Ovarian serous surface papillary borderline tumor: characteristic imaging features with clinicopathological correlation

Abstract

Objective:

To retrospectively evaluate the clinical, pathological, and imaging features of ovarian serous surface papillary borderline tumors (SSPBTs).

Methods:

Imaging features were analyzed, including the tumor size, laterality, tumor spread and the presence of ascites. Morphological nature (encased normal ovary, characteristics on MRI) and contrast enhancement (increased flow on Doppler ultrasound) were also evaluated. Clinical and pathological features, such as tumor markers (CA 125), treatment methods, outcomes on follow-up, and surgical staging, were analyzed.

Results:

10 tumors in 5 patients were evaluated. Mean largest tumor diameter was 45 mm. All patients had bilateral involvement and ascites. 40% of all patients showed peritoneal implants. 20% of patients evaluated had lymph node metastasis. These patients showed grossly normal ovaries that were encased in or surrounded by irregular solid tumors. They had a mostly hyperintense papillary architecture with hypointense internal branching on T₂ weighted MR images (90%). Contrast enhancement and serum levels of CA 125 were elevated in all cases evaluated. All patients underwent radical surgery, and 80% of patients evaluated were of low stage. No recurrence occurred, during follow-up.

Conclusion:

SSPBT of the ovary, which has a good prognosis, should be considered as a diagnosis for patients who have bilateral enhancing irregular solid masses with papillary architecture and internal branching, and encasing normal-appearing ovaries.

Advances in knowledge:

Serous surface papillary borderline tumor of the ovary is unique and has characteristic features. Knowledge of this specific ovarian tumor and radiological suspicion can have important implications for the patient to facilitate management including fertility-preserving surgery.

Introduction

Ovarian serous tumors are classified as benign (60%), borderline (15%), or malignant (25%) according to their biological nature.^{1, 2} With a low malignant potential, borderline tumors histologically demonstrate mitotic activity and nuclear atypia that are intermediate to those of their clearly benign and overtly malignant counterparts, but without stromal invasion.^{2, 3} The clinical importance of borderline tumors is that they have a better prognosis than that of their malignant counterparts,^{4, 5} even with intraperitoneal dissemination or lymph node metastasis.^{4, 6} The reported 5- and 10-year survival rates are 95 and 85%, respectively.^5–8 Because of the younger age of onset, earlier stage at presentation, and better prognosis, patients can undergo conservative surgery with unilateral salpingo-oophorectomy to preserve fertility.^{3, 5,8,9} Accurate characterization using pre-operative imaging is important in surgical planning.

The serous subtype of epithelial tumors is classified as cyst-forming, surface proliferative, or stromal proliferative, based on macroscopic proliferation. Accordingly, there are three types of serous tumors: serous cystic tumors, serous surface papillary tumors, and serous cystadenofibromas.² Serous surface papillary tumors are a distinct subtype of serous tumors of the ovary, in which the tumor is confined to the ovarian surface, and the normal size and shape of the ovary are preserved. They display a surface proliferative pattern and papillary excrescences on their surface.²

Only a few imaging studies regarding ovarian serous surface papillary borderline tumors (SSPBTs), a subtype of serous borderline tumors, have been reported.^2,3,9–14 The purpose of this study was to evaluate the clinical, pathological, and imaging features of five ovarian SSPBTs.

Methods and Materials

This retrospective study was approved by our institutional review board. The need for informed consent was waived.

Patient population

Five patients with pathologically confirmed ovarian SSPBTs were identified by a computerized search (November 2009 to July 2015) of the pathology databases from two hospitals (blinded for review). The mean patient age was 38.4 years (range, 27‒60 years). We retrospectively reviewed medical records.

Ultrasound, CT, and MRI examinations

Four examinations, both transvaginal and transabdominal ultrasound, in two of our patients were performed. Color or power Doppler ultrasound was used in both patients. Two CT examinations were performed in one of our patients. 10 MRI examinations were performed among our 5 patients.

Thus, the following imaging studies were available for analysis: 4 ultrasound examinations for 4 tumors in 2 patients, 2 CT examinations for 2 tumors in 1 patient, and 10 MRI examinations for 10 tumors in 5 patients.

