Primary peritoneal mucinous cystadenocarcinoma mimicking possible recurrent ovarian mucinous cystadenoma: coincidental pathology or a spectrum of disease?

Abstract

We present the case of a 32-year-old woman with a previous surgical history of benign mucinous cystadenoma resected in 2012 who underwent magnetic resonance cholangiopancreatography following her second pregnancy in 2020. This demonstrated a large cystic mass in the left subdiaphragmatic space. Histopathology confirmed a well-differentiated primary peritoneal mucinous cystadenocarcinoma displaying cells of a Mullerian origin. We subsequently discuss the aetiology of these conditions separately and explore the possibility of a connection between the two regarding origin or potential malignant transformation that may otherwise have occurred coincidentally in this young patient. We also acknowledge a paucity of evidence regarding subsequent management strategies.

Background

Mucinous tumours are defined by the presence of epithelial cells containing intracytoplasmic mucin. Mucinous ovarian tumours may resemble those of the endocervix, gastric pylorus or bowel,¹ and represent the second most common type of surface epithelial–stromal tumour of the ovary, accounting for approximately 15% of all ovarian tumours.² Similar to their serous ovarian counterparts, mucinous tumours are divided into benign, borderline and invasive categories, depending on the degree of stromal invasion. Approximately 80% of mucinous ovarian tumours are benign (cystadenomas), with the remainder split between borderline and invasive carcinomas (cystadenocarcinoma).

Even though mucinous cystadenomas are relatively common ovarian tumours, primary mucinous tumours of the peritoneum and retroperitoneum are exceedingly rare. There have been approximately 50 cases of primary retroperitoneal cystadenocarcinoma described in international literature, with most patients diagnosed when lesions grow large enough to exert mass effect or cause obstruction.³ To our knowledge, there has been only one other case reported of a primary peritoneal mucinous adenocarcinoma, evidencing the absolute scarcity of this histology subtype.⁴ Evolving research suggests that mucinous ovarian tumours, along with peritoneal tumours with similar mucinous histology, represent a spectrum of disease originating from the Mullerian compartment, collectively forming adenocarcinomas of Mullerian origin,¹ with the Federation of Gynaecology and Obstetrics (FIGO) staging classifications having recently been revised to reflect this.⁵

Here we report on the case of a 32-year-old woman who presented on two separate occasions, once in 2012 with a benign mucinous cystadenoma of the ovary and subsequently in 2020 with a subdiaphragmatic mass confirmed to be a mucinous tumour arising from the peritoneum itself, specifically a well-differentiated primary cystadenocarcinoma. While the coincidental appearance of these pathologies is curious in itself, there are theories that suggest a link between the diagnoses, which we will discuss.

Case presentation

The patient has no previous medical history of note and a previous surgical history of laparoscopic left salpingo-oophorectomy for a 35 cm benign mucinous cystadenoma in 2012 (figure 1). The patient has no significant medical or surgical family history.

Figure 1.

Coronal CT scan showing a large benign mucinous cystadenoma diagnosed and surgically removed in 2012.

The patient is 6 months postnatal, having had vaginal deliveries of monochorionic diamniotic twins, spontaneously labouring at 34 weeks and 4 days on a background of diet controlled gestational diabetes mellitus as well as obstetric cholestasis and pre-eclampsia. Antenatal obstetric ultrasound had been normal, though an abdominal ultrasound performed 2 days postnatally demonstrated a soft tissue mass in the gallbladder wall thought to either be adherent sludge or a gallbladder polyp (figure 2).

Figure 2.

Gallbladder ultrasound demonstrating soft tissue attenuation within the fundus.

The patient was discussed with the gastroenterology team who suggested magnetic resonance cholangiopancreatography 3 months into the postnatal period. This reported a 130×110×110 mm cystic lesion in the left subdiaphragmatic space, displacing the spleen inferiorly (figure 3). The mass was thin walled and contained multiple internal separations with restricted diffusion, as well as an ill-defined soft tissue component medially, which demonstrated some mild contrast enhancement. Posterior to the well-rounded mass, there was also reported to be multiple locules of the same lesion abutting the diaphragm and peritoneum, intimately related to but not arising from the spleen. The stomach was seen separately. There was once again noted to be a 10 mm lenticular-shaped filling defect in the fundal posterior wall aspect of the gall bladder, which was thought to be either sludge or a polyp.

Figure 3.

T2-weighted coronal and axial MRI demonstrating the left upper quadrant mass.

