Abstract
This is a case of a 73-year-old woman who first presented in 2020 with a fullness in her abdomen. After several thorough investigations and unforeseen complications, the fullness was diagnosed as a granulosa cell tumour. In 2003, she had been diagnosed with a granulosa cell tumour of the ovary. Complete excision was performed, however she was not given a follow-up appointment after the procedure. This case highlights the importance of frequent follow-up of these tumours, the high recurrence rate, the severe complications which may result and the awareness of possible variations in this tumour’s histologic appearance.
Keywords: gynecological cancer, pathology, general surgery, healthcare improvement and patient safety
Background
The case describes the presentation of a granulosa cell tumour (GCT) and the implications of its recurrence. Since this patient was never followed-up after her initial GCT was excised, it recurred insidiously, and severe life-threatening complications ensued. Throughout the 17 years before she presented to clinic, the tumour progressively enlarged and went unnoticed by the patient and her close relatives. In fact, it was only noticed during a routine abdominal examination. We therefore believe this case is valuable to other health professionals as it stresses the importance of adequate follow-up to avoid a delay in presentation and to improve prognosis. Furthermore, this case highlights the importance of the histopathologist being aware of the previous histological diagnosis to avoid misdiagnosis.
Case presentation
Madam MCZ is a 73-year-old woman who is a known case of hypertension, hypothyroidism, gastro-oesophageal reflux disease and asthma and who had a total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH +BSO) in 2003 for a right ovarian GCT.
She presented to the outpatient clinic on 20 April 2020 with fresh bleeding per rectum. She also complained of prolapsing haemorrhoids. She denied weight loss and anorexia. On examination of her abdomen, a vague fullness was noted in the suprapubic region. On digital rectal examination (DRE), a haemorrhoid was visualised and an extraluminal pelvic mass was palpated. The haemorrhoid was the probable cause of bleeding.
Three large abdominal masses and lymphadenopathy adjacent to the external iliac vessels were seen on CT.
The histology report from an ultrasound (US)-guided percutaneous biopsy (first biopsy) of the largest mass described a fasciculated spindle cell smooth muscle lesion which in the absence of mitosis or necrosis, was considered compatible with leiomyomatosis.
In view of the mass at the distal sigmoid colon, a colonoscopy was performed on 15 July 2020. This confirmed that the tumour was indeed of extraluminal origin.
She was discussed at the lower gastrointestinal multidisciplinary team (MDT) meeting where a consensus was reached to refer to the oncologist for a trial of goserelin injections.
The oncologist referred her to cardiology due to an incidental finding of atrial fibrillation where she was subsequently started on warfarin.
She presented to casualty 10 days later with a sudden onset of left-sided abdominal pain, profuse vomiting and a decrease in appetite. On examination, her abdomen was soft, distended and tender in the left lower quadrant. A large mass was palpated in the lower abdomen. On DRE, the extraluminal mass was once again palpable at the anterior aspect of rectum.
An urgent CT TAP (thorax, abdomen and pelvis) was performed which showed progression of the disease, when compared with the previous CT scan.
She was admitted to hospital due to the finding of a blood disorder crisis which was managed with the disseminated intravascular coagulation regime.
Her case was rediscussed at the MDT. Due to a drop in haemoglobin, it was likely that the masses were bleeding. Additionally, the histology report of her TAH +BSO became available, describing the mass in her right ovary as a GCT. In consensus, they agreed to review the transcutaneous biopsy and perform additional immunostainings to assess for the possibility of a recurrent GCT.
On 3 October 2020, the patient agreed to proceed with the surgery after considering the risks and complications of the procedure. On opening, a large amount of blood was seen obscuring the masses. The middle mass was noted to be cystic and was infiltrating the adjacent retroperitoneal region. The parts of the mass which could be excised were removed and sent for histology and for frozen section (second biopsy).
