Abstract
Background/Aim: Histological changes induced by neoadjuvant chemotherapy (NAC) have rarely been reported in histological subtypes other than ovarian high-grade serous carcinoma (HGSC). Case Report: We report a 49-year-old woman whose tumors in interval debulking surgery (IDS) specimen exhibited prominent papillary growth pattern and severe nuclear pleomorphism due to NAC. In the initial microscopic examination, ovarian HGSC was the most likely candidate; however, immunostaining results were not compatible with HGSC. We detected areas resembling mucinous cystadenoma and borderline tumor, and finally diagnosed this case as ovarian mucinous carcinoma. Conclusion: Although the tumor mimicked histologically HGSC, its clinical features differed from those of advanced-stage HGSC. It is important for pathologists to recognize NAC-induced histological changes, be aware of the diagnostic mimics and pitfalls, and to identify the correct histological subtype by considering the patient’s previous history, clinical features, preoperative imaging findings, and histological features.
Ovarian carcinoma is the most lethal gynecological malignancy and the fifth most common cause of cancer-related deaths in women (1-6). It is classified into type 1 tumors, which include low-grade serous carcinoma, endometrioid carcinoma, clear cell carcinoma, and mucinous carcinoma, and type 2 tumors, which include high-grade serous carcinoma (HGSC), based on molecular and clinicopathological differences (7). Most ovarian carcinomas are HGSCs (8-11). Since most HGSCs are diagnosed in the advanced stage, HGSC patients have a poor prognosis despite surgery and chemotherapy. There has been a recent shift towards increased use of neoadjuvant chemotherapy (NAC) and interval debulking surgery (IDS) for patients with advanced-stage HGSC, where achieving optimal cytoreduction with primary debulking surgery is difficult (12, 13). NAC has been associated with increased rates of optimal debulking, no residual disease after debulking surgery, and decreased surgical morbidity and mortality (14).
While HGSC is the most frequent histological subtype, ovarian mucinous carcinoma has an incidence of approximately 3% (15, 16). However, primary mucinous carcinoma of the ovary is more common in Korea (17). It appears to evolve in a stepwise fashion from the benign epithelium and mucinous cystadenoma through a preinvasive lesion to carcinoma (18); its prognosis is better in early stage and worse in advanced stage than the prognosis of HGSC patients, which is mainly due to inadequate response to platinum-based chemotherapy (19).
Ovarian carcinoma cells display various morphological alterations induced by NAC. In a morphological and immunohistological study regarding post-chemotherapy ovarian carcinomas (20), eight of the 16 cases exhibited significant morphological differences between pre- and post-chemotherapy tissue samples. Tumor cells affected by chemotherapy were characterized by both nuclear and cytoplasmic alterations. Nuclear alterations included severe pleomorphism with hyperchromasia, membrane irregularity, and chromatin clumping or smudging. Degenerative chromatin patterns were observed, and bizarre multinucleated cells were frequently noted. Cytoplasmic alterations exhibited voluminous change, exuberant clear cell and foamy changes, deeply eosinophilic intracytoplasmic inclusions, and micro- and macrovacuoles. Although tumor cells showed high-grade cytological atypia, the nuclear-to-cytoplasmic ratio was low. Moreover, there were foci of coagulative tumor cell necrosis, fibrosis, hemorrhage, and degeneration (21). In contrast, the immunophenotype of post-chemotherapy tumors was similar to that of pre-chemotherapy tumors. If the ovarian tumor presents a significant response to NAC and displays morphological alterations, it is difficult for pathologists to determine the histological type of post-chemotherapy tumors.
Since NAC is generally administered to HGSC patients, NAC-induced histological changes in histological types other than HGSC have rarely been reported. We recently experienced an ovarian mucinous carcinoma showing NAC-induced histological changes in interval debulking surgery (IDS) specimens. Since the tumor showed severe nuclear pleomorphism and prominent papillary growth patterns, there was a possibility of being mistaken for HGSC. Immunostaining was performed for accurate determination of the histological type. The pathological features of ovarian mucinous carcinoma after NAC have not been detailed previously; thus, we aim to provide a clinicopathological description of these tumors. Pathologists should be aware of the resultant morphology because NAC is likely to be used more frequently in the management of ovarian carcinoma.
