Association between preoperative statin use and survival after pylorus-preserving pancreaticoduodenectomy for pancreatic or periampullary cancer: a propensity score-matched cohort study

Introduction

As the cornerstone of dyslipidemia management, statins [3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors] play a critical role in the primary and secondary prevention of cardiovascular diseases (CVD) (1,2). Beyond their lipid-lowering and cardioprotective effects, statins exhibit various additional biological actions, including anti-inflammatory, immunomodulatory, and endothelial-stabilizing effects (3,4). These pleiotropic properties may influence tumor biology and contribute to improved postoperative recovery (5). Statins can induce apoptosis, inhibit angiogenesis, and suppress tumor growth and metastasis in several malignancies, including pancreatic and lung cancers (6,7). In addition, statins may modulate the tumor microenvironment and inhibit major oncogenic signaling cascades, thereby exerting potential anticancer effects.

Despite these biological mechanisms, clinical evidence regarding the association between perioperative statin use and cancer outcomes remains inconsistent. Several observational studies have suggested lower cancer-related and all-cause mortality among statin users (8), whereas randomized controlled trials and meta-analyses have failed to confirm such benefits (9-11). Moreover, most prior studies have been confined to patients with advanced or metastatic disease.

Pancreatic cancer is a key solid tumor of interest because of its aggressive biological behavior and persistently poor survival even after curative resection (12). Pylorus-preserving pancreaticoduodenectomy (PPPD) is the standard surgical procedure for resectable periampullary or pancreatic malignancies, yet postoperative morbidity and mortality remain considerable (13,14). Given the anti-inflammatory and endothelial-stabilizing effects of statins, their perioperative use may potentially influence both short- and long-term survival after surgery. However, evidence on the impact of preoperative statin therapy on postoperative outcomes in resectable pancreatic or periampullary cancer remains limited.

The evidence regarding statin therapy and pancreatic cancer outcomes remains mixed. Although some observational data have suggested a lower incidence of pancreatic ductal adenocarcinoma among statin users, evidence for a survival benefit is inconsistent, and recent large cohort studies have reported no association between statin use and survival in pancreatic cancer (15,16). Nevertheless, studies evaluating the association between preoperative statin use and postoperative outcomes after PPPD are scarce, and none have comprehensively analyzed both early and long-term survival in this setting. Therefore, we aimed to investigate the association between preoperative statin use and postoperative outcomes in patients undergoing PPPD for periampullary or pancreatic cancer, comparing short-term (1-year) and long-term (5-year and overall) survival and disease-free survival (DFS) before and after propensity score (PS) matching. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0469/rc).

Methods

Patients

This retrospective cohort study was approved by the institutional review board of Asan Medical Center (protocol No. 2024-1090), and the requirement for written informed consent was waived owing to the retrospective design. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. From January 2012 to September 2016, we identified adult patients at our center who received PPPD as a surgical treatment for pancreatic or periampullary malignancies. Patients were identified through a comprehensive search of the institutional electronic medical record (EMR) system. Patients who underwent PPPD for pathologically confirmed periampullary or pancreatic malignancy during the study period were included. Patients were excluded if they met any of the following conditions: aged <18 years or >80 years; presence of another active malignant disease; history of emergency surgery; or incomplete or missing demographic, laboratory, or medication data. The final study population, established after these exclusions, was used for subsequent PS matching and outcome analyses (Figure 1).

Figure 1 Study flowchart. PPPD, pylorus-preserving pancreaticoduodenectomy.

Data collection and variable definitions

Clinical information was extracted retrospectively from the institutional EMR system. The collected dataset encompassed patients’ demographic, clinical, laboratory, intraoperative, and statin-related characteristics.

