According to Oppedal et al.1 and Carbone and Micheau,2 lymphoepithelioma is an undifferentiated, malignant, epithelial tumour of the nasopharynx that is distinguished histologically by its prominent lymphoid infiltrate, suggestive of malignant lymphoma.1
Tumours that are histologically similar to lymphoepithelioma of the nasopharynx have been found elsewhere in the body, such as the stomach,3 lung,4 cervix5 and urinary bladder.6 These types of tumours have been called lymphoepithelioma-like carcinoma (LELC), and fewer than 100 cases of have been reported since they were first described in 1991.
Because of the rarity of LELC of the urinary bladder, practitioners may be unfamiliar with the biological behaviour. This article reviews the literature to describe the features of LELC of the urinary bladder.
Carcinoma of the urinary bladder is the most common malignancy involving the urinary tract system. Urothelial (transitional cell) carcinoma is the most common histological type in Europe and the USA, where it accounts for 90% of all carcinomas of the urinary bladder.7 In other parts of the world, non-urothelial carcinomas are more frequent.7 Much less commonly, urothelial carcinomas can arise in the renal pelvis, ureter or urethra.7
Several systems have been utilized to grade and classify neoplasms of the urinary bladder. The classification system proposed by the World Health Organization (WHO) in 1972 distinguished papillomas from ‘grade I, II and III’ papillary transitional cell carcinomas.8
WHO, in collaboration with the International Society of Urologic Pathologists, published a consensus classification system for urothelial neoplasms in 1998.9 A subsequent study validated the clinical significance of this scheme and, in 2004, it was accepted as the standard scheme of classification.10
Based upon the 2004 classification, urothelial carcinoma has been classified as both low and high grade, depending on the degree of nuclear anaplasia and architectural abnormalities. Invasive urothelial carcinoma is high grade, with the exception of, for example, nested and tubular variants.
Urothelial carcinomas of the urinary bladder are either non-invasive or invasive. Non-invasive urothelial lesions are divided into flat or papillary subgroups. Flat non-invasive urothelial lesions include dysplasia and carcinoma in situ (CIS) whereas non-invasive papillary lesions of the urinary bladder can be benign (e.g. urothelial papilloma, inverted papilloma) but may have malignant potential [e.g. papillary urothelial neoplasm of low malignant potential (PUNLMP)] or contain non-invasive low-grade papillary urothelial carcinoma (LGPUC) or high-grade papillary urothelial carcinoma (HGPUC). PUNLMP, LGPUC and HGPUC usually present with haematuria and are more common in men, with a mean age at diagnosis in the 60s.7
Invasive urothelial carcinoma, which is also referred to as transitional cell carcinoma, characteristically invades beyond the basement membrane. Urothelial carcinoma has a propensity for multidirectional differentiation. The terminology ‘variant’ was coined to describe a distinctively different histomorphological phenotype of a certain type of neoplasm. In 2004, the spectrum of microscopic forms of urothelial carcinoma was expanded to encompass a number of unusual histological variants. It is important for clinicians to be aware of these unusual patterns in order to avoid misinterpretations of symptoms and a wrong diagnosis. Urothelial carcinomas quite often contain elements of squamous or glandular differentiation which are more commonly found in high-grade and high-stage lesions.11–13
The WHO classification has recognized a number of different histological subtypes of urothelial carcinoma:
Squamous differentiation is present in approximately 20% of urothelial carcinomas of the urinary bladder and in 44% of tumours of the renal pelvis. The diagnosis of true squamous differentiation requires evidence of keratinization, intercellular bridges or both, and this should be obvious on routine haematoxylin and eosin staining. The presence of squamous differentiation does not appear to affect prognosis, although these tumours may be less likely to respond to chemotherapy or radiotherapy.10
Glandular differentiation is present in approximately 6% of cases, and has been defined as the presence of true glandular spaces within the urothelial component of the tumour. It could consist of tubular or enteric glands with mucin secretions. It has been said that intracytoplasmic mucin does not indicate glandular differentiation and that the presence of glandular differentiation and mucin positivity in urothelial carcinomas does not affect the prognosis for any given stage.10
Nested variants of urothelial carcinoma are an uncommon, but aggressive, neoplasm with a deceptive benign appearance, closely mimicking von Brunn’s nests infiltrating the lamina propria. Some authors have reported that the useful characteristic features for recognizing this lesion as malignant include the presence of muscle invasion, irregularly invading crowded nests and the tendency for increasing atypia in the deeper portions of the lesion. It has also been reported that there is a marked male predominance and that, despite treatment, 70% of the patients die within 3–4 years of diagnosis.14,15
Microcystic variants of urothelial carcinoma rarely exhibit a striking microcystic pattern. The cysts may range from microcystic up to 1–2 mm in diameter and they may contain necrotic material or pale-pink secretions. The cyst lining may be absent, flattened or urothelial and may exhibit differentiation towards mucinous cells. It has been reported that the cytology of the microcystic variant of urothelial carcinoma is bland and that this variant quite often coexists with the nested variant and is unrelated to primary adenocarcinoma of the urinary bladder.7 There is no apparent striking biological significance associated with this pattern of urothelial carcinoma.16
Micropapillary variants of urothelial carcinoma are rarely encountered. Some authors have found that these tumours characteristically tend to be high grade, presenting at an advanced stage, and they are associated with a poor prognosis.17–21 There is a male predominance and the median age at diagnosis is 70 years.7 It has been stated that the micropapillary variant of urothelial carcinoma resembles serous carcinoma of the ovary and the surface of the tumours shows slender, delicate papillary and villiform processes, often without a central vascular core.7 The invasive component characteristically exhibits minute nests of cells or delicate papillae contained within tissue retraction spaces, simulating lymphatic spaces. These tumours invariably invade muscle and are associated with a high incidence of metastasis. Furthermore, a large micropapillary component has been shown to be a significant adverse prognostic factor.7
Sarcomatoid variants of urothelial carcinoma encompass all biphasic malignant neoplasms that exhibit morphological and/or immunohistochemical evidence of both mesenchymal and epithelial differentiation. About 0.6% of all bladder carcinomas are sarcomatoid carcinomas and carcinosarcomas. Molecular studies strongly suggest a monoclonal origin of both the epithelial and mesenchymal components in sarcomatoid carcinoma and carcinosarcoma.7 A number of authors22–26 have reported that the mesenchymal component, which is most frequently observed in sarcomatoid variants of urothelial carcinoma, is an undifferentiated high-grade spindle cell neoplasm and the most common heterologous element is osteosarcoma, followed by chondrosarcoma, rhabdomyosarcoma and leiomyosarcoma. Nodal and distant organ metastases at diagnosis are common and three-quarters of patients with a sarcomatoid variant of urothelial carcinoma die as a result of their cancer within 48 months.
Malignant cells in lymphoma-like and plasmacytoid variants of urothelial carcinoma resemble those of malignant lymphoma or plasmacytoma. Few cases of these types of variants of urothelial carcinoma have been reported, and most of them contained high-grade urothelial carcinoma. A number of authors6,27–30 have reported that the histological features of these variants characteristically exhibit dyscohesive oval to round malignant cells in a loose or myxoid stroma and that the tumour cells have eccentrically placed nuclei and abundant eosinophilic cytoplasm. The differential diagnosis of these tumours includes lymphoma (plasmacytoid type), multiple myeloma, metastatic carcinoma, paraganglioma, neuroendocrine carcinoma and rhabdomyosarcoma. The prognosis of lymphoma-like and plasmacytoid-like carcinomas is poor and related to the pathological stage; most patients who present with locally advanced disease will experience progression to metastatic disease and survival is usually short.
Few cases of urothelial carcinoma resembling lymphoepithelioma of the nasopharynx have been reported7 and, to the knowledge of the authors, fewer than 75 cases have been reported in the literature. The tumours have shown predominance in males and tend to occur in late adulthood. Unlike lymphoepithelioma of the nasopharynx, the association between the lymphoepithelioma-like variant of urothelial carcinoma and Epstein–Barr virus (EBV) has so far not been documented, either by means of immunohistochemistry or by in situ hybridization.31–33 The epithelial tumour cells tend to be large and are arranged in syncytia, with individual undifferentiated neoplastic cells displaying indistinct cell borders, pleomorphic vesicular nuclei, prominent nucleoli and numerous mitoses. The background of LELC of the urinary bladder consists of a prominent lymphoid infiltrate, and the differential diagnosis includes malignant lymphoma, poorly differentiated urothelial carcinoma with lymphoid stroma and poorly differentiated squamous cell carcinoma. In limited biopsy material, florid chronic cystitis could be confused with LELC.7 Some authors have reported that, when focal LELC only is present, then the proportion of LELC should be reported as the behaviour of these tumours is similar to that of conventional urothelial carcinoma of the same grade and stage.7 Additionally, the lymphoepithelioma-like component is responsive to chemotherapy.7 LELC of the urinary bladder accounts for 0.4–1.3% of all carcinomas of the urinary bladder and mainly occurs in adults with a mean reported age at diagnosis of 67–69 years (range 52–84 years). LELC of the urinary bladder occurs more frequently in men (75% of cases), and these tumours are commonly found in the dome, posterior wall and trigone of the bladder. Some of the microscopic descriptive characteristics of LELC of the urinary bladder include the following: (a) it is a subtype of urothelial carcinoma that resembles lymphoepithelioma of nasopharynx but is EBV negative; (b) the lymphoepithelioma-like component should be > 50% to establish a diagnosis; (c) it is usually a muscle-invasive tumour; (d) there is often a coexisting urothelial carcinoma and urothelial CIS; (e) the tumour cells may be obscured by chronic inflammation; (f) the tumour may have markedly pleomorphic epithelial component; (g) non-neoplastic cells in the tumour are a mixture of B- and T-lymphocytes, eosinophils and plasma cells; and (h) the tumour contains undifferentiated cells arranged in syncytial sheets with minimal cytoplasm, prominent nucleoli and numerous mitoses, as well as lymphocytes (Figures 1, 2 and 3 show haematoxylin and eosin-stained specimens that illustrate typical morphological features of LELC of the urinary bladder). It is recommended that the percentage of lymphoepithelioma-like areas should be reported histologically in order to decide whether the tumour is the pure type, predominant type or the focal type. This is important for planning and choosing a treatment as there are known differences in the outcome depending on subtype. Cytological examination of urine specimens from patients with such carcinoma of the urinary bladder may reveal (a) single cells, alone or mixed with inflammatory cells, or (b) large tumour cells with high nuclear to cytoplasmic ratio, vesicular chromatin and prominent nucleoli. Immunohistochemical studies of LELC of the urinary bladder may reveal positive staining for cytokeratin (CK) 7, AE1/AE3, often staining positive for epithelial membrane antigen (EMA), p53 and also B- and T-lymphocyte markers. Staining for EBV may be negative.
A 79-year-old woman was referred to hospital as she was suffering from haematuria and recurrent urinary tract infections. Her general and systematic examinations were normal but computed tomography (CT) revealed a filling defect on the left side of her urinary tract, suggestive of a bladder tumour. The rest of her urinary tract was normal and there was no evidence of any lymph node involvement or metastasis within her abdomen and pelvis.
Cystoscopy revealed two tumours on the left side of the urinary bladder that were treated by transurethral resection. A biopsy of the tumour base/muscle was taken at the end of the tumour resection. The patient’s postoperative recovery was unremarkable and she was discharged home.
The results of histological examination of the tumour specimen and the muscle biopsy were reported as follows:
Sections of the tumour showed a poorly differentiated tumour of the urinary bladder that comprised syncytial arrangements of malignant epithelial cells with prominent nucleoli intermixed with inflammatory cell infiltrates. In other places, the epithelial component of the tumour had a trabecular arrangement and very occasional lymphoid aggregates with germinal centres (see Figures 1–3). The tumour showed areas of necrosis, and the epithelial cells of the tumour stained positive for CK7 (see Figure 4). There was focal positive staining for CK20 (see Figure 6), 34BE12 (see Figure 7) and p63 (see Figure 8). Stain for prostate-specific antigen (PSA) and prostatic acid phosphatase were negative. Immunohistochemical staining for leucocyte common antigen (LCA) highlighted the inflammatory cell infiltrates in the lesion (see Figure 5). The overall features of the tumour were in keeping with LELC of the urinary bladder. The carcinoma had infiltrated the lamina propria and the detrusor muscle but there was no evidence of lymphovascular invasion and there was flat urothelium displaying CIS.
