Ovarian clear cell carcinoma

Incidence/mortality

In the United States, it is estimated that there will be 21,410 new diagnoses of ovarian cancer and 13,770 deaths from ovarian cancer. For these women with newly diagnosed ovarian cancer, estimated five-year survival for all stages and histologies is approximately 49%. The most recent average annual age-adjusted incidence of ovarian cancer for all Surveillance, Epidemiology, and End Results (SEER) areas is 10.9 per 100,000 women, and the average age-adjusted mortality rate is 6.7 per 100,000 women .

Ovarian clear cell carcinoma represents 5% to 10% of ovarian cancers in the United States; however, there are known geographical and racial differences in incidence. For example, there is a much higher incidence of ovarian clear cell carcinoma in Japan, Taiwan, and Korea, where ovarian clear cell carcinoma represents 15% to 25%, 19%, and 10% of ovarian cancers, respectively. In the United States, Asian/Pacific Islanders have the highest incidence rates of ovarian clear cell carcinoma (1.0 per 100,000 women), which is about twice that of women in other racial groups. Environmental and genetic factors likely play a role in development of ovarian clear cell carcinoma but reasons for variation of incidence among racial groups are unclear. Additionally, it is unknown why ovarian clear cell carcinoma is more likely to present in younger women with a median age of 55 years, compared to the more common high-grade serous counterpart, where the median age of presentation is 64 years.

Ovarian clear cell carcinoma is a rare ovarian malignancy; as a result, clinical trials are not powered to make conclusions for ovarian clear cell carcinoma despite known differences in response to treatment. Recommendations for surveillance and treatment of ovarian clear cell carcinoma are nonspecific and follow recommendations for the most common histology of ovarian cancer, high-grade serous. In this chapter we will highlight the known distinct features of ovarian clear cell carcinoma and treatment recommendations.

Etiology/risk factors

Similar to endometrioid ovarian carcinoma, ovarian clear cell carcinoma is often associated with or arises from endometriosis. Women with endometriosis have a significantly elevated age-adjusted incidence rate ratio for ovarian clear cell carcinoma (2.29, 95% CI, 1.24–4.20). A metaanalysis combining data from 13 ovarian cancer case–control studies found that self-reported endometriosis significantly increased the risk of ovarian clear cell carcinoma (OR 3.05, 95% CI, 2.4–3.8). On final pathology, endometriosis is identified in up to 51% of ovarian clear cell carcinoma cases. The association of endometriosis is further supported by the protective nature of tubal ligation against the development of ovarian clear cell carcinoma. The Ovarian Cancer Cohort Consortium found that tubal ligation and hysterectomy were associated with a significantly reduced risk of ovarian clear cell carcinoma (RR 0.35, 95% CI, 0.18–0.69; RR 0.57, 95% CI, 0.36–0.88, respectively). Theoretically, occlusion of the tubes or hysterectomy could prevent retrograde menstruation and subsequent development of endometriosis. However, the effect of endometriosis may confound these results. Most women undergo tubal ligation after childbearing is complete and women with endometriosis have higher rates of infertility. The group who did not undergo tubal ligation may have a disproportionate number of patients with endometriosis.

Endometriosis with cytologic atypia or complex hyperplasia (atypical endometriosis) is the most likely precursor to clear cell ovarian cancers. Studies have found that similar gene mutations are detected in ovarian clear cell carcinoma and adjacent atypical endometriosis. Chronic inflammation associated with endometriosis has also been identified as a potential risk factor for development of cancers, involving the intensive release of cytokines and infiltration of immune cells.

In addition to endometriosis, patients with germline mutations associated with BRCA1/2 or Lynch syndrome also have an increased risk of ovarian clear cell carcinoma. In a study of 1119 BRCA1/2 associated ovarian cancers, 2% were of clear cell histology. In Lynch syndrome, a Swedish/Danish cancer registry found that ovarian cancer subtypes differed from the sporadic population, with 35% endometrioid and 17% clear cell histology. Therefore, as in all epithelial ovarian cancers, women with ovarian clear cell carcinoma should be recommended to undergo genetic testing.

