Trial of low-dose aspirin + metformin in colon cancer (ASAMET)

Epidemiological studies and cardiovascular prevention trials have shown that low-dose aspirin can inhibit colorectal cancer incidence and mortality, including inhibition of distant metastases. Metformin has also been associated with decreased colorectal cancer incidence and mortality in meta-analyses in diabetic patients. Recent studies have shown that aspirin is an inhibitor of mTOR and an activator of AMPK, targeting regulators of intracellular energy homeostasis and metabolism, and that the combination of aspirin and metformin, another AMPK activator, has a striking additive effect on AMPK activation and mTOR inhibition, with increased autophagy in colon cancer cell lines. We propose a randomized, placebo-controlled, 2x2 biomarker trial of aspirin and metformin to test the activity of either agent alone and their combination on a set of surrogate biomarkers of colorectal carcinogenesis. After surgery and completion of adjuvant chemotherapy (if applicable) n=160 patients with stage I-III colon cancer will randomly be assigned in a four-arm trial to either aspirin, 100 mg day, metformin 850 mg BID, their combination, or placebo for one year. The primary endpoint biomarker is the change in the immunoistochemical expression of nuclear factor kappa-B (NF-kB), a key transcription factor in inflammation and colon carcinogenesis, in the normal mucosa of proximal and distal colon obtained with multiple biopsies in two paired colonoscopies one year apart. Additional biomarkers will include: 1) the genomic profile of candidate genes and pathways in tissue biopsies by genome wide gene expression, 2) the IHC expression of tissue pS6K, p53, beta-catenin, PI3K and 3) the measurement of circulating IL-6, CRP and VEGF. These biomarkers have been selected to elucidate the impact of the interventions on molecular/cellular mechanisms in epithelial tissue or circulation that characterize risk of cancer recurrence and metastasis. Moreover, their modulation by the preventive interventions may provide important clues for a subsequent phase III, cancer incidence reduction trial. The Austrian Breast & Colorectal Cancer Study Group (ABCSG) will participate to this trial by setting up a recruitement site, recruiting 40 patients (Principal Investigator: Univ. Prof. Dr. Gnant, Medical University of Vienna). All necessary labours to conduct the Austrian part of the clinical trial in high quality will be done by ABCSG project management, monitoring and regulatory affairs department. This includes submission at ethics committees and authorities, registration of the study, generation of essential documents, SAE reporting, on site monitoring and source data verification, logistics and medication supply, as well as surveillance of project progress. ABCSG will provide the Clinical Data Management System MACRO, Interface TRIALDATAPORT for the whole trial, including programming of the clinical database e.g. electronic Case Report Forms (eCRFs) for web-based data entry and helpdesk for all partners. To guarantee accurate reflection of the endpoint related topics within the eCRFs, an ABCSG statistician is closely involved in the database set-up phase. MACRO, Interface TRIALDATAPORT is an electronic data capture system developed by InferMed Ltd for running mono- and multi-centre clinical trials. The softwares intuitive, interactive tools for eCRF generation and electronic data capture simplify study set-up, reduce time to database lock and help to succeed in increasingly complex clinical environment. The Software is designed to be user-friendly, simple to operate and usable by any member of the clinical study team. The webbased data entry module enhances navigation between eForms (=eCRF pages), visits and subjects and provides visual feedback and status indicators to the data entry user. Furthermore, the Interface TRIALDATAPORT provides on-line support e.g. immediate validation of values as they are entered by the user. Using the Software MACRO, Interface TRIALDATAPORT makes it easy to enter and monitor data and to run reports. The Data Management System MACRO, Interface TRIALDATAPORT is designed to support the requirements of all relevant regulatory bodies including the internationally recognized ICH Good Clinical Practice and FDA 21 CFR Part 11. An ABCSG statistician is also involved at the time of analysis and will perform a validation analysis on the primary end point of the trial. All steps of statistical analyses will be performed in close interaction with the international leading partner of the study.

This randomized, placebo-controlled trial is based on the results of previous studies, which have shown that low-dose aspirin can, among others, inhibit colorectal cancer and reduce the mortality rate associated with the disease. Furthermore, meta-analyses of data from diabetic patients indicate that metformin has also been associated with decreased colorectal cancer incidence and mortality. In the ASAMET trial, which sponsor is Ente Ospedaliero Ospedali Galliera (EOG) the efficacy of aspirin and metformin - alone and in combination - using a set of biomarkers in relation to colorectal carcinogenesis will be tested. According to the original protocol 160 patients with a stage I-III colorectal cancer are planned to be enrolled after a surgery, in this 4-arm study and to be treated with 100 mg aspirin/day and 850 mg metformin twice daily or one agent each in combination with placebo or placebo only for 1 year. The primary endpoint is the change in NF-B expression in noncancerous mucosa of the proximal and distal colon. In addition to one study site in Italy, one site in Ljubljana, Slovenia and three in Austria (Wels, Vienna, Innsbruck) were opened for patient enrollment. Due to the limited patient population, the recruitment goal could not be achieved within the predefined time. Therefore, a protocol amendment is currently under preparation (i.e. change of study design from double-blind to open-label). The final analysis will be performed on the data acquired from both study phases: double- blind and open-label.