CHICAGO - A chest x-ray after the surgical treatment of melanoma, as suggested in the National Comprehensive Cancer Networks guidelines, does not detect recurrence at levels sufficient to justify its use, according to an analysis from the Sunbelt Melanoma Trial.
In the review of more than 1,200 patients who had yearly chest x-rays following surgical treatment for melanoma, fewer than 0.1% of the x-rays showed pulmonary metastases that were amenable to resection, Dr. Russell E. Brown said at the annual meeting of the Central Surgical Association.
This finding has implications for patients' exposure to radiation, and, according to one member of the audience, for health care costs as well. "I think in 2010, with health care reform, we all need to look very carefully at what these follow-up criteria are and what we're using resources for," said Dr. Thomas Howard, professor of surgery at Indiana University, Indianapolis. "These are powerful data, and they really speak to the inadvisability of having a chest x-ray for everyone."
The National Comprehensive Cancer Networks Medical Practice Guidelines in Oncology (v.2.2009) describes the "option" of a chest x-ray - as well as a CBC and a lactate dehydrogenase (LDH) measurement - every 6-12 months in patients with melanoma stage IB-IV and no evidence of disease.
However, physicians do not necessarily view such guidelines as optional, said Dr. Brown of the University of Louisville (Ky.). "Recommendations such as these lead physicians to feel obligated to use the chest x-ray during follow-up," he noted.
The Sunbelt Melanoma Trial was a prospective, randomized trial of 3,619 patients (aged 18-70 years) at 79 centers in the United States and Canada. The trial evaluated the role of completion lymph node dissection or high-dose interferon alpha-2b in patients with melanoma that was staged by sentinel lymph node biopsy. All patients were clinically node negative, with primary tumors at least 1 mm thick. Those who had positive nodes in sentinel lymph node biopsy underwent completion lymphadenectomy. All patients also had yearly follow-up chest x-rays.
This analysis from the University of Louisville studied a subset of 1,235 patients whose follow-up comprised both a chest x-ray and other exams, including blood tests and evaluation of disease state and recurrence pattern. Within the subset, 1,025 patients (83%) had no recurrence, and 210 (17%) had recurrence. Sites of recurrence were locoregional in 108 patients (51%), distant in 74 (35%), and in the lung only in 28 (13%).
The subset of 1,235 patients had 4,218 chest x-rays, of which 38 (0.9%) were true positives and 3,593 (85.2%) were true negatives.
"An overwhelming 3,593 true negatives," said Dr. Brown. Of the 38 positives, 35 had widely disseminated metastases.
Only 3 (0.07%) of the 4,218 chest x-rays showed pulmonary metastases that were amenable to resection, said Dr. Brown. He invited the audience to extrapolate, pointing out that 1,406 chest x-rays were required to find a single resectable pulmonary metastasis. "Or conversely, 412 patients that you screen yearly for 74 months," he said.
Dr. Brown noted that the ideal screening test is noninvasive, low cost, and widely available, and has high sensitivity and specificity. It also must have potential for therapeutic benefit. A chest x-ray has "terrible" sensitivity and little therapeutic potential, he said.
Discussant Dr. Margo Shoup of Loyola University Medical Center in Maywood, Ill., asked Dr. Brown whether these findings have altered his practice in Louisville.
"Practice in multidisciplinary settings varies widely," said Dr. Brown. "Among our surgeons, we do not routinely perform chest x-ray imaging, LDH, or really anything other than physical exams. We rely heavily on patient education, and we see our patients about every 6 months for the first 1-5 years," he said. "Our medical colleagues use PET and CT liberally, without proven benefit, especially with the lack of effective systemic therapies."
Dr. Brown disclosed that the Sunbelt Melanoma Trial was supported by a grant from Schering Oncology/Biotech.