The impact of preoperative use of calcium channel blockers on outcomes of patients undergoing esophagectomy: a propensity score-matched cohort study

In this study, we compared the effects of using preoperative CCBs on perioperative outcomes, cancer recurrence and overall survival in patients undergoing esophagectomy. A retrospective cohort study was performed on patients who underwent esophagectomy at the Sun Yat-Sen University Cancer Center (n=2415, 2009-2013). Univariate and multivariate logistic regression analyses were performed to assess the perioperative outcomes, while recurrence-free survival and overall survival were assessed using Kaplan-Meier survival estimates and compared using a multivariate Cox proportional hazards regression, adjusted with propensity scores. and perioperative outcome variables. The results are reported as percentages and odds ratios (ORs) with 95% confidence intervals (CIs). The univariable association between recurrence-free survival or overall survival and preoperative CCB use was assessed with Kaplan-Meier survival estimates, and the groups were compared with log-rank tests and with univariable Cox proportional hazards regression. Multivariate Cox proportional hazards regression analysis adjusting for propensity score was performed to assess the associations between preoperative CCB use and recurrence-free survival and overall survival. The results are reported as percentages and hazards ratios (HRs) with 95% CIs. All reported P values were 2-sided, and P < 0.05 was considered statistically significant.

between the CCB and non-CCB groups (HR: 1.425, 95%CI: 0.989-2.053, P=0.054). A history of chronic lung disease, hypertension, and preoperative use of beta-blockers affected the recurrence-free survival of patients after esophagectomy.

CONCLUSION
Preoperative CCBs use was associated with shorter overall survival but did not affect recurrence-free survival or the postoperative complications for patients undergoing esophagectomy.

Background
Hypertension is a highly prevalent disease that affects approximately 30-45% of the general population (1) and antihypertensive medications are among the most commonly prescribed medications. The beneficial therapeutic effect of antihypertensive medications on controlling blood pressure has been well established in previous studies (2). Moreover, the association between antihypertensive medications and the risk of cancer, such as breast cancer (3), renal cell carcinoma (4), and prostate cancer (5), has been a concern for nearly 50 years (6). Recently, the potential of antihypertensive medications to affect cancer progression has received increasing interest (7). Some epidemiological studies have shown that beta-blockers might be associated with longer survival and reduced mortality in patients with non-small-cell lung cancer, colorectal cancer, and prostate cancer (8)(9)(10). Esophageal cancer ranked fourth in cancer mortality in China, with an estimated 375,000 deaths from esophageal cancer in 2015 (11). The prevalence of cardiovascular diseases is consistently higher in older adults (12), and esophageal cancer also affects elderly people more often; therefore, antihypertensive medications is commonly used to treat these comorbidities among esophageal cancer patients (13).
However, study on esophageal cancer had never been considered. Thus, determining whether antihypertensive medications affect the prognoses of esophageal cancer patients is highly desirable.
Calcium channel blockers (CCBs) are one of the most commonly prescribed cardiovascular medications for patients with hypertension and coronary heart disease, accounting for more than 30% of antihypertensive medications used (14). Patients will continue to take such medications during oncology treatment, but the long-term safety of these medications has been questioned. Studies on the association between CCBs usage and cancer prognosis or risk remain controversial. CCBs may inhibit apoptosis to promote tumor cell proliferation, which is a theory that has been supported by several in vitro studies and animal studies (15)(16)(17). However, Jan et al. found that CCBs enhanced apoptosis of prostate cancer cells and might have a protective effect on prostate cancer (5). Some studies showed that CCB usage significantly increased cancer mortality (18,19), while other studies did not find this relationship (14). To obtain further data on the possible influence of CCBs usage on cancer outcomes, we conducted an observational study with 2415 patients who underwent esophagectomy from 2009 to 2013 with data in the Sun Yet-Sen University Cancer Center Database.

