Abstract
Background/Aim: Although several epidemiological studies have indicated that statins may have antitumor properties, the effect of statins on patient survival after curative resection of gastric cancer is unknown. The aim of the present study was to determine whether statin use could improve long-term outcomes after radical gastrectomy. Patiens and Methods: We conducted a matched case–control study of 65 statin users and 176 non-users who underwent radical gastrectomy for stage II and III gastric cancer from January 2006 to December 2009. Results: No significant differences were found in recurrence-free survival (RFS) or overall survival (OS) between statin users and non-users. However, subgroup analysis showed that patients who used statins for more than six months had more favorable outcomes than non-users or those who used statins for less than six months [adjusted hazard ratio of death (non-users as reference); statin use <6 months: 2.405, 95% confidence interval (CI)=1.056-5.477 and statin use >6 months: 0.168, 95% CI=0.032-0.881, p=0.006]. Conclusion: Statin use did not improve RFS or OS after curative resection of stage II or III gastric cancer in the overall study population. However, statin use of more than six months was associated with increased survival.
Inhibitors of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase, known as statins, are widely used for the treatment of lipid disorders, especially hypercholesterolemia. Statins are cholesterol-lowering medications that are effective in the primary and secondary prevention of heart attack and stroke (1, 2). Statins inhibit the conversion of HMG-CoA to the cholesterol precursor mevalonate, which is the rate-limiting step in cholesterol biosynthesis. Mevalonate is also a precursor for other isoprenoids, including farnesyl pyrophosphate and geranylgeranyl pyrophosphate, which are required for post-translational modification of proteins involved in cell growth, including rat sarcoma (RAS) oncogene and RAS homolog (RHO) oncogene (3, 4). Consequently, statins are hypothesized to have anti-neoplastic actions, possibly through inhibition of RAS signaling. Several experimental and observational studies that were carried out to test this hypothesis reported that statin use was associated a reduced risk for various types of cancers (5-14). However, a meta-analysis and additional observational studies reported either no association or even an increased cancer incidence with statin use (15-19). Regarding gastric cancer, several studies reported a non-significant inverse relationship between statin use and gastric cancer risk (20-22), and a recent case control study reported that statins appear to reduce the risk of gastric cancer in a dose-dependent manner (23). The reasons for the varying results are unclear but may relate to methodological issues, including heterogeneous patient populations, small sample sizes, variable durations of statin exposure, and short follow-up periods (24). Beyond studies assessing the chemopreventive role of statins, the relationship between statin use and outcomes of gastric cancer is unknown.
The aim of the present study was, therefore, to investigate whether statin use could improve long-term outcomes after radical gastrectomy in patients with stage II and III gastric cancer.
Patients and Methods
We retrospectively reviewed the medical records of a consecutive series of 928 patients with stage II or III gastric cancer [according to the American Joint Committee on Cancer staging manual, sixth edition (25)] who underwent curative resection at the Yonsei University, Seoul, Korea from January 2006 to December 2009. The standard operation for gastric cancer was total or subtotal gastrectomy with D2 lymph node dissection in accordance with the rules of the Japanese Research Society for Gastric Cancer (26). Curative resection was defined as complete macroscopic and microscopic removal of the tumor. We found that 65 of the 928 patients had been prescribed statins for hypercholesterolemia of our hospital during the period of January 2006 to August 2012. For each case, we identified two or three control patients who were matched by sex, age (±5 years), tumor stage, and time of operation (±3 months). Our study population thus consisted of 65 statin users (cases) and 176 non-users (controls).
Baseline characteristics classified by statin treatment.
Clinical characteristics, underlying disease, adjuvant chemotherapy, recurrence-free survival (RFS), and overall survival (OS) were analyzed retrospectively. We calculated the Charlson comorbidity index (CCI) (27) to predict the effects of underlying conditions on the outcomes of cancer treatment. RFS was defined as the interval from the operation date to the date of documented recurrence, death from any cause, or last follow-up. Recurrence was confirmed by imaging studies including ultrasonography, computed tomography, gastrointestinal tract radiography, and endoscopy. OS was defined as the interval from the operation date to the date of death from any cause or last follow-up. Clinicopathological factors included histology, Lauren classification, pathological tumor stage, lymph node stage, tumor size, and lymphovascular and perineural invasion. Data regarding all statin prescriptions were extracted from the electronic medical record database. Statin users were defined as patients who received at least one prescription for any statin (atorvastatin, rosuvastatin, simvastatin, pitavastatin, fluvastatin, or pravastatin) at any time from January 2006 to August 2012. Patient features and clinical characteristics were analyzed by two-sample t-test for continuous variables and χ2 statistics for categorical variables. RFS and OS were estimated by using Kaplan–Meier curves and compared using the log-rank test. A stratified Cox regression model was used to estimate the simultaneous effect of potential confounders. Analyses were performed using SAS statistical software (version 9.2, SAS Institute, Cary, NC, USA); a value of p<0.05 was considered significant.
