients after tamoxifen therapy.
Clinical trials have shown that treatment with tamoxifen reduces the risk of breast carcinoma recurrence.1,2However, tamoxifen is associated with a number of adverse events that result from its partial estrogen receptor (ER) agonist action such as higher rates of venous thromboembolic events and severe hypertriglyceridemia.3–6One study showed that the plasma triglyceride (TG) levels were significantly increased 1.38-fold in breast cancer patients with hypertriglyceridemia after tamoxifen treatment in comparison to the baseline measurements.7Although the increase remained within the reference range for most patients, severe hypertriglyceridemia was observed in 4 patients after 6 months.7Our previous report suggested that one of the molecular events might be linked to the APOE genotype.8When receiving 6-month tamoxifen therapy, plasma TG levels were reduced by 27.2% in breast cancer patients with APOE4 allele, whereas no significant changes were observed in the APOE4-negative patients.8In addition, tamoxifen was reported to induce an elevation, reduction, or no change in high-density lipoprotein–cholesterol (HDL-C) levels.9High TG and low HDL-C levels were associated with insulin resistance and metabolic syndrome, which were positively correlated with coronary heart disease (CHD).10Hadaegh et al.10showed that the prevalence of metabolic syndrome in subjects with TG/HDL-C of 6.9 or higher reached 63.6% versus 3.0% in those with TG/HDL-C less than 2.8. These findings suggest that an increased risk for cardiovascular events is an important safety consideration in the overall tamoxifen treatment of breast cancer.11–13
Approximately 25% of invasive human breast tumors have amplification of the epidermal growth factor receptor ErbB-2 (HER2) gene and/or overexpression of HER2 protein, which reduces responsiveness to antiestrogen therapy.14The HER2 protein is involved in cell division, differentiation, and apoptosis and is frequently amplified in breast tumors.15The HER2 protein is a tyrosine kinase that phosphorylates downstream serine/threonine kinases in response to growth factor stimulation.16,17These downstream kinases phosphorylate ER in the presence of estrogen or tamoxifen. Notably, crosstalk between ER and HER2 signaling pathways is one of the mechanisms for resistance to endocrine therapy in breast cancer.18In addition, both HER2 and its dimerization partner, epidermal growth factor receptor (EGFR), have been found to be associated with lipid rafts, which are membrane microdomains that are enriched in cholesterol and sphingolipids. Membrane cholesterol provides a dynamic environment that facilitates the lateral mobility of EGFR and HER2 in the membrane and promotes ligand stimulation.19
Retrospective studies have observed that HER2 overexpression, even when ER is positive, may lead to a reduced response to tamoxifen therapy.14,18The HER2 I655V polymorphism (ATC/isoleucine to GTC/valine) in the transmembrane region was associated with increased risk of breast cancer in Chinese women younger than 45 years and in women with a first-degree family history of the disease.20Therefore, to delineate the role that HER2 I655V has on the response of lipid metabolism to tamoxifen, we determined the plasma lipid concentrations before and after a 6-month tamoxifen therapy regimen in breast cancer patients.
SUBJECTS AND METHODS
Study Subjects
A total of 34 premenopausal and postmenopausal patients with breast cancer (aged 36–68 years) were recruited from the Fong Yuan Hospital and the Changhua Christian Hospital in Taiwan. After the surgical treatment, patients received the tamoxifen therapy (10 mg, twice daily) for 6 months. This study was approved by both the institutional review board of the Fong Yuan Hospital and the institutional review board of the Changhua Christian Hospital. Patients with unexplained uterine bleeding, preexisting endometrial cancer, thyroid function abnormalities, diabetes mellitus, or other endocrinopathies were excluded from this study. Patients receiving chemotherapy, radiotherapy, or other medications (eg, β-blockers, diuretics, or corticosteroids) that were likely to interfere with the drug under study or with the serum lipid profiles before beginning tamoxifen treatment were also excluded. None of the patients had distant metastasis, bilateral breast cancer, or overlapping cancer at the beginning of tamoxifen therapy. The HER2 genotypes and clinical data including age, body mass index (BMI), lipid profiles, menopause status, family history of breast cancer, pathologic disease, ER status, progesterone receptor status, and HER2 status in the breast cancer patients before beginning tamoxifen treatment were shown in Tables 1 and 2. Breast cancer patients were classified into 2 groups according to HER2 genotypes: A-allele (A/A, n = 22) and G-allele (A/G and G/G, n = 12). Allele frequencies of the variant alleles were 80.9% for the A-allele and 19.1% for the G-allele (Table 1). The sample size is in Hardy-Weinberg equilibrium proportions. The G-allele frequency in Taiwanese patients in our study (19.1%) is higher than that in the Chinese patients (15.8%),20whereas the frequency of the A/G and G/G genotypes is 35.3%, falling within a range from 21.7% in Chinese populations to 40% in white populations.20This study was designed to evaluate the relationship between HER2 I655V polymorphism (ATC/isoleucine to GTC/valine) and fasting plasma lipid profiles after tamoxifen treatment for 6 months.
