Osteopontin, Macrophage Migration Inhibitory Factor and Anti-Interleukin-8 Autoantibodies Complement CA125 for Detection of Early Stage Ovarian Cancer

Abstract

Early detection of ovarian cancer promises to reduce mortality. While serum CA125 can detect more than 60% of patients with early stage (I-II) disease, greater sensitivity might be observed with a panel of biomarkers. Ten protein antigens and 12 autoantibody biomarkers were measured in sera from 76 patients with early stage (I-II), 44 patients with late stage (III-IV) ovarian cancer and 200 healthy participants in the normal risk ovarian cancer screening study. A four-biomarker panel (CA125, osteopontin (OPN), macrophage inhibitory factor (MIF), and anti-IL-8 autoantibodies) detected 82% of early stage cancers compared to 65% with CA125 alone. In early stage subjects the area under the receiver operating characteristic curve (AUC) for the panel (0.985) was significantly greater (p < 0.001) than the AUC for CA125 alone (0.885). Assaying an independent validation set of sera from 71 early stage ovarian cancer patients, 45 late stage patients and 131 healthy women, AUC in early stage disease was improved from 0.947 with CA125 alone to 0.974 with the four-biomarker panel (p = 0.015). Consequently, OPN, MIF and IL-8 autoantibodies can be used in combination with CA125 to distinguish ovarian cancer patients from healthy controls with high sensitivity. Osteopontin appears to be a robust biomarker that deserves further evaluation in combination with CA125.

Keywords: CA125; IL-8 autoantibodies (IL-8 AAb); early detection; macrophage migration inhibitory factor (MIF); osteopontin (OPN); ovarian cancer.

Figure 1

Serum levels of osteopontin (OPN), macrophage inhibitory factor (MIF) anti-IL-8 autoantibody and CA125 are elevated in the discovery set from ovarian cancer patients and healthy controls. (A) OPN antigen; (B) MIF antigen; (C) anti-IL-8 autoantibodies (AAb) and (D) CA125 antigen. In the upper panels, each symbol represents the average of duplicate serum samples from individual early stage ovarian cancer cases (green), late stage ovarian cancer cases (red) or controls (blue). The middle panels contain a scatter plot where the blue line represents the median value in early and late stage case and control groups, respectively. The red dashed lines in each plot represent the cut-off value at 98% specificity for OPN, MIF and anti-IL-8 AAb, and cut-off value of 35 U/mL for CA125. The lower panel is a table displaying the sensitivity for early and late stage ovarian cancer cases and controls near 98% specificity.

Figure 2

Serum levels of OPN, MIF anti-IL-8 autoantibody and CA125 are elevated in the validation set from ovarian cancer patients and healthy controls. (A) OPN antigen; (B) MIF antigen; (C) anti-IL-8 autoantibody and (D) CA125 antigen. In the upper panels, each symbol represents the average of duplicate serum samples from individual early stage ovarian cancer cases (green), late stage ovarian cancer cases (red) or controls (blue). The middle panels contain a scatter plot where the blue line represents the median value in early and late stage case and control groups, respectively. The red dashed lines in each plot represent the cut-off value at 98% specificity for OPN, MIF and anti-IL-8 AAb, and cut-off value of 35 U/mL for CA125. The lower panel is a table displaying the sensitivity for early and late stage ovarian cancer cases and controls near 98% specificity.

Figure 3

Receiver operating characteristic (ROC) curves for a combination of CA125, OPN, MIF and anti-IL-8 autoantibody biomarkers. (A) Early stage cases of discovery set. (B) Late stage cases of discovery set. (C) Early stage cases of validation set. (D) Late stage cases of validation set. A combination of CA125, OPN, MIF and IL-8 AAb levels had a significantly greater AUC (0.985) than did CA125 alone in early stage cases of discovery set (AUC = 0.885, p < 0.001). Compare to CA125 alone (AUC = 0.947), the combination also made a significant difference (AUC = 0.974, p = 0.015) in early stage cases of validation set.