Relationship between statin therapy and long-term outcomes after curative gastrectomy.
Results
Baseline characteristics. Baseline characteristics of the 65 statin users (cases) and 176 matched controls are shown in Table I. Significant differences were observed with respect to tumor differentiation (p=0.014), Lauren classification (p=0.006), and CCI score (p<0.01). Other patient and tumor characteristics were similar between cases and controls.
Post-gastrectomy survival outcomes according to statin therapy. Kaplan–Meier curves of recurrence-free survival (A) and overall survival (B) of statin users and non-users after a median follow-up of 3.5 years. Recurrence-free survival (C) and overall survival (D) of users of statin for less than six months and for more than 6 months after a median follow-up of 3.5 years.
Relationship between statin use and cancer recurrence and death. After a median follow-up of 3.5 years, 23 (35.6%) statin users and 53 (30.1%) controls experienced recurrence of gastric cancer; however, this difference was not significant (p=0.435). In addition, cases and controls had similar RFS (log-rank, p=0.087; Figure 1A) and OS (p=0.221; Figure 1B). Similar outcomes were observed after adjusting for other predictors of cancer recurrence (Table II). Compared with controls, statin users had similar RFS [adjusted hazard ratio (aHR) for cancer recurrence or death=1.518, 95% confidence interval (CI)=0.761-3.03; p=0.236]. Furthermore, statin use was not associated with a significant improvement in OS (aHR for death=1.169, 95% CI=0.57-2.397; p=0.669).
Baseline characteristics of patients according to duration of statin use. We divided the statin users into two groups according to the duration of statin use (<6 months or >6 months). Table III compares the demographic characteristics and selected medical conditions of controls with the two groups of statin users. Patients who used statins for more than six months had the highest rate of intestinal-type gastric cancer (Lauren classification; p=0.039) and the highest CCI score (p<0.001). Other potentially prognostic patient and tumor characteristics were similar among the three patient groups.
Baseline characteristics of subgroups according to duration of statin therapy.
Relationship between long-term statin use and gastric cancer recurrence. Short-term statin use (<6 months) was associated with the highest recurrence rate (51.4%), followed by non-use (30.1%) and statin use >6 months (14.3%; p=0.005). The patient groups also differed according to RFS (log-rank p<0.001; Figure 1C) and OS (log-rank p<0.001; Figure 1D). Results of Cox regression analysis (Table IV) also showed that statin use for less than six months was associated with a more than three-fold increased recurrence risk compared with no use (p<0.008). Whilst long-term statin use was associated with better long-term survival (p=0.006). Taken together, our results show that statin use for longer than six months was associated with increased OS, and statin use for less than six months was associated with lower RFS and OS rates.
The association between duration of statin therapy and long-term outcomes after curative gastrectomy.
Discussion
In the present study, we found that statin use after curative resection of stage II or III gastric cancer did not significantly reduce the risk of recurrence or death in the overall study population. However, subgroup analysis comparing controls with two groups of statin users (duration of statin use <6 months or >6 months) firstly revealed significantly different recurrence rates and survival outcomes. Specifically, long-term statin use (>6 months) was associated with increased OS (aHR for death=0.168, 95% CI=0.032-0.881; p=0.006), suggesting that long-term statin use has a survival benefit for patients who undergo curative gastrectomy. This result is consistent with the assumed mechanism of action for statins, which suggests that statin use may reduce the incidence of gastric cancer. Our study provides some evidence for the antitumor activity of statins with regard to the postoperative period after curative resection of gastric cancer. The details of this effect have not been fully-elucidated.