Methods
We performed a retrospective cohort study involving patients with esophageal cancer (n = 2415) who underwent esophagectomy at the Sun Yat-Sen University Cancer Center from November 2009 to July 2013. The follow-up period ended in May 2017. The study was reviewed and approved by the Sun Yat-Sen University Cancer Center review board. The study protocol was granted exemption from written informed consent (not human subjects research) by the Sun Yat-Sen University Cancer Center review board.
Only the records of patients with esophageal cancer where esophagectomy was indicated were included in the analysis. Patients were excluded if the surgery was performed in other hospitals, if the operation was palliative, if the final diagnosis was a benign tumor, if tumor metastases were present, if neoadjuvant chemoradiation was provided, or if other systemic tumors existed. Of the original 2415 patients, 1272 patients met the inclusion criteria and were divided into 2 groups: used (CCB group, n = 162) or did not use (n = 1110) preoperative CCB (non-CCB group) (Fig. 1).
We collected follow-up data from the patients' medical records, hospital surgery database or pathology database (or both); after a letter of introduction was sent to the patients, telephone contact was established with the patients or their families. Independent investigators prospectively collected the data on each patient. We obtained the demographics, oncologic characteristics, and operative characteristics of all patients, including age, sex, preoperative medications, tumor site, pathological stage, clinical stage, transfusion, postoperative complications including cardio-cerebral events and infection from thoracotomy, and 30-day all-cause mortality.
The major outcomes included recurrence-free survival and overall survival. Recurrence was defined as radiologic evidence of local recurrence or distant metastatic disease.
Recurrence-free survival was calculated from the date of the operation to the date of recurrence. For patients without a record of recurrence, recurrence-free survival was defined as the time between the date of the operation and the date of the last follow-up or the date of death. Overall survival was calculated from the date of the operation to the date of death from any cause.
Because this was an observational study, a propensity score-adjusted analysis was performed to control for selection bias as a result of nonrandom assignment to the 2 groups. Patients in the CCB and non-CCB groups were matched using the propensity-score matching method (20), which was carried out using R software version 2.12.1. The propensity score was calculated with consideration for all baseline variables except for history of hypertension, as shown in Table 3. We used the forward procedure, and variables were included up to a limit of a monotonized p-to-enter value of < 0.2. Then, we applied 1:1 nearest-neighbor matching without replacement to ensure that conditional bias was minimized. A caliper width of 0.001 resulted in the best trade-off between homogeneity and minimal loss of sample size. Small absolute values in standardized differences (< 25%) were assumed to support the assumption of balance between the treatment groups (20).

Results
In total, 1272 patients who underwent esophagectomy were divided into two groups: those who used (n = 162, 12.7%) or did not use (n = 1110, 87.3%) preoperative CCBs (non-CCB users). The distribution of classes of preoperatively used CCBs is that nifedipine was the most commonly used active ingredient (61.7%), followed by amlodipine (19.1%), nimodipine (9.9%), felodipine (4.9%) and nitrendipine (4.3%). The comparisons of patient demographic and clinical data between the two groups are illustrated in Table 1. Most characteristics, oncologic and therapeutic variables were not significantly different between the two groups. However, the patients using CCBs preoperatively were older and had a more frequent history of chronic lung disease and hypertension than those who did not use CCBs preoperatively. Propensity-score matching analysis successfully created 155 pairs of patients. The baseline characteristics of these patients are shown in Table 2. Apart from a history of hypertension, all other baseline characteristics were balanced between the two groups, and all of the standardized differences were < 25%.   Overall survival: The overall mortality rate before (  Recurrence-Free Survival: The recurrence rate of the CCB group was higher than that of the non-CCB group (43.8% vs 35.3%, P=0.035) before matching. However, the difference in recurrence rate between the two groups was not statistically significant after matching (43.2% vs 32.9%, P = 0.061). There was a significant difference between the CCB and non-CCB groups in recurrence-free survival (P =0.025, log-rank test) (Fig. 2C), with an unadjusted estimated HR of 1.330 (95% CI, 1.032-1.713) in the entire cohort. After propensity matching, the use of preoperative CCBs was not associated with a significant difference in recurrence-free survival (P = 0.054, log-rank test) (Fig. 2D)