Taking the patient’s previous surgical history into consideration, concerns were raised regarding a potential recurrence or malignant transformation of the previously resected ovarian cyst. Tumour markers were reported as normal (HCG/AFP/LDH/CA125/CA153/CEA) with the exception of CA19-9 (raised at 277). A CT scan of the chest/abdomen/pelvis further characterised the mass, which contained slender septations and an 8 mm medial mural nodule (figure 4). It was reported to overlap, indent and displace the spleen and was separate to the liver, stomach and pancreas. There was no convincing peritoneal or separate solid organ abnormality. The radiological diagnosis was suggested as a likely benign cystic peritoneal tumour, with differentials including primary peritoneal serous borderline or benign multicystic mesothelioma. The mass was felt to be amenable to complete surgical resection based on multidisciplinary team (MDT) recommendation.

Figure 4.

Coronal and axial CT demonstrating the left upper quadrant mass.

Outcome and follow-up

Under the care of the gynaecological oncology team, the patient underwent laparotomy with mobilisation of the descending, transverse and splenic flexure of the colon with opening of the lesser sac, liver mobilisation, resection of left upper quadrant mass, left diaphragmatic peritoneum, spleen, appendix, gall bladder and omentum with repair of diaphragmatic defect. Intraoperative rupture of the mass was noted, which resulted in spillage of contents. A sample was sent to cytology with further solid contents sent to histology. Gynaecological organs present appeared unremarkable. Postoperative recovery was unremarkable.

Histopathology subsequently confirmed a well differentiated primary mucinous cystadenocarcinoma on a background of benign and borderline mucinous tumour (figure 5).

Figure 5.

Low power and close-up views of the resected cystadenocarcinoma.

Macroscopically, the resected cyst contents were of a solid mucinous consistency with possible epithelial lining and focal whitish nodular areas representing calcification. Microscopically, cysts contained mucinous glands of intestinal and Mullerian morphology. A slight villous architecture was noted with mild to moderate complex inward branching and glands showing back-to-back architecture. The nuclei showed some atypia with a degree of stratification and hyperchromasia, open chromatin and small nucleoli. The surrounding stroma showed spindle cells resembling ovarian stroma. Fibrosis and hyalinisation were also seen in places with glands showing expansile growth and focal infiltration. There was no malignant infiltration of the spleen, diaphragm or pleura. No lymphovascular invasion was seen. The gall bladder, appendix, omentum and falciform ligament showed no evidence of malignancy.

Pathologically, this was felt to be a new and separate process from the previous ovarian cyst and not related to the gastrointestinal tract. Immunohistochemistry showed that the tumour cells were diffusely positive for CK7, PAX8, CA19-9 and CK8/18 with focal positivity for p16 and focal strong positivity for CK20, CEA and CA125 (figure 6). There was some weak focal positivity for CDX-2, while oestrogen and progesterone receptors were negative. Karyotyping was not performed.

Figure 6.

CA-125 staining confirming the Mullerian origin of the resected tumour.

With this in mind, findings were suggestive of a primary peritoneal mucinous cystadenocarcinoma of Mullerian origin. Peritoneal washings sent to cytology were reported as negative with no malignant cells seen. Taking into consideration this unusual pathology as well as the patients previous surgical history, following much consideration, the final recommendation from MDT was for referral to medical oncology for consideration of adjuvant chemotherapy treatment.

The patient was advised to continue penicillin V antibiotic prophylaxis lifelong, as is typical after splenectomy, and a follow-up plan was made for 3 months following the end of adjuvant chemotherapy—six cycles of single-agent carboplatin. Gynaecology tumour markers were requested to be completed in advance of future clinic appointments. A CT scan performed 6 months postoperatively showed no evidence of residual or recurrent disease, no ascites and no peritoneal infiltration. The patient will remain under review by the gynaecology oncology team and be seen on a 6-monthly basis.

Discussion

We have reported on the unusual case of a patient treated for a benign mucinous cystadenoma in 2012, who subsequently re-presented in 2020 with a low-grade primary mucinous cystadenocarcinoma arising from the peritoneal lining. On review of the literature, limited data exist documenting recurrent mucinous cystadenoma,⁶ let alone cases of a benign cystadenoma followed by a cystadenocarcinoma years later. While this case may simply reflect an interesting but coincidental occurrence of two distinct pathologies in the same patient, we will consider the individual pathologies and theories discussed in recent literature to ascertain if a possible link could be established as well as management strategies.