Although statins are approved only for the treatment of lipid disorders, there is mounting evidence for their having an antitumor effect. Statins inhibit cholesterol biosynthesis by inhibiting HMG-CoA. This depletes mevalonate, a precursor of cholesterol, leading to a reduction in activity of the RAS protein, which is involved in cell differentiation and proliferation (4). A significant proportion of tumors in the gastrointestinal tract carry mutations in the Kirsten rat sarcoma viral oncogene (KRAS). Statins may have the potential to inhibit RAS activation, thereby restoring normal cell growth. There is also evidence to suggest that statins trigger apoptosis of gastrointestinal cancer cells (28), inhibit angiogenesis (29), or target mechanisms involved in the metastatic spread of cancers (30). Several observational studies have evaluated the relationship between statin use and gastric cancer. In a case–control study that used information from the drug dispensing records of residents in eight cities in the Netherlands (20), incidental diagnoses of cancer were matched with at least four controls, and exposure to statins before the index date was compared. Statin use did not appear to reduce the risk of gastric-cancer (adjusted risk estimate: 0.88; 95% CI=0.36-2.15). In another case–control study, conducted using the General Practice Research Database, all first-time recorded diagnoses of cancer were matched to at least five controls, and the effect of statin exposure was examined (22). The prevalence of gastric cancer was 3.2% for current statin users and 6.4% for non-users, but the reduction in relative risk of gastric cancer was not significant (0.4, 95% CI=0.1-1.3). In another epidemiological study using social insurance data from Finland, the incidence of gastric cancer was lower in users of any statin, but again this difference was not significant (21). However, a recent population-based case–control study (337 gastric cancer cases and 1,348 controls) reported that statins may reduce the risk of gastric cancer (23). Ever-use of any statin was associated with a significant decrease in gastric cancer risk (odds ratio=0.68, 95% CI=0.49-0.95), with a trend toward decreasing gastric cancer risk with increasing cumulative dose (χ2 for linear trend=7.42, p=0.006). This was the first study to show a chemopreventive effect of statins in gastric cancer.
Although several studies have demonstrated a chemopreventive role for statins, no study had reported a relationship between statin use and long-term outcomes after curative gastrectomy. One recent study reported a relationship between statin use and outcomes of colon cancer (31) but found that statin use during and after adjuvant chemotherapy was not associated with improved outcomes in patients with stage III colon cancer. However, our study is the first as far as we are aware to show a positive relationship between statin use and outcomes in patients with stomach cancer after gastrectomy. One of the strengths of our study is that statin users and non-users were matched by sex, age, tumor stage, and operation month. In addition, the analysis was adjusted for several factors that could influence outcomes. Although several studies have investigated the relationship between recurrence patterns, time-to-recurrence, and clinicopathological features, only pathological stage was found to predict outcomes (32-34). In this study, we assessed a number of factors, including tumor differentiation, Lauren classification, tumor location, lymphovascular invasion, perineural invasion, and tumor size, in addition to depth of invasion and lymph node metastasis. We also used the CCI, a measure of patient comorbidity (27, 35). A recent review reported that no gold standard exists for measuring comorbidity in the context of cancer (27); however, no data were shown regarding gastric cancer. Nevertheless, the CCI is the most extensively studied comorbidity index and has been used to assess patients with breast, lung, colorectal, urological, cervical, head and neck, and hematological cancers (36). We found that statin users had higher CCI (0.91±1.01) than statin non-users (0.20±0.49). In the subgroup analysis, CCI were higher for patients who used statins for more than six months (1.00±1.05) than for those who used statins for less than 6 months (0.81±1.00) or not at all (0.2±0.49). Because the CCI may influence outcome after gastrectomy, we adjusted for these differences in the RFS and OS analyses. The most interesting result of our study was that statin use of more than six months was associated with a survival benefit compared with non-use and short-term statin use. This finding suggests that long-term statin use has chemopreventive effects. In our study, patients who used statins for less than six months had increased tumor recurrence and mortality. Although we adjusted for underlying conditions by using the CCI, confounders such as metabolic syndrome or other risk factors that could influence the outcomes may have affected our results.
Several limitations of the present study should be noted. Firstly, although we adjusted for several factors potentially associated with outcomes after curative gastrectomy, our analysis did not include other potentially confounding variables. Secondly, various statins were prescribed to the patients in our study, but we were unable to analyze the risks associated with each statin because of the relatively small number of individuals taking each drug. It has been hypothesized that only lipophilic statins can inhibit tumor development, and hydrophilic statins would be expected to promote tumor development (37). Finally, the mean follow-up was 3.5 years, which may not provide with sufficient time to evaluate the effect of statin use after gastrectomy. In a study evaluating the relationship between statin use and colon cancer recurrence and survival (median follow-up=6.5 years), duration of statin use (≤2 years, 3-5 years, or ≥6 years) was not associated with outcomes (31).
In conclusion, statin use did not improve RFS and OS after curative resection of stage II or III gastric cancer in our overall patient population. However, subgroup analysis revealed that statin use of more than six months was associated with increased OS, providing some evidence for the antitumor effect of statins after curative gastrectomy in patients with gastric cancer. Experimental studies are needed to determine the mechanisms underlying this antitumor effect, and prospective randomized studies are necessary to confirm the benefits of statin use after curative gastrectomy in patients with gastric cancer.
- Received November 22, 2013.
- Revision received December 9, 2013.
- Accepted December 10, 2013.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