Sections from the muscle biopsy showed fragments of lamina propria with detrusor muscle infiltrated by the LELC; however, there was no evidence of lymphovascular invasion.
In summary, based on the histological appearances of the tumour and muscle biopsy, a histological diagnosis of pT2 LELC of the urinary bladder was made. Subsequent staging CT of her thorax, abdomen and pelvis did not reveal any lymphadenopathy or metastasis.
This patient's case was discussed at a multidisciplinary team meeting, following which she was seen in the clinic, where her further treatment options were discussed with her; these included radical cystectomy and radical radiotherapy. The patient opted for radical radiotherapy and treatment was instituted.
Eight months following her initial diagnosis of a bladder tumour, the patient was admitted because she had symptoms of lethargy. CT revealed right-sided hydroureter, but there was no evidence of a recurrent tumour in the urinary bladder. There was also no evidence of any significant lymphadenopathy or metastasis and, because of the right-sided ureteric obstruction, a right nephrostomy was made and an antegrade right ureteric stent was inserted.
Nine months after transurethral resection of her tumour, the patient underwent her first check cystoscopy, which revealed a reduced bladder capacity and inflamed-looking bladder without any evidence of recurrent tumour. Bimanual examination during the cystoscopy did not reveal any abnormal bladder masses. The patient underwent a number of bladder biopsies and the histological results revealed chronic cystitis and no evidence of carcinoma. She was discharged and registered for a review cystoscopy after 3 months (12 months after her initial transurethral resection of the bladder tumour).
Holmang et al.34 and Amin et al.35 reported that carcinomas of the urinary bladder with a lymphoepithelioma-like component constitute < 1% of all carcinomas of the urinary bladder. There are some preliminary lines of information that suggest that pure or predominant LELCs have a relatively favourable outcome.
Amin et al.35 reported the results of their study as follows:
Three patients with pure LELC of the bladder were treated by transurethral resection of bladder tumour and chemotherapy only. After an average follow-up of 47.6 months, there was no evidence of disease.
Five patients with predominant lymphoepithelial-like cell carcinoma of bladder also showed no evidence of disease after an average follow-up of 12.6 months.
Three patients with a focal (≤ 50%) LELC had died by the time of follow-up; two deaths were directly tumour related.
Holmang et al.34 reported that, out of nine patients with stage T2M0–T3M0 carcinoma of the urinary bladder who were treated with locoregional therapy, six with either pure or predominant LELC pattern were alive and had no evidence of disease after a median follow-up of 4 years (range 1–18 years); the remaining three patients, with a focal variant, died as a result of their disease 9–68 months after treatment.
Lopez-Beltran et al.36 reported the results of their study as follows:
At a mean follow-up of 33 months (range 21–47 months), three patients with pure LELC of the urinary bladder were still alive. Two of these patients had been treated with transurethral resection of bladder tumour and chemotherapy while the third patient underwent cystectomy.
Of the patients with predominant LELC of the urinary bladder, 66% were disease free and 33% had died by the time of follow-up.
All four patients who had focal LELC died of their disease.
The results of these studies indicate that pure or predominant LELC is associated with afavourable prognosis.
Tamas et al.31 reported their study of 28 cases of LELC of the urinary bladder, one case in the renal pelvis and one in the urethra and the tumours were classified as ‘pure’ or ‘mixed’. The authors reported that the mean age of the patients was 67.6 years and 21 patients were male (70%). Fifteen cases (50%) were pure and the remaining cases were mixed with other patterns of carcinoma, including invasive urothelial carcinoma (n = 10), invasive adenocarcinoma (n = 3) and squamous cell carcinoma (n = 2). The surface demonstrated CIS in six cases, non-invasive high-grade papillary urothelial carcinoma in three cases and in situ adenocarcinoma in one case. In 19/30 (63%) cases, there was a heavy lymphocytic infiltrate, and with regard to the findings of the histological examination of the tumours, the remaining 11/30 (37%) cases had a mixed inflammatory infiltrate. None of the 26 cases tested positive for EBV RNA by in situ hybridization. Seven cases presented with tumour stage T1 (23%), 14 with stage T2 (47%), seven with stage T3 (23%) and two with stage T4 (7%). The treatment consisted of radical cystectomy in 13 out of 30 cases (43%), partial cystectomy in 4 out of 30 cases (13%), nephrectomy in one case (3%) and transurethral resection, often followed by radiotherapy or chemotherapy, in 12 cases (40%).
The mean follow-up for patients without progression was 31 months. Eight of the 27 patients followed up (29.6%) experienced tumour recurrence, with seven of these patients displaying metastases. In patients treated by radical cystectomy, the 5-year actuarial recurrence-free risk was 59% (62% and 57% for pure and mixed cases, respectively). According to Tamas et al.,31 the prognosis for patients with LELC, whether in pure or mixed form, when treated with cystectomy, is similar to that of patients with ordinary urothelial carcinoma. Of the three patients with pure disease treated by chemotherapy, two were free of disease at 4 and 65 months and the third experienced disease recurrence at 17 months. The authors speculated that, given the association of LELC with urothelial carcinoma in 47% of their cases, and its propensity for multifocality, partial cystectomy in patients with these tumours would be ill advised. The main results from the Tamas et al.31 study are:
No significant differences in recurrence-free survival were observed between patients in the ‘pure’ and ‘mixed’ groups following cystectomy for muscle-invasive disease.
Two of three patients with pure LELC, who were treated by chemotherapy, were free of disease at 4 and 65 months' follow-up. However, the third patient developed recurrent disease at 17 months.
In the two patients who underwent transurethral resection of bladder tumour alone, there was no evidence of disease at 38 and 48 months.