The para-rectal lesion (lower mass) was not explored due to the patient’s poor condition and probable malignant nature of the tumour. The upper mass was excised and sent for histology. During the procedure, she had to be cardioverted multiple times due to fast atrial fibrillation. She was bleeding from multiple sites and the decision was taken to abort surgery.
The histopathologist stated that the frozen section appeared to be in keeping with a GCT recurrence.
She was referred to the oncologists who reviewed her and started her on letrozole. They requested further studies on the histological specimen, one of which was oestrogen sensitivity. The tumour was noted to weakly sensitive to oestrogen which implied that letrozole was unlikely to work. She was subsequently switched to palliative chemotherapy (carboplatin and paclitaxel).
Investigations
The first CT TAP, performed on 21 June 2020 (figure 1), showed the presence of a large heterogeneous mass in the pelvis (middle mass) which was partially cystic and partially solid and measured 12.6 cm × 12.3 cm × 13.5 cm, a heterogeneous mass in the left upper quadrant (upper mass) which measured 5 cm × 8 cm × 5 cm and a third mass lying para-rectally/distal sigmoid colon (lower mass) measuring 3.7 cm × 4.9 cm. Enlarged lymph nodes were also noted along the external iliac vessels bilaterally.
Figure 1.

Serial CT TAP scans showing progression of the three abdominal/pelvic masses, including before and after surgical treatment. Arrows point to the tumours.
All tumour and hormonal markers were within normal limits except LDH, which was slightly elevated.
The initial histological diagnosis of the US-guided percutaneous biopsy (first biopsy) favoured a solitary fibrous tumour due to the presence of spindle cells which were noted to be arranged in a fascicular manner (figure 2). However, the immunoprofile of the tumour cells prompted the histopathologist to an alternate diagnosis compatible with a leiomyoma. This diagnosis was supported by the following immunohistochemistry: a strong diffuse desmin positivity with CD34, pancytokeratin, S100 and oestrogen negativity, absence of mitosis and necrosis, a Ki-67 index between 1% and 5% and only a few cells displaying p53 staining (figures 3 and 4).
Figure 2.
The first biopsy sample showing vaguely fasciculated spindle cells with angulated blood vessels (H&E ×200 magnification).
Figure 3.
Desmin immunostaining of the same US-guided biopsy sample (×100 magnification).
Figure 4.
Biopsy sample at H&E ×400 magnification showing cell characteristics in comparison to epithelial cells of figure 6.
When the patient presented acutely to the emergency department on 29 September, a CT TAP was performed, which when compared with the previous CT, revealed an increase in size of; the pelvic mass (16.1 cm × 12.4 cm × 13.1 cm) the mass in the left abdominal cavity (13 cm × 7 cm × 6 cm) and the para-rectal mass (7.6 cm × 6.5 cm). Minimal free fluid was noted in the abdomen. No changes were seen in the thorax.
The tumour sample (second biopsy) sent perioperatively for frozen section was found to be compatible with a recurrent GCT. After surgical debulking, the rest of the tissue excised was also submitted for histological analysis. It demonstrated a typical diffuse solid pattern of an adult GCT. It was composed of epithelial cell sheets with ovoid nuclei and nuclear grooves (figures 5 and 6). These cells displayed focal keratin positivity, strong calretinin positivity and extensive inhibin and CD 56 positivity. They were essentially oestrogen negative. Areas with spindle cells showing strong desmin positivity and lacking keratin or calretinin expression were also seen embedded in the epithelial component, and they corresponded to the tumour stroma (figure 7). In retrospect, these were the cells forming fascicles in the percutaneous biopsy (first biopsy). Eventually, it was noted that in both biopsies, the stromal cells expressed strong CD56 positivity and very focal inhibin staining. A very low Ki67 index with no necrosis was also observed.
Figure 5.
Main tumour mass shows typical appearance of granulosa cell tumour with diffuse pattern (H&E ×100 magnification).
Figure 6.
Main mass specimen showing the typical cell characteristics of GCT (epithelial cell sheaths with ovoid nuclei and nuclear grooves) at H&E ×400 magnification. GCT, granulosa cell tumour.