Gross

The gross appearance of ovarian clear cell carcinoma is variable with cut surface being entirely solid, entirely cystic or solid and cystic. The entirely cystic tumors are grossly not distinguishable from other benign ovarian cysts.

Microscopic findings

Histologically ovarian clear cell carcinomas are heterogeneous with papillary ( Fig. 4.2 ), tubulocystic, and solid patterns. The papillae of ovarian clear cell carcinomas are lined by cells with clear cytoplasm that have characteristic hobnail appearance, lining a hyalinized fibrovascular core. The tubulocystic pattern ( Fig. 4.3 ) is composed of numerous dilated cysts with markedly flattened epithelium that may impart a “benign” appearance on low power examination. However, on high power review, the prominent cytologic atypia is usually apparent. In ovarian clear cell carcinoma, the nuclei are variably atypical and usually round without significant pleomorphism, though bizarre cells may be seen in a subset of cases. Prominent “cherry-red” nucleoli, though not specific, are typically seen in ovarian clear cell carcinomas. While majority of ovarian clear cell carcinomas have typical cytoplasmic clearing, in some cases the cells have eosinophilic cytoplasm, and can mimic endometrioid or serous carcinoma. The mitotic activity is usually low. Ovarian clear cell carcinoma has been shown to have significant interobserver variability, even among gynecologic pathology experts, as both endometrioid adenocarcinoma and high-grade serous carcinoma can have prominent clear cell changes.

Ancillary testing

The currently available immunohistochemical (IHC) stains used to establish a diagnosis of ovarian clear cell carcinoma, are unfortunately neither sensitive nor specific, limiting their utility. The tumors are positive for PAX-8, hepatocyte nuclear factor 1 beta (HNF-1B) ( Fig. 4.4 ), napsin A ( Fig. 4.5 ), and α-methylacyl-coenzyme A racemase (AMACR, p504S), while negative for WT-1. Estrogen and progesterone receptors (ER and PR) are typically negative in ovarian clear cell carcinomas; however, focal and or diffuse weak staining can be seen in a subset of tumors. HNF-1B, frequently used to confirm ovarian clear cell carcinoma is highly sensitive but not specific, as it can be expressed in both endometrioid and serous carcinomas. Conversely, AMACR and napsin-A are more specific for a diagnosis of ovarian clear cell carcinoma but their low sensitivity results in them not being helpful in many cases.

Differential diagnosis

The differential diagnosis of clear cell carcinoma is quite vast and encompasses epithelial, germ cell and sex cord tumors, as well as metastatic tumors with clear cytoplasm. As mentioned earlier both endometrioid carcinoma and serous carcinoma can mimic ovarian clear cell carcinoma. IHC stains can show overlapping expression among ovarian clear cell carcinomas, serous carcinoma and endometrioid carcinoma, making a definite diagnosis challenging in some cases. This is further compounded by the lack of sensitivity and specificity of the available immunomarkers used to confirm a diagnosis of clear cell carcinoma. In typical cases endometrioid carcinoma and serous carcinoma will have hormone receptor expression with absent staining for HNF-1B, and napsin-A. p53 staining has limited utility in distinguishing serous carcinoma from clear cell carcinoma in isolation, as the former can rarely show wild-type staining and the latter can show p53 overexpression. An important differential is yolk sac tumor (YST), which can have histologic features that are virtually indistinguishable from clear cell carcinoma. Young age of presentation, elevated serum alpha-fetoprotein (AFP) and presence of Schiller-Duval bodies would favor YST. YSTs are positive for SALL-4, and AFP while negative for CK7, while ovarian clear cell carcinomas will show the opposite staining pattern. The solid variant of clear cell carcinoma can mimic steroid cell tumor and metastatic renal cell carcinoma. Steroid cell tumors and ovarian clear cell carcinomas have completely different immunoprofiles that will facilitate the correct diagnosis. Metastatic clear cell renal cell carcinoma (RCC) can be more challenging due to overlapping immunomarker expression such as PAX-8, napsin-A, and HNF-1B. Presence of a renal mass and positivity for RCC-specific markers will help in making the correct diagnosis.

