Rubin S C, Hoskins W J, Hakes T B, Markman M, Reichman B S, Chapman D, Lewis J L
Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY.
Am J Obstet Gynecol. 1989 Mar;160(3):667-71. doi: 10.1016/s0002-9378(89)80054-7.
Serum CA 125 levels and surgical findings were prospectively compared in 96 secondary laparotomies performed on patients with epithelial ovarian cancer. All patients had documentation of an elevated CA 125 level (greater than 35 U/ml) at a time when ovarian cancer was present, and thus the tumors were known to be "marker positive." Operation was performed in 45 patients who were clinically free of disease and in 51 patients in whom there was clinical evidence of disease. At the time of operation, 29 patients had normal CA 125 levels; persistent disease was documented in 18 (62%) of these. Of the patients with normal CA 125 levels at the time of operation, those with persistent disease had a significantly higher mean CA 125 level (16.9) than those with no disease detected (8.8, p = 0.001). At exploration, cancer was found in 84 patients. There was a correlation between the maximum diameter of the largest residual tumor mass and the accuracy of the CA 125 level as follows: microscopic to 1 cm disease, 55% accuracy; greater than 1 to 5 cm disease, 80% accuracy; greater than 5 cm disease, 92% accuracy (p = 0.013). There was no correlation of CA 125 accuracy with the patient's age, number of months from initial diagnosis, tumor stage, grade, or cell type, or the highest-ever level of CA 125. Of the 84 patients with tumor found at exploration, 66 had elevated CA 125 levels, yielding a sensitivity of 78.5%. There were 12 patients with no tumor found at exploration; 11 of these had normal CA 125 levels. The one patient who had an elevated CA 125 level subsequently had a recurrence; the corrected specificity is thus 100%. An elevated CA 125 level is an accurate predictor of persistent disease. Most of these patients will have gross tumor present. The accuracy of the CA 125 level in detecting disease is related to the amount of tumor present. In our population, the predictive value of an elevated CA 125 level was 100%; the predictive value of a normal CA 125 level was 38%.
对96例上皮性卵巢癌患者进行二次剖腹手术,前瞻性地比较了血清CA 125水平与手术结果。所有患者在患卵巢癌时均有CA 125水平升高(大于35 U/ml)的记录,因此这些肿瘤被认为是“标志物阳性”。对45例临床无疾病的患者和51例有疾病临床证据的患者进行了手术。手术时,29例患者CA 125水平正常;其中18例(62%)记录有持续性疾病。手术时CA 125水平正常的患者中,有持续性疾病的患者平均CA 125水平(16.9)显著高于未检测到疾病的患者(8.8,p = 0.001)。探查时,84例患者发现有癌症。最大残留肿瘤块的最大直径与CA 125水平的准确性之间存在如下相关性:微小至1 cm疾病,准确性为55%;大于1至5 cm疾病,准确性为80%;大于5 cm疾病,准确性为92%(p = 0.013)。CA 125的准确性与患者年龄、自初次诊断以来的月数、肿瘤分期、分级或细胞类型,或CA 125的最高水平均无相关性。在探查时发现有肿瘤的84例患者中,66例CA 125水平升高,敏感性为78.5%。探查时未发现肿瘤的患者有12例;其中11例CA 125水平正常。1例CA 125水平升高的患者随后复发;校正后的特异性为100%。CA 125水平升高是持续性疾病的准确预测指标。这些患者大多数会有肉眼可见的肿瘤。CA 125水平在检测疾病方面的准确性与存在的肿瘤量有关。在我们的研究人群中,CA 125水平升高的预测价值为100%;CA 125水平正常的预测价值为38%。