Development of multiplexed bead-based immunoassays for the detection of early stage ovarian cancer using a combination of serum biomarkers

Abstract

CA125 as a biomarker of ovarian cancer is ineffective for the general population. The aim of this study was to evaluate multiplexed bead-based immunoassay of multiple ovarian cancer-associated biomarkers such as transthyretin and apolipoprotein A1, together with CA125, to improve the identification and evaluation of prognosis of ovarian cancer. We measured the serum levels of CA125, transthyretin, and apolipoprotein A1 from the serum of 61 healthy individuals, 84 patients with benign ovarian disease, and 118 patients with ovarian cancer using a multiplex liquid assay system, Luminex 100. The results were then analyzed according to healthy and/or benign versus ovarian cancer subjects. When CA125 was combined with the other biomarkers, the overall sensitivity and specificity were significantly improved in the ROC curve, which showed 95% and 97% sensitivity and specificity, respectively. At 95% specificity for all stages the sensitivity increased to 95.5% compared to 67% for CA125 alone. For stage I+II, the sensitivity increased from 30% for CA125 alone to 93.9%. For stage III+IV, the corresponding values were 96.5% and 91.6%, respectively. Also, the three biomarkers were sufficient for maximum separation between noncancer (healthy plus benign group) and stage I+II or all stages (I-IV) of disease. The new combination of transthyretin, and apolipoprotein A1 with CA125 improved both the sensitivity and the specificity of ovarian cancer diagnosis compared with those of individual biomarkers. These findings suggest the benefit of the combination of these markers for the diagnosis of ovarian cancer.

Figure 1. Scatter plots of concentrations of CA125, transthyretin, and apolipoprotein A1.

(A) healthy controls, benign ovarian disease and ovarian cancer patients (B) tumor stages in ovarian cancer patients (C) different histological subtypes in ovarian cancer patients. P value over a group denotes statistical significance of differences between each group and healthy controls. Each ovarian cancer subjects was compared with healthy controls. N, healthy controls; BE, benign ovarian disease; C, clear cell; E, endometrioid; G, granulosa cell; M, mucinous; S, serous; O, other.

Figure 2. ROC (receiver operator characteristic) discriminating ovarian cancer patients from benign ovarian disease patients.

The curves shown were obtained by processing quantified raw data by SigmaPlot 12.0 version software and the sensitivity/specificity values were predicted from the area under the curves and the calculated data. ROC curves for CA125, transthyretin, and apolipoprotein A1 alone: (A) benign patients versus patients with ovarian cancer; (B) benign patients versus patients with stages I to II ovarian cancer; (C) benign patients versus patients with stages III to IV ovarian cancer.

Figure 3. ROC discriminating ovarian cancer patients from healthy controls or benign ovarian disease patients using the three-biomarker panel.

The curves shown were obtained by processing quantified raw data by SAS 9.1 version software and the sensitivity/specificity values were predicted from the area under the curves and the calculated data. ROC curves for CA125 alone and the three-biomarker panel: (A) healthy controls versus patients with ovarian cancer. The overall difference in AUCs between the three-biomarker panel and CA125 alone was statistically significant (P<0.0001); (B) healthy controls versus patients with stages I to II ovarian cancer (P<0.0001); (C) healthy controls versus patients with stages III to IV ovarian cancer (P = 0.043); (D) benign patients versus patients with ovarian cancer. (P = 0.49); (E) benign patients versus patients with stages I to II ovarian cancer (P = 0.19); (F) benign patients versus patients with stages III to IV ovarian cancer (P = 0.045).

Figure 4. ROC discriminating ovarian cancer patients from healthy controls plus benign ovarian disease patients using the three-biomarker panel.

ROC curves for CA125 alone and the three-biomarker panel: (A) healthy controls plus benign patients versus patients with ovarian cancer. (P = 0.012); (B) healthy controls plus benign patients versus patients with stages I to II ovarian cancer (P = 0.014); (C) healthy controls plus benign patients versus patients with stages III to IV ovarian cancer (P = 0.52).

Figure 5. ROC discriminating ovarian cancer patients from healthy controls or benign ovarian disease patients using CA125, the two and three-biomarker panels.

ROC curves for CA125 alone, the two and three-biomarker panels: (A) healthy controls versus patients with ovarian cancer; (B) healthy controls versus patients with stages I to II ovarian cancer; (C) healthy controls versus patients with stages III to IV ovarian cancer; (D) benign patients versus patients with ovarian cancer; (E) benign patients versus patients with stages I to II ovarian cancer; (F) benign patients versus patients with stages III to IV ovarian cancer.