Demographic and clinical variables included age, sex, height, weight, and body mass index (BMI), as well as comorbid conditions such as diabetes mellitus (DM), hypertension (HTN), and CVD. The use of antihypertensive and lipid-lowering medications, including angiotensin-converting enzyme (ACE) inhibitors, angiotensin II receptor blockers (ARBs), calcium channel blockers, and diuretics was also recorded. The American Society of Anesthesiologists (ASA) physical status classification was used to assess preoperative physical condition. Data on neoadjuvant and adjuvant chemotherapy or radiotherapy were obtained from oncology records. Cancer-related variables comprised tumor location (pancreatic head, distal bile duct, ampulla of Vater, or duodenum) and tumor-node-metastasis (TNM) stage. Pathologic prognostic factors were also collected, including resection margin status (R0/R1), lymph-node ratio (the ratio of metastatic to examined lymph nodes), tumor differentiation grade (well/moderate vs. poorly/undifferentiated), and the preoperative serum carbohydrate antigen 19-9 (CA19-9) level (dichotomized at 37 U/mL). All tumors were staged according to the eighth edition of the American Joint Committee on Cancer TNM classification system for pancreatic malignancies (17).

Laboratory variables included preoperative white blood cell count, hemoglobin concentration, platelet count, prothrombin time and international normalized ratio, serum sodium, potassium, aspartate aminotransferase, alanine aminotransferase, albumin, and creatinine levels. Intraoperative variables comprised the type of surgical approach (open or laparoscopic PPPD), vascular resection, total operative time, the lowest recorded systolic, diastolic, and mean blood pressures, total volumes of crystalloids and synthetic colloids administered, intraoperative red blood cell (RBC) transfusion, urine output, and inotropic agent use.

Statin-related variables included the presence or absence of statin therapy, intensity of statin treatment (low, moderate, or high), the specific type of statin prescribed (atorvastatin, pitavastatin, pravastatin, rosuvastatin, or simvastatin), and the lipophilicity class of each agent (lipophilic or hydrophilic) based on physicochemical properties. Combination therapy with ezetimibe was also documented. Preoperative statin use was defined as a documented prescription of any statin prior to surgery, confirmed through electronic medical and inpatient prescription records. Information on medication use was obtained from the institutional EMR system, which includes outpatient and inpatient prescriptions as well as home medications recorded at the time of hospital admission.

Postoperative outcomes were determined from electronic medical and insurance records. Mortality was assessed at multiple time points (30-day, 90-day, 180-day, 1-year, 5-year, and overall), calculated from the date of surgery to death or last follow-up. DFS was defined as the time from surgery to the first occurrence of cancer recurrence or all-cause death, whichever occurred first (18). Patients without an event were censored at the date of last follow-up. The cause of death was classified as cancer-related, surgical/perioperative, or cardiovascular. Cause-of-death classification was restricted to in-hospital deaths, for which medical records were available; the cause of out-of-hospital deaths was not classified.

Primary and secondary endpoints

The primary endpoint of this study was the comparison of postoperative 5-year mortality between patients with and without preoperative statin use before and after PS matching. The association between statin exposure and this primary outcome was further evaluated in multivariable analyses. Secondary endpoints included 1-year mortality, overall mortality during the entire follow-up period, and DFS at 1 year, 5 years, and overall. Associations between preoperative statin use and each secondary endpoint were also assessed before and after matching.

Statistical analysis

Continuous variables were summarized as means ± standard deviations or as medians with interquartile ranges, depending on the distribution of data, and categorical variables are expressed as counts and percentages. Groups were compared using the Student’s t-test or Mann-Whitney U test for continuous variables and the chi-square or Fisher’s exact test for categorical variables, as appropriate. To reduce potential confounding, PS matching was performed using nearest-neighbor matching without replacement at a 1:2 ratio, targeting the average treatment effect on the treated with a caliper width of 0.2 of the standard deviation of the logit of the PS. The PS model included age, sex, BMI, DM, HTN, CVD, ASA physical status, use of ACE inhibitors, ARBs, calcium channel blockers, and diuretics, tumor location, TNM stage, laparoscopic surgery, vascular resection, white blood cell count, hemoglobin, platelet count, international normalized ratio, sodium, potassium, aspartate and alanine aminotransferase, albumin, creatinine, resection margin status, CA19-9, tumor grade, and lymph-node ratio. Postoperative and adjuvant variables were not included because they follow surgery and could act as mediators rather than confounders. To evaluate the effectiveness of the matching process, absolute standardized mean differences (SMD) were calculated for all covariates. An SMD below 0.10 was adopted as the threshold for achieving a negligible imbalance between the groups.

