Soluble MUC1 and serum MUC1-specific antibodies are potential prognostic biomarkers for platinum-resistant ovarian cancer

Abstract

MUC1 (CA15-3) and MUC16 (CA125) tumor-associated antigens are upregulated in ovarian cancer and can be detected in patients' sera by standardized tests. We postulated that increased MUC1 and MUC16 antigens augment antibody responses in platinum-resistant ovarian cancer patients and that the frequency and intensity of these responses can be used as immune biomarkers of treatment response and disease outcome. We measured MUC1 and MUC16 tumor expression by immunohistochemistry (IHC), assessed serum antigenic levels and quantitated circulating antibodies by ELISA in a cohort of 28 ovarian cancer patients with platinum-resistant or platinum-refractory ovarian cancer, and treated with intraperitoneal (IP) interleukin-2 (IL-2). MUC1 and MUC16 were overexpressed in tumor samples and showed differential distribution profiles. Serum MUC1 (CA15-3) measurements were elevated in all patients and significantly correlated with increased risk of death (P = 0.003). MUC1-specific IgM and IgG anitbodies were found in 92 and 50% of cases, respectively. Patients with progressive disease had higher mean anti-MUC1 IgG than responders at both early (P = 0.025) and late (P = 0.022) time points during IP IL-2 treatment. Anti-MUC1 IgM antibodies inversely correlated with overall survival at both early (P = 0.052) and late (P = 0.009) time points. In contrast to MUC1, neither soluble MUC16 nor MUC16-specific antibodies were significantly associated with clinical response or overall survival in this study. Increased serum MUC1 and high anti-MUC1 antibody levels are prognostic for poor clinical response and reduced overall survival in platinum-resistant or platinum-refractory ovarian cancer.

Fig. 1

MUC1 and MUC16 expression in healthy tissues and ovarian tumors according to histology. Immunohistochemical staining for MUC1 (clone HMPV, 1:400) and MUC16 (clone X325 1:500) were performed on all available formalin-fixed, paraffin-embedded tissue biopsies (n = 19). MUC1- or MUC16-expressing cells are shown in brown. a MUC1 and MUC16 are undetectable on the normal ovarian surface epithelium (OSE, top left and right panels, respectively). The endometrial glands express low levels of both MUC1 and MUC16 on the apical side, facing the lumen of ducts (lower left and right panels, respectively). b Increased MUC1 and MUC16 expression in ovarian epithelial tumors, according to histology. c Immunohistochemical staining score based on area (percentage of positive-stained cells): 0, no staining; 1+, <10% of tissue with positive staining; 2+ , 10–50% of tissue with positive staining; 3+, >50% of tissue with positive staining. All tumor sections were scored for MUC1 and MUC16, and summarized data are shown in Table 1

Fig. 2

Frequency and amplitude of MUC1- and MUC16-specific antibodies by ELISA. Serum samples from n = 28 patients were run in triplicate and average values were plotted. Early (divot) and late (narrow horizontal) IgM (left panels) and IgG (right panels) antibody response to MUC1 (a and b) and CA125 (c and d). The y axis shows the average delta OD (optical density) measurements (OD sample-OD control) at 405 nm

Fig. 3

MUC1 antibody responses by treatment response. Patients (n = 24) were classified as responders and non-responders according to criteria described in “Materials and methods”. Kruskal–Wallis tests were used to detect significant differences in mean early and late IgM (a, b) and IgG (c, d) antibody measurements by ELISA among responders and non-responders. Neither anti-MUC1 early (a) or late (b) IgM antibody responses showed significant differences between responders and non-responders. Non-responders have higher early mean IgG than responders (a, P = 0.025) and higher late mean IgG than responders (b, P = 0.022). Using statistical analysis software (SAS), schematic box-plots were made for significant associations