SAN FRANCISCO – Regular aspirin use did not improve outcomes among patients with colorectal cancers with mutations of the PIK3CA gene, suggest new data from the largest study yet of aspirin use in this patient population.
Regular aspirin use was reported by 26% of the 185 patients studied, researchers reported at the annual Gastrointestinal Cancers Symposium sponsored by the American Society of Clinical Oncology.
Users and nonusers were statistically indistinguishable with respect to a variety of recurrence- and survival-related outcomes, according to first author Dr. Nishi Kothari of the H. Lee Moffitt Cancer Center and Research Institute in Tampa.
"Our collaborative study did not validate overall or cancer-specific survival benefits associated with aspirin in PIK3CA-mutant patients across all stages, despite having a larger data set of patients" than earlier studies. "We were also not able to validate the recurrence-free survival benefits associated with aspirin in PIK3CA patients with stages II and III colorectal cancers."
Dr. Nishi Kothari
The earlier health professionals study (N. Engl. J. Med. 2012;367:1596-606)and adjuvant VICTOR (Vioxx in Colorectal Cancer Therapy: Definition of Optimal Regime) trial (J. Clin. Oncol. 2013;31:4297-305) suggested that PIK3CA mutation, found in about 15% of colorectal cancers, is a biomarker for treatment benefit from aspirin.
Such research has been prompted by the known impact of constitutive PI3K signaling on cyclooxygenase-2 (COX-2) activity and prostaglandin E2 synthesis, blocking tumor cell apoptosis, according to Dr. Kothari. By inhibiting COX-2, aspirin may restore apoptosis.
Factors that may explain the studies’ differing results include differing patient populations, more advanced-stage disease in the new study, a relatively smaller share of right-sided primaries (which are associated with poorer survival) among patients with metastases, inclusion of patients with mutations of PIK3CA in exons 9 and 20 (which have an unclear impact on outcomes), and shorter follow-up.
"At this point, we have positive as well as negative results regarding the predictive value of PIK3CA for aspirin therapy. To help resolve this issue, we will contribute this data to an individual patient meta-analysis" encompassing all three studies, Dr. Kothari said. "To further explore the validity of PIK3CA as a predictive biomarker, a prospective randomized study design should be used."
"As we move into an era of next-generation sequencing, we need to consider which somatic mutations should be studied ... As we identify more mutations in PIK3CA and put together larger patient cohorts, prospective work on aspirin as targeted therapy can begin to evaluate outcomes by mutation," she added.
In the new study, the investigators performed targeted exome sequencing of tumors from patients with stage I to IV colorectal cancer treated at the Moffitt Cancer Center and Royal Melbourne Hospital to identify the cohort with PIK3CA mutations.
The primary colorectal cancer tumor was right sided in 107 patients, left sided in 77 patients, and of unknown location in 1 patient.
With a median follow-up of 46 months, aspirin users and nonusers had statistically indistinguishable overall survival and cancer-specific survival, reported Dr. Kothari. The findings remained the same in a multivariate analysis that included potential confounders.
Furthermore, there was no benefit in terms of recurrence-free survival among patients with stage II and III disease.
There was a trend toward better overall survival for aspirin users among patients with stage IV disease in univariate analysis (hazard ratio, 0.40; P = .06) but not in multivariate analysis.
"Of note, we did find a statistically significant worsened survival with right-sided cancers in this stage IV population," Dr. Kothari pointed out (HR for left vs. right, 0.43; P = .037). "The improved survival with aspirin use seen initially in our stage IV univariate analysis might thus be due to a decrease in incidence of right-sided cancers."
Dr. Kothari disclosed no relevant conflicts of interest.