To evaluate survival outcomes, we estimated the risk ratio (RR) and risk difference (RD) with corresponding 95% confidence intervals (CIs) for DFS and mortality at 30-day, 90-day, 180-day, 1-year, 5-year, and overall intervals. These metrics were assessed both in the pre-matching and post-matching cohorts, with chi-square, Fisher’s exact, or Wald tests employed for intergroup comparisons based on data distribution. The Kaplan-Meier curve was used to analyze 1-year, 5-year, and overall survival as well as DFS, and differences between statin users and nonusers were assessed using the log-rank test. Cox proportional hazards models were applied to estimate hazard ratios (HR) and 95% CIs for associations between preoperative statin use and postoperative outcomes, including mortality and DFS at 1 year, 5 years, and over the entire follow-up period. For each outcome, a single pre-specified multivariable model with a fixed covariate set [age, sex, laparoscopic surgery, vascular resection, tumor location, TNM stage, DM, HTN, CVD, ASA physical status, hemoglobin, albumin, resection margin status, CA19-9, tumor grade, and lymph-node ratio (per 0.1 increment)] was applied, rather than variable selection based on univariable significance, to avoid model-selection bias. The fixed covariate set was used for all multivariable Cox models. Crude models included statin use only, and PS-matched models estimated the association within the matched cohort. The proportional hazards assumption was examined using scaled Schoenfeld residuals (global test). As a sensitivity analysis to address tumor heterogeneity, we repeated the matching and Cox analyses in the subgroup with pancreatic ductal adenocarcinoma. Because a few covariates retained an absolute SMD above 0.10 after matching, we performed an additional sensitivity analysis in which the Cox models were refitted within the matched cohort with further adjustment for these residually imbalanced covariates (CVD, resection margin status, and ASA physical status class), yielding additionally adjusted estimates.

All analyses were conducted using the R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria) with validated routines for matching, balance diagnostics, survival analysis, and visualization. Statistical significance was predefined as a two-tailed P value of less than 0.05.

Results

Study cohort and baseline characteristics

Among 771 patients initially screened, 53 were excluded according to predefined criteria, resulting in a final cohort of 718 patients (Figure 1). Of these, 130 (18.1%) received preoperative statin therapy, whereas 588 (81.9%) did not. After PS matching, 116 statin users and 193 nonusers were included in the matched cohort.

Table 1 summarizes the baseline characteristics, perioperative findings, and statin-related variables before and after PS matching. Before matching, the statin group was significantly older and had a higher prevalence of comorbidities, including diabetes and HTN, as well as CVD (27.7% vs. 3.7%, P<0.001), along with more frequent use of antihypertensive medications (all P<0.001). Pathologic factors (margin status, lymph-node ratio, tumor grade, and CA19-9) were comparable between groups before matching. Following PS matching, these baseline imbalances were successfully mitigated, with absolute SMDs for most covariates falling below 0.1; CVD (SMD 0.21), resection margin status (SMD 0.12), and ASA physical status class (SMD 0.12) remained modestly imbalanced. Statin therapy was predominantly moderate-intensity, with atorvastatin being the most common agent, while approximately 5% of users received combination therapy with ezetimibe.