Molecular findings

In ovarian clear cell carcinoma, the most common molecular genetic alterations are somatic inactivating mutations of AT-rich interactive domain 1A (ARID1A) , activating mutations of phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha ( PIK3CA), and deletion of phosphatase and tensin homolog ( PTEN) . ARID1A mutations are present in approximately 50% of ovarian clear cell carcinomas, but there are also reports of its presence in the adjacent foci of atypical endometriosis. This finding suggests that ARID1A mutations represent an early molecular event in ovarian clear cell carcinomas, and that endometriosis may be a precursor lesion for these tumors.

Differential diagnosis

Diagnosis of ovarian clear cell carcinoma is difficult due to the diversity of clinical presentation, nonspecific symptoms and endometriosis-associated symptoms. In contrast to the more common high-grade serous ovarian cancer, patients with ovarian clear cell carcinoma often present with unilateral, sometimes large adnexal masses and more commonly present with earlier stage disease. The differential diagnosis for an adnexal mass is quite large, including but not limited to ectopic pregnancy, physiologic/ functional cysts, polycystic ovaries, serous/mucinous cystadenomas, germ cell tumors, sex cord stromal tumors, peritoneal inclusion cysts, fibroids, endometriosis, tubo-ovarian abscess, appendiceal/diverticular abscess, colon cancer, and/or metastasis from a different primary. The workup for an adnexal mass must keep this broad differential diagnosis in mind.

Signs and symptoms

Ovarian clear cell carcinoma typically presents as a pelvic mass, often large pelvic mass. In early-stage ovarian clear cell carcinoma, initial presentation is often an asymptomatic adnexal mass discovered incidentally during workup for another condition. Some may present with pelvic pain and pressure due to a growing, typically unilateral, adnexal tumor. Clinical presentation of patients with advanced ovarian cancer varies, and symptoms are often nonspecific, such as fatigue, loss of appetite, early satiety, nausea, and anorexia, and may overlap with symptoms of concurrent endometriosis. Symptoms in advanced-stage ovarian clear cell carcinoma may be due to ascites, pleural effusions and peritoneal carcinomatosis. Because symptoms overlap with more common disorders (e.g., endometriosis, menopause, constipation, irritable bowel syndrome) diagnosis may be delayed.

Ovarian clear cell carcinoma tumors range up to 30 cm in diameter, with a mean of 13 to 15 cm. As the mass grows, it may cause bulk symptoms. Urinary complaints are common including urinary urgency and frequency. As these tumors expand posteriorly, they can compress the colon, leading to constipation and pain. Pelvic and abdominal symptoms include bloating, and diffuse, dull, constant abdominal pain. Adnexal masses may also present with acute onset abdominal pain, which may be associated with rupture, torsion, or bowel obstruction.

Deep venous thrombosis (DVT) may develop due to the hypercoagulability associated with advanced cancer, or the tumor pressing on the pelvic veins. Of the epithelial ovarian cancers, ovarian clear cell carcinoma are most commonly associated with DVT which occur in up to 42% of cases. Ovarian clear cell carcinoma expresses high levels of tissue factor, a transmembrane protein that is associated with hypercoagulability. Complaints of lower extremity swelling/pain, shortness of breath, and pleuritic chest pain must be evaluated thoroughly due to elevated risk of DVT and pulmonary embolism. Hypercalcemia is observed at a higher frequency in ovarian clear cell carcinoma compared to other ovarian cancers due to production of parathyroid hormone-related protein.

