Introduction

Neuroendocrine neoplasms (NENs) of the urinary bladder represent less than 1% of all malignancies in this site and are mainly represented by neuroendocrine carcinoma (NEC), whereas well-differentiated neuroendocrine tumors (NETs) are only anecdotally reported [1]. A significant proportion of NECs of the urinary bladder contains a non-neuroendocrine component, mostly represented by urothelial carcinoma and, more rarely, by squamous cell carcinoma or adenocarcinoma, and can be designated as mixed neuroendocrine/non-neuroendocrine neoplasms (MiNENs) in analogy to similar neoplasms arising in the digestive system [2]. Among vesical NECs, small cell NECs (SCNECs) are more frequently diagnosed than large cell NEC (LCNEC) [2, 3].

Here, we present a case of a MiNEN of the urinary bladder in which the neuroendocrine component, represented by a LCNEC, underwent complete pathological regression after neoadjuvant chemotherapy, while the non-neuroendocrine portion persisted and spread to metastatic sites.

Case history

A 49-year-old man was referred to the Urology Department for self-limiting painless gross hematuria in March 2018. Urinary cytology was positive for malignant epithelial neoplastic cells. Contrast-enhanced computerized tomography (CECT) showed a 46-mm-wide lesion located on the dome of the bladder (Fig. 1). Transurethral resection of the bladder (TURB) was then performed, and the specimen was sent to the Pathology service. A diagnosis of MiNEN composed of LCNEC and adenocarcinoma of the bladder was signed out. Computed tomography of the brain, chest, and abdomen did not show metastatic disease. The patient received 3 cycles of neoadjuvant chemotherapy (methotrexate, vinblastine, adriamycin, and cisplatin—MVAC).

Fig. 1
figure 1

Contrast-enhanced computerized tomography (CECT) of the bladder: Contrast-enhanced computed tomography revealed a 46-mm lesion on the dome of the bladder, with concomitant thickening of the bladder walls

Full size image

Radical cystoprostatectomy combined with the removal of pelvic and obturator lymph nodes was performed and a muscle-invasive poorly differentiated adenocarcinoma was reported, with no evidence of residual LCNEC. Three magnetic resonance imaging (MRI) scans of the abdomen were performed for clinical re-staging in January, May, and September 2019, respectively, without any evidence of relapse or metastatic disease.

In late November 2019, a growing lump on the penis and right epididymis was biopsied, revealing a poorly differentiated adenocarcinoma, without a neuroendocrine component. Emasculation was performed. After 2 years and 2 months after initial diagnosis, the patient is alive with ultrasonographic evidence of residual metastatic disease in inguinal lymph nodes.

Materials and methods

Morphology and immunohistochemistry

Tissue samples obtained from the different specimens (i.e., TURB, radical cystoprostatectomy, and percutaneous biopsy of the epididymis) were fixed in buffered formalin and routinely processed to paraffin wax. Five-micrometer-thick sections were routinely stained with hematoxylin and eosin and Alcian-PAS stain.

The immunohistochemical study was performed on additional 3-μm-thick sections using prediluted ready-to-use vials of the antibodies listed in Table 1 with an automated immunostainer (BenchMark Ultra, Ventana Roche Diagnostics) and standardized protocols (Ventana OptiView DAB IHC Detection Kit).

Table 1 Antibodies used for immunohistochemical analysis
Full size table

Review of the literature

The Pubmed database of the National Center for Biotechnology Information (NCBI) of the U.S. National Library of Medicine was searched using the following string “large cell neuroendocrine carcinoma [AND] urinary bladder.” All articles written in English were included. For each article, the reported cases were identified and, for each case, the following parameters were considered: age, sex, symptoms, presence of non-neuroendocrine component, immunophenotype, treatments, and outcome.

Results

Morphology and immunohistochemistry

The TURB specimen was entirely processed for microscopical analysis. Most of the specimens (70% of the total neoplastic volume) featured muscle-infiltrating neoplastic proliferation with organoid architecture, showing zonal necrosis (Fig. 2a). Neoplastic cells had moderately abundant, lightly eosinophilic cytoplasm, large vesicular nuclei, and focally prominent eosinophilic nucleoli. Apoptotic bodies were abundant and mitotic index was 40/10 high-power fields (HPFs) (Fig. 2b). Immunostains (Fig. 2c–h) were positive for Synaptophysin, Chromogranin A, CD56, CK Cam5.2, and, focally, for CK20 and TTF1. CDX2, GATA3, and p63 were negative. Intense cytoplasmic and nuclear p16 signal was also present, as well as p53 hyperexpression, whereas Rb1 expression was lacking. Ki67-related proliferative index was 85%.

