Epidemiology of non-steroidal anti-inflammatory drugs and cancer.

Abstract

There is evidence that aspirin--and apparently other NSAIDs--may be protective agents against cancer in the gastrointestinal tract. These effects are particularly well documented in the colon and rectum. Even considered in isolation, the observational data regarding colorectal neoplasia are quite strong, and the reality of a protective effect is buttressed by clinical trial data showing that aspirin prevents sporadic adenomas. Furthermore, the NSAIDs sulindac celecoxib have actually led to the regression of existing colorectal polyps in patients with FAP. Clearly, NSAIDs have the potential to suppress carcinogenesis in the large bowel. Observational data suggesting inverse associations of NSAIDs with cancers of the stomach and esophagus have emerged from several case-control studies and a few cohort analyses. In some studies the findings display features often associated with causal relationships, for example decreasing risks with increasing doses or duration of use. Nonetheless, the data currently do not support a secure conclusion that NSAIDs protect against these malignancies. The relevant data are not nearly as extensive as those for the colorectum, and case-control investigation of these upper gastrointestinal sites may be particularly delicate. It is conceivable that early symptoms of cancer (or of pre-invasive lesions) may have discouraged NSAID use in the cancer patients, creating the appearance of a protective association of the drugs with the risk of these malignancies. More extensive observational data particularly from cohort studies would be desirable to confirm the existing findings and clarify the doses and durations of use required for an effect. Clinical trial investigation might also be practical for pre-neoplastic endpoints, or--in carefully selected populations--perhaps with cancer as the focus. There are only relatively limited data available regarding the effect of NSAIDs on cancer of the pancreas. However, the studies that have investigated this malignancy have reported indications that NSAIDs may have a protective effect. The effects of NSAIDs on cancers outside the gastrointestinal tract are not clear. Some investigations suggest that NSAID use, particularly aspirin, is inversely associated with risk of cancers of the breast or ovary, but several well-done studies have not seen these associations, and the observations could have been due to bias or confounding. Findings regarding prostate cancer are similarly conflicting. The urinary tract is one organ system in which several studies have reported an increased cancer risk in association with NSAID use. Nonetheless, the effects remain unclear. There is only limited available information regarding carcinoma of the bladder, and no firm conclusions can be drawn at this point. More extensive data have been generated regarding the effect of NSAIDs--largely salicylates--on renal cell carcinoma or cancer or the renal pelvis and ureter. Although some studies have reported increased risks, there are also findings suggesting no association. It is particularly difficult for observational studies to ascertain with confidence the true effects of aspirin because of the suspected relationship of these cancers with use of phenacetin and perhaps acetaminophen. Further data--particularly from careful and large cohort studies--would be important to clarify these issues. As a body of research, the findings discussed here from epidemiological studies and clinical trials have begun to clarify the effect of NSAIDs on carcinogenesis in various organs in humans. There is clear potential for protective effects at several anatomic sites. Even for the colorectum, however, it is probably premature to now begin to use these drugs widely for cancer prevention. To reach that point, a weighing of the risks and benefits of the drugs needs to be made, together with a judgement regarding the benefits of alternative means of prevention. For colorectal cancer, for example, aspirin may provide only limited benefit over regular colonoscopy [95, 96]. Nonetheless, with the increased understanding of the clinical effects of NSAIDs on cancer, the development of effective chemoprevention with these drugs appears to be a real possibility.