This article is more than 5 years old. Information may no longer be current.
Aspirin improved survival in colon cancers expressing HLA class I antigen
Patients with colon cancer whose tumors expressed HLA class I antigen demonstrated improved survival when they received aspirin for at least 2 weeks after their diagnosis, according to results of a cohort study.
The analysis included 999 patients, 18.2% (n=182) of whom received a prescription for aspirin for at least 14 days after their colon cancer diagnosis. The median duration of aspirin prescriptions was 30 days, and the mean number of prescriptions was 12 (range, 1-220).
Of 936 evaluable patients, 643 (66.8%) had tumors that expressed HLA class I antigen, whereas 320 (33.2%) did not. Aspirin use was comparable between patients whose tumors did and did not express HLA class I antigen (19% vs. 18%).
Among patients whose tumors expressed HLA class I antigen, aspirin use was associated with improved OS (RR=0.53; 95% CI, 0.38-0.74). However, aspirin users whose tumors did not express HLA class I antigen did not demonstrate a survival benefit (RR=1.03; 95% CI, 0.66-1.61).
Aspirin use conferred similar OS benefits among patients whose tumors had strong prostaglandin endoperoxide synthase 2 (PTGS2) expression (RR=0.68; 95% CI, 0.48-0.97) and those with weak PTGS2 expression (RR=0.59; 95% CI, 0.38-0.97).
Researchers observed a significant association between OS and aspirin use post-diagnosis among patients with wild-type PIK3CA tumors (RR=0.55; 95% CI, 0.40-0.75); however, researchers observed no association between OS and aspirin use among patients with PIK3CA-mutated tumors (RR=0.73; 95% CI, 0.33-1.63).
The findings suggest aspirin’s benefits in this setting must be weighed with the potential risks, Alfred I. Neugut, MD, PhD, of the department of medicine at the New York Presbyterian Hospital and at Columbia University Medical Center, wrote in an invited commentary.
“While aspirin has significant data demonstrating its benefits as a chemopreventive agent for colon cancer, its adverse effects (gastrointestinal tract bleeding, intracranial hemorrhage, gastritis) outweigh its benefits in the prevention setting (eg, giving aspirin to 2,000 average-risk adults might prevent one colorectal cancer at the expense of multiple cases of gastrointestinal tract bleeding [or] stroke),” Neugut wrote. “However, in the setting of patients with stage III colon cancer, who have significant risks for mortality, the risks of a daily aspirin seem minor relative to the putative gains in mortality.”
For more information:
- Neugut AI. JAMA Intern Med. 2014;doi:10.1001/jamainternmed.2013.14544.
- Reimers MS. JAMA Intern Med. 2014;doi:10.1001/jamainternmed.2014.511.
Disclosure: The researchers report employment or consultant roles with and honoraria from the Aspirin Foundation, Bayer and the PHARMO Institute.
Perspective
Back to Top
Aspirin targets cyclooxygenases (COX) and results in inhibition of prostaglandin synthesis. Regular aspirin use impacts on colon polyp formation and the evolution of colorectal cancer. Population-based studies as well as clinical trials indicate that aspirin, and in some cases other NSAIDs, results in significant reductions in polyp formation in familial adenomatous polyposis, the incidence of colorectal cancer in patients with Lynch syndrome, population-wide colorectal cancer mortality (30% to 50%), and the risk of colon cancer mortality after curative treatment.
Studies attempting to identify the mechanism of effect of aspirin have implicated overexpression of COX-2 as a predictive biomarker for an aspirin preventive effect. Subsequent studies, including the VICTOR trial and the pooled analysis of the Nurses’ Health and Health Professionals Follow-up Study, have indicated that only patients with tumors with activating mutations in pI3 kinase (PIK3CA) had a reduced risk of colorectal cancer with aspirin compared with no benefit for wild-type PIK3CA. PIK3CA mutation may impact downstream on COX-2 by upregulation or enhanced expression.
Two large data sets recently released, including the study from the Netherlands by Reimers and colleagues published in JAMA, now question the role of PIK3CA mutation and aspirin effects in colon cancer. In nearly 1,000 patients evaluated from the Eindhoven Cancer Registry who underwent surgery for colon cancer from 2002 to 2008, paraffin-embedded tissue was evaluated for expression of HLA class I antigen, COX-2, microsatellite instability and PIK3CA mutation. Aspirin use was evaluated by medical record review, as aspirin was apparently only available by prescription. Aspirin use after diagnosis (present in 18.2% of the population) correlated with improved OS. HLA class I antigen expression correlated with a survival benefit with aspirin, whereas loss of HLA class I antigen expression correlated with no aspirin effect. In contrast to other published results, a survival benefit for aspirin was associated with low COX-2 expression, and a survival benefit was seen only in patients with PIK3CA wild-type cancers. The sample size for deaths in patients with mutant PIK3CA precluded statistical analysis.
Another recent study by Kothari and colleagues from Moffit Cancer Center, presented at the Gastrointestinal Cancers Symposium in January, also questions the relation of PIK3CA mutation and a benefit for aspirin use in colon cancer. More than 1,000 patients from the Royal Melbourne Hospital in Australia and nearly 500 patients from Moffit were studied, including patients with stage IV disease, and 12% had PIK3CA kinase mutation. Across the entire cohort, there was no association with aspirin use and improved survival, and there was no correlation with PIK3CA mutation and aspirin effect.
The failure of these two studies, in particular the Eindhoven registry, to confirm a predictive benefit for PIK3CA mutation and aspirin use in colon cancer indicates that further study of the role of this biomarker in colon cancer is required. A planned pooled analysis of the Kothari study with the VICTOR trial and the Nurses’ and Health Professionals Follow-up Study will hopefully shed greater light on the question. Ultimately, controlled retrospective studies evaluating the efficacy of aspirin therapy in PIK3CA-mutant patients, and other biomarker-identified patient subsets, may be required. Current practice is to recommend regular aspirin use to patients with Lynch syndrome, and to consider it for use in patients after curative treatment of colorectal cancer.