Fig. 2
figure 2

Neuroendocrine carcinoma in vesical biopsy: Low (a, hematoxylin-eosin, × 50) and intermediate (b, hematoxylin-eosin, × 200) magnification showing solid, trabecular, and insular growth of large neoplastic cells. Zonal necrosis is also present. Immunohistochemical stains show positivity for general neuroendocrine markers (synaptophysin (c) and chromogranin A (d)). Ki67 proliferation index is very high (e) and tumor cells show hyperexpression of p16 (f) and p53 (g), whereas Rb1 expression is lost (h) (immunoperoxidase, × 200)

Full size image

The residual 30% of the total neoplastic volume was composed of an adenocarcinoma (Fig. 3), which was partially admixed with the former, but showed a tendency to be located in the most superficial layers of the bladder mucosa. Mitotic index was 4/10 HPFs. Immunostains for Synaptophysin, Chromogranin A, CD56, CEA, and p63 were negative, whereas those for CK Cam5.2, CK20, and GATA3 were diffusely positive and CDX2 was zonally expressed. Scattered cells were positive for TTF1. Rb1 was focally positive, while p16 and p53 had the same expression pattern as the neuroendocrine component. The final diagnosis was of muscle-invasive primary urinary bladder MiNEN, composed of LCNEC (70%) and moderately differentiated adenocarcinoma (30%).

Fig. 3
figure 3

Adenocarcinoma in vesical biopsy: Low (a, hematoxylin-eosin, × 20) and high (b, hematoxylin-eosin, × 400) magnification of papillary and gland-like structures of neoplastic cells with polarized nuclei

Full size image

The radical cystoprostatectomy specimen did not show, at gross evaluation, any residual neoplastic mass in the bladder. Microscopically, an estimated 90% of the vesical wall showed fibrosis and chronic inflammation with giant-cell granulomas. In the remaining 10%, residual poorly differentiated adenocarcinoma was present, showing discohesive atypical cells with signet-ring-like and lipoblast-like features (Fig. 4). p63 and, focally, GATA3 were positive, but TTF1, CDX2, Chromogranin A, Synaptophysin, and Rb1 were absent. No residual LCNEC was identified.

Fig. 4
figure 4

Vesical bladder surgical specimen: Poorly differentiated adenocarcinoma composed of discohesive signet-ring-like cells (a, hematoxylin-eosin, × 630), with intense positivity for Alcian blue (b, AB-PAS stain, × 630)

Full size image

In the percutaneous needle biopsy of the epididymis, poorly differentiated adenocarcinoma infiltrating fibromuscular tissue was seen (Fig. 5). Heterogenous positivity for GATA3 and p63 and negative stains for Chromogranin A, Synaptophysin, CD56, CD138, and PSA were observed. No evidence of LCNEC was found. The same morphological and IHC characteristics were observed in the specimen obtained from emasculation.

Fig. 5
figure 5

Epididymal biopsy: Poorly differentiated adenocarcinoma infiltrating with an “Indian file” pattern (a, hematoxylin-eosin, × 200), immunoreactive for GATA 3 (b, immunoperoxidase, × 200)

Full size image

Review of the literature

We identified 25 articles published between 1986 and 2020, reporting a total of 41 cases of LCNEC of the urinary bladder (Table 2) [4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28]. The male-to-female ratio was 36:5 and patients’ age at diagnosis ranged from 20 to 84 years, with a median of 61 years. Specifically, 23 cases (56.1%) were pure LCNEC, 7 cases (17.1%) were a combined SCNEC/LCNEC [20, 23], 1 case (2.4%) had sarcomatous components [8], and 10 cases (24.4%) showed epithelial non-neuroendocrine components. Overall, the amount of the epithelial non-neuroendocrine components was small: in two cases, it was reported to account for less than 2% and less than 5%, respectively [6, 20]; in the remaining cases, a descriptive report was given (i.e., “evidence of,” [9] “some foci of,” [13] or “minor contributions of” [16] epithelial non-neuroendocrine component).

Table 2 Published cases of large cell neuroendocrine carcinoma (LCNEC) of the urinary bladder
Full size table

Surgery and chemotherapy were the most frequently adopted treatments. Neoplasms were frequently muscle invasive, with or without fat infiltration, and commonly metastatic to regional lymph nodes. Outcomes were quite varied and based on follow-ups of different lengths.

Discussion

Our case is a rare example of what can be called a true MiNEN of the urinary bladder, as two morphologically distinct components, intimately admixed, one neuroendocrine and the other non-neuroendocrine, were evident, both morphologically and immunohistochemically. In addition, this case is strictly adherent to the criteria used for digestive MiNENs [3], as each component represented at least 30% of tumor mass. In contrast, in previously reported cases of mixed vesical LCNECs, only a minor non-neuroendocrine component was detected [6, 9, 13, 16, 20]. Indeed, the adoption of a 30% cutoff is not based on clinical evidence, but rather it was arbitrarily introduced to avoid overestimating the biological relevance of focal cells with a divergent differentiation, which would be unlikely to influence the overall prognosis [29]. Nevertheless, as it has been underlined elsewhere [2, 29], we believe that minor, but morphologically recognizable, neoplastic components with divergent differentiation must be recorded in the pathological report, above all when they are morphological high-grade, because they still may influence prognosis and need a specific management.

LCNECs of the urinary bladder are exceptionally rare tumors, with only 41 cases reported in the literature (Table 1). Given their rarity, the exclusion of vesical metastatic disease from an unknown primary site is of paramount importance. Clinical and radiological information is pivotal in this task, as immunohistochemical markers have poor reliability in the identification of the primary sites of NECs [30]. In our case, the application of immunohistochemistry for general neuroendocrine markers on the TURB sample highlighted the neuroendocrine component of the vesical neoplasm and allowed to negatively detect the non-neuroendocrine component. As expected, transcription factors (TTF1 and GATA3) were not useful in confirming or denying the vesical origin of the LCNEC.

The origin of NECs of the urinary bladder has been tentatively explained by different theories, including the possible derivation of the neoplastic neuroendocrine clone from a common multipotential cancer stem cell shared with non-endocrine carcinoma or its development from normal or metaplastic neuroendocrine cells of the urothelial mucosa [31, 32]. Recently, Chang and colleagues provided an elegant demonstration that the genomic alterations present in NECs of the urinary bladder more closely resemble urothelial carcinoma than small cell lung cancers, suggesting an organ-specific rather than a cell type–specific mechanism of cancerogenesis for NECs [33]. This model also explains the pathogenesis of vesical MiNENs and gives details on the molecular pathways involved. In the case of our patient, the non-neuroendocrine component was an adenocarcinoma with heterogeneous morphology across different specimens. It is conceivable that this change in morphology is related to intratumor heterogeneity, possibly enhanced by the selection of a previously unwitnessed neoplastic clone by neoadjuvant chemotherapy.

The morphological heterogeneity of MiNENs is mirrored by their variable prognosis, which, at least in digestive MiNENs including a NEC, seems to be driven by the high-grade neuroendocrine component and to be comparable to that of pure NECs [2]. Intriguingly, in our patient, the adenocarcinomatous dyscohesive component was revealed to be the most aggressive part of the MiNEN, persisting after neoadjuvant chemotherapy and giving rise to metastatic localizations. In contrast, the NEC component responded well to chemotherapy and did not recur. In fact, the M-VAC regimen administered to our patient was specifically chosen on the basis of the histopathological report on the diagnostic biopsy. The importance of a specific therapeutic approach to the NENs of the urinary bladder relies on their clinicopathological features [30, 34] and has been strengthened by molecular studies. Indeed, the latest consensus on the molecular subtypes of muscle-invasive bladder cancer identifies a neuroendocrine-like class featuring TP53 and RB1 gene mutations, poorly differentiated neuroendocrine morphology, poor survival, and sensitivity to platinum-based chemotherapy, similar to NECs affecting other organs [35, 36]. On the basis of this case, it should be considered that, at least in the urinary bladder, non-neuroendocrine cancer can play a pivotal role in the determination of life quality and prognosis even in the setting of the NEC-including MiNEN.

In summary, we have reported a rare, if not unique, case of LCNEC of the urinary bladder, admixed with a high-grade carcinomatous component, for which we endorse the term of MiNEN. The correct diagnosis on the preoperatory biopsy allowed the administration of a platinum-based neoadjuvant polychemotherapy to the patient, which was followed by the complete pathological response of the LCNEC component, which did not recur in metastatic sites.