Polymorphism Analysis
Genomic DNA was extracted from whole blood using the Viogene isolation kit (Viogene, Taipei, Taiwan). HER2 codon 655 genotyping was performed according to the method of Akisik and Dalay.21The polymorphisms (HER2 I655V, rs1136201) were analyzed using polymerase chain reaction (PCR)-restriction fragment length polymorphism analysis. The amplicon was generated with the following PCR primers: forward, 5′-AGCCCTC TGACGTCCAT-3′; and reverse, 5′-GCAGCAGTCTCCGCA-3′. The PCR product was subsequently digested with the BsmA1 (New England Biolabs, Ipswich, MA) restriction enzyme. Restriction fragments were separated on a 2% agarose gel.
Plasma Lipids Analysis
Blood samples were obtained from each patient after a 12-hour overnight fast for the determination of lipids and lipoproteins before tamoxifen treatment (baseline) and after 6 months of tamoxifen treatment. Plasma concentrations of total cholesterol (TC), low-density lipoprotein–cholesterol (LDL-C), HDL-C, and TG were measured by a Beckman Coulter SYNCHRON LX20 PRO analyzer (Brea, CA).
Statistical Analysis
Plasma lipid parameters, which included TC, TG, LDL-C, HDL-C, TC/HDL-C ratio, and LDL-C/HDL-C ratio, were measured before and after tamoxifen treatment and were compared using the paired t test and repeated-measures of analysis of variance (ANOVA). A one-way ANOVA was used to compare the differences between the HER2 codon 655 genotypes. P < 0.05 was considered statistically significant. Data are shown as mean ± SD. Data were analyzed using SPSS 15.0 for Windows (SPSS Inc., Armonk, NY).
RESULTS
Baseline Characteristics in Study Participants
The baseline clinical characteristics and HER2 codon 655 genotypes of study participants are shown in Table 2. We found no significant difference in age, BMI, TC, TG, LDL-C, and HDL-C as determined by one-way ANOVA between the carriers of the A-allele (A/A genotype) and G-allele (A/G and G/G genotypes) before beginning tamoxifen treatment.
Effects of HER2 Codon 655 Genotypes on Plasma Lipids After Tamoxifen Treatment
The effects of tamoxifen before (baseline) and after the 6-month treatment regimen on plasma lipids are shown in Table 3. Individual paired t tests were performed to determine differences between specific HER2 codon 655 genotypes in response to tamoxifen treatment. After tamoxifen treatment, we found that A-allele carriers had significant decreases in plasma TC (P = 0.041), LDL-C (P < 0.001), and HDL-C (P = 0.012; Table 3) concentrations. We also observed that G-allele carriers had significant decreases in plasma LDL-C (P < 0.001) and HDL-C (P < 0.001) concentrations and an increase in the TC/HDL-C ratio (P = 0.027; Table 3). Furthermore, we used ANOVA for repeated measures to test the effects of the HER2 codon 655 genotypes on the plasma lipid parameters. G-allele carriers had a lower LDL-C/HDL-C ratio than A-allele carriers did (P = 0.016; Table 3). However, no significant difference in the TG/HDL-C ratio was found between the groups with the A and G alleles (P = 0.551).
Changes of Plasma Lipids Between A- and G-Allele Carriers
One-way ANOVA was used to analyze the changes of plasma TC, LDL-C, HDL-C, and TG from baseline to the sixth month levels separated according to HER2 codon 655 genotypes (Fig. 1). When we compared the changes in lipid concentrations after 6-month tamoxifen treatment, G-allele carriers had a greater reduction in HDL-C concentrations than A-allele carriers did (P = 0.039; Fig. 1). No significant difference in the TC, LDL-C, and TG changes was observed between both genotypes (Fig. 1).
DISCUSSION
Our study shows for the first time that the HER2 I655V polymorphism influences the response of plasma HDL-C concentrations and the LDL-C/HDL-C ratio in breast cancer patients treated with tamoxifen. Patients with the AG/GG genotypes (G-allele carriers) had a greater reduction in HDL-C concentration in response to tamoxifen therapy compared with the AA genotype (A-allele carriers). After 6-month tamoxifen treatment, HDL-C decreased by 8.7% (mean percent change from baseline) in A-allele carriers, whereas in G-allele carriers, the reduction was 20.3%. The reduction in plasma HDL-C levels could be partially explained by the tamoxifen treatment. However, G-allele carriers showed a greater HDL-C reductive effect in response to tamoxifen treatment. Previous reports have shown that the HER2 signaling pathways could influence the response to the tamoxifen therapy in breast cancer patients.14,18The isoleucine-to-valine change in the HER2 codon 655 is known to affect stabilization of hydrophobic protein domains such as transmembrane domains.22,23The I655V variant can excessively stabilize the HER2 active dimeric state due to substitution of the bulk side chain of isoleucine with a smaller one of valine, thus allowing tighter transmembrane helix packing.23This can result in an increased activity of protein tyrosine kinase24and influence estrogen action.25Through this effect, the HER2 codon 655 valine allele may improve the estrogen action of tamoxifen and result in lowering HDL-C levels in breast cancer patients.
The decrease in HDL-C should be considered as having an important effect on the lipid profile because low levels of HDL-C are associated with an increased risk of CHD. Epidemiological studies have revealed that an increment in HDL-C level of 1 mg/dL was associated with a decrease in CHD risk of 2% to 3%.26Indeed, HDL-C was found to be a CHD risk factor of approximately equal magnitude to LDL-C.27In our study, both LDL-C and HDL-C levels were decreased in G-allele carriers, and G-allele carriers had a lower overall LDL-C/HDL-C ratio than A-allele carriers did. The LDL-C/HDL-C ratio is a clinically important cardiovascular risk factor. The decrease in LDL-C levels of G-allele carriers may help to counter the negative impact of the decrease in HDL-C.28,29However, HDL was reported to help the endothelium to maintain its structural integrity and selective permeability for biomolecules.30Therefore, on the basis of our findings that G-allele carriers manifested a larger reduction in HDL-C levels after tamoxifen treatment, breast cancer patients on tamoxifen therapy may subsequently develop an increased risk for CHD. However, the present study was not designed to assess the long-term effect of HER2 codon 655 genotypes on CHD risk in breast cancer patients undergoing tamoxifen therapy. Therefore, this finding should be assessed in subsequent and larger studies.
In addition, serum HDL-C level was inversely related to the risk of postmenopausal breast cancer in overweight (BMI = 25–30 kg/m2) and obese women.31Epidemiological studies have revealed that the peak of age distributions for East Asian breast cancer patients occurs in the range from 40 to 50 years, contrasting with the incidence occurring at older than 50 years in Western patients.32According to the age stratum analysis, there was a 2.24-fold increased risk in the early-onset (<45 years old) breast cancer in Taiwanese women with the G-allele.33Our study shows that the mean (SD) age of female breast cancer patients in Taiwan (45.5 [9.0] years) is younger than that reported in American white women (58.0 [9.8] years).34There are only 7 of the 34 postmenopausal patients in our study. Therefore, we did not focus on postmenopausal status in this study. Nevertheless, the effects of HER2 codon 655 genotypes on lipid parameters after tamoxifen treatment were also analyzed by an analysis of covariance. After adjustment for baseline values of each lipid parameter, age, and BMI, we found that plasma levels of TC, LDL-C, and HDL-C did not differ significantly between HER2 codon 655 genotypes after tamoxifen therapy (data not shown). Results from animal studies and clinical trials suggested that tamoxifen-induced hypertriglyceridemia may be dose or genotype related.7Our previous study showed that breast cancer patients with the APOE4 allele had favorable effects on lowering the plasma TG level after tamoxifen therapy.8However, no significant differences in the TG changes were observed between HER2 A and G genotypes after tamoxifen treatment (Fig. 1). Because the serum lipid profile is influenced by many factors such as genotypes, dietary fats, lifestyles, and environmental factors, after adjusting for baseline values of TG, age, and BMI, TG levels did not differ significantly between both genotypes (P = 0.144). The mean change from baseline in A-allele is 34.2% (between −55.7% and 250.0%) and that in G-allele is 15.5% (between −53.9% and 208.8%). These findings suggest that HER2 I655V polymorphism was not related to high plasma TG levels.
Finally, there was no conclusive evidence for a relationship between the overexpression of HER2 protein and the HER2 genotypes in this study. We found that only 5 (15%) of the 34 patients were positive for HER2 expression. It has been known that HER2 oncogenic signaling caused by gene amplification or activating mutation provides survival advantages for tumor growth.35One possible explanation for this observation is that the HER2-positive rates (15%) in our study was similar to previous reports which showed HER2-positive rates have trended below 20%, with most investigators currently reporting that the true positive rate is in the range of 15% to 20%.35The majority (approximately 80%) of HER2 screening is done by immunohistochemistry, with results of 0 and 1 considered negative, 2 considered equivocal, and 3 considered positive.36Therefore, whether HER2 gene mutation at codon 655 can directly respond to HER2-positive status, however, has not been validated until now. Another possible explanation is that other biologic pathways improved by tamoxifen are not directly linked to HER2 protein overexpression, whereas they are correlated to HER2 genotypes. The I655V variant may improve the estrogen action of tamoxifen, and the membrane ER can in turn activate several growth factor pathways that can then activate and augment nuclear ER functions.18Apparently, further studies will be needed to prove bidirectional crosstalk between ER and growth factor pathways such as EGFR, insulin-like growth factor I receptor, and phosphoinositide 3-kinase in breast cancer.
In conclusion, our results indicate that there are no significant differences between AG/GG and AA genotypes in the effects of tamoxifen on TC, LDL-C, and TG concentrations. However, patients with the AG/GG genotypes had a greater reduction in HDL-C concentration in response to tamoxifen therapy compared with those with the AA genotype. Therefore, we suggest that the polymorphism of the HER2 codon 655 should be taken into consideration also for preventing cardiovascular events in breast cancer patients under tamoxifen therapy.