Analysis of imaging features

Imaging features in our five patients were evaluated by consensus between two abdominal radiologists (with 14 years and 15 years of clinical experience in the interpretation of body images, respectively).

The following general imaging findings were analyzed from the available data (cross-sectional high-quality objective imaging modalities such as MRI or CT had priority over ultrasound): tumor size, laterality (unilateral or bilateral), and tumor spread and presence of ascites. Tumor size was assessed as the largest tumor diameter, measured along three dimensions. Tumor spread was analyzed in terms of peritoneal implants or deposits, lymph node metastasis, and hematogenous metastasis. Peritoneal implant or deposit positivity was considered if any nodule or calcification was observed on the surface of the intraperitoneal organ or parietal peritoneum on MRI or CT.² Lymph node metastasis positivity was considered if the short axis of the lymph node was > 6 mm on MRI or CT.²

Ultrasound-specific imaging findings were analyzed in terms of the morphological nature and relationship of the tumor with the normal ovary (tumor encasing or surrounding the normal ovary or not).^{9, 12,13} Doppler ultrasound findings were also evaluated with regard to intralesional vascularity, with subjective assessment of the presence or absence of intralesional vessels according to the International Ovarian Tumor Analysis (IOTA) color score (1: no blood flow, 2: minimal flow, 3: moderate flow, 4: highly vascular with marked blood flow).¹⁵

CT- or MRI-specific imaging findings were analyzed for the following features: morphological nature, relationship of the tumor with normal ovary (tumor encasing or surrounding the normal ovary or not), and contrast enhancement (present or not). On T₂ weighted MRI images, tumors were classified in terms of papillary architecture and internal branching, papillary architecture and internal branching with cysts, and cysts with mural nodules.²

Analysis of clinical features

The analyzed clinical features included tumor markers (CA 125 level), treatment methods (radical or conservative surgery), and outcomes after >12 months follow-up (disease-free, tumor recurrence, or metastasis).

Analysis of pathological features

Surgically resected specimens were examined and reviewed. The imaging features were correlated with the surgical and pathological features based on medical records with regard to tumor morphology, invasion, and metastasis on gross and microscopic assessment. The surgical International Federation of Gynecology and Obstetrics (FIGO) stages were determined.

Results

Imaging features

10 tumors in 5 patients were assessed. The imaging features are summarized in Table 1. The mean largest tumor diameter was 45 mm (range, 30–67 mm). All patients had bilateral involvement and ascites. Two (40%) had a peritoneal implant or deposit. One (20%) showed lymph node metastasis. No tumors showed hematogenous metastasis.

Table 1.

Imaging feature analysis of SSPBT (10 tumors in 5 patients)

Parameter	No. (%)
Generalimagingfeatures
Bilateral	5/5 (100%)
Ascites	5/5 (100%)
Peritoneal implant or deposit	2/5 (40%)
Lymph node metastasis	1/5 (20%)
Ultrasound features
Solid tumor encased normal ovaries	4/4 (100%)
Intralesional vascularity	4/4 (100%)
CT features
Solid tumor suspicious encased normal ovaries	2/2 (100%)
Enhancement	2/2 (100%)
MRI features
Solid tumor encased normal ovaries	10/10 (100%)
Papillary architecture and internal branching	9/10 (90%)
Enhancement	10/10 (100%)

SSPBT, serous surface papillary borderline tumor.

On ultrasound, all four tumors evaluated showed grossly normal ovaries that were encased in or surrounded by irregular solid tumors (Figure 1a and b). All four tumors examined by color or power Doppler ultrasound demonstrated minimal or moderate degrees of intralesional vascularity (IOTA score 2 or 3) (Figure 1c).

Figure 1.