Primary endpoints

Table 2 compares postoperative 5-year mortality between the statin and non-statin groups before and after PS matching. Before matching, 5-year mortality rates were 56.5% in the non-statin group and 64.6% in the statin group (P=0.09). After matching, 5-year mortality did not differ significantly (non-statin 56.5% vs. statin 63.8%; RR, 1.13; 95% CI, 0.94–1.36; P=0.21). Table 3 shows the univariable and multivariable Cox regression analyses for factors associated with 5-year mortality in the overall cohort. In the multivariable analysis, preoperative statin use was not independently significantly associated with 5-year mortality (adjusted HR, 0.96; 95% CI, 0.72–1.27; P=0.77). Independent predictors of higher 5-year mortality included male sex (HR, 1.33; 95% CI, 1.07–1.66; P=0.01), higher TNM stage (stage II vs. I: HR, 1.51; 95% CI, 1.08–2.10; P=0.02), vascular resection (HR, 1.43; 95% CI, 1.12–1.84; P=0.004), R1 resection margin (HR, 1.36; 95% CI, 1.09–1.70; P=0.007), elevated CA19-9 (HR, 1.33; 95% CI, 1.07–1.64; P=0.009), poorly/undifferentiated grade (HR, 1.35; 95% CI, 1.01–1.80; P=0.04), and higher lymph-node ratio (HR per 0.1 increment, 1.24; 95% CI, 1.15–1.34; P<0.001).

Secondary endpoints

Table 2 summarizes postoperative outcomes according to preoperative statin use, including both mortality and DFS at 1 year, 5 years, and overall follow-up. Before matching, 1-year (15.8% vs. 10.8%, P=0.14) and overall mortality rates (65.1% vs. 69.2%, P=0.37) did not differ significantly between the non-statin and statin groups, and similar nonsignificant differences were observed for DFS outcomes. After matching, 1-year mortality was numerically lower in statin users (11.2% vs. 18.7%) but the difference was not statistically significant (RR, 0.59; 95% CI, 0.34–1.10; P=0.08). No significant differences were observed in 5-year mortality, overall mortality, or any DFS endpoint. Overall mortality was 67.2% in statin users and 63.7% in nonusers (RR, 1.06; 95% CI, 0.89–1.24; P=0.53). The cumulative incidence of DFS events likewise did not differ (statin vs. non-statin: 32.8% vs. 39.4% at 1 year, 71.6% vs. 64.2% at 5 years, and 75.0% vs. 69.4% overall). In Cox models (Table 4), the 1-year association was non-significant in both the crude (HR, 0.67; 95% CI, 0.38–1.18; P=0.16) and PS-matched analyses (HR, 0.59; 95% CI, 0.31–1.13; P=0.11), reaching significance only in the multivariable model (HR, 0.45; 95% CI, 0.24–0.86; P=0.02). No significant associations were found for 5-year mortality (multivariable HR, 0.96; PS-matched HR, 1.09), overall mortality (multivariable HR, 0.92; PS-matched HR, 1.03), or DFS at any time point.

Early postoperative mortality was very low and similar between groups [30-day, 1/193 (0.5%) vs. 0/116 (0.0%); 90-day, 4/193 (2.1%) vs. 2/116 (1.7%); 180-day, 11/193 (5.7%) vs. 5/116 (4.3%)]. Among the 54 in-hospital deaths for which the cause could be ascertained, 50 (92.6%) were cancer-related and 4 (7.4%) were surgical/perioperative; no cardiovascular deaths occurred (Table S1).

In the PDAC-restricted sensitivity analysis (n=320; matched n=109), preoperative statin use was not associated with mortality at any time point, with null DFS results (Table S2), consistent with the main analysis. In the covariate-adjusted sensitivity analysis that accounted for the residually imbalanced covariates (CVD, resection margin status, and ASA class), the estimates were materially unchanged and remained nonsignificant for every endpoint; for 1-year mortality, the PS-matched HR shifted modestly from 0.59 to 0.51 (95% CI, 0.26–1.01; P=0.053) but did not reach statistical significance, whereas 5-year mortality, overall mortality, and all DFS endpoints remained close to unity (Table S3).

