PI3KCA Could Predict Aspirin Response in Colorectal Cancer

September 26, 2013
Anna Azvolinsky

In the VICTOR trial, patients with PI3KCA-positive colorectal cancer who took aspirin had a lower cancer recurrence compared to those with a wild-type PI3KCA gene.

A new study, by researchers from the VICTOR trial (Vioxx in Colorectal Cancer Therapy: Definition of Optimal Regime), tested whether treatment with rofecoxib (Vioxx) or aspirin was effective for colorectal patients whose tumors harbored a mutation in PI3KCA, finding that PI3KCA may be a predictive biomarker of aspirin therapy. Patients with PI3KCA-positive colorectal cancer who took aspirin had a lower colorectal cancer recurrence compared with those with a wild-type PI3KCA gene. No evidence of a greater benefit for PIK3CA-positive patients from rofecoxib treatment was found compared with placebo. The study is published in the Journal of Clinical Oncology.

Patients in the VICTOR trial had stage II or III colorectal cancer and were randomized to either rofecoxib or placebo. All patients had previously been treated with therapy that included surgery and may have included radiotherapy and chemotherapy. Patients took rofecoxib for a median of 7.4 months before study treatment was discontinued and rofecoxib was removed from the market. Patients who had been taking aspirin during randomization or those who started aspirin daily at follow-up were classified as aspirin users in the study.

Of 896 patients in the VICTOR trial for whom PI3KCA mutation status was known, 792 patients had PI3KCA wild-type colorectal cancer and 104 (11.6%) had PI3KCA-mutated colorectal cancer.  Among patients in the PI3KCA-mutated tumor group, 23 of 90 patients who did not use aspirin had a recurrence of disease. Zero of the 14 patients who took aspirin regularly had a recurrence. Among patients who were PI3KCA wild-type, 151 of 681 non-aspirin users (22.1%) recurred while  22 of 111 aspirin users recurred (19.8%). The adjusted hazard ratio for recurrence-free survival in the non-aspirin group was 0.94 (P = .786) and 0.11 in the aspirin user group (P = .027).

Among PI3KCA-mutated patients, those who were aspirin users had better survival compared with those who did not use aspirin regularly, but the results did not reach statistical significance.

No difference in colorectal cancer recurrence was found in PI3KCA-mutated patients treated with rofecoxib compared with placebo (P = .66), and no effect on recurrence-free survival was shown (P = .473).

Aspirin has been shown in observational studies to reduce the rate of colorectal cancer development (see “Role of Aspirin in Colorectal Cancer Prevention and Treatment”). Other studies suggest that aspirin may even be effective as a treatment of colorectal cancer. A recent study provided evidence that the approximately 15% of patients with PI3KCA-mutated colorectal cancers derive the most benefit from aspirin therapy. Because taking nonsteroidal anti-inflammatory drugs (NSAIDs) or aspirin can lead to side effects such as gastrointestinal bleeding in some patients, identifying a specific subpopulation of patients who will benefit the most from such treatment will provide a rationale for treating these patients with aspirin or an NSAID.

While the current study had a relatively small number of PI3KCA-mutated colorectal cancer patients, the results warrant further study and provide support for PI3KCA as a predictive biomarker. The authors conclude that these results “suggest that prospective evaluation of aspirin as adjuvant therapy in this tumor molecular subtype is merited.”