ASPIRIN AND COLORECTAL CANCER SURVIVAL

August 11, 2009 — Regular aspirin use appears to improve survival in patients with nonmetastatic colorectal cancer, particularly those with tumors that overexpress cyclooxygenase (COX)-2.

According to the report, which appears in the August 12 issue of JAMA, the regular use of aspirin after a diagnosis of colorectal cancer was associated with a significant reduction in risk for both colorectal-cancer-specific mortality (log-rank P = .02) and overall mortality (log-rank P = .03).

The overall 5-year survival for patients using aspirin regularly was 88%, compared with 83% for those who did not. Ten-year survival rates were 74% for regular aspirin users and 69% for those who were not. The hazard ratio (HR) associated with regular aspirin use after diagnosis, compared with no use, was 0.71 (95% confidence interval [CI], 0.53 - 0.95) for colorectal-cancer-specific mortality and 0.79 (95% CI, 0.65 - 0.97) for overall mortality.

The influence of aspirin use on survival was seen across all disease stages. In a separate analysis of patients with stage II or III disease, the HR was 0.72 (95% CI, 0.53 - 0.99) for colorectal-cancer-specific mortality and 0.82 (95% CI, 0.66 - 1.04) for overall mortality.

No Clinical Recommendations Yet

However, routine use of aspirin or related agents as a cancer therapy cannot be recommended just yet. "These results are observational and, in the absence of a clinical trial, we can't make a clinical recommendation," said study author Andrew Chan, MD, MPH, assistant professor of medicine at Harvard Medical School in Boston, Massachusetts. "But it does provide compelling data to do a randomized controlled trial."

A growing body of evidence suggests that aspirin and other related nonsteroidal anti-inflammatory drugs might exert a chemopreventive effect, as previously reported by Medscape Oncology. "The data do support what we know about the mechanism of action of aspirin and the anti-cancer effect," Dr. Chan said in an interview. "Randomized trials have shown that aspirin can prevent precursor lesions."

Even though an overall clinical recommendation cannot be made at this point, individual patients could benefit from aspirin use. "Based on the result of the study, and based on results from previous trials, it is reasonable for individual patients to discuss aspirin use with their physicians," he said.

Randomized placebo-controlled trials have shown that aspirin use, as well as the use of celecoxib and rofecoxib, can significantly lower the risk for adenoma in patients at high risk for colorectal cancer. Aspirin might prevent colorectal neoplasia, at least in part, by inhibiting COX-2, which promotes inflammation and cell proliferation and is overexpressed in the majority of human colorectal tumors, the authors note. In some studies, the overexpression of COX-2 has been associated with a poorer prognosis, but this has not been consistent across all trials.

It remains unclear whether aspirin use can improve survival in patients who have already been diagnosed with colorectal cancer, even though aspirin has been found to inhibit tumor growth and metastases and to prolong survival in animal models. To address the issue, Dr. Chan and colleagues evaluated the association between aspirin use and colorectal cancer diagnosis using data from patients with nonmetastatic (stage I, II, and III) colorectal cancer who were participating in 2 large prospective cohort studies.

Improved Cancer-Specific and Overall Survival

A total of 1279 individuals (840 from the Nurses' Health Study and 439 men from the Health Professionals Follow-Up Study) were included in the analysis. All of the study participants were enrolled prior to a diagnosis of colorectal cancer and were followed until June 1, 2008.

There were fewer deaths among those who took aspirin, and regular use of aspirin after diagnosis was associated with a significant reduction in risk for both colorectal-cancer-specific mortality (log-rank P = .02) and overall mortality (log-rank P = .03).

At a median follow-up of 11.8 years, there were 193 (35%) total deaths and 81 (15%) colorectal-cancer-associated deaths among the 549 patients who regularly used aspirin. This compared with 287 (39%) total deaths and 141 (19 %) colorectal-cancer-specific deaths among 730 participants who did not use aspirin on a regular basis.

Among the 719 participants who did not use aspirin prior to their diagnosis, the researchers observed that aspirin use after their cancer diagnosis was associated with a HR for colorectal-cancer-specific mortality of 0.53 (95% CI, 0.33 - 0.86) and for overall mortality of 0.68 (95% CI, 0.51 - 0.92). Conversely, individuals who used aspirin prior to being diagnosed with colorectal cancer and who continued taking it after their diagnosis did not have a significant reduction in colorectal-cancer-specific survival (HR, 0.89; 95% CI, 0.59 - 1.35) or overall survival (HR, 0.95; 95% CI, 0.71 - 1.28).

Strongest Association With High COX-2 Expression

The inverse association between aspirin use and a lower risk for colorectal-cancer-specific mortality appeared to be strongest among individuals with primary tumors that overexpressed COX-2, compared with those with weak or absent expression (HR, 0.39 [95% CI, 0.20 - 0.76] vs 1.22 [95% CI, 0.36 - 4.18]).

It is possible that other COX-2 inhibitors might have a similar effect, explained Dr. Chan. "The effect seems to be related to the inhibition of COX-2, and that mechanism is shared, so it is possible that celecoxib might have a similar benefit, although we don't know if it will be of the same magnitude," he said. "It has been shown to prevent precancerous polyps, so there are reasons to believe that impact on survival might be similar to aspirin."

But for both agents, the risk for adverse events has to be considered. Aspirin is associated with gastrointestinal irritation and bleeding, and although COX-2 inhibitors, such as rofecoxib and celecoxib, have less gastrointestinal toxicity than aspirin, they have been associated with cardiovascular toxicity, Dr. Chan pointed out.

"Our study provides a compelling rationale [with which] to understand how these agents work and to develop medications that have anticancer properties but fewer side effects," he added.

New Biomarker?

An accompanying editorial hones in on the COX-2 finding.

"The specificity of the response of colorectal cancers to aspirin for patients in whom tumors overexpressed COX-2 suggests that this potential future treatment comes with its own ready-made predictive biomarker," writes editorialist Alfred I. Neugut, MD, PhD, professor of medicine and epidemiology at Columbia University in New York City.

In the near future, COX-2 expression may become "a standard predictive marker and aspirin may become standard adjuvant therapy in the management of colorectal cancer," he suggests.

This would follow a recent major trend in oncology, which has seen a number of predictive biomarkers used in clinical practice. In breast cancer, he notes, hormone receptors and ERBB2 status have been used successfully to determine the use of hormonal therapy and trastuzumab. More recently, KRAS mutations in colorectal cancer have attained a similar status as predictors of response to cetuximab and panitumumab, and BRAF mutations are soon likely to attain a similar status.

Dr. Chan agrees that the current study is another example in which a molecular marker predicts outcome. "That's where we are headed, as we know that colorectal cancer is not a single disease but one that has specific molecular subtypes," he said. "If we can explore our understanding of molecular subtypes, then we can use that information to tailor treatment."

More studies are needed, however, to see if it makes sense to use COX-2 as a marker. "We're just scratching the surface on this, and most likely we are going to be looking at a panel of markers," he added. "We then need to assess which ones are the critical markers, and COX-2 may well be on that list."

The study was supported by grants from the National Cancer Institute and National Institutes of Health. The authors and editorialist have disclosed no relevant financial relationships.

JAMA. 2009;302(6):649-659.