MIAMI BEACH – Head and neck cancer patients treated with radiation should be screened routinely for carotid artery stenosis, investigators recommended at the annual meeting of the American Society for Radiation Oncology.
Among 225 patients who had received radiation and were screened, an estimated 18% had significant asymptomatic stenosis (50% or greater narrowing) of one or both carotid arteries 3 years after treatment said Dr. Jennifer Dorth, a resident in radiation oncology at Duke University Medical Center in Durham, N.C.
"We recommend screening for head and neck cancer patients given that there are high rates of stenosis as well as high rates of progression of stenosis," she said.
Factors significantly associated with risk for stenosis included Framingham risk factors (smoking history, hypertension, hyperlipidemia, diabetes mellitus, cardiovascular/peripheral vascular disease, and atrial fibrillation) and radiation dose.
The investigators retrospectively reviewed outcomes of asymptomatic, disease-free head and neck cancer patients who had received radiation with curative intent to the neck. The patients were screened with carotid Doppler ultrasound at or after the 1-year follow-up visit, and this was repeated every 2-3 years. Patients with ultrasound evidence of 50% or greater stenosis were referred to vascular surgery.
The study identified 225 patients, 139 of whom had received intensity-modulated radiation therapy (IMRT), with the dose calculated separately for each side of the neck. Because of the separate treatment planning, the investigators analyzed the data by creating two separate models: one looking at all patients, and the other looking at 278 treatments in the 139 patients with IMRT.
In each model, about 85% of patients had stage III or IV disease and about 58% had cancer in the oropharynx, followed in order of frequency by the larynx, oral cavity, nasopharynx, or other sites.
A total of 33 patients had stenosis in 51 arteries. The median time between completion of radiation therapy and the last follow-up screening was 2 years. The median time to stenosis was 3 years.
Actuarial estimates of carotid artery stenosis were 2% at 1 year, 6% at 2 years, and 18% at 3 years.
In univariate analysis, factors associated with stenosis included male gender (P = .02), hypertension (P = .003), vascular disease (P less than .001), and Framingham score (P less than .001).
In the multivariate model looking at all patients, each Framingham risk factor was associated with a near doubling of stenosis risk (hazard ratio 1.8, P = .0003). In the model focusing on the IMRT population (adjusted for Framingham score), only radiation dose was significantly associated with stenosis (HR 1.07/Gy, P = .02).
Of the 33 patients with stenosis, 8 had no further follow-up imaging, 8 had stable stenosis, and 17 had progressive stenosis, 2 of whom had a cerebrovascular event. Eight patients with progressive stenosis received medical management only, and nine went on to surgery (three endarterectomies and six stent placements).
"Of the nine patients who underwent surgical management, there was a high rate of restenosis in 30% of patients at a year median follow-up, and this is consistent, unfortunately, with other series looking at rates of restenosis," Dr. Dorth said.
The study was internally funded. Dr. Dorth reported having no relevant financial disclosures.