After 5 years, aspirin therapy reduced the incidence of cancer death by nearly 40% (OR 0.63, CI 0.49-0.82, P=0.0005), although high-dose aspirin (>300 mg/day) was associated with a reduction in overall mortality after only 3 years. Because the number of cancers was small, only the reduction in death from colorectal cancer or lymphoma reached statistical significance. Aspirin therapy also reduced the incidence of cancer (OR 0.88, CI 0.80-0.98, P=0.017), the effect first appearing after 3 years of treatment and becoming stronger after 5 years. Interestingly, with longer duration of follow-up, the effect of aspirin on cancer incidence increased while its effect on prevention of vascular events diminished. The composite outcome of major vascular events, cancer, and extracranial bleeds was lower in subjects allocated to aspirin, indicating an overall favorable risk-benefit ratio.
Effects of Aspirin on Cancer Metastasis
That the effect of aspirin on cancer mortality is greater than its effect on cancer incidence implies that aspirin may slow the progression of cancer, which prompted the second of the three recent studies.13 This meta-analysis of data from five large randomized controlled trials of aspirin in the prevention of vascular events, which included a total of 17,285 subjects, recorded metastasis at initial diagnosis and during follow-up in all subjects with a new diagnosis of cancer. During a mean follow-up of 6.5 years, allocation to aspirin reduced the incidence of new cancers. Among the 1,101 cases of cancer that occurred during the trials, aspirin reduced the risk of definite distant metastases by about one-third (HR 0.64, CI 0.48-0.84, P=0.001). Reductions in incidence of metastasis to the lung, liver, and brain were found, as well as reductions in the incidence of malignant brain tumors and metastasis to multiple sites. Cancers were less likely to be metastatic at diagnosis and during follow-up in the aspirin-treated groups than in the placebo-treated groups (OR 0.59, CI 0.44-0.78, P=0.0003). The reduction in the risk of metastasis occurred in adenocarcinomas but not in other cancers. Death from cancer and death from any cause were both lower in the aspirin-treated groups. Although the number of individual cancers was low, case fatality in colorectal and prostate cancer, two of the three most common adenocarcinomas, was significantly reduced by aspirin therapy. This was the first study to show that aspirin prevents distant metastases and suggests that, in addition to its potential for preventing cancer, aspirin may one day find a role in cancer treatment.
Validity of Observational Studies
The chief limitation of these and other studies linking aspirin use with reductions in cancer incidence and mortality is their observational design. Large, prospective, controlled trials lasting 10 or more years are impractical, so the Oxford researchers set out to determine whether the results of case-control and cohort studies are reliable estimates of the effects of aspirin on cancer prevention.14 They pooled data from 150 case-control and 45 cohort studies on the association between use of aspirin or nonsteroidal anti-inflammatory drugs (NSAIDs) and risk of all types of cancer and compared the results to those from randomized trials of aspirin for the prevention of vascular events. Despite substantial heterogeneity in study design among the observational studies, meta-analysis of their results found a reduction in colorectal cancer incidence after 20 years similar to that seen in the randomized trials. The effects seemed to be equally strong for use of aspirin or NSAIDs, although the number of studies using NSAIDs was small. Incidence of esophageal, gastric, biliary, and breast cancer were also reduced by regular aspirin use. Thus, there is good reason to believe that observational studies can be relied on to paint an accurate picture of the potential of aspirin to prevent cancer.