A 27-year-old female patient with bilateral ovarian SSPBTs with ultrasound, MRI, and histopathological features (reprint from reference 12 with permission). (a) and (b) Gray-scale transvaginal ultrasound for both adnexal masses (a: right, b: left) showed lobulated heterogeneous echoic masses (arrows) encasing or surrounding a presumably normal ovarian follicle (arrowheads). (c) Doppler ultrasound of the right ovary showed increased vascular flow within the mass. The IOTA color score is 3 (moderate flow). (d) Coronal T₂ weighted MRI of the pelvis showed bilateral ovarian masses that were entirely multilobulated with hyperintense papillary architecture on the surface (arrow) and hypointense internal branching patterns (arrowheads). The masses involved the surface of both ovaries, but the otherwise normal-appearing ovaries with multiple follicular cysts and a distorted shape were discriminated clearly. (e) Gross findings, obtained via laparoscopy showed exophytic proliferation (arrows) from the surface of the right ovary. (f) Photomicrograph of a histological specimen showing a papillary serous borderline tumor with branching. The fibrous papillae were covered by proliferating epithelial cells. Low-grade cytological atypia is noted. Stromal invasion was absent, but stromal edema can be seen under the epithelium (hematoxylin–eosin staining, original magnification ×100). IOTA, International Ovarian Tumor Analysis; SSPBTs, serous surface papillary borderline tumors.

On CT, the two tumors evaluated showed solid ovarian masses that were suspected of surrounding the normal ovary. The two tumors also showed contrast enhancement (Figure 2a).

Figure 2.

A 47-year-old female patient with bilateral ovarian SSPBT with CT and MRI features. (a) Enhanced CT of the pelvis showed solid bilateral adnexal masses with homogeneous enhancement (arrows). Normal ovaries observed within the masses were not definite, but suspicious. (b) Axial T₂ weighted MRI of the pelvis showed bilateral ovarian masses that were entirely multilobulated with hyperintense papillary architecture on the surface (arrows) and hypointense internal branching patterns (arrowheads). SSPBTs, serous surface papillary borderline tumors.

On MRI, all 10 tumors were irregular solid tumors encasing or surrounding the grossly normal ovary. On T₂ weighted images, papillary architecture and internal branching were noted in nine (90%; Figures 1d, 2b, 3a) of the 10 tumors; papillary architecture and internal branching with cysts were noted in one (10%). All 10 tumors showed contrast enhancement (Figure 3b).

Figure 3.

A 47-year-old female patient with bilateral ovarian SSPBTs with MRI features. (a) Axial T₂ weighted MRI of the pelvis showed bilateral ovarian masses that were entirely multilobulated with hyperintense papillary architecture on the surface (arrows) and hypointense internal branching patterns (arrowheads). (b) Enhanced axial T₁ weighted MRI showed the same enhancement pattern in the bilateral ovarian masses (arrows) as that in the ovaries. SSPBTs, serous surface papillary borderline tumors.

Clinical features

The clinical features are summarized in Table 2.

Table 2.

Clinical and pathological features analysis of SSPBT (five patients)

Parameter	No. (%)
Clinical features
Elevated serum CA 125	5/5 (100%)
Radical surgery	5/5 (100%)
Conservative surgery	0/5 (0%)
Adjuvant chemotherapy	2/5 (40%)
Disease-free on follow-up	5/5 (100%)
Tumor recurrence or metastasis on follow-up	0/5 (0%)
Pathological features
Stage I	3/5 (60%)
Stage II	1/5 (20%)
Stage III	1/5 (20%)

SSPBT, serous surface papillary borderline tumor.

The serum level of CA 125 (normal, <35 U ml⁻¹) was elevated in all five cases. The mean level was 132.56 (range, 87.32‒242) U ml^–1.

All five patients underwent radical surgery, such as bilateral salpingo-oophorectomy in combination with hysterectomy with peritoneal washing, omentectomy, multiple peritoneal biopsies, and complete resection of all macroscopic suspected lesions. Two (40%) of the five patients underwent adjuvant chemotherapy. On follow-up, all five patients were disease-free, even at least 24 months.

Pathological features

The pathological features are summarized in Table 2. Histopathological diagnosis revealed bilateral involvement in all five patients that were detected on pre-operative imaging. Gross findings revealed multiple reddish, foamy nodules presenting as exophytic growths from the ovarian surface (Figure 1e). The final histopathological findings were numerous micropapillae lined by one or two layers of cuboidal cells with nuclei, suggesting a diagnosis of SSPBT (Figure 1f).