Case Report
This study (2021-01-008) was reviewed and approved by the Institutional Review Board of Samsung Medical Center (Seoul, Republic of Korea). A 49-year-old woman presented with abdominal distension, with no previous gynecological history. Abdominopelvic magnetic resonance imaging revealed a huge cystic mass in the right ovary with several solid components along the inner aspect of multilocular cystic lesions. The mass measured 16 cm in the greatest dimension (Figure 1A). Although small, multiple peritoneal nodular thickening, radiologically suggestive of peritoneal seeding, was observed in the right paracolic gutter and hepatic dome, with no discrete mass formation. No significant lymphadenopathy was observed in the pelvic and para-aortic regions, but there were some enlarged lymph nodes in the mesentery and mid-abdomen, suspicious for metastatic lymphadenopathy. The serum cancer antigen (CA) 125 levels were elevated up to 297.5 U/ml. The ovarian mass was too huge for performing colonoscopy and esophagogastroduodenoscopy for preoperative clinical work-up. Under the preoperative impression of ovarian carcinoma, the patient received three cycles of preoperative NAC with paclitaxel and carboplatin for two months.
After NAC, abdominopelvic computed tomography revealed a huge ovarian cystic mass, measuring 19 cm (Figure 1B). The number of solid components contained in the mass decreased significantly compared to that before receiving NAC but did not disappear completely. The greatest dimension of the post-NAC mass increased because of the increase in the number of necrotic debris and fluid inside cystic lesions. The extent of suspected peritoneal seeding also diminished; however, the possibility of viable metastatic tumors was likely to remain. The enlarged lymph nodes observed in the mesentery and mid-abdomen were not observed after NAC. By combining these imaging findings, it was determined that the partial resolution of ovarian carcinoma remained. Post-NAC levels of serum CA 125 were normalized (25.0 U/ml). IDS was performed for the residual ovarian carcinoma.
Intraoperatively, a 19-cm right ovarian cystic mass was identified. The left ovary and salpinx, uterus, and omentum were unremarkable. The pelvic and abdominal peritoneum did not show a remarkable lesion. There were a few foci of fibrous adhesion between the uterine serosa and posterior cul-de-sac peritoneum and rectosigmoid colon, but those appeared to be due to peritoneal endometriosis rather than metastasis. No lymph node enlargement was noted. Total hysterectomy with bilateral salpingo-oophorectomy, total omentectomy, bilateral pelvic lymph node dissection, and intraoperative peritoneal washing cytology were performed.
Gross examination of the resected specimen revealed a huge cystic mass of the right ovarian, the wall of which was irregularly thickened. The cyst was filled with serous fluid. The inner surfaces of the cystic lesions possessed variable-sized, well-formed, solid papillae; some of them were fused to each other, measuring up to 3.0 cm (Figure 1C). Histologically, the tumor exhibited predominantly papillary architecture (Figures 2A and B). At low-power magnification, the tumor showed a very similar appearance to HGSC (Figure 2C). Several papillary structures were observed with the dilated cystic lumina, but no overt destructive infiltration into the cyst wall was identified. The tumor cells displayed very large and markedly atypical nuclei (Figure 2D and E) and voluminous eosinophilic granular cytoplasm (Figure 2F). Floating inflammatory cells, fibrin, and necrotic debris were present between the papillae (Figure 2G). Mitotic figures were frequently noted and were counted to 24 per 10 high-power fields. Atypical mitotic figures were also occasionally detected (Figure 2H). Single or multiple conspicuous nucleoli were easily identifiable. Intranuclear inclusions (Figure 2I) and intracytoplasmic microvacuoles were present. Most tumor cells displayed low nucleus-to-cytoplasm ratio. Although intracytoplasmic mucins were observed in some tumor cells, we found small amounts of intraluminal and intracytoplasmic mucins and infrequent cytoplasmic clearing in HGSCs (22). HGSC was diagnosed based on dominant papillary configuration, severe nuclear pleomorphism, and brisk mitotic activity.
