Immunohistochemical evaluation of epithelial ovarian carcinomas identifies three different expression patterns of the MX35 antigen, NaPi2b

Abstract

Background: To characterize the expression of the membrane transporter NaPi2b and antigen targeted by the MX35 antibody in ovarian tumor samples. The current interest to develop monoclonal antibody based therapy of ovarian cancer by targeting NaPi2b emphasizes the need for detailed knowledge and characterization of the expression pattern of this protein. For the majority of patients with ovarian carcinoma the risk of being diagnosed in late stages with extensive loco-regional spread disease is substantial, which stresses the need to develop improved therapeutic agents.

Methods: The gene and protein expression of SLC34A2/NaPi2b were analyzed in ovarian carcinoma tissues by QPCR (n = 73) and immunohistochemistry (n = 136). The expression levels and antigen localization were established and compared to the tumor characteristics and clinical data.

Results: Positive staining for the target protein, NaPi2b was detected for 93% of the malignant samples, and we identified three separate distribution patterns of the antigen within the tumors, based on the localization of NaPi2b. There were differences in the staining intensity as well as the distribution pattern when comparing the tumor grade and histology, the mucinous tumors presented a significantly lower expression of both the targeted protein and its related gene.

Conclusion: Our study identified differences regarding the level of the antigen expression between tumor grade and histology. We have identified differences in the antigen localization between borderline tumors, type 1 and type 2 tumors, and suggest that a pathological evaluation of NaPi2b in the tumors would be helpful in order to know which patients that would benefit from this targeted therapy.

Keywords: Monoclonal antibody; NaPi2b expression; Ovarian cancer; Radiotherapy.

Fig. 1

Boxplots illustrating the level of SLC34A gene expression. a) level of expression in relation to the different histologies, with significant differences in expression between mucinous and both clear cell and serous (**P < 0.01), and significant differences between endometrioid and serous tumors (*P < 0.05,); b) in relation to grade and in c) to stage. d) Boxplot illustrates the correlation between the staining of NaPi2b and the gene expression of SLC34A, Pearson correlation r = 0.302*

Fig. 2

Illustrative images of the staining intensities and the distribution of the different intensities among the samples. a) Representative images demonstrating the different staining intensities Upper left: no staining = 0, serous (highly differentiated stage I), Upper right: 1 = weak staining (endometrioid poorly differentiated, stage II). Lower left: 2 = moderate staining (endometrioid poorly differentiated stage III. Lower right: 3 = strong staining, serous poorly differentiated stage III). b) Malignant and borderline tumor samples divided in scored staining intensity. c) Bars illustrating the samples divided into histology and how the staining intensities were distributed within their histological group

Fig. 3

Results presented in relation to tumor type 1 and 2. a) Upper panel present the tumors according to type and intensity of staining, b) The tumors are grouped according to type and presented by the pattern they present

Fig. 4

Illustration presenting the three indicated staining patterns and their distribution among the samples. a) Illustration of the three characteristic MX35 staining patterns identified among the samples; Pattern a, the target was located in the cell membranes of the cells close to the surface of the tumor (serous borderline, stage I). Pattern b, MX35 staining in all of the tumor cell membranes over the entire tumor (clear cell highly differentiated, stage I). Pattern c, includes the tumors with a mixed staining pattern including staining of membranes in addition to cytoplasmic staining spreading inside the as well (serous poorly differentiated stage I). The right panel shows images with higher magnification to give a more detailed view of the three patterns. b) The distribution of the tumor samples, malignant and borderline, between the three patterns. c) The tumors divided based on pattern, presented according to their histology