There is no evidence that screening for symptoms aids in recognizing ovarian cancers at earlier stage or improves detection with use of CA-125 or ultrasound. Early referral to a gynecologic oncologist for patients with symptomatic, persistent or growing adnexal masses, ascites, or evidence of abdominal/ distant metastases is imperative.

Physical exam findings

Physical exam findings are similar for benign and malignant adnexal masses. It is nearly impossible to identify a malignant mass based on clinical exam alone. A comprehensive exam is essential, including evaluation of extremities, lungs, supraclavicular nodes, and breasts. Lower extremity edema, tenderness, or a palpable cord is suggestive of a DVT. Abdominal exam should include palpation and percussion. If the tumor produces a large amount of ascites, the abdominal cavity may be distended, causing increased abdominal circumference and marked discomfort. Ascites can also cause dyspnea, as the lower lungs are compressed by abdominal distension.

On pelvic exam, a bimanual and rectovaginal exam should be performed. Visually, a cystocele may indicate the presence of ascites. During pelvic exam, the clinician should note the size, borders, mobility, and location of the mass. Invasive cancers are typically large, fixed, and have irregular borders. Exam may also reveal involvement of the parametrium or nodularity of the rectovaginal septum, also known as a Blumer’s shelf.

Tumor markers

The most studied tumor marker associated with epithelial ovarian cancer is CA-125. Pretreatment CA-125 levels are elevated in the majority of patients with ovarian clear cell carcinoma, but levels do not predict clinical outcome. A retrospective case–control study of 375 patients with ovarian clear cell carcinoma, found that pretreatment CA-125 levels were not associated with relapse-free survival (RFS) or overall survival. Pretreatment CA-125 is normal in 30% of early-stage and 13% of advanced-stage ovarian clear cell carcinoma. The majority of ovarian clear cell carcinomas are diagnosed at early stage so elevated CA-125 levels are not a reliable marker of malignancy in these patients ( Table 4.1 ).

For advanced-stage ovarian cancers, CA-125 values are highest for patients with serous histology with a median of 2000 U/mL; patients with clear cell histology have lower CA-125 values with a median of 154 U/mL (normal < 35 U/mL). CA-125 normalization after initiation of chemotherapy is correlated with overall survival and is likely a surrogate for inherent chemosensitivity. A retrospective review of seven previously reported GOG phase three trials in patients with stage III/IV ovarian clear cell carcinoma found that changes in CA-125 at the end of treatment compared to baseline can serve as an indicator of progression-free survival (PFS) and overall survival.

It should be noted that elevated CA-125 levels may be false positives. For example, other gynecologic conditions that cause peritoneal or pelvic inflammation can elevate CA-125 levels, including endometriosis. Meanwhile, conditions that lead to ascites, such as cirrhosis or heart failure may also elevate the CA-125. Overall, tumor markers can be useful in monitoring treatment and surveillance of ovarian clear cell carcinoma, but clinical history remains integral to diagnosis.

Imaging

Imaging findings associated with epithelial ovarian cancer include partially solid and cystic components to the mass, irregularity of borders, and fixed location. The three main imaging modalities used for ovarian clear cell carcinoma are ultrasound, CT scan, and MRI. Imaging findings particular to ovarian clear cell carcinoma include unilocular or large cysts with solid protrusions into the cavity. Due to association with endometriosis, large endometriomas with solid components are highly suspicious for ovarian clear cell carcinoma. On CT scan, ovarian clear cell carcinoma margins are typically smooth, solid protrusions are rounded and few in number with high-attenuated cystic or necrotic portions. On T1-weighted MRI images, signal intensity varies from low to very high. Imaging can help identify patients in whom aggressive initial surgical cytoreduction may not be the most beneficial primary treatment option, as they have a lower chance of optimal cytoreduction. Preoperative imaging can also identify disease in unresectable locations.