Kruslin et al.37 reported a case of LELC of the urinary bladder in a 70-year-old man who presented with a 1-month history of painless haematuria. He underwent transurethral resection of his bladder tumour twice and the diagnosis following histopathological examination of the resected tumour was transitional cell carcinoma. However, histopathological examination of a third biopsy showed tumour tissue with typical syncytial growth pattern of atypical, large, epithelioid cells with ill-defined cytoplasmic borders, prominent nucleoli and numerous mitoses. A prominent lymphocytic infiltrate was observed as a component of the tumour. Immunohistochemistry of the specimen showed a positive reaction for CK and EMA and negative results for LCA, CD3, CD20, CD68 and PSA. The lymphoid infiltrate was an admixture of T- and B-cells and the tumour had invaded the muscle wall. The patient underwent chemotherapy by local application of doxorubicin hydrochloride (Adriablastine) following his last surgery. At a 10-month follow-up, no tumour recurrence was detected.
Guresci et al.38 reported the case of a 90-year-old man with haematuria who underwent transurethral resection of an undifferentiated bladder tumour that showed prominent lymphoid infiltration on light microscopy. Immunohistochemical examination of the specimen revealed positive staining of tumour cells for CK, EMA and CK-20. This case was reported because of its rarity and to show that LELC of the urinary bladder can affect a patient of advanced years.
According to Porcaro et al.,39 LELC is typically diagnosed in the sixth to eighth decades of life (mean age at diagnosis 68 years) and constitutes 0.4–1.3% of all bladder carcinomas.
Lopez-Beltran et al.36 stated that LELC has a strong association with EBV as well as cancers at other anatomical sites. The authors investigated the possible oncogenic role of EBV in 13 cases of LELC using in situ hybridization and EBV-encoded latent membrane protein 1. Their results failed to demonstrate the presence of EBV in any LELC cases.36 Their observations were similar to those of Amin et al.,35 which, in the opinion of Mayer et al.,40 would be suggestive of an alternative aetiology for LELC in the urinary bladder, such as an exaggerated host immune response.
Mayer et al.40 reported that LELC has a typical morphological appearance, comprising tumour cells with a syncytial growth pattern and poorly defined cytoplasmic borders. The atypical cells have large, prominent nucleoli and frequent mitoses.40 A dense cellular infiltrate is present in LELC, consisting of mature lymphocytes that are often admixed with plasma cells and histiocytes.40
A number of authors6,34,35 have stated that the diagnosis of LELC is supported by immunohistochemical staining that shows positivity to markers of the lymphoid cells that constitute the inflammatory infiltrate, the atypical neoplastic cells and the urothelial component of the tumour.6,36
Amin et al.35 stratified the tumours according to a classification system based upon the percentage of LELC pattern within the tumour: pure (100%), predominant (≥ 50%) and focal (< 50%).
Some authors6,35 have claimed that the differential diagnosis of LELC of the urinary bladder includes chronic cystitis, poorly differentiated transitional cell carcinoma of bladder with a prominent lymphoid stromal infiltrate and malignant lymphoma of the urinary bladder.
According to Porcaro et al.,39 LELCs of the urinary bladder are usually muscle-invasive by the time of diagnosis (90% are either stage pT2 or pT3). Porcaro et al.39 reviewed 43 cases of primary lymphoepithelioma of the urinary bladder that were reported in the English literature between 1991 and 2002, including their own reported case. They found that the majority of patients (93%) had undergone either transurethral resection of the bladder tumour or radical cystectomy for local control. Porcaro et al.39 reported that the overall population of patients included 31 males and 12 females. The average age was 68.4 years (range 52–84 years) and the histological subtype was pure in 17 cases (39.5%), predominant in 16 (37.2%) and focal in 10 (23.3%). The mean follow-up was 37.7 months (range 0–216 months) and the overall reported outcomes were that 26 patients (60.5%) did not have any evidence of disease, 11 died of disease (25.6%), one (2.3%) was alive with metastases and four (9.3%) died of diseases unrelated to LELC.
Porcaro et al.39 calculated an overall survival rate of 71%. The mean follow-up was 48.1 months in the group with pure LELC, 32 months in the group with predominant LELC and 30.3 months in the group with focal LELC. Thirteen (81%) patients with LELC had no evidence of disease, compared with 13 (82%) with predominant LELC and 0 (0%) with focal LELC. One patient (6%) with pure LELC died as a result of the disease, as did one (6%) with predominant LELC and nine (90%) with focal LELC. Two patients (13%) with pure LELC died of disease unrelated to their tumour, as did one (6%) with predominant LELC and one (10%) with focal LELC. One patient (6%) with predominant LELC was alive with metastases. The overall survival rate was 93% in pure LELC group, 93% in the predominant LELC group and 0% in focal LELC group. Porcaro et al.39 drew the following conclusions:
There are currently no clear-cut guidelines for the treatment of LELC of the urinary bladder.
Treatments performed include deep transurethral resection of the bladder tumour and partial or radical cystectomy with or without adjuvant treatment, including systemic chemotherapy.
The prognosis is favourable for patients with the pure or predominant forms with a diploid DNA pattern but very poor for patients who present with a focal form of the disease.
Bladder salvage therapy by transurethral resection of the tumour alone or accompanied by adjuvant systemic chemotherapy may be a reasonable option for patients with pure or predominant LELC, whereas radical surgery with adjuvant systemic therapy may be indicated for focal muscle-invasive LELC of the urinary bladder.
Mayer et al.40 reported that there a lack of clarity regarding how patients are stratified for selection of a particular treatment modality and they believe that tumour variability makes it impossible to directly compare two treatment modalities even in adjusted and matched groups. However, aggregated data from Porcaro et al.39 would suggest that oncological outcome, as defined by disease-specific survival (DSS) rate, is similar irrespective of the type of treatment. In their study, the DSS rate was 93% in the group with predominant LELC, of whom 44% underwent transurethral resection of bladder tumour and 50% underwent radical cystectomy. In the focal LELC group, of whom 40% underwent transurethral resection of bladder tumour and 60% had radical cystectomy, the DSS was 0%. Porcaro et al.39 identified three patients who underwent partial cystectomy for LELC of the urinary bladder: two patients had no evidence of disease at 36 and 72 months' follow-up and the third patient had stage T3 pure LELC and had died as a result of the tumour at 41 months' follow-up.
