Popis: |
Despite considerable research on cancer treatments and preventatives, poor outcomes in cancer patients are common. The vital search for effective cancer drugs often begins in the laboratory, where unfortunately the effects of a drug in humans cannot be perfectly modelled. Epidemiology can play a vital role in determining the real world efficacy of a drug currently used for other purposes before clinical trials begin. This thesis therefore used primarily laboratory evidence to identify potential anti-cancer uses for existing common drugs. The drugs and cancers studied were; tricyclic antidepressants and both incidence and survival in a number of cancer types, particularly glioma; aspirin and colorectal cancer survival; and angiotensin converting enzyme (ACE) inhibitors and hepatocellular carcinoma (HCC) incidence. A series of studies using The General Practice Research Database as a data source assessed any potential associations: A case-control study for tricyclic antidepressant use and cancer incidence; cohort studies to examine mortality in colorectal cancer and glioma in relation to tricyclic use, and for colorectal cancer mortality in aspirin users; and a case-control study in relation to ACE inhibitor use and HCC. A strong, cancer type specific, dose and time dependant protective effect was found for the incidence of glioma and colorectal cancer. This led to a further study examining mortality for these cancer types in tricyclic users. While no significant protective effects in all-cause mortality of tricyclic users were found, a larger study could still find such an effect in glioma. For aspirin and colorectal cancer mortality, a small but significant reduction in mortality was observed, though these effects were not entirely consistent throughout the study. There were no significant associations found between ACE inhibitors and HCC. These findings contribute to the knowledge of the anti-cancer effectiveness of these drugs, and may assist in designing future clinical studies. |