Abstract
Perivascular epithelioid cell tumours, better known as PECOMAs, are a very uncommon pathological finding. In English medical literature it has rarely been reported. In the genitourinary system, mostly urinary bladder, the incidence is as low as 10 cases described since 2003 until now. In this case report, we present a urinary bladder PECOMA with a detailed pathological description and a review of literature.
Background
This case will be the 11th described in the registry of bladders perivascular epithelioid cell tumours (PECOMAs).
Case presentation
The patient was a 54-year-old man, previously healthy, heavy smoker (50 pack-years) who presented with gross painless haematuria on May 2012. He denied any previous lower urinary tract symptoms, fever or chills. The haematuria subsided on its own with conservative management (increased fluid hydration).
Investigations
Workup included an ultrasound and MRI of the kidneys and bladder that showed a suspicious right anterior wall bladder tumour (figure 1).
Figure 1.

MRI pelvis showing a right anterior bladder wall tumour.
Differential diagnosis
The differential diagnosis included leiomyoma/leiomyosarcoma (ruled out by positive HMB45 staining), sarcomatoid carcinoma (ruled out by negative cytokeratin stains) and inflammatory myofibroblastic tumour (ruled out by positive smooth muscle markers, HMB45 and negative ALK).
Treatment
The patient underwent a transurethral resection of a bladder tumour in an outside hospital. The pathology read outside was consistent with a smooth muscle tumour. Immunohistochemical stains were not performed. The patient was then scheduled for a right partial cystectomy at our institution. The final pathological diagnosis was Spindle/epithelioid neoplasm with smooth muscle and melanocytic differentiation consistent with PECOMA. Histological features favour benign/low-grade behaviour.
Outcome and follow-up
Follow-up until April 2013 by cystoscopy and CT of the chest/abdomen/pelvis with intravenous and PO contrast showed a disease-free survival with no urological sequel from the partial cystectomy operation.
The segment of resected bladder measured 5×3×3 cm. It contained a well-circumscribed mass measuring 3×2.5×1.5 cm which was tan white, friable involving the entire thickness of the bladder wall. Histologically, the tumour involved the entire thickness of the muscularis propria and extended to the perivesicular fat. The tumour predominately had a pushing border but was focally infiltrative. It was composed of a monotonous population of spindle cells with focal epithelioid areas. The spindle tumour cells had elongated nuclei with blunt edges, abundant eosinophilic cytoplasm and showed mild-to-moderate cytological atypia with occasional mitosis (1 mitosis in 30 HPFs). The epithelioid cells had a nested pattern and clear cytoplasm. Focal areas showed granular brown pigment positive for iron stain. The tumour had numerous thin-walled blood vessels. There was no necrosis or vascular invasion seen (figure 2).
Figure 2.

(A) Low-power view of urinary bladder PECOMAs showing focal infiltrative growth pattern (H&E×40). (B) The tumour shows a monotonous population of spindle cells with elongated nuclei and abundant eosinophilic cytoplasm (H&E×200). (C) Admixture of epithelioid and spindle patterns (H&E×200).
Immunohistochemically, the tumour cells demonstrated strongly positive for HMB45 (melanocytic marker) and smooth muscle actin (SMA), MyoD1 and caldesmon (smooth muscle markers) they failed to stain with other melanocytic markers (Melan A, S100), epithelial markers (CKAE1/AE3, CK7, CK20) GIST (CD117), neuroendocrine (synaptophysin) and inflammatory myofibroblastic tumour (ALK) markers (figure 3). CD34 was negative in tumour cells but highlighted numerous vascular channels. The proliferation index (Ki67) was 10%. All stains were performed with a positive and negative control.
Figure 3.

Immunohistochemical profile: the tumour cells demonstrated strong and diffuse cytoplasmic positivity for HMB45 (A), smooth muscle actin (B) and caldesmon (C) (original magnification ×200).
Discussion
The majority of primary bladder tumours are predominantly malignant tumours originating from the urothelium. However, there are several non-urothelial bladder tumours that can originate from any cell lineage. The latter include postoperative spindle cell nodule, inflammatory myofibroblastic tumour, leiomyoma/leiomyosarcoma, haemangioma, neurofibroma and perivascular epithelioid cell tumours, known as ‘PECOMA’.