Figure 7.
Main mass specimen at ×100 magnification showing strong desmin positivity of granulosa cell tumour stroma hidden within epithelial component.
Treatment
She was initially started on goserelin injections to attempt to shrink the misdiagnosed leiomyoma.
Surgical intervention was chosen as the definitive treatment modality since, due to the patient’s poor renal function, a CT angiography with embolisation was contraindicated.
Following discharge from hospital, she was reviewed by the oncologists in clinic who started her on letrozole. This was then switched to palliative chemotherapy.
Discussion
GCTs originate from the sex cord stromal cells of the ovary and are considered as being potentially malignant. They are rare tumours and account for 2%–5% of all ovarian neoplasms. They usually progress slowly and tend to have a more favourable prognosis. They most commonly occur in women of a reproductive age and post-menopausal women.1
Late recurrences of the GCT are well documented. Increasing age, advanced initial stage, rupture of tumour, high mitotic index, nuclear atypia and lymphatic invasion are all factors which increase the risk of recurrence following surgery. Molecular markers, for instance p53 and Ki-67 overexpression, have also been associated with a higher risk of recurrent disease.2 Since GCTs commonly recur, it is paramount to keep following-up these patients. The European society of medical oncology (ESMO) and British Journal of Cancer have emphasised the importance of long-term follow-up of GCTs postoperatively.3 4
GCTs run an indolent course. The rate of recurrence increases with time and thus follow-up with adequate history taking, examination and measurement of tumour markers for such patients is required for early recognition of recurrent disease. They may recur many years after initial diagnosis, as occurred in this case.2
Follow-up of GCT patients consists of standard gynaecological examinations including US and blood tests for at least 3–5 years.2 The ESMO recommend follow-up every 3 months for the first 2 years, every 6 months in the third year, then yearly.3 The gynaecological examination and transvaginal US are limited in the detection of relapses outside the pelvis. Serum markers such as inhibin, 17-B oestradiol and anti-Mullerian hormone (AMH) are thus commonly utilised as part of the standard follow-up scheme.2
What distinguishes GCTs from other ovarian tumours is their tendency for late recurrence. Recurrence is thought to occur from peritoneal seeds which initiate between the GCT and a surrounding structure within the abdomen and pelvis. Following excision of the primary tumour, these seeds remain within the abdominal or pelvic cavity and grow slowly, eventually presenting as a mass within the abdomen or pelvis.5
Recurrence most commonly occurs in the pelvis; however, it may also occur in the upper abdomen.1 Most early stage GCT patients experience an excellent outcome, although a third of patients will eventually develop a tumour relapse and death occurs in 50%–80% of these patients.6
GCTs often secrete oestradiol which in postmenopausal women (such as this patient) may result in endometrial hyperplasia or endometrial adenocarcinoma. Patients with these conditions may present with postmenopausal bleeding. On the other hand, females of reproductive age may present with heavy menstrual bleeding, amenorrhoea or intermenstrual bleeding. Rarely, patients may present with infertility in view of inhibin secretion.1
GCTs tend to grow significantly and thus may present with abdominal pain. They are also very vascular and a haemorrhagic rupture of the tumour into the abdominal cavity may result, in which the patient presents with an acute abdomen, as this female presented.1
Usually, when examining a patient with a GCT, a palpable mass is felt in the abdomen or pelvis, as was palpated in this patient.1
Surgery is the mainstay of treatment and is required to confirm the diagnosis and stage of tumour. In postmenopausal women, a TAH and BSO is classically carried out, while in younger women who wish to preserve fertility, a unilateral salpingo-oophorectomy may be carried out. The latter may only occur in women with disease confined to one ovary and stage A1 GCT disease.1
Postoperative treatment depends on the stage of disease. Stage 1 GCTs have an excellent prognosis with the majority not requiring adjuvant treatment. Patients with more advanced disease are suitable candidates for adjuvant treatment such as chemotherapy and radiotherapy.1 In this case, the final adjuvant treatment of choice was carboplatin and paclitaxel. However, a consensus for optimal treatment of advanced disease and disease recurrence has not yet been reached.