Post-operative staging, according to the FIGO criteria, was I in three (60%), II in one (20%), and III in one (20%) of the five patients.

Discussion

In this series, SSPBTs presented as soft tissue masses with surface papillary projections and characteristic imaging features, in which the ovaries appeared grossly normal but were covered or encased with tumor deposits confined to the surface. Ovarian tumors with rich solid tissue on imaging are generally considered malignant.^2,11–13 According to this principle, an ovarian SSPBT might be diagnosed as a highly malignant tumor, because it forms entirely solid masses.^{12, 13} The important finding that differentiates SSPBTs from other malignant lesions is the presence of normal ovaries, or at least some normal ovarian tissue, completely or partially surrounded by the solid tumor (Figures 1‒3).^{9, 12,13} Thus, awareness of this disease entity is very important in the diagnosis and determination of appropriate treatment options.⁹ Prior knowledge of the tumor type and the presence of normal ovaries, or at least some normal ovarian tissue, is also likely to facilitate attempts to conserve the ovaries during surgery.⁹

There are specific merits associated with each imaging modality in the evaluation of SSPBTs (Table 1). Such primary ovarian imaging features were demonstrated by ultrasound or MRI rather than by CT. They were also well-correlated with the histopathological findings. However, secondary features, such as peritoneal implants, were better demonstrated by MRI or CT than by ultrasound.

The presence of abundant papillary projections is associated with serous borderline tumors as opposed to mucinous and benign or malignant tumors.¹³ Papillary projections are best identified on T₂ weighted images. They appear as intermediate- to high-signal structures with low-signal internal branching.^2,10–13 The MRI findings of hypointense signals of the internal branching architecture on T₂ weighted images were correlated with fibrous stalks on histopathology.^2,3,10–13 Papillary projections may restrict on diffusion-weighted imaging and enhance on post-contrast imaging.¹³ Where the high-signal papillary projections are related to the surface of the ovary without any cystic component, it suggests a diagnosis of SSPBT.¹² Tanaka et al² described it as a sea anemone and taken hallmark of SSPBT. Within the tumors, normal-appearing ovaries containing multiple follicles could be identified clearly. As the primary imaging modality for gynecological disease, ultrasound was useful for diagnosing ovarian SSPBTs, as was MRI. On ultrasound, ovarian SSPBTs appeared as irregular solid lesions, surrounding normal ovarian parenchyma. CT showed poorly demarcated, solid ovarian masses, but visualization of normal ovaries within the masses was difficult because of limited contrast resolution.^{2, 10,13} CT is mainly a staging and monitoring tool rather than a characterization technique.^{2, 13}

In this series, all tumors evaluated showed contrast enhancement on CT or MRI, and increased flow on Doppler ultrasound within the solid masses. Doppler ultrasound showed minimal-to-moderate increased vascularity, as indicated by IOTA scores of 2 or 3.^{9, 11,12,14} A previous study has shown that the rate of detection of intratumoral blood flow on Doppler ultrasound in borderline tumors was similar to that in malignant neoplasms (borderline vs malignant: 90% vs 92%).¹¹

CA 125 is an important tumor marker in distinguishing benign and malignant adnexal masses.^{16, 17} However, serum CA 125 is raised in roughly half of serous borderline ovarian tumor patients, especially in Stage I patients.^{5, 13,17} The serum CA 125 level was variable but was increased consistently in all patients evaluated in this series. The range of the CA 125 level overlaps between SSPBTs and serous surface papillary carcinomas, and CA 125 level alone is thus not sufficient to differentiate the two entities. CA 125 level has a role in the follow-up of borderline ovarian tumor patients.^{4, 5}

Serous surface papillary tumors are classified as benign adenomas, borderline tumors and malignant carcinomas. Different imaging features may aid in the classification of serous surface papillary tumors. Serous surface papillary carcinomas are confined to the ovarian surface or show focal, minimal invasion. Because of the characteristic growth pattern of this carcinoma, both ovaries are likely to be of normal size and shape,^{10, 18} despite extensive peritoneal carcinomatosis. That is, the carcinoma shows extensive peritoneal masses with a large amount of ascites without definite masses in normal-sized ovaries.¹⁸ Less commonly, SSPBTs may show more prominent ovarian masses than peritoneal masses.^{10, 12}