A review of the literature revealed only 10 cases of PECOMA involving the bladder reported in the English literature. Perivascular epithelioid cell tumours (PECOMAs) are defined as a family of mesenchymal tumours composed of histologically and immunohistochemically distinctive perivascular epithelioid cells that do not have a normal anatomic homologue.1 They are mesenchymal neoplasms that show dual differentiation with both melanoma and smooth muscle characteristics. This is demonstrated pathologically by a spindle and epithelioid tumour with immunohistochemical positivity for melanoma markers (HMB45, Melan A tyrosinase and microphthalmia transcription factor) in addition to smooth muscle markers (SMA, MSA, calponin and caldesmon).
Initially described by Pea and colleagues as unusual cells in both angiomyolipoma and clear cell sugar tumour of the lung.1 They are also now recognised as the cell of origin in lymphangiomyomatosis and unusual clear cell PECOMAs of the uterus, pancreas, thigh, heart and other sites.2 PECOMAs are associated with tuberous sclerosis complex; the PECOMA cells can show loss of heterozygosity in the TSC1 and/or TSC2 region which play a role in the regulation of the Rheb/mTOR/p70S6K signal pathway.3––5
PECOMAS have been described in variety of sites starting from the base of skull down to the thigh.5 6 PECOMAs of the bladder are very rare. With only 10 cases reported in the English medical literature (table 1).
Table 1.
All bladder PECOMAs reported in literature
| Author | Age/gender | Pathology | Immunohistochemistry |
|---|---|---|---|
| Kalyanasundaram (2005) | 19/F | Perivascularly arranged cells with granular eosinophilic fluffy cytoplasm, round-to-oval nuclei and prominent nucleoli | Positive for HMB45, negative cytokeratin, vimentin, S-100 |
| Pan (2003) | 33/F | Clear to eosinophilic, epithelioid and spindled cells arranged in fascicles or packets | Positive for HMB45 and smooth muscle actin but negative for S-100 protein, Melan-A, desmin and pan-cytokeratin |
| Parfitt (2006) | 48/M | Epithelioid but occasionally spindled, with abundant cytoplasm that varied from eosinophilic and granular to clear | Positive for HMB45 (cytoplasmic), Melan-A (cytoplasmic) and smooth muscle actin (membranous, smooth muscle myosin heavy chain, desmin and CD117− S100 protein, cytokeratin (AE1/AE3, 8/18), vimentin, muscle-specific actin, myoglobin, CD31, CD34 and WT-1 |
| Weinreb (2005) | 39/M | Pleomorphic eosinophilic cells containing larger nuclei and prominent nucleoli, focal single cell necrosis and isolated mitotic figures | Positive for HMB45, S-100, smooth muscle actin, MSA |
| Sukov (2009) | 36/M | Clear to partially eosinophilic cytoplasm, epithelioid and spindled cells | Positive for smooth muscle actin, HMB45, tyrosinase negative, Melan-A and desmin |
| Sukov (2009) | 37/M | Large epithelioid cells with clear to finely granular, eosinophilic cytoplasm. A small component showing spindle cell | Negative for HMB45 and negative for high-molecular weight cytokeratin, pankeratin, vimentin, synaptophysin, chromogranin, CD56, Melan-A, S-100, inhibin and smooth muscle actin |
| Sukov (2009) | 26/F | Fascicles of spindled cells with eosinophilic cytoplasm intermixed with epithelioid cells with clear cytoplasm arranged in a nested pattern | Negative for smooth muscle actin and vimentin, HMB45 tyrosinase negative for Melan-A, S-100, c-kit, muscle-specific actin and cytokeratin |
| Pianezza (2008) | 24/F | Spindle cells, the cytoplasm was eosinophilic or clear | Positive for HMB45, S-100, caldesmon and actin positive. Negative for MART-1, CD34, K903, pankeratin, calretinin, CD99, ALK-1, C-Kit and desmin |
| Shringarpure (2012) | 39/M | Spindled and epithelioid clear cell neoplasm | Positive for HMB45, vimentin, S-100, MSA, smooth muscle actin, negative cytokeratin and desmin |
| Kyrou (2012) | 44/F | Bundles of uniform cells with variably clear or eosinophilic cytoplasm. In many areas the cells appeared to be epithelioid, whereas focally they were more spindled | Positivity for protein S-100 ,HMB45 negative for EMA, keratin and desmin |
This rare tumour has a large pathological differential diagnosis which includes all spindle cell proliferations. The most common spindle cell neoplasm in the bladder is the sarcomatoid variant of urothelial carcinoma which is similar to the usual urothelial carcinomas except that it exhibits spindle cell morphology. This diagnosis is readily made by demonstrating origin from the urothelium and by positive immunohistochemical staining for cytokeratins. Other non-epithelial spindle cell proliferations include leiomyoma and leiomyosarcoma, malignant melanoma, clear sarcoma, inflammatory myofibroblastic tumour, postoperative spindle cell nodule and schwannoma. Immunohistochemistry is invaluable in arriving at the correct diagnosis. PECOMAs demonstrate positive immunostaining for markers of both myoid and melanocytic differentiation.7
PECOMAs are malignant tumours and bladder PECOMAs are no exception; their clinical behaviour varies widely from complete remission to local, distal metastasis and even death. PECOMAs are thus clinically ubiquitous tumour. Their presentation is similar to other bladder tumour and preoperative suspicion is almost impossible based on symptoms or imaging. In case of bladder PECOMAs, surgical resection represents the only curative option. This treatment modality was extrapolated from extragenitourinary PECOMAs since chemotherapy or radiation therapy have been inefficient in the different clinical trials.