There have been multiple theories put forward to explain the development of mucinous carcinomas in general terms, which include the adenoma–carcinoma sequence (whereby a tumour undergoes malignant transformation from benign epithelium to invasive carcinoma), germ cell origin (although most mucinous carcinomas have no teratoma components), mucinous metaplasia of surface epithelium or an association with endometriosis (whereby cells are usually endocervical or Mullerian mucinous tumours). Cobb et al¹ have recently elaborated on the adenocarcinoma of Mullerian origin theory, whereby they discuss the growing body of evidence suggesting that epithelial ovarian, peritoneal and tubal cancers are not distinct entities but rather reflect a spectrum of disease originating in the Mullerian component.¹ Essential to the understanding of the theory is an appreciation of the origin of the Mullerian system. Ovarian surface epithelium is derived from coelomic epithelium during fetal development, which itself is derived from the mesoderm and consists of the epithelial lining of the body cavity. It also covers what will become the peritoneal lining and the area that will subsequently develop into the gonadal structures. While the reproductive organs and peritoneum do originate from their own distinct pathways, Mullerian epithelium, ovarian surface epithelium and the peritoneal epithelium have an intricate developmental relationship. Such understanding of the development of the ovarian and peritoneal epithelium and of the above theory may provide a potential link for the development of the two mucinous tumours in our patient, indicating in fact that they reflect a spectrum of the same disease process. Indeed, the serum biomarkers noted in the supporting literature to be beneficial in the detection of adenocarcinomas of Mullerian origin include CA125, CA19-9 and CEA, and positive immunohistochemistry staining for CK7 and PAX-8, all of which were positive in the excised peritoneal tumour of our patient. Support for the adenocarcinoma of Mullerian origin theory is becoming increasingly well recognised, and as such, the FIGO classification has been recently revised to reflect this supporting evidence. As well as providing an understanding of the developmental origins of mucinous lesions, it is thought that the identification of adenocarcinoma of Mullerian origin will guide appropriate treatment options.

Mucinous cystadenoma, resected from our patient in 2012, accounts for 80% of mucinous ovarian tumours. Symptoms include abdominal pain and distension, an enlarging pelvic mass and occasionally hormonal manifestations. They often present as a large, multiloculated cystic mass, with a mean size of 18 cm, although tumours can be very large and fill the abdominal and pelvic cavity.¹ Benign mucinous cystadenomas have an excellent prognosis but are frequently excised both for histological confirmation and due to large mass effect. As they are confined to the ovary, intact oophorectomy or fertility sparing conservative cystectomy is performed. No further procedure is required, although close inspection of the abdomen and pelvis should be undertaken intraoperatively. Recurrence of mucinous cystadenoma is very rare after complete excision,⁷ particularly if there is no spillage of cyst contents.

Primary peritoneal cystadenocarcinoma, particularly of a mucinous subtype and referring to the cystic mass lesion recently diagnosed in our patient in 2020, is exceedingly rare.^8–10 There have been less than 25 reported cases of a retroperitoneal mucinous cystadenocarcinoma,⁴ and only 1 other case report to our knowledge of a cystic peritoneal mass with mucinous histology. Wang et al⁴ describe the case of a 49-year-old woman with an intraperitoneal cystic mass proven to be primary peritoneal adenocarcinoma, where CA-125 was raised and immunohistochemistry was positive for CK-7, CK-20 and CDX2. Not only are these identical to the makers expressed by ovarian mucinous carcinomas of the ovary, but also they mirror those of our patient’s peritoneal tumour. The origin of such a cystic mass lesion is thought to stem from the embryological development of the primitive mesothelium shared by the ovarian and peritoneal epithelium.

In the case of our patient, staging presented a number of challenges. This was not only in view of tumour location but also in relation to intraoperative spillage of cyst contents, as well as concerns raised regarding primary versus recurrent pathology. In view of a histopathologically potential gynaecological (Mullerian) origin, when comparisons were drawn to a cyst rupture in the case of ovarian cancer, this would usually necessitate adjuvant treatment prior to definitive surgical management, whereby complete surgical removal of such a tumour is recommended due to risk of infection, recurrence or metastasis.³ The most commonly used chemotherapeutic treatment for primary serous papillary peritoneal carcinoma (the most commonly reported histological subtype of primary peritoneal carcinoma) includes platinum-based chemotherapeutic agents, a strategy based on proven effectiveness in serous ovarian carcinoma. This was the treatment modality used by Wang et al⁴ for treatment of a primary mucinous cystic tumour of the peritoneum and by the gynaecological oncology team for treatment of our patient for the same pathology. Overall, treatment of peritoneal cystic malignancy occurs in a manner similar to that of ovarian carcinoma, although chemotherapy resistance is more of a concern for mucinous compared with serous subtypes⁹; hence, the potential benefits of adjuvant treatment in our patient’s case became less clear. The subtypes of primary peritoneal carcinoma are biologically unique and likely to have a unique response to therapy. Unfortunately, due to the absolute rarity of case reports of primary peritoneal mucinous carcinoma, we can only postulate about the appropriateness of treatment and its efficacy.

Despite encouraging research, an established and unequivocal link between mucinous peritoneal and ovarian carcinomas, as we present here, is sadly lacking at this time due to the relative rarity of these histological subtypes and the reliance on case reports for much of the data. Mucinous tumours that demonstrate Mullerian characteristics are an uncommon subtype of mucinous tumour,^{11 12} and while the theories linking the occurrence of such rare tumours are interesting and compelling, there are just not enough data at present to suggest with confidence that the two tumours discovered in our patient are in some way linked. Our pathologists felt assured that the mucinous peritoneal cystadenocarcinoma excised in 2020 was a new and separate process from the previous ovarian cyst removed in 2012, and therefore this case most likely demonstrates two very interesting but coincidental diagnoses in our young patient.