According to Dinney et al.,41 LELC of the urinary bladder has been successfully treated with various systemic chemotherapy regimens such as adjuvant therapy after transurethral resection of bladder tumour, radical cystectomy or as primary multiagent chemotherapy. The most favourable outcome reported by Dinney et al.41 was associated with cisplatin-based systemic chemotherapy. Two out of three patients who were treated with primary systemic chemotherapy had no evidence of disease recurrence at 5 and 6 years' follow-up; however, in the case of the third patient there were insufficient follow-up data to assess the patient's response to chemotherapy.
Mayer et al.40 found a DSS rate for all grades of LELC of the urinary bladder, irrespective of adjuvant therapy use, of 71% (with a mean follow-up of 38 months). They extrapolated from the review by Porcaro et al.39 that the DSS rates for patients with well-differentiated, pure LELC and predominant LELC were equal, at 93%. This was despite the fact that 88% of the pure group had received adjuvant therapy, in comparison with only 25% of the predominant group (mean follow-up 48 months vs. 32 months, respectively). Porcaro et al.39 reported that the DSS rate in the focal LELC group was 0%; however, only 30% of patients in this group had received adjuvant chemotherapy.
Zukerberg et al.6 postulated that the favourable outcome associated with pure and predominant LELC of the urinary bladder could be explained by the strong host immune response to the tumour cells that have been incited by the inflammatory infiltrate. They also suggested that the inflammatory reaction results in the development of irritative bladder symptoms at an early stage, which leads to the detection of the tumour at a relatively early stage, when it is more amenable to local treatments.
According to Mayer et al.,40 the encouraging survival figures for pure and predominant LELC of the urinary bladder, the more favourable prognosis for patients with these forms of the disease compared with patients with transitional cell carcinoma (as reported by Lopez-Betran et al.36) and the sensitivity of these variants of LELC to systemic cisplatin-based chemotherapy regimens have led some authors, including Lopez-Bertran et al.36 and Amin et al.,35 to suggest that these types of LELC should be treated with transurethral resection of the tumour, with or without systemic chemotherapy, rather than by radical cystectomy. The more aggressive nature of focal LELC of the urinary bladder means that patients suffering from this form of the disease may be best managed with radical cystectomy alongside adjuvant systemic chemotherapy. In view of the rarity of LELC of the urinary bladder, there is no accepted opinion on the approach to its management.
Porcaro et al.39 suggested that endoscopic resection of LELCs should be followed by cystoscopy every 3–6 months for the initial 2 years and on a yearly basis thereafter. Kassouf et al.42 suggested that more frequent cystoscopies, combined with regular radiological imaging for a longer period, may be required in patients with LELC of the urinary bladder who have undergone partial cystectomy for the treatment of muscle-invasive disease, in view of the notable rates of fatal late recurrences.
In 2009, Fadare et al.43 concluded that LELC of the urinary bladder is often mixed with conventional transitional cell carcinoma and/or other histotypes. They also claimed that determination of the morphological purity of LELC by the pathologist at the biopsy stage is clinically relevant as there is some anecdotal evidence suggesting that pure or predominant LELC may be comparatively chemosensitive and have a favourable prognostic profile, which may offer the possibility of effective therapy without bladder resection. In addition, they stated that the precise degree of cellular pleomorphism present in a pure LELC is not clear.
These authors reported a case of conventional pure LELC that showed a two- to sixfold variation in nuclear size, including multinucleated tumour cells, in a localized area about 25% of the volume of the overall tumour. The pleomorphic areas were set in the same lymphoplasmacytic infiltrate as their conventional counterparts and also displayed cellular syncytia. Fadare et al.43 performed a detailed immunophenotypic comparison of the conventional and pleomorphic areas but did not find any significant difference between the two areas in the overall lymphoplasmacytic or histiocytic density, lymphocytic CD4/CD6 ratio or lymphoplasmacytic kappa/lambda ratio. Likewise, both were found to display similar qualitative and quantitative staining indices for p53, ki67, CK AE1/AE3 and p16(INKa). They also found that the cytoplasm of scattered cells was beta (β)-catenin positive exclusively in the pleomorphic areas; nevertheless, these cells were notably larger than the cells in the conventional areas. Both components were immunohistochemically negative for Human Melanoma Black-45, CD1a, oestrogen receptor, EBV, CD117, D2-40, CD56, CK20 and chromogranin. Clinicopathological analysis of a series of cases was required to establish if there was any significance to nuclear pleomorphism in LELC. However, the phenotypic similarity between the two areas in this case, the intimate admixture of the pleomorphic cells with the lymphoplasmacytic infiltrate and their syncytial pattern of growth all suggested that pure LELC may display marked nuclear pleomorphism and that this may not be a valid basis for removing a case from the ‘pure’ group.
In 2004, Izquierdo-Garcia et al.44 stated that LELC of the bladder was uncommon and, at that time, only 49 cases had been described in the English literature, none of which had been studied for the expression of p53 protein. They reported that they studied three muscle-infiltrating cases of this tumour by means of immunohistochemsitry, in situ hybridization and polymerase chain reaction (PCR). They found that the three cases were positive for epithelial markers and negative for lymphoid antigens in the tumoral syncytial areas. The intensive infiltrate of small cells was negative for epithelial and positive for lymphoid markers. This population was mainly made up of cytotoxic T-lymphocytes and positive for TIA-1. p53 protein was intensively positive in more than 90% of the epithelial component nuclei and was negative in the lymphoid cells. PCR study did not show mutations on p53, but both lymphocytes and epithelium were negative for EBV markers, such as the latent membrane protein and EBV-encoded RNA. The prognosis was found to be very good after radiotherapy and chemotherapy, preserving the bladder despite the muscle infiltration. The authors concluded that the presence of an intense cytotoxic T-lymphocyte population may be related to this good prognosis. Both p53 protein status and T-lymphoid population had not been previously studied in cases of bladder LELC.