Figure 2 shows Kaplan-Meier survival curves for 1-year, 5-year, and overall survival in the crude (full) and PS-matched cohorts. In the unadjusted analyses, no statistically significant differences in survival were observed between the statin and non-statin groups (log-rank P=0.16 for 1-year survival, P=0.34 for 5-year survival, and P=0.58 for overall survival). In the PS-matched cohort, survival did not differ significantly at any time point (log-rank P=0.10, 0.58, and 0.84, respectively). Figure S1 presents Kaplan-Meier curves for DFS at 1 year, 5 years, and overall follow-up. In the unadjusted analyses, DFS did not differ significantly between statin users and nonusers (log-rank P=0.16 for 1-year, P=0.72 for 5-year, and P=0.81 for overall DFS). In the PS-matched cohort, the DFS curves likewise showed no significant separation at any time point (log-rank P=0.27, 0.60, and 0.70, respectively), consistent with the nonsignificant associations observed in the Cox regression models.

Figure 2 Kaplan-Meier curves for 1-year, 5-year, and overall survival according to preoperative statin use. Panels show unadjusted (crude) survival in the full cohort for (A) 1-year survival (log-rank P=0.16), (B) 5-year survival (log-rank P=0.34), and (C) overall survival (log-rank P=0.57), and survival in the propensity score-matched cohort for (D) 1-year survival (log-rank P=0.10; PS-matched HR, 0.59; 95% CI, 0.31–1.13), (E) 5-year survival (log-rank P=0.58; PS-matched HR, 1.09; 95% CI, 0.80–1.49), and (F) overall survival (log-rank P=0.84; PS-matched HR, 1.03; 95% CI, 0.76–1.39). CI, confidence interval; HR, hazard ratio; PS, propensity score.

The proportional hazards assumption was assessed using global Schoenfeld residual tests (Table S4). It was satisfied for 1-year mortality and all DFS endpoints, but not for 5-year or overall mortality; the corresponding long-term HRs were therefore interpreted as time-averaged estimates.

Discussion

In this retrospective study, we examined the association between preoperative statin therapy and perioperative, short-term, and long-term outcomes in patients undergoing PPPD for pancreatic or periampullary cancer. Preoperative statin use was not associated with long-term survival or oncologic outcomes. Five-year mortality, overall mortality, and DFS did not differ between statin users and nonusers in unadjusted, multivariable, or PS-matched analyses. A lower 1-year mortality was observed among statin users, but this difference was not significant in the unadjusted or PS-matched comparisons and emerged only after multivariable adjustment, and it was not accompanied by any improvement in 1-year DFS. Taken together, these findings indicate that preoperative statin use was not significantly associated with short-term or long-term survival or oncologic outcomes after PPPD.

Several biological mechanisms have been proposed by which statins might transiently influence the early postoperative course. Apart from their cholesterol-lowering and cardioprotective roles, statins possess multiple pleiotropic effects—such as anti-inflammatory, antithrombotic, and endothelial-stabilizing actions—that may modulate the host’s physiological response to surgery (4,19). By dampening systemic inflammation, enhancing endothelial function, and stabilizing microcirculatory perfusion, statins could mitigate perioperative cardiovascular stress and inflammatory injury (20,21). By blocking the mevalonate pathway, statins interfere with the isoprenylation of small GTPases, including RAS and RHO. This disruption subsequently impairs critical oncogenic processes such as cellular proliferation, angiogenesis, and metastatic spread (22,23). However, because the 1-year mortality difference was not statistically significant in the unadjusted or PS-matched analyses and was not observed for DFS, these mechanisms remain speculative and cannot be confirmed.

The absence of a durable survival advantage in our cohort mirrors the results of prior randomized trials and meta-analyses, which have yielded inconsistent or null findings regarding the anticancer potential of statins across tumor types (9,10). Although several population-based and observational studies have shown improved overall or cancer-specific survival in statin users (24,25), such associations often attenuate after adjustment for confounders, including comorbidities and concurrent cardiovascular medications (26,27). In pancreatic cancer, the aggressive natural history of the disease, high recurrence rates, and frequent discontinuation of chronic medications during oncologic therapy may explain the lack of a persistent protective effect over time.