An alternative treatment for recurrence and advanced disease is hormonal treatment. The advantage of hormonal therapy over chemotherapy and radiotherapy is that it is better tolerated by patients and is associated with less side effects and thus can be used for longer. The use of hormonal treatment arises from the fact that oestrogen stimulates proliferation of granulosa cells by increasing their responsiveness to FSH, therefore, by blocking the effect of oestrogen, an antiproliferative environment is created.2
Teoh et al discovered that from a total of 22 GCTs, all the tumours were progesterone receptor positive. It was noted that continuous progesterone causes downregulation of progesterone receptors therefore progestins, such as megestrol acetate, have an antitumour effect.7
Various hormonal treatment options have been tried, such as tamoxifen, aromatase inhibitors, goserelin and megestrol acetate, however, they have shown varied success rates and results have proven to be short-lived.2
In this case, two types of hormonal treatments were attempted: goserelin and letrozole. Letrozole is a non-steroidal aromatase inhibitor. Studies by Korach et al and Freeman et al have shown that in recurrent GCTs, letrozole caused remission ranging from 12 to 54 months. Such treatment has also resulted in a reduction in size of the tumour, a fall in inhibin levels and an overall improvement in performance status. However, few cases had a complete response to such treatment.2
This patient was started on letrozole following debulking surgery. However, once further histological studies were carried out, it was stopped in view of the weak oestrogen sensitivity of the tumour, therefore chemotherapy was chosen as the optimal form of treatment. In the study carried out by Teoh et al, only 32% of the tumours were oestrogen receptor positive.7
Goserelin injections were also used initially in this case. GCTs express FSH receptors which have been shown to promote the growth of GCTs, thus blocking gonadotropins will block the stimulatory effect on the granulosa cells. Some studies, such as those carried out by Martikainen et al8 reported a short-lived partial remission when goserelin was administered to a woman with metastatic, platinum resistant GCT, however, other studies have described no response to GnRH agonists. A prospective study conducted by Fishman et al involved six patients who had recurrent or persistent GCTs and were on leuprolide acetate therapy. This study demonstrated a 40% partial response rate and stable disease in the remaining 60%.1
Another form of treatment described in literature is antiangiogenesis. Clinical trials have demonstrated that some GCTs responded to antiangiogenesis therapies due to the important role that VEGF and VEGFR2 play in GCT pathology.6 However, vascular density varies widely between GCTs, and this may affect the response to such treatment.9
GCTs are functional tumours as they produce hormones. These are used as tumour markers for detection of disease presence and recurrence.2
GCTs usually secrete oestradiol. No association is noted between oestradiol levels and recurrence or progression of disease, thus, limiting its role as a possible tumour marker. Moreover, 30% of patients with GCTs lack oestradiol secretion. This may be due to the lack of theca cells within the tumour stroma. Oestradiol may therefore be useful in monitoring some patients postoperatively, however, it is not sensitive enough to be used as a reliable tumour marker in monitoring disease recurrence.2
AMH and inhibin are more reliable tumour markers for disease progression and recurrence. Granulosa cells physiologically secrete inhibin and AMH, therefore, both hormones should be low in postmenopausal women. High levels of these hormones in such patients imply neoplastic disease. Both tumour markers have a high sensitivity (92%–93%) and specificity (82%–83%) to GCTs and high levels are strongly associated with disease burden and are reliable indicators of disease relapse.6
Studies have shown that investigating both tumour marker levels at diagnosis is superior to investigating the inhibin level alone since inhibin is not secreted by all GCTs (absence of inhibin correlates to poorer tumour differentiation and worse outcome) and is also secreted by other classes of ovarian tumours. Recommendations advocate the use of either AMH or inhibin for follow-up, depending on their respective levels at diagnosis. Serum inhibin levels rise 0.9–2.8 years before clinical features of recurrence are noted, whereas, serum AMH levels have a lead-time of 3.4 years.6
The microscopic appearance of ovarian GCTs is highly variable. These tumours contain sex cord derived epithelial elements mixed with mesenchymal elements.10–13 They are grouped into adult and juvenile categories based on distinctive microscopic appearance and age of the patient.