Yaqoob et al.45 reported the case of a 55-year-old woman who presented with a 5-month history of burning micturition and gross haematuria. Ultrasonography of her renal tract returned normal results, but cystoscopy revealed a 4 cm × 4 cm sessile mass in the right posterolateral wall of the bladder invading deep into the bladder muscle. This was completely resected endoscopically and macroscopic examination of the specimen revealed multiple irregular fragments of tissue measuring 5 cm in aggregate. Microscopic examination of the specimen revealed sheets and clusters of large epithelial cells that had moderate amounts of cytoplasm. The lymphoid background consisted of mature lymphocytes mixed with plasma cells and abundant mitotic figures were observed. The nuclei were round to oval in shape and vesicular with prominent nucleoli. Glycogen was identified in the epithelial cells and the tumour cells were strongly positive for epithelial markers (CK MNF, CK cam 52, CK AE1/AE3) and negative for lymphoid markers (LCA CD45RD, Pan BCD20 and CD79a, Pan T, UCHL1) and the vimentin protein. p53 was positive in 80–85% of the nuclei of the epithelial cells and negative in the lymphoid cells. The overlying urothelium showed no dysplasia and the tumour was deeply infiltrated into the detrusor muscle. A diagnosis of LELC of the urinary bladder was given based on the above markers.
Trabelsi et al.46 reported a 58-year-old man who presented with a 2-week history of macroscopic haematuria. Cystoscopy revealed a 4 cm × 4 cm sessile mass in the bladder and pathological examination of the resected specimen was consistent with the diagnosis of poorly differentiated carcinoma infiltrating the muscle of the bladder; therefore, a radical cystoprostatectomy was performed. Macroscopic examination of the specimen revealed a well-demarcated tumour in the dome of the bladder which measured 2 cm × 3 cm × 1 cm. Microscopic examination revealed an undifferentiated carcinoma with a lymphoid stroma and the tumour cells were characterized by a syncytial growth pattern in a dense lymphoid stroma. Immunohistochemically, most of the tumour cells were found to be positive for EMA and CK and the surrounding cellular infiltrate was a mixture of B- and T-lymphocytes. Hybridization to EBV-encoded RNA was negative and the patient was alive and disease free 8 months after the diagnosis.
In 2005, Chen et al.47 reported two patients who were treated for LELC of the bladder at the age of 57 and 76 years. Both patients presented with haematuria and the first patient was diagnosed with pT1 G3 transitional cell carcinoma at presentation. He underwent three transurethral resections for pT2b G3 pure LELC tumour recurrences and proceeded to radical cystectomy. The second patient presented with a PT2 G3 transitional cell carcinoma with squamoid and LELC component and was treated with radical cystectomy. The patients were alive at 20 and 18 months' follow-up, respectively. The authors reported that the LELC component in both tumours was indistinguishable from lymphoepithelioma of the nasopharynx and was characterized by nests of undifferentiated malignant cells with ill-defined cytoplasmic borders, prominent nucleoli and numerous mitoses growing in a syncytial pattern. The neoplastic cells showed intense positivity for CK (AE1/AE3) and the lymphoid infiltrate comprised mixed B- and T-cells (CD3 and CD79a) with many admixed CD68-positive macrophages.
Cai and Parwani,48 in 2008, claimed that LELC of the urinary bladder is a rare variant of high-grade urothelial carcinoma. They reported urine cytological findings in two cases of this rare entity, the diagnosis of which was confirmed by histological examination of the resected tumours. The cytological features that were found by Cai and Parwani48 included large tumour cells with high nuclear to cytoplasmic ratios, vesicular chromatin and prominent nucleoli presented as single cells or interspersed with inflammatory cells. They stated that the differential diagnoses of the cytological findings included otherwise typically high-grade urothelial carcinoma, reactive urothelial cells and rarely large cell lymphoma. They concluded that the rarity of the tumour cells may pose a diagnostic challenge when presented with a urine specimen.
Singh et al.49 stated that LELC of the urinary bladder is a rare malignant neoplasm of the urinary bladder that can histologically mimic lymphoma, poorly differentiated invasive transitional cell carcinoma or poorly differentiated squamous cell carcinoma with a lymphoplasmacytic background. Singh et al.49 reported the case of a 65-year-old man who presented with a history of recurrent painless haematuria of several weeks' duration. Cystoscopy revealed a sessile ulcerated lesion in the right posterolateral wall of the bladder that measured approximately 3 cm in diameter. CT revealed a soft tissue tumour mass in the right posterolateral aspect of the urinary bladder that extended to the right ureteric orifice and right-sided hydroureter. There was no evidence of perivesical spread of the tumour or metastasis; therefore, transurethral resection of the tumour was attempted. The pathologist reported that the biopsy results indicated LELC with infiltration into the muscularis propria. A radical cystectomy was performed and the cystectomy specimen measured 8.5 cm × 7.4 cm. On macroscopic examination of the cut specimen, an ulcerated tumour mass was found at the right posterolateral wall of the bladder that measured 3 cm × 3 cm × 1.5 cm. Microscopic examination of the specimen revealed diffuse sheets and cords of undifferentiated cells with large pleomorphic nuclei and coarse chromatin; many of the nuclei contained prominent nucleoli. The cytoplasm was moderate with poorly defined borders and the background showed dense lymphoplasmacytic infiltration with occasional eosinophils. The tumour showed transmural infiltration that did not invade the perivesical adipose tissue. There was no evidence of an in situ component in any of the sections that were examined. The entire tumour was of similar morphology, without any associated component of conventional urothelial type carcinoma or any other subtype, such as squamous cell carcinoma or adenocarcinoma. The tumour cells on immunohistochemistry were found to be positive for pan-CK, CK7, CK8 and EMA, but negative for chromogranin, vimentin and desmin. A diagnosis of pure LELC with evidence of muscular invasion was made. The patient had an uneventful postoperative recovery and was symptom free 12 months after the operation.