Overall, although statins may exert certain tumor-suppressive effects, their influence in biologically aggressive malignancies such as pancreatic cancer appears limited. The observed short-term association may be more consistent with enhanced perioperative physiological resilience than with sustained oncologic control. Consistently, the lack of significant differences in DFS between groups further supports the notion that statins do not provide a meaningful long-term oncologic benefit in this population. Prior investigations in surgical populations have reported similar early survival advantages associated with preoperative statin use, likely through improved vascular function, preserved microcirculatory integrity, and modulation of inflammatory pathways (28,29). Although two previous surgical series have examined the impact of statins in patients undergoing pancreatoduodenectomy, both failed to demonstrate a significant survival benefit (30,31). These results, consistent with our findings regarding long-term mortality, suggest that statins may have only a limited or modest oncologic effect after curative resection. However, both studies were limited by relatively small sample sizes, the absence of PS adjustment, and the lack of detailed evaluation of short-term outcomes.

Regarding 1-year mortality, the difference between groups was not statistically significant in the unadjusted comparison or the PS-matched analysis, reaching significance only after multivariable adjustment. This suggests that the small early survival difference may largely reflect statistical adjustment for baseline factors such as age and comorbidity rather than a genuine therapeutic effect, particularly because it did not appear in the unadjusted comparison or the PS-matched analysis and was not mirrored in DFS. Any potential physiological benefit of statins in the early postoperative period should be interpreted cautiously. Possible mechanisms such as anti-inflammatory or endothelial-stabilizing effects remain speculative and cannot be confirmed within the limits of this retrospective dataset.

This study has several limitations. First, its retrospective observational design precludes any causal inference, and residual confounding cannot be fully eliminated despite adjustment and PS matching. In particular, statin use is a marker of cardiovascular risk and overall healthcare engagement; although we matched on CVD and the recorded comorbidities, the indication for statin therapy (primary vs. secondary prevention), the severity of CVD, and unmeasured factors such as the healthy-adherer effect, socioeconomic status, health literacy, frailty, and nutritional status could not be captured. To address this residual post-matching imbalance, we refitted the Cox models within the matched cohort with additional adjustment for CVD, resection margin status, and ASA class; the estimates were essentially unchanged and remained nonsignificant, supporting the robustness of the findings against residual confounding by these measured factors. Second, the duration of preoperative statin therapy and postoperative continuation after discharge could not be determined from the available records, limiting assessment of dose–response or maintenance effects. Because postoperative continuation of statin therapy could not be reliably determined, the observed association should be interpreted as reflecting preoperative statin exposure rather than the effect of sustained perioperative statin treatment. Third, although preoperative statin exposure was identified using outpatient prescriptions, inpatient medication orders, and home medication lists recorded at admission, the EMR-based approach could not definitively distinguish between newly initiated and chronic use. Fourth, the cause of death could be ascertained only for in-hospital deaths; because our institution is a tertiary referral center, many patients are followed or die at local hospitals after discharge, and the cause of these out-of-hospital deaths could not be reliably determined, precluding a complete cause-specific mortality analysis. Fifth, the cohort combined four periampullary/pancreatic histologies with differing prognoses; although tumor location was included in matching and adjustment and a PDAC-restricted analysis was consistent, residual heterogeneity may persist and the PDAC subgroup was underpowered. Sixth, postoperative pancreatic fistula could be identified only through reoperation records, which capture severe cases and underestimate its true incidence, and it was therefore not analyzed as a discrete outcome. In addition, adjuvant therapy was recorded as received rather than completed; because completion is the prognostically meaningful measure and is itself influenced by early mortality, the possibility of reverse causation cannot be excluded. Seventh, the proportional hazards assumption was not fully met for long-term mortality, although the PS-matched analyses yielded consistent null associations, supporting the robustness of the main findings. Eighth, the relatively small sample size of statin users and the limited number of events may have reduced the statistical power to detect modest differences, particularly in long-term survival analyses. Finally, because this study was conducted at a single tertiary center, validation in larger and multi-institutional cohorts is necessary to confirm generalizability.

