Sample size and Population
This study is designed as a monocentric observational prospective study. In a recent study \[9\] authors found that the use of low-dose aspirin before and after a diagnosis of pancreatic cancer reduces of 32% the risk of recurrence (Hazard ratio HR=0.68, p\<0.01). On the basis of this study and considering that the effect of other drugs as chemopreventive agentsw will be studies, the estimate required sample size to achieve 90% power to detect at least 28% reduction in a hazard of the "drug users" group, by using a two-sided 0.05-level log-rank test, is 400. Therefore, from February 2019 to February 2022, 400 patients with a diagnosis of pancreatic ductal adenocarcinoma at any stage meeting the following inclusion criteria are expected to be enrolled. Median follow-up is estimated to be 24 months after the first disease diagnosis.
Data collection methods
Anamnestic, clinical and pathological data, included data on the aspirin, statins, metformin, angiotensin converting enzyme (ACE)-inhibitors and beta-blocking agents assumption will be collected during the first visit with the surgeon. A database managed by a dedicated data manager will be created to collect and analyse data. The PI will be responsible of the data security.
Statistical analysis
Association between variables will be assessed using the Chi Squared test (or Fisher's exact text where appropriate) for categorical variables and the Spearman's correlation for scale variables. DFS and OS will be estimated using Kaplan-Mayer method and Log Rank tests will be used to evaluate the difference between survival curves. The impact of aspirin, statins, metformin, angiotensin converting enzyme (ACE)-inhibitors and beta-blocking agents on the risk of recurrence will be estimated using Cox regression models. Variables resulting significant (p value \<0.05) at univariate analysis or variables which are known prognostic/risk factors will be included in the multivariable regression models. A p value of \<0.05 will be considered statistically significant. Statistical analysis will be conducted using SPSS v23 (IBM, Armonk, New York, USA) and R v3.3.0 (https://cran.r-project.org).