Williamson et al.50 stated that an investigation of immunohistochemical and molecular characteristics of bladder LELC is limited and the pathogenesis and biological behaviour of these tumours are controversial. Williamson et al.50 examined the clinicopathological features of LELC of the urinary tract using light microscopy, immunohistochemistry for CK7, CK20, 34βE12, p53, p63, alpha (α)-methylacyl-coenzyme A racemase, thyroid transcription factor-1, EBV latent membrane protein-1 and CD30, and in situ hybridization for human papillomavirus (HPV) and UroVysion fluorescence in situ hybridization (FISH). The authors reported that they identified tumours from 34 patients (male to female ratio of 2.8:1) ranging in age from 54 to 84 years (mean 70 years), the largest series to date. Williamson et al.50 identified urothelial CIS in 50% of patients. 34βE12 (75%), CK7 (57%) and p6 (53%) were frequently positive in tumour cells; however, thyroid transcription factor-1 and CD30 were consistently negative, with weak positivity in a single case. UroVysion FISH showed frequent chromosomal abnormalities similar to those of urothelial carcinoma. The authors also reported that in tumours with concurrent urothelial, squamous, sarcomatoid and glandular components, identical FISH abnormalities were noted in both areas. Williamson et al.50 further reported that in situ hybridization for HPV and immunostaining for EBV were negative in all studied lesions. Five patients with pure or predominant LELC tumours who were treated with transurethral resection and followed by chemotherapy were alive without evidence of the disease at 2–5 years. In contrast, two patients with less than 50% LELC morphology who were treated similarly died from the tumour or from distant metastasis. Williamson et al.50 stated that:
Urinary tract LELC is a rare histological variant of urothelial carcinoma.
The frequent presence of UroVysion FISH abnormalities, urothelial CIS and p53 positivity by immunohistochemistry in cases of urinary tract LELC suggests a similar pathogenesis to high-grade invasive urothelial carcinoma.
In contrast to typical urothelial carcinoma, CK20 is frequently negative in LELC.
Pure or predominant LELC may be treated with transurethral resection and chemotherapy. Nevertheless, a large-scale study with long-term follow-up is required to better understand the biological behaviour of urinary bladder LELC.
Pantelides et al.51 described LELC of the urinary bladder as a rare variant that can occur in a pure form or in conjunction with transitional cell carcinoma. They also claimed that, owing to the scarcity of reported cases, the optimum treatment is yet to be defined, although the benefits of chemotherapy are increasingly recognized. Pantelides et al.51 reported the case of a 64-year-old man who presented with a 2-month history of haematuria. A non-smoker for the previous 30 years, he was otherwise asymptomatic with no significant occupational risk factors. Flexible cystoscopy revealed a mass within the bladder neck but CT showed no evidence of metastases or lymphadenopathy. The patient underwent a transurethral resection of bladder tumour and subsequent histological examination of the specimen revealed a high-grade T2 LELC with no conventional transitional cell carcinoma present. The patient received four cycles of gemcitabine- and platinum-based chemotherapy. Gemcitabine was given at 1 g/m2 on days 1 and 8 of a 21-day cycle and cisplatin was given at 70 mg/m2 on day 1 and was replaced by carboplatin AUC5 on day 1 for cycles 2–4 when the patient experienced ototoxicity. Bladder biopsies from a repeat cystoscopy after the third cycle of treatment revealed only minimal non-specific chronic inflammation with no remaining malignancy. He was free of disease at 6 months' follow-up.
According to Serrano et al.,52 LELC has been described in multiple locations but only 55 cases in the bladder have been reported. The authors described the 56th such case in 2008 and reviewed all published cases, with the aim of defining their characteristics and obtaining a therapeutic and prognostic guide applicable to this disease. They reviewed the epidemiological characteristics, treatments and outcomes of the 56 published cases (40 males and 16 females with a mean age of 69 years). Histological subtypes were determined following the classification of Amin et al.35 Nineteen cases were of the pure subtype (33.9%), 20 the predominant subtype (35.7%) and 11 the focal subtype (19.6%); there was no indication of histology in six cases (10.7%). With regard to tumour stage, six cases (10.7%) were T1 stage, 32 (57.1%) were T2 stage and 17 (30.4%) were T3 stage. Regarding treatment, 58.9% underwent transurethral resection, 35.7% radical cystectomy and 5.4% partial cystectomy while 42.9% of patients received no adjuvant treatment, 30.4% received chemotherapy and 19.6% received radiotherapy. Overall survival was 67.9% and disease-free survival 64.3% with a mean and median follow-up of 34.5 and 25 months respectively.
Serrano et al.52 reported that 84.3% of patients with disease of the pure subtype, 100% of those with the predominant subtype and 76.7% with the focal subtype presented with stage T2/T3 infiltration and that 78.9% of patients with the pure form, 45% with the predominant form and 45.5% with the focal form underwent transurethral resection of the tumours. The majority (83%) of patients with the pure subtype received adjuvant treatment, whereas 60% of patients with the predominant type and 63% of those with the focal type received no adjuvant therapy. DFS among patients with stage T2/T3 disease was 87.5% for the pure subtype, with a median follow-up of 39 months, 75% for the predominant type with a median follow-up of 22 months and 0% for the focal type, with a median follow-up of 18 months. Serrano et al.52 concluded that no specific therapeutic protocol can be established for patients with bladder LELC, even taking into consideration the apparent good outcome of the pure and predominant subtypes and the bad outcome of the focal subtype. It seems that transurethral resection may be a good alternative in selected patients with pure or predominant histology, even with infiltrative stages. However, radical treatment with cystectomy and adjuvant treatment seems to be the best choice for the focal subtypes.
Chen et al.53 reported two cases in 2003. The first patient, a 75-year-old man, underwent a radical cystoprostatectomy for a muscle-invasive pure LELC with perivesical soft tissue invasion. He received cisplatin-based chemotherapy and was healthy without any evidence of tumour recurrence at 26 months' follow-up, based on CT. The second patient, a 63-year-old man, underwent transurethral resection of a pT1 LELC involving the dome of the urinary bladder. He received adjuvant chemotherapy in view of suspicion of extravesical tumour spread and characteristics suggestive of high-grade tumour. The tumour later recurred and as a result the patient underwent transurethral resection of the recurrent tumour; histological findings were consistent with high-grade (G3) superficial LELC. He received instillations of intravesical bacillus Calmette–Guérin and cystoscopy 10 months postoperatively revealed no evidence of tumour recurrence.
A summary of the reported experiences gained by a number of authors in relation to the diagnosis and management of LELC of the urinary bladder is provided in Table 1.
Lymphoepithelioma-like carcinoma of the urinary bladder is an extremely rare tumour. There is no consensus of opinion on the optimum treatment in view of the rarity of the disease entity. Nevertheless, there is evidence to suggest that patients with pure or predominant LELC of the urinary bladder should undergo transurethral resection with or without adjuvant cisplatin-based systemic chemotherapy and patients suffering from focal LELC should be treated with radical cystectomy with adjuvant cisplatin-based systemic chemotherapy. Follow-up after bladder-preserving treatment should include regular check cystoscopies, as are required for transitional cell carcinoma of the urinary bladder.
|Authors||Number of patients||Types of tumour||Treatment||Result||Follow-up|
|Amin et al.35||3||Pure LELC||Transurethral resection of bladder tumour and chemotherapy||No recurrence all alive||Average 47.6 months|
|5||Predominant LELC||One had transurethral resection of bladder tumour and chemotherapy and four had different types of treatment||No recurrence all alive||12.6 months|
Focal LELC < 50%
Tumours were all poorly differentiated, 10 were muscle-invasive and one was non-muscle invasive
|Different types of treatment||Three died of tumour||Follow-up for all mean 38 months (Range 9 to 72 months)|
|Holmang et al.34||6||Pure or predominant LELC||Locoregional therapy||All six alive and well with no tumour recurrence||1–18 years|
|3||Focal||Locoregional therapy||All died||9 to 68 months|
|Lopez-Beltran A et al.36||3||Pure LELC||Two had transurethral resection of bladder tumour and chemotherapy and one underwent cystectomy||All three alive||Mean 33 months; range 21–47 months|
|6||Predominant LELC and concurrent TCC||66% alive and disease free; 33% died of the disease||Mean 33 months; range 21–47 months|
|4||Focal LELC||All died of their disease||Mean 33 months; range 21–47 months|
|Tamas et al.31||
One in renal pelvis and one in urethra (21 male 70%)
17 (60.7%) pure11 mixed with other types
10 with urothelial carcinoma; three with invasive adenocarcinoma; two with squamous cell carcinoma; six with surface CIS; three with non-invasive high-grade papillary urothelial carcinoma and one with in situ adenocarcinoma
Seven at stage T1 (23%); 14 at stage T2 (47%), seven at stage T3 (23%) and two at stage T4 (7%)
47% LELC associated with urothelial carcinoma and multifocality (thus partial cystectomy not recommended)
|Radical cystectomy in 13 out of 30 cases (43%); partial cystectomy in 4 out of 30 (13%); nephrectomy in one case (3%) and transurethral resection followed by radiotherapy or chemotherapy in 12 out of 30 cases (40%)||Eight of 27 patients with follow-up data (29.6%) had tumour recurrence with seven patients showing metastases. For cases treated by radical cystectomy, the 5-year actuarial recurrence-free risk was 59% (62% and 57% for pure and mixed cases, respectively). Of the three pure patients treated by chemotherapy, one was disease free at 4 months and one at 65 months; however, in the third the tumour recurred at 17 months||The mean follow-up for patients without progression was 31 months|
|Kruslin et al.37||1||Pure LELC. Previously had transurethral resection of bladder tumour for transitional cell carcinoma and recurred for a third time into a pure form of LELC||Transurethral resection of bladder tumour and chemotherapy||Alive and well with no recurrence||10 months|
|Porcaro et al.39 identified three cases and reviewed 43 literature cases||3||Muscle invasive||Partial cystectomy||Two were alive but one had died by follow-up||
36 months and 72 months follow-up (alive patients)
41 months (patient that had died)
Mean 37.7 months; range 0–216 months for all
17 cases pure
16 cases predominant
10 cases focal
|Details not in our source of information||
13 (81%) alive with no disease but one (6%) died of the disease
Two (13%) died of unrelated causes (DSS 93%)
13 (82%) alive with no disease and one (6%) died of disease (DSS 93%)
One (6%) died of unrelated causes and one (6%) was alive with metastasis (DSS 0%)
None was disease free and nine (90%) died of disease, one (10%) died from an unrelated cause
Mean for pure group 48.1 months
Mean for predominant group 32 months
Mean for focal group 30.3 months
|Dinney et al.41||2||Muscle-invasive LELCs (details of subtype not available from our source)||Cisplatin-based chemotherapy||Two alive at 5 and 6 years||5 and 6 years|
|1||Muscle-invasive LELCs (details of subtype not available from our source)||Cisplatin-based chemotherapy||Insufficient follow-up data, outcome not available||5 and 6 years|
|Yaqoob et al.45||1||Pure muscle-invasive LELC||Transurethral resection of bladder tumour||Case was reported shortly after transurethral resection of bladder tumour. Outcome was not included in case report||No follow-up data|
|Trabelsi et al.46||1||G3 pure muscle-invasive LELC||Radical cystectomy||Alive and disease free||8 months|
|Chen et al.47||2||
One was previously G3pT1b transitional cell carcinoma that had recurred three times and the most recent was LELC
The second patient had a transitional cell carcinoma G3pT2b with squamoid component and LELC
|Radical cystectomy||Two alive and free of tumour at follow-up||
|Singh et al.49||1||Pure muscle-invasive LELC||Radical cystectomy||Alive and well||12 months|
|Williamson et al.50||5||Pure or predominant LELC||Transurethral resection of tumour plus systemic chemotherapy||All alive between 2 and 5 years||2–5 years|
|2||Focal LELC||Transurethral resection of tumour plus systemic chemotherapy||Both died of their disease||The first patient died of the disease at 39 months and the second patient was alive with lung metastases at 34 months|
|Pantelides et al.51||1||Pure muscle-invasive (T2) LELC||Transurethral resection of bladder tumour plus four cycles of gemcitabine- and platinum-based chemotherapy||Alive and well||6 months|
|Chen et al.53||2||Pure muscle-invasive (T3/4) LELC||
One patient underwent radical cystoprostatectomy for the pure muscle-invasive tumour that was invading the perivesical soft tissue
The second patient underwent transurethral resection of bladder tumour for PT1 pure LELC. There was suspicion of extravesical tumour spread; therefore, adjuvant chemotherapy as given. The patient developed G3 recurrent superficial LELC and had further transurethral resection of bladder tumour and intravesical bacillus Calmette–Guérin
|One patient was alive and well at follow-up with no recurrence and the second also showed no recurrence at follow-up||The first patient had a follow-up at 26 months and the second patient at 10 months|