Conclusion

We have described an unusual case of a young woman who presented on two occasions, 8 years apart, with separate mucinous tumours. The benign mucinous cystadenoma of the ovary and the peritoneal cystadenocarcinoma are ultimately likely to be unrelated, given the time between diagnoses, pathological data and lack of further disseminated disease. However, this case provides intriguing data that collectively may increase our understanding of mucinous tumour subtypes, particularly those developing in the peritoneum for which there are so few documented cases. The complex development of primitive mesothelium in multiple locations, including the peritoneum and ovarian surface epithelium, continues to inform interesting cases such as this one, and ultimately such an understanding informs treatment outcomes, which is vital in such a rare disease. Future cross-sectional imaging and review of tumour markers in our patient can be used to ensure she remains disease-free after treatment.

Patient’s perspective.

In 2012, I ended up in A+E with really bad stomach pains and an enlarged stomach. After scans, a large cyst on my left ovary was discovered and later removed via keyhole surgery. After around 6 weeks’ recovery, I was back to work and my only concern was that it may cause problems in conceiving in the future. This wasn’t the case as in 2017 when I gave birth to a baby boy.

In 2019, we decided to try again for another baby. I fell pregnant and was very shocked to discover at the first scan that it was monochorionic diamniotic twins. I was monitored closely during the pregnancy due to it being MCDA. I was diagnosed with gestational diabetes and cholestasis, and it was decided at 34 weeks that I would be induced as there was a concern about pre-eclampsia. I went into natural labour and gave birth naturally to both twins within 10 min of each other. The twins were taken to special care and spent 10 days in the hospital to help them with breathing and tube feeding. I spent 5 days in hospital due to my blood not being right.

I was then sent for a ultrasound at my community hospital to check my liver and gall bladder. This scan showed that my liver was ok but my gall bladder needed further investigation. I was then sent for an MRI scan where the cyst was discovered. I then had a CT scan to have another look at the cyst.

In October 2020, I was booked in to have an operation to remove the cyst. This caused a rollercoaster of emotions as I knew it was the right thing to do but I had a 3 year old and 5-month-old twins, which I would not see for around a week. I had been a stay-at-home mum since having my son.

On the day of the operation, my surgeon explained the procedure to me. I had decided to have my gall bladder removed at the same time as there was still something showing on the scans. It was also explained to me that my spleen would most definitely need removing and it was advised that I had my appendix out as well. This was a lot to take in and I was on my own due to COVID-19 restrictions.

After the surgery, I spent 6 days in hospital. The care I received was amazing; however, due to the COVID-19 restrictions, I was only allowed one visitor per day at an allocated time. The hardest part of my stay in was not seeing my children. Although I did video calls with them each day, it’s not quite the same.

I was so happy to get home, but the happiness soon turned to frustration as I am a very independent person and don’t like to have to rely on others. It was really hard not being able to pick the babies up, especially when they were crying. The only way I could hold one of them was if someone put them on my lap, but I knew that if I tried to do things too soon, this would put my recovery back and possibly result in a hernia.

Around 6 weeks after the operation, I had a follow-up appointment with my surgeon. I went to the appointment by myself as once again we were in a lockdown. I was told at the appointment that there were cancerous cells in the cyst. This took awhile to sink in and I don't think it really did until I left the appointment and told my husband. Once again this caused a rollercoaster of emotions, but I’m trying to stay positive and my children give me a reason and focus.

I have now started my treatment and I’m due to have my second dose today. So far, the side effects haven't been too bad, I’ve had a couple of days where I feel a bit tired and lack energy.

The hardest part with all of this has been the lockdown and COVID-19 restrictions, as it means I haven't been able to have the full help and support from friends and family.

Learning points.

• Primary peritoneal mucinous tumours are vanishingly rare but, along with mucinous ovarian tumours, may represent a spectrum of disease originating from the Mullerian compartment.

• Malignant transformation of benign cystadenomas to cystadenocarcinoma, with or without metastatic disease, may account for the subsequent presentation of a peritoneal mucinous cystadenocarcinoma in our patient following prior removal of a mucinous cystadenoma of the ovary, in line with the adenocarcinoma of Mullerian origin theory.

• Ultimately, there are too few case reports of this rare tumour and in particular the mucinous subtype to establish a confident link in this case.

• As such, there remain a number of challenges regarding the appropriate staging and adjuvant treatment of cystic peritoneal mucinous tumours, and further investigation into this histological variant is required to determine optimal management.

Ethics statements

Patient consent for publication

Obtained.