In the comparison between this series (Group I, 10 tumors in 5 patients) to the previously reported cases (Group II, 33 tumors in 17 patients),^{2,9–11,13,14} many parameters were similar between the two groups. Most patients had bilateral involvement (Group I vs Group II: 100% vs 94%) and ascites (Group I vs Group II: 100% vs 86%). The characteristic imaging features - enhancing irregular solid masses with encasing normal-appearing ovaries (Group I vs Group II: 100% vs 100%) - and good prognosis on follow-up imaging (Group I vs Group II: 100% vs 92%) were almost the same. On the other hand, one prominent and important parameter that differed in the analysis of imaging features analysis was tumor spread (peritoneal implants: Group I vs Group II: 40% vs 56%; lymph node metastasis: Group I vs Group II: 20% vs 50%). These might reflect different FIGO stages in terms of pathological features between the two groups (Stage I or II: Group I vs Group II: 80% vs 44%). Higher-stage disease tended to show of tumor spread more commonly. All six Stage III cases in both this series and previously reported cases had peritoneal implants or lymph node metastasis. Another parameter differing in analysis of clinical features was the treatment methods. In this series, all patients underwent radical surgery (Group I vs Group II: 100% vs 71%), and no patient underwent conservative surgery (Group I vs Group II: 0% vs 29%).

Ludovisi et al⁹ reported one case of successful pregnancy with vaginal term birth and one case of a recurrent right serous borderline tumor with laparoscopic right salpingo-oophorectomy in 6 months follow-up among three conservative surgery patients. The rate of recurrence is higher after conservative surgery (10% to 20%, up to 35% vs approximately 5% for radical surgery).^{19, 20} If either both ovaries are involved, or the mass has the potential to be malignant, it is essential for a pre-operative discussion of patient preferences and options regarding conservative or radical surgery.¹⁴ When conservative management is not feasible because of massive bilateral ovarian tumor involvement, at least the uterus can be preserved for eventual transfer of frozen embryos obtained before radical surgery.⁵

There are limitations that need to be considered in interpreting these findings. First, our study was a retrospective consensus review. Second, our study included a small number of patients, whose findings may not reflect the full spectrum of imaging features in these tumors. To overcome this, we evaluated the clinical, pathological, and imaging features of ovarian SSPBTs (10 tumors in 5 patients) and compared our results with the features of previously reported cases (33 tumors in 17 patients). Third, SSPBT features were not compared with those of serous surface papillary carcinomas. Furthermore, statistical analyses were not performed. Fourth, no standardization of imaging modalities (ultrasound, CT, and MRI) was conducted in our study, because it was retrospective; thus, we combined patients with and those without available data for the individual imaging modalities. The available number of parameters evaluated in previously reported cases, such as lymph node metastasis, CA 125 levels, adjuvant chemotherapy, follow-up data, and post-operative staging, also differed. Lastly, the follow-up periods (>1 year) in this study is a relatively limited value with these kinds of tumors. This is caused by the same periods with previous reports for comparison and all five patients in this study were disease-free at least 2 years. Long-term surveillance is recommended to document and treat late recurrences.^4–6

Regardless, we believe that our study makes a useful contribution to the literature. To our knowledge, there have been few reports on the imaging, clinical, and pathological features of SSPBTs. Furthermore, previous studies have only evaluated specific imaging features separately, such as those of ultrasound and MRI. Our study analyzed and correlated all available imaging features from ultrasound, MRI, and CT. Second, we compared this series (10 tumors among 5 patients) to the previously reported cases (33 tumors among 17 patients).^{2,9–11,13,14} Third, and most importantly, this series may have implications for patient management, such as appropriate conservative surgical methods and good outcomes.

In conclusion, SSPBTs, which have a good prognosis, should be considered in the diagnosis of patients who have bilateral enhancing irregular solid masses with papillary architecture and internal branching, and encasing normal-appearing ovaries.