Use of CA-125 to define progression of ovarian cancer in patients with persistently elevated levels.

Abstract

To determine an accurate definition for progression of ovarian cancer in patients with a persistently elevated serum CA-125. A retrospective analysis was performed on 300 patients with epithelial ovarian carcinoma with at least one measurement of CA-125. The date of progression according to clinical or radiologic criteria was ascertained in the 88 patients with persistently elevated CA-125 levels (> 23 U/mL). This was compared with the date of progression according to CA-125, defined as the date on which the CA-125 level first increased to >or= twice its nadir level, confirmed by a second sample also >or= twice the nadir. Eighty of the 88 patients had evidence of progression by both standard and CA-125 criteria, giving a sensitivity of 94%. In six of these patients, no sample was taken to confirm CA-125 doubling. In 13 patients, CA-125 doubling occurred after the date of clinical progression. Only one patient had a false-positive prediction of progression according to CA-125; the patient died as a result of a myocardial infarct before evidence of clinical progression. In patients whose CA-125 level decreases to normal after chemotherapy, a doubling from the upper limit of normal has been shown to predict progression. In those with persistently elevated levels, doubling of CA-125 from its nadir level has now been shown to accurately define progression. If confirmed, these CA-125 criteria should be used as additional end points in clinical trials.