Currently, low-grade and high-grade serous carcinomas are thought to represent two distinct pathways of ovarian carcinogenesis, rather than opposite ends of severity along a single trajectory of tumor progression. Recent studies have convincingly demonstrated that morphological differences between these tumors are a manifestation of their underlying biological and genetic disparity. Briefly, low-grade carcinomas evolve along type I pathway and represent relatively indolent neoplasms that arise in a stepwise fashion from well-characterized precursor lesions. High-grade carcinomas are aggressive, genetically unstable neoplasms that arise through type II pathway. However, it remains an open issue whether some high-grade serous carcinomas arise from low-grade serous carcinomas that follow type I pathway [3, 4, 21].
The proposed dualistic model has important implications for early detection and targeted treatment. Current screening approaches, namely pelvic examinations, CA 125 levels and transvaginal ultrasound are reasonable for low-grade carcinomas, but are not likely to be sufficiently beneficial for high-grade carcinomas. Although the management of these two groups is currently identical, the growing body of evidence suggests that low-grade serous carcinomas are not as responsive as high-grade serous carcinomas to conventional chemotherapy with platinum and taxane agents [22–24].
A better understanding of the molecular pathogenesis of low-grade serous carcinomas would lead to rational evaluation of new targeted agents for the treatment of this disease. Reports point towards a high frequency of KRAS and BRAF mutations in low-grade OSCs, making this pathway an attractive therapeutic target by interfering with its downstream effectors [25, 26]. The preliminary promising results of a phase II clinical trial evaluating AZD6244 (selumetenib), an inhibitor of MEK-1/2, have been reported .
We report our findings of the immunohistochemical expression of p53, MAPK, topoII alpha and Ki67, and molecular analysis for KRAS and BRAF mutations in the OSCs.
p53 is a tumor suppressor gene located on the short arm of chromosome 17, involved in regulation of cell growth . Despite compelling evidence for the central role of the p53 pathway in human neoplasia, the assessment of p53 status in clinical samples remains unanswered, with confusing and often contradictory literature reports. Methodological differences in the interpretation of the staining results in different studies further contribute to the confusion [18, 28–30]. While the correlation between p53 mutational status and immunohistochemical expression is suboptimal, differences in the immunoexpression of p53 in low- and high-grade carcinomas can be diagnostically useful.
There have been a few studies investigating the p53 immunoreactivity in low- and high-grade OSCs [28, 31]. In these studies, the extent of immunoexpression was significantly different between low- and high-grade carcinomas (16.7% vs. 53.6%, and 18% vs. 64% of cases exhibited strong staining intensity). Our study confirms significantly higher p53 immunoexpression in the high-grade group (85.7% high-grade carcinomas showed >10% positive cells). In a study by Mishra et al.  22.2% of low-grade samples scored as 0 and 1+ (negative or <10% positive cells). Unlike Mishra’s study, none of our low-grade samples showed more than 10% of p53 immunopositive nuclei.
There is no definitive proof that all low-grade OSCs arise in a stepwise fashion from well-characterized precursor lesions and it is possible that some do not. Likewise, on rare occasions, a low-grade OSC may transform into a high-grade neoplasm . Several studies have shown that, in rare cases, low- and high-grade serous tumors do coexist and/or high-grade serous carcinomas share similar gene expression profile as low-grade carcinomas [14, 33].
Therefore, we upgraded the basic morphology and p53 immunoexpression with added MAPK, topo II alpha and Ki67 analysis.
MAPK is a downstream target of the RAS, RAF and MAP/ERK kinases, and is crucial for transduction of growth signals from several key growth factors, cytokines and proto-oncogenes. Mutations (including KRAS and BRAF) or overexpression of upstream components in signal transduction cascades, lead to constitutive activation of MAPK pathway . Because of the frequent KRAS or BRAF mutations in serous tumors that follow type I pathway , we examined whether there would be a differential immunoexpression of activated MAPK in our low- and high-grade group.