We performed immunostaining for paired box 8 (PAX8), Wilms’ tumor 1 (WT1), estrogen receptor (ER), progesterone receptor (PR), p53, and p16 using the most representative section (Figure 3A). PAX8 was focally positive with moderate-to-strong intensity (Figure 3B). The tumor cells were negative for WT1 (Figure 3C), and they were also negative for both ER and PR (Figure 3D). p53 expression was completely absent in the entire tumor tissue (Figure 3E), indicating a nonsense frameshift mutation of the TP53 gene. Patchy p16 expression was observed throughout the tumor (Figure 3F). The lack of WT1, ER, and PR expression excluded the possibility of HGSC and endometrioid carcinoma. Following immunostaining, we carefully reviewed hematoxylin and eosin (H&E)-stained slides. In some areas of the lining epithelium, we observed benign mucinous epithelium (Figures 4A-C) and epithelial proliferation (Figures 4D and E), resembling mucinous borderline tumors. Several foci showing intraepithelial carcinoma were also present (Figure 4F). We considered that these lesions were overlooked during the initial microscopic examination. The presence of a morphological spectrum of ovarian mucinous tumors (mucinous cystadenoma, mucinous borderline tumor, and intraepithelial carcinoma associated with the mucinous borderline tumor) and the lack of WT1 expression was supportive for the diagnosis of mucinous carcinoma. The final diagnosis was primary ovarian mucinous carcinoma of the expansile invasive pattern showing an unusual growth pattern.
Discussion
In addition to ovarian carcinoma, NAC has been utilized for the management of other malignancies, including carcinomas of the breast, lung, stomach, and rectum (23-27); its effects on the histomorphology of these tumors have been well documented (12, 13, 20, 21). Chemotherapy-induced morphological changes in the ovary have been described in several studies. The alterations seem to parallel those observed in other malignancies, including the presence of multinucleated giant cells, stromal fibrosis, inflammation, hemorrhage, and coagulative necrosis. Nuclear alterations include enlargement with irregularities of the membrane, chromatin smudging and clumping, multinucleation, and degenerative chromatin appearance. Cytoplasmic changes include vacuolization, eosinophilic inclusion, clear to foamy change, and granularity. However, previous studies on the histomorphological changes of mucinous carcinoma induced by NAC were difficult to find because of the low prevalence of the tumor itself and the fact that most NACs were performed for HGSC patients.
A prominent papillary architectural pattern is not commonly observed in ovarian mucinous carcinoma. The mucinous carcinoma showing expansile (confluent) invasion pattern is histologically characterized by complex and extensive proliferation of mucin-producing epithelial cells, accompanied by little or no intervening stroma. Although papillary or villoglandular growth patterns can be observed in focal areas, those patterns do not constitute the dominant portion of the tumor tissue. In our case, most of the tumor tissue showed intraluminal papillary proliferations lining variable-sized cystic spaces instead of exhibiting typical histology of mucinous carcinoma. Since mixed solid and cystic tumor masses composed of papillary structures of various sizes are the typical histological features of HGSC, it is not unusual that HGSC was the most likely candidate for our case until the immunostaining results were known; it is thought that this type of confusion is likely to occur during ovarian carcinoma diagnosis. Severe nuclear pleomorphism coexisting with prominent papillary configuration, multifocal necrosis, fibrosis, and old hemorrhage were sufficiently supportive of NAC-induced histological changes observed in HGSC. However, since WT1 expression was negative, after ruling out the possibility of HGSC, a thorough review of H&E slides revealed a benign-to-borderline degree of mucinous epithelial proliferation in some areas of the cystic lining. Although the lesion was small, it was overlooked during the initial microscopic examination. The presence of the morphological spectrum of ovarian tumors from mucinous cystadenoma to intraepithelial carcinoma associated with mucinous borderline tumor were reasonably supportive of mucinous carcinoma diagnosis. Although intracytoplasmic mucins can be observed in a small number of tumor cells in HGSCs, such cases do not show the typical morphology of benign mucinous epithelium and mucinous borderline tumors.