The adult variant demonstrates a wide variety of patterns, which are usually admixed. One pattern which predominates in most cases is the diffuse pattern. It contains sheets of characteristic epithelial cells extending in non-apparent stroma. Call-Exner bodies which are a heralding feature of GCTs are only rarely seen. In fact, they were not seen in this case.
The epithelial component of GCTs comprises cells with pale nuclei, nuclear grooves and indistinct cytoplasm. This component is invariably calretinin positive, FOXL2 positive, mostly inhibin positive and show pancytokeratin positivity in a punctate pattern. Other positive markers are CD56, CD99, S-100 and CD10. The epithelial component is almost invariably desmin negative.11 14 15
The tumour stroma is more peculiar and shows CD56 and variable inhibin positivity. It shows features of ovarian stroma origin and may express strong desmin positivity, as seen in this case, although variable findings are presented in the literature.13 15–17 It is usually paucicellular and fibrosed though rare cases with a malignant stromal component are described. Sometimes the stromal component may contain theca cells, Leydig cells or cells with hepatocytic features.11
High-grade transformation of the epithelial component is rarely seen in the primary ovarian tumour or in recurrences. p53 staining was noted to be involved in high-grade transformation.18
Learning points
.
Granulosa cell tumours must be followed up regularly, even after total excision.
Multiple biopsies should be taken from the lesion to overcome the heterogeneity of the histology.
A proper medical laboratory system should be readily available to trace the previous histology reports and to provide patients with the utmost care.
Despite being unable to totally excise the tumours, the patient’s quality of life was still noted to have improved significantly postoperatively since debulking reduced the pressure on surrounding structures.
Low molecular weight heparin was considered the best mode of anticoagulation to reduce this patient’s risk of bleeding while also reducing her risk of thrombotic events.
Footnotes
Contributors: TA assumed the role of guarantor and came up with the idea of writing the article. She formed part of the caring firm and was also involved in the patient’s management. She helped in gaining the patient’s consent to submit the case report to the BMJ and helped in researching literature and composing the case report. MG formed part of the caring firm and was also involved in the patient’s management. She helped in gaining the patient’s consent to submit the case report to the BMJ and helped in researching literature and composing the case report. She also helped in obtaining and editing the CT scan slices used for the report. DF also formed part of the caring firm. He helped in gaining the patient’s consent to submit the case report to the BMJ and helped in researching literature and composing the case report. He also formatted the report according to the BMJ guidelines. DB reviewed the histological samples and gave the final diagnosis of granulosa cell tumour. He contributed to the write-up by researching papers regarding granulosa cell tumour histology and providing photos and descriptions of the slides. All the authors are affiliated with Mater Dei Hospital in Malta.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Ethics statements
Patient consent for publication
Consent obtained directly from patient(s).