Conclusions

In conclusion, preoperative statin use was not associated with improved 5-year survival or overall long-term survival, which were the primary outcomes of this study. A lower 1-year mortality was observed only after multivariable adjustment and was not significant in the unadjusted or PS-matched analyses, nor was it accompanied by any DFS benefit. These findings do not support preoperative statin use to improve survival or oncologic outcomes after PPPD for pancreatic or periampullary malignancy.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0469/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0469/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0469/prf

Funding: This research was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean Government (MSIT) (grant number: RS-2026-25497186).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0469/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The institutional review board of Asan Medical Center (protocol No. 2024-1090) approved this study. The need for informed consent from individual patients was waived owing to the retrospective nature of the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Liao JK. Effects of statins on 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibition beyond low-density lipoprotein cholesterol. Am J Cardiol 2005;96:24F-33F. [Crossref] [PubMed]
2. Yusuf S, Bosch J, Dagenais G, et al. Cholesterol Lowering in Intermediate-Risk Persons without Cardiovascular Disease. N Engl J Med 2016;374:2021-31. [Crossref] [PubMed]
3. Diamantis E, Kyriakos G, Quiles-Sanchez LV, et al. The Anti-Inflammatory Effects of Statins on Coronary Artery Disease: An Updated Review of the Literature. Curr Cardiol Rev 2017;13:209-16. [Crossref] [PubMed]
4. Oesterle A, Laufs U, Liao JK. Pleiotropic Effects of Statins on the Cardiovascular System. Circ Res 2017;120:229-43. [Crossref] [PubMed]
5. Ahmadi M, Amiri S, Pecic S, et al. Pleiotropic effects of statins: A focus on cancer. Biochim Biophys Acta Mol Basis Dis 2020;1866:165968. [Crossref] [PubMed]
6. Duarte JA, de Barros ALB, Leite EA. The potential use of simvastatin for cancer treatment: A review. Biomed Pharmacother 2021;141:111858. [Crossref] [PubMed]
7. Spampanato C, De Maria S, Sarnataro M, et al. Simvastatin inhibits cancer cell growth by inducing apoptosis correlated to activation of Bax and down-regulation of BCL-2 gene expression. Int J Oncol 2012;40:935-41. [Crossref] [PubMed]
8. Ren QW, Yu SY, Teng TK, et al. Statin associated lower cancer risk and related mortality in patients with heart failure. Eur Heart J 2021;42:3049-59. [Crossref] [PubMed]
9. Hague WE, Simes J, Kirby A, et al. Long-Term Effectiveness and Safety of Pravastatin in Patients With Coronary Heart Disease: Sixteen Years of Follow-Up of the LIPID Study. Circulation 2016;133:1851-60. [Crossref] [PubMed]
10. Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016;388:2532-61. [Crossref] [PubMed]
11. Yuan M, Han S, Jia Y, et al. Statins Are Associated with Improved Survival of Patients with Gastric Cancer: A Systematic Review and Meta-Analysis. Int J Clin Pract 2022;2022:4938539. [Crossref] [PubMed]
12. Stoop TF, Javed AA, Oba A, et al. Pancreatic cancer. Lancet 2025;405:1182-202. [Crossref] [PubMed]
13. Hüttner FJ, Fitzmaurice C, Schwarzer G, et al. Pylorus-preserving pancreaticoduodenectomy (pp Whipple) versus pancreaticoduodenectomy (classic Whipple) for surgical treatment of periampullary and pancreatic carcinoma. Cochrane Database Syst Rev 2016;2:CD006053. [Crossref] [PubMed]
