Elevated free plasma DNA is a reliable indicator of recurrent esophageal cancer, more so than carcinoembryonic antigen (CEA), and is able to detect recurrent disease in most patients before clinical evidence emerges. These findings come from a study by researchers at the University of Southern California, published in the July issue of the Journal of the American College of Surgeons.
The results are only "a beginning," Dr. Farzaneh Banki told Reuters Health. "It's not a screening tool, not yet."
Plasma DNA was measured using polymerase chain reaction in 44 healthy control subjects and preoperatively in 45 patients with esophageal cancer, with follow-up measurements in 21. Serum CEA was measured preoperatively in 31 of the patients and postoperatively in 14 of the 21 patients who had sequential plasma DNA measured.
At the time of surgery, 39 patients had localized cancer and 6 had disseminated disease; 7 developed disseminated disease after surgical resection.
A CEA level of 5.0 ng/mL was used as a cut-off, and that for DNA was 19 ng/mL — the 95th percentile value of the healthy control subjects in the study.
In detecting unresectable esophageal cancer, plasma DNA showed 100% sensitivity but only 22% specificity, for a positive predictive value (PPV) of 19% and a negative predictive value (NPV) of 100%. The PPV and NPV for CEA were 40% and 89%, respectively.
In detecting recurrent esophageal cancer postoperatively, plasma DNA showed 100% specificity, sensitivity, NPV and PPV. CEA showed 100% specificity and PPV, but only 33% sensitivity and 67% NPV.
The report cautions that elevated plasma DNA and/or CEA do not rule out resectability of esophageal cancer. In addition, it noted that although a normal plasma DNA level postoperatively rules out recurrent disease, "a normal serum CEA level does not exclude recurrent disease."
Dr. Banki told Reuters that free plasma DNA is generally a marker of cell death and that it is elevated, although not as much as in this study, among marathoners and smokers. It is unknown at this point, she added, whether plasma DNA is from tumor cells or from normal tissue that has been displaced by a tumor.
J Am Coll Surg 2008;207:30-35.
Reviewed by Ramaz Mitaishvili, MD