A full clinical or genetic workup for tuberous sclerosis is not necessary in case of bladder PECOMAs, since no association has been reported in the medical literature unlike cases of extraurogenital PECOMAs. PECOMAs commonly involve mTOR activating mutation. Many hypothesis that mTOR inhibitors, such as sirolimus, might benefits patients clinically especially in locally invasive and metastatic diseases. Until now, no randomised control trial has been conducted to prove the effectiveness of mTOR inhibitors as a treatment modality in PECOMAs. Its use remains based on case reports; physician should not omit the high cost burden when choosing this treatment modality option.
Learning points.
Owing to the rarity of perivascular epithelioid cell tumours (PECOMAs) in general and their extreme rarity in the bladder reliable criteria to determine their biological behaviour are lacking. Folpe et al proposed criteria for prediction of malignant behaviour in PECOMA. It included size greater than 5 cm, infiltrative growth, high cellularity, high nuclear grade, necrosis, vascular invasion and mitotic rate greater than or equal to 1/50 high power fields.
Tumours with only nuclear pleomorphism (‘symplastic’ features) or size greater than 5 cm are classified as having uncertain malignant potential.7
Wide surgical resection is the main treatment of PECOMAs, as most of tumours are malignant.
Up to the date this review was written, no role for chemotherapy and/or radiotherapy in treatment of PECOMAs was confirmed. The use of mTOR inhibitors (ie, sirolimus) needs to be consolidated by randomised controlled trials.
The pathological behaviour and prognosis has only been extrapolated from existing and reported cases. We advise on close follow-up by cystoscopy every 6 months and a yearly CT of the abdomen and pelvis. The optimal follow-up, timing and modality, will not be possible, except after a significant number of reported cases and analysis of their natural course of disease.
Footnotes
Competing interests: None.
Patient consent: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Bonetti F, Martignoni G, Colato C, et al. Abdominopelvic sarcoma of perivascular epithelioid cells. Report of four cases in young women, one with tuberous sclerosis. Mod Pathol 2001;2013:563–8 [DOI] [PubMed] [Google Scholar]
- 2.Fletcher CD, Unni KK, Mertens F. Pathology & genetics: tumours of soft tissue and bone. World Health Organization, 2002 [Google Scholar]
- 3.Folpe AL, Kwiatkowski DJ. Perivascular epithelioid cell neoplasms: pathology and pathogenesis. Hum Pathol 2010;2013:1–15 [DOI] [PubMed] [Google Scholar]
- 4.Kwiatkowski D. Tuberous sclerosis: from tubers to mTOR. Ann Hum Genet 2003;2013:87–96 [DOI] [PubMed] [Google Scholar]
- 5.Martignoni G, Pea M, Reghellin D, et al. PECOMAs: the past, the present and the future. Virchows Arch 2008;2013:119–32 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pan C, Jong Y, Chai C, et al. Comparative genomic hybridization study of perivascular epithelioid cell tumor: molecular genetic evidence of perivascular epithelioid cell tumor as a distinctive neoplasm. Hum Pathol 2006;2013:606–12 [DOI] [PubMed] [Google Scholar]
- 7.Tazelaar HD, Batts KP, Srigley JR. Primary extrapulmonary sugar tumor (PEST): a report of four cases. Mod Pathol 2001;2013:615–22 [DOI] [PubMed] [Google Scholar]