14. Huang L, Jansen L, Balavarca Y, et al. Stratified survival of resected and overall pancreatic cancer patients in Europe and the USA in the early twenty-first century: a large, international population-based study. BMC Med 2018;16:125. [Crossref] [PubMed]
15. Saito K, Nakai Y, Sasaki T, et al. Impact of Statin Use on Survival in Patients With Unresectable Pancreatic cancer Receiving Gemcitabine Plus Nab-Paclitaxel: A Multicenter Retrospective Study. Pancreas 2025;54:e107-13. [Crossref] [PubMed]
16. Mavros MN, Ladak F, Chan KKW, et al. The association between cardiovascular medication use and survival in patients with advanced pancreatic adenocarcinoma. Ther Adv Med Oncol 2026;18:17588359261449094. [Crossref] [PubMed]
17. Amin MB, Greene FL, Edge SB, et al. The Eighth Edition AJCC Cancer Staging Manual: Continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging. CA Cancer J Clin 2017;67:93-9.
18. Oba K, Paoletti X, Alberts S, et al. Disease-free survival as a surrogate for overall survival in adjuvant trials of gastric cancer: a meta-analysis. J Natl Cancer Inst 2013;105:1600-7. [Crossref] [PubMed]
19. German CA, Liao JK. Understanding the molecular mechanisms of statin pleiotropic effects. Arch Toxicol 2023;97:1529-45. [Crossref] [PubMed]
20. Greenwood J, Mason JC. Statins and the vascular endothelial inflammatory response. Trends Immunol 2007;28:88-98. [Crossref] [PubMed]
21. Khan S, Huda B, Bhurka F, et al. Molecular and Immunomodulatory Mechanisms of Statins in Inflammation and Cancer Therapeutics with Emphasis on the NF-κB, NLRP3 Inflammasome, and Cytokine Regulatory Axes. Int J Mol Sci 2025;26:8429. [Crossref] [PubMed]
22. Nam GH, Kwon M, Jung H, et al. Statin-mediated inhibition of RAS prenylation activates ER stress to enhance the immunogenicity of KRAS mutant cancer. J Immunother Cancer 2021;9:e002474. [Crossref] [PubMed]
23. Liu C, Chen H, Hu B, et al. New insights into the therapeutic potentials of statins in cancer. Front Pharmacol 2023;14:1188926. [Crossref] [PubMed]
24. Vell MS, Loomba R, Krishnan A, et al. Association of Statin Use With Risk of Liver Disease, Hepatocellular Carcinoma, and Liver-Related Mortality. JAMA Netw Open 2023;6:e2320222. [Crossref] [PubMed]
25. Chang WT, Lin HW, Lin SH, et al. Association of Statin Use With Cancer- and Noncancer-Associated Survival Among Patients With Breast Cancer in Asia. JAMA Netw Open 2023;6:e239515. [Crossref] [PubMed]
26. Emilsson L, García-Albéniz X, Logan RW, et al. Examining Bias in Studies of Statin Treatment and Survival in Patients With Cancer. JAMA Oncol 2018;4:63-70. [Crossref] [PubMed]
27. Abrahami D, Hudson M, Suissa S. Statins and lower mortality in rheumatic diseases: An effect of immortal time bias? Semin Arthritis Rheum 2021;51:211-8. [Crossref] [PubMed]
28. London MJ, Schwartz GG, Hur K, et al. Association of Perioperative Statin Use With Mortality and Morbidity After Major Noncardiac Surgery. JAMA Intern Med 2017;177:231-42. [Crossref] [PubMed]
29. Collard CD, Body SC, Shernan SK, et al. Preoperative statin therapy is associated with reduced cardiac mortality after coronary artery bypass graft surgery. J Thorac Cardiovasc Surg 2006;132:392-400. [Crossref] [PubMed]
30. Joliat GR, Gaspar-Figueiredo S, Labgaa I, et al. Impact of preoperative statin medication on long-term outcomes after pancreatoduodenectomy for ductal adenocarcinoma: an international multicentric cross-sectional study. J Cancer Res Clin Oncol 2023;149:17007-14. [Crossref] [PubMed]
31. Karavias D, Thomas P, Koh A, et al. Statin therapy does not influence the outcome of patients undergoing surgery for pancreatic cancer. ANZ J Surg 2020;90:1671-6. [Crossref] [PubMed]