References
- 1.Schumer ST, Cannistra SA. Granulosa cell tumor of the ovary. J Clin Oncol 2003;21:1180–9. 10.1200/JCO.2003.10.019 [DOI] [PubMed] [Google Scholar]
- 2.Kottarathil VD, Antony MA, Nair IR, et al. Recent advances in granulosa cell tumor ovary: a review. Indian J Surg Oncol 2013;4:37–47. 10.1007/s13193-012-0201-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ray-Coquard I, Morice P, Lorusso D, et al. Non-epithelial ovarian cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Annals of Oncology 2018;29:iv1–18. 10.1093/annonc/mdy001 [DOI] [PubMed] [Google Scholar]
- 4.Mangili G, Ottolina J, Gadducci A, et al. Long-term follow-up is crucial after treatment for granulosa cell tumours of the ovary. Br J Cancer 2013;109:29–34. 10.1038/bjc.2013.241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Singh-Ranger G, Sharp A, Crinnion JN. Recurrence of granulosa cell tumour after thirty years with small bowel obstruction [Internet], 2004. Available: http://www.issoonline.com/content/1/1/4 [DOI] [PMC free article] [PubMed]
- 6.Färkkilä A, Haltia U-M, Tapper J, et al. Pathogenesis and treatment of adult-type granulosa cell tumor of the ovary. Ann Med 2017;49:435–47. 10.1080/07853890.2017.1294760 [DOI] [PubMed] [Google Scholar]
- 7.Teoh D, Freedman R, Soliman PT. Nearly 30 years of treatment for recurrent granulosa cell tumor of the ovary: a case report and review of the literature. Case Rep Oncol 2010;3:14–18. 10.1159/000274590 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Martikainen H, Penttinen J, Huhtaniemi I, et al. Gonadotropin-Releasing hormone agonist analog therapy effective in ovarian granulosa cell malignancy. Gynecol Oncol 1989;35:406–8. 10.1016/0090-8258(89)90089-9 [DOI] [PubMed] [Google Scholar]
- 9.Mills AM, Chinn Z, Rauh LA, et al. Emerging biomarkers in ovarian granulosa cell tumors. Int J Gynecol Cancer 2019;29:560–5. 10.1136/ijgc-2018-000065 [DOI] [PubMed] [Google Scholar]
- 10.Young RH. Ovarian sex Cord–Stromal tumors: reflections on a 40-year experience with a fascinating group of tumors, including comments on the seminal observations of Robert E. Scully, MD. Arch Pathol Lab Med 2018;142:1459–84. 10.5858/arpa.2018-0291-RA [DOI] [PubMed] [Google Scholar]
- 11.Young RH. Ovarian sex cord-stromal tumours and their mimics. In: Pathology. Vol. 50. Elsevier B.V, 2018: 5–15. [DOI] [PubMed] [Google Scholar]
- 12.He H, Luthringer DJ, Hui P, et al. Expression of CD56 and WT1 in ovarian stroma and ovarian stromal tumors. Am J Surg Pathol 2008;32:884–90. 10.1097/PAS.0b013e3181609d59 [DOI] [PubMed] [Google Scholar]
- 13.Kanthan R, Senger J-L, Kanthan S. The multifaceted granulosa cell tumours-myths and realities: a review. ISRN Obstet Gynecol 2012;2012:1–12. 10.5402/2012/878635 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lastarria D, Sachdev RK, Babury RA, et al. Immunohistochemical analysis for desmin in normal and neoplastic ovarian stromal tissue. Arch Pathol Lab Med 1990;114:502–5. [PubMed] [Google Scholar]
- 15.Bakula-Zalewska E, Danska-Bidzinska A, Kowalewska M, et al. Uterine tumors resembling ovarian sex cord tumors, a clinicopathologic study of six cases. Ann Diagn Pathol 2014;18:329–32. 10.1016/j.anndiagpath.2014.08.009 [DOI] [PubMed] [Google Scholar]
- 16.Roth LM, Czernobilsky B. Perspectives on pure ovarian stromal neoplasms and tumor-like proliferations of the ovarian stroma. Am J Surg Pathol 2011;35:e15–33. 10.1097/PAS.0b013e31820acb89 [DOI] [PubMed] [Google Scholar]
- 17.Zhao C, Vinh TN, McManus K, et al. Identification of the most sensitive and robust immunohistochemical markers in different categories of ovarian sex cord-stromal tumors. Am J Surg Pathol 2009;33:354–66. 10.1097/PAS.0b013e318188373d [DOI] [PubMed] [Google Scholar]
- 18.Fashedemi Y, Coutts M, Wise O, et al. Adult granulosa cell tumor with high-grade transformation: report of a series with FOXL2 mutation analysis. Am J Surg Pathol 2019;43:1229–38. 10.1097/PAS.0000000000001296 [DOI] [PubMed] [Google Scholar]






