Long-term oncological outcomes of ovarian serous carcinomas with psammoma bodies: a novel insight into the molecular pathogenesis of ovarian epithelial carcinoma

Abstract

A two-tier system in which ovarian epithelial carcinomas are subdivided into type I and type II tumors has been proposed on the basis of recent molecular pathogenesis findings. Type I tumors, unrelated to tumor protein p53 (TP53) mutations, show favorable prognosis in a slow step-wise process, whereas type II tumors, related to TP53 mutations, contribute to poor prognosis. Ovarian serous carcinomas with excessive psammoma bodies behave like type I tumors. However, their etiology and prognostic significance remain obscure. The objective of the present study was to evaluate the characteristic features and potential relevance of psammoma bodies to the clinical outcome of 44 patients with serous carcinomas with long-term follow-up. The 5- and 10-year survival rates were significantly different between the serous carcinomas with less than 5% area of psammoma bodies and those at least 5% area (P < 0.01). All tumors with at least 5% area were both diploid and immunohistochemically negative for TP53 mutations. All patients with these tumors, including eight with International Federation of Gynecology and Obstetrics (FIGO) stages III or IV disease, survived more than 5 years and their 10-year survival rate was 76%. In multivariate analysis using clinical parameters, the apparent existence of psammoma bodies was an indication to view serous carcinomas as type I tumors with long-term survival. Our results suggested that the formation of psammoma bodies is associated with increased apoptotic tumor cell death related to normal TP53 function. The pathological findings of psammoma bodies might contribute to the consideration of pathogenesis and to the development of prognostic prediction rules for serous carcinomas.

Figure 1

Life‐table analysis of all 44 patients with ovarian serous carcinoma (Kaplan–Meier).

Figure 2

Photomicrograph of ovarian serous carcinomas with psammoma bodies. High‐grade serous ovarian carcinomas that had no psammoma bodies were classified as PB0 tumors (a) and those with rare psammoma bodies were classified as PB1 (b). Low‐grade carcinomas with 5–50% areas of psammoma bodies were classified in PB2 tumors (c). A histologically low‐grade psammocarcinoma of the ovary, showing greater than 75% areas of psammoma bodies and bland serous epithelium with moderate nuclear atypia, was classified as PB3 tumors (d) (hematoxylin–eosin stain: ×10, inset: ×40).

Figure 3

Immunohistochemical examination for tumor protein p53 (TP53). Immunohistochemical examination for TP53 showed strongly positive stain in a PB0 tumor (a) and a PB1 tumor (b), but negatively positive stain in a PB2 tumor (c) and weakly positive stain in a PB3 tumor (d) (methyl green stain: ×10).

Figure 4

Immunohistochemical findings for macrophages and apoptosis. Immunohistochemical examination of psammocarcinoma of the ovary showed that CD68‐positive macrophages mainly coincide with the areas of psammoma bodies in a PB3 tumor (a). Similarly, single‐stranded DNA was almost exclusively expressed in areas of tumor cells around the psammoma bodies in a PB3 tumor (b) (methyl green stain: ×20).

Figure 5

Life‐table analyses of the patients with ovarian serous carcinoma in relationship to psammoma bodies (Kaplan–Meier, log rank). The cumulative differences were significant between patients with PB1 and patients with PB2 tumors (P < 0.05), and between patients with PB0 and patients with PB2 tumors (P < 0.01) (a). Moreover, there were significant differences between patients with PB0 and PB1 tumors and patients with PB2 and PB3 tumors (P < 0.01) (b).

Figure 6

Life‐table analysis of the patients with ovarian serous carcinoma in relationship to tumor protein p53 (TP53) mutational status (Kaplan–Meier, log rank). The cumulative difference among the curves was significant (P < 0.01).

Figure 7

Life‐table analysis of the patients with ovarian serous carcinoma in relationship to the histological grade (low‐grade vs high‐grade) (Kaplan–Meier, log rank). The cumulative difference among the curves was significant (P < 0.01).

Figure 8

Life‐table analysis of the patients with ovarian serous carcinoma in relationship to International Federation of Gynecology and Obstetrics (FIGO) stage (Kaplan–Meier, log rank). The cumulative difference among the curves was significant (P < 0.05).

Figure 9

DNA histograms from flow cytometric analysis in ovarian serous carcinomas. Diploidy was defined as a single G0/G1 peak on the DNA histogram (a). Aneuploidy was defined as a distinct additional G0/G1 peak with a small G2/M peak on the DNA histo‐gram (b).

Figure 10

Life‐table analysis of the patients with ovarian serous carcinoma in relationship to DNA‐ploidy (Kaplan–Meier, log rank). The cumulative difference among the curves was significant (P < 0.01).

Figure 11

Multivariate analysis by type I quantification method; relationship between overall survival time and clinicopathological prognostic factors. Not only the grade of psammoma bodies (PB grade) but also classical prognostic factors had a significant influence on overall survival time (P < 0.01). FIGO, International Federation of Gynecology and Obstetrics.