Discussion
Overall survival From this observational study of 2415 patients with esophageal cancer, we found that the preoperative use of CCBs was significantly associated with a poor outcome in overall survival. Compared with the non-CCB group, the relative risk for overall mortality was were more likely to die from cancer than those prescribed beta-blockers (adjusted HR: 1.406, 95% CI: 1.334-1.482, P < 0.001) (19).
In the past decade, CCBs have been suggested to affect the survival outcomes in cancer patients by interfering with biological processes such as apoptosis or cell differentiation (15). These results have been supported by several pieces of experimental evidence. For example, nifedipine was found to promote the proliferation and migration of breast cancer cells (17 A history of smoking cigarettes and a highly advanced tumor stage are the other two risk factors that affect overall survival in this study and are well known for a poor prognosis (27). In addition, in our study, the type of operation to dissect three lymph nodes and postoperative chemotherapy could improve overall survival in patients who underwent esophagectomy, which is consistent with the meta-analysis conducted by Pasquali et al. (28).

Recurrence-free Survival
Although no association between the use of CCBs and recurrence-free survival of patients who underwent esophagectomy was observed in our study, surprisingly, beta-blockers played a role in prolonging the recurrence-free survival of esophageal cancer patients. A cohort study conducted in Norway with 3561 prostate cancer patients with high-risk or metastatic disease showed that beta-blocker use was associated with reduced prostate cancer mortality and reduced prostate cancer-specific mortality (29). Another large population-based cohort study of patients with colorectal cancer in Germany indicated that beta-blocker use was associated with prolonged overall survival and colorectal cancer-specific survival in stage IV patients (10). Overall, these findings are in accordance with ours (7,10,29). Our study showed that a history of chronic lung disease was associated with poor recurrence-free survival, which could be because patients with a history of chronic lung disease were more likely to be complicated with other diseases and poor physical conditions (30).
In this study, hypertension did not affect the overall survival of patients with esophageal cancer but was an independent risk factor for recurrence-free survival when propensityscore matching was applied. Several studies have reported that regardless of the use of antihypertensive drugs, hypertension was associated with a modestly increased risk of cancer and mortality from cancer (31,32). Dyer et al. suggested that the potential of hypertension was a risk for cancer in patients with a blood pressure increase (i.e., relative risk of 2.2 with blood pressure > 148 mmHg and 3.05 with blood pressure > 160 mmHg) (33). However, the reports on the association between hypertension and cancer had inconsistencies. A cohort study found no association between blood pressure and subsequent cancer incidence or mortality (34). The organs most frequently cited regarding the association between hypertension and cancer incidence or mortality were the kidney (34), colon (33), breast (35) and endometrium (36), but the esophagus was seldomly mentioned in previous studies.

Postoperative Outcomes
Esophagectomy has historically been the primary curative treatment for esophageal cancer (37). The incidence of postoperative complications varies from 28-36% among different health centers (38). In this study, no significant association was found between CCBs usage and postoperative outcomes. However, as other studies that clinical stage III at diagnosis and operation type were found to be related to poor overall postoperative complications (39,40).

Limitation
This study also has some limitations. First, this study was constrained by being retrospective in nature. Second, as a single-center observational study, avoiding selection bias is difficult. However, applying propensity-score matching was the best solution in this study and could balance covariates between the two groups to make the analyses comparable to a quasi-randomized experiment. Third, since the majority of the CCBs users were on nifedipine, the results should be generalized to other CCBs with caution.
Conclusion that patients with esophageal cancer who received CCBs preoperatively have shortened overall survival after esophagectomy compared to those who did not receive CCBs.
However, there is no association between CCBs usage and recurrence-free survival or the incidence of postoperative complications for patients undergoing esophagectomy.

Ethics approval and consent to participate
The study was reviewed and approved by the Sun Yat-Sen University Cancer Center review board. The study protocol was granted exemption from written informed consent (not human subjects research) by the Sun Yat-Sen University Cancer Center review board.