Nucleocytoplasmic distribution of MAPK is a pivotal point in regulation of its downstream targets. Dual phosphorylation of MAPK on tyrosine and threonine occurs in the cytoplasm. Activated MAPK must translocate into the nucleus to phosphorylate nuclear targets. Active form freely diffuses as a monomer through nuclear pores, homodimerizes and enters the nucleus via a carrier-free/nuclear pore-independent mechanism or interacts with the nuclear pore complex for entry. The nucleus has been proposed to act as an “anchoring and inactivating center” were signal must be terminated by dephosphorylation . We found nuclear and cytoplasmic MAPK in almost all positive samples, which is consistent with previous reports [19, 35]. We did not find any difference in localization of positive staining between low- and high-grade group.
In the present study we stated that the immunoexpression of activated MAPK was significantly higher in low-grade as compared to high-grade serous carcinomas. Although the literature on MAPK immunoexpression in serous ovarian tumors is quite limited, our results support findings reported by Hsu et al. .
We compared the findings of KRAS mutational analysis with active MAPK immunoreactivity. In this study, frequency of KRAS mutation was significantly higher in low-grade as compared to the high-grade group. Interestingly, none of our OSC samples had BRAF mutation. Similar findings were reported by Wong et al. , who detected BRAF mutation in only 2%, and KRAS mutation in 19% of low-grade OSCs. In contrast to our study, they did not detect KRAS or BRAF mutations in their high-grade group. We detected positive MAPK immunoexpression in some low- and high-grade samples with wild-type KRAS, suggesting that activation of MAPK pathway is not ultimately related to KRAS or BRAF mutations.
Seven morphologically high-grade samples (11.7%) showed KRAS mutation, characteristic for type I pathway and p53 immunopositivity, hallmark of type II pathway. However, due to the low number of cases, we refrain from giving a definitive answer to open issues and urge further investigation.
According to our results, unlike the ones of Hsu et al. , MAPK immunostaining was not sufficiently sensitive, nor specific, to precisely predict the KRAS mutational status of the tumor. However, MAPK immunostaining appears to be quite reliable in ruling out a KRAS mutation if the staining is negative.
Immunohistochemical expression of topoII alpha in ovarian carcinomas has been demonstrated in several studies, but the results of these studies are difficult to compare because the methodology and criteria for evaluation varied greatly [8, 9, 37, 38]. According to studies on OSCs performed by Brustmann [8, 38], the topoII alpha labeling index (LI) increased with mitotic activity (P < 0.0004), tumor grade (P = 0.0303), FIGO stage (P = 0.0076) and indicates poor prognosis (P = 0.0182). To the best of our knowledge, no study compared different topoII alpha immunoexpression with regard to proposed dualistic model of ovarian serous carcinogenesis. Based on our results, we report a significantly higher topoII alpha expression in the high-grade group compared to the low-grade group (P = 0.001).
As expected, we identified a significant difference between Ki67 immunoexpression in the low-grade and the high-grade group. The results of our study are in broad agreement with previous studies by O’Neill et al.  and Mishra et al. . Both groups have shown a lower Ki67 proliferation index in low-grade compared to high-grade OSCs.
The distinction between low- and high-grade serous carcinoma may occasionally be a differential diagnostic problem. Some high-grade serous carcinomas have been shown to mimic low-grade serous carcinomas architecturally. Many of these carcinomas have grade 2 nuclear atypia . Our results indicate that morphologically problematic serous carcinomas with markedly elevated Ki67 proliferation index and positive topoII alpha immunoexpression, are more likely to follow the type II pathway and these markers could be a useful additional tool in distinguishing the low- and high-grade groups of OSCs, along with nuclear atypia and mitotic count.
The findings of our study mostly support the proposed dualistic model of ovarian carcinogenesis. However, morphological examination combined with immunohistochemistry and molecular analyses reveal rare intersections between type I and type II tumorigenic pathway.