The reason for the dominance of the papillary pattern in our case, which is not typical for mucinous carcinoma, is presumed to be because the residual viable tumor tissue showed an unusual growth pattern after NAC-induced histological alterations. While most of the residual carcinoma cells showed high-grade nuclear atypia, benign mucinous cystadenoma and mucinous borderline tumor components maintained their typical morphology after NAC, which indicates the possibility of a difference between the malignant and non-malignant components in terms of response to NAC.
Although our case showed histological features similar to HGSC, the clinicopathological features that can exclude the possibility of HGSC are as follows: First, variable-sized cystic lesions were observed in the tumor tissue, and destructive infiltration was not detected. Second, although small, benign mucinous cystadenoma was observed with mucinous borderline tumor components. Third, WT1, which is expressed in almost all the cases of HGSCs, was negative (28). Fourth, HGSC cases are mostly presented in an advanced stage (9, 10, 19), whereas our case showed a huge mass involving the unilateral ovary only, without surface extension. No evidence of nodal involvement, distant metastases, and peritoneal seeding was observed other than the primary ovarian tumor. Although there are some cases of ovarian mucinous carcinoma with extensive peritoneal carcinomatosis, most mucinous carcinoma cases are presented in the early stage (29).
We performed immunostaining for the differential diagnosis. Since there was no tumor sample before NAC treatment, the immunophenotype could not be compared between pre- and post-NAC tissue samples. However, previous studies have reported that the immunophenotype of post-chemotherapy ovarian carcinomas was very similar to that of untreated tumors, illustrating that WT1, ER, p53, and p16 expression may be of value in determining the histological type (20). In our case, the lack of WT1 expression excludes the possibility of HGSC. PAX8, a sensitive marker for diagnosing the carcinomas of Mullerian origin, is expressed in approximately 50% of ovarian mucinous carcinomas (30, 31). The fact that PAX8 was positive in our case supports the findings of the primary ovarian tumor. The mutant pattern of p53 expression is a characteristic finding observed in approximately 95% of HGSCs but can also be observed in about two-thirds of ovarian mucinous carcinomas (32).
In summary, we described the clinicopathological characteristics of ovarian mucinous carcinoma after NAC treatment. Ovarian mucinous carcinoma showing the prominent papillary architectural pattern and severe pleomorphism could be mistaken for HGSC. The histological features were very similar to those of HGSC, indicating NAC-induced morphological changes; however, their clinical features were different from those of advanced-stage HGSCs. The fact that there was no nodal or distant metastasis or abdominopelvic peritoneal seeding other than primary ovarian tumor was a finding that was suitable for mucinous carcinoma presented in the early stage during its discovery. However, as in our case, when the peritoneal metastatic disease accompanied by a huge mass is suspected in the initial image, the diagnosis of mucinous carcinoma will be difficult, and NAC may likely be performed. Additionally, as the frequency of applying NAC for the management of advanced ovarian carcinoma will increase, it is important for pathologists to recognize NAC-induced histological changes and be aware of the diagnostic mimics and pitfalls. To establish an accurate diagnosis, particularly identifying the correct histological subtype, is one of the most crucial factors in desining an appropriate treatment plan for the patient. It is important to determine the histological type by considering the patient’s previous history, clinical features, preoperative imaging findings, and histological features.
Acknowledgements
This research was supported by the research grant funded by Lunit Inc. (Seoul, Republic of Korea) and by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2018R1D1A1B07047640).
Footnotes
Authors’ Contributions
All Authors made substantial contributions to the conception and design of the study; the acquisition, analysis, and interpretation of the data; drafting of the article; critical revision of the article for important intellectual content; and the final approval of the version to be published.
Conflicts of Interest
The Authors declare that they have no conflicts of interest in relation to this study.
- Received January 22, 2021.
- Revision received February 2, 2021.
- Accepted February 3, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.