Introduction

Primary adenocarcinoma of the urinary bladder is an uncommon tumor accounting for 0.5–2% of all bladder malignancies. The diagnosis of primary adenocarcinoma of the urinary bladder could be challenging, especially in small biopsy specimens [1, 2]. Histologically, primary adenocarcinoma manifests as pure adenocarcinoma with various morphologies including enteric, mucinous, or mixed pattern, with or without extracellular mucin [3]. The presence of any urothelial carcinoma component will exclude the diagnosis of primary adenocarcinoma of the urinary bladder.

Mutations in the TERT promoter have been implicated in carcinogenesis of 60–80% of urothelial carcinomas but the molecular pathways for primary adenocarcinoma of the urinary bladder are not well defined [4,5,6,7,8,9]. In this study, we evaluated TERT promoter mutations in 46 cases of primary adenocarcinoma of the urinary bladder and in 30 cases of urothelial carcinoma with glandular differentiation to gain further insight into the molecular carcinogenesis and the possible link to conventional urothelial carcinoma.

Materials and methods

Patients

Specimens from 46 patients with nonurachal adenocarcinoma of the urinary bladder and from 30 patients with urothelial carcinoma with glandular differentiation were retrieved from the archives of participating institutions. The diagnoses of primary adenocarcinoma of the urinary bladder and of urothelial carcinoma with glandular differentiation were based on the criteria in the World Health Organization Classification of Tumors of the Urinary System and Male Genital Organs [3]. All slides were reviewed retrospectively, and the diagnoses were confirmed. None of the patients with primary adenocarcinoma of the urinary bladder had history of prostate cancer or radiation therapy. Additionally, 24 cases of nephrogenic adenoma, eight cases of villous adenoma, 31 cases of florid cystitis glandularis, and 20 cases of intestinal metaplasia of the bladder were included in the TERT promoter mutation analysis.

This research was approved by the Institutional Review Boards of participating institutions in accordance with each Institutional Committee for the Protection of Human Subjects.

High-resolution fluorescence melting curve analysis

Genomic DNA was prepared from microdissected tissue areas. Approximately 1000–5000 cells were microdissected from 5 µm histological sections of formalin-fixed, paraffin-embedded tissue. Normal tissue microdissected from the same specimen was used as control samples for each patient. The DNA extraction was performed using the QIAamp DNA FFPE tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. DNA was measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA).

TERT promoter mutation detection was carried out using a LightCycler 2.0 real-time polymerase chain reaction (PCR) system (Roche Diagnostics, Basel, Switzerland), as previously described [10,11,12]. Forward and reverse primers were designed to amplify DNA products encompassing C228T and C250T hotspot mutations in the TERT promoter 124 and 146 base pairs upstream of the ATG starting site, respectively. Real-time PCR was performed in 20 µl reaction mixtures containing 10 ng of genomic DNA, 0.1 µM forward and reverse primers, 0.2 µM of each nucleotide, and 1.5 units of OneTaq DNA polymerase in GC Buffer (New England Biolabs, Ipswich, MA, USA). The PCR included one cycle denaturation at 95 °C for 6 min and 45 cycles of denaturation at 94 °C for 20 s, annealing at 53 °C for 20 s, extension at 68 °C for 40 s. The primer and probe sequences were as following: TERT forward primer 5′-CGGGCTCCCAGTGGATTC-3′; TERT reverse primer 5′-CAGCGCTGCCTGAAACTC-3′; TERT C228T donor probe 5′-CGACCCCTCCCGGGTCCCCG-Flc-3′; TERT C228T acceptor probe 5′-LC640-AGCCCCTTCCGGGCCCTCCCAG-phos-3′; TERT C250T donor probe 5′-GCGCGGACCCCGCCCCGTC-Flc-3′; TERT C250T acceptor probe 5′-LC640-CCCCTTCCGGGTCCCCGGCCC-phos-3′.

Melting analyses were performed in the original capillary with one cycle of 95 °C for 1 min, 40 °C for 2 min, and then ramped up to 70 °C at 0.1 °C/second. Melting curves of the PCR products were monitored by plotting the changes in fluorescence emission at 640 nm that occurred by gradual heating of the PCR product–probe complex [10]. The melting curve was analyzed by LightCycler version 2.0 software. A single melting curve peak with a melting temperature within a ± 2.5 °C range of the predicted melting point was interpreted as representing a pure, single amplicon. If two peaks were present and both peaks fell within ± 2.5 °C of the predicted melting temperatures for mutant and wildtype, the amplicon was interpreted as containing a mutant cell population and a wildtype population. Serial dilution tests with known positive DNA demonstrated that a mutant cell population >5% was consistently detected.

Statistical analysis

Student t, Pearson Chi-square, and Fisher exact tests were used in the statistical analysis of clinicopathological parameters and TERT promoter mutation status. Statistical significance was defined as P < 0.05 and all P values were two-tailed.

Results

Patient demographics and pathological characteristics

TERT promoter mutation status was analyzed in 46 primary nonurachal adenocarcinoma specimens including 16 cystectomy and 30 transurethral resection samples. Table 1 summarizes the patient characteristics and clinical findings.

Table 1 Clinicopathologic characteristics and TERT promoter mutation status in 46 cases of urinary bladder adenocarcinoma.
Full size table

Of the 46 cases of primary adenocarcinoma analyzed, 26 patients were male and 20 were female (M:F ratio, 1.1:1), with a median age of 64 years (range, 39–92 years) (Table 1). At the time of diagnosis, there were 26 (56%) pT1, six (13%) pT2, 10 (22%) pT3, and four (9%) pT4 cases. All pT1 lesions were diagnosed from transurethral resection specimens. In all pT1 cases, muscular propria invasion was not present. Among the 46 primary adenocarcinomas of the urinary bladder, 17 (37%) were enteric, 11 (24%) were mucinous, and 18 (39%) were mixed adenocarcinoma variants (Table 1).

Of the 30 cases of urothelial carcinoma with glandular differentiation analyzed, 26 patients were male and four were female (M:F ratio, 5.2:1), with a median age of 73 years (range, 57–88 years). Most patients were over 60 years of age (80%). The cohort included 14 cystectomy and 16 transurethral resection cases. At the time of diagnosis, there were six (20%) pT1, 19 (63%) pT2, and five (17%) pT3 cases (Table 2).

Table 2 Clinicopathologic characteristics and TERT promoter mutation status in 30 cases of urothelial carcinoma with glandular differentiation.
Full size table

Compared to primary adenocarcinoma of bladder, urothelial carcinoma with glandular differentiation had a high male-to-female ratio (5.2:1 vs 1.1:1; P = 0.0021) and higher median ages (73 vs 64 years; P = 0.0065), respectively.

TERT promoter mutation in the primary adenocarcinoma of urinary bladder

The prevalence of TERT promoter mutations C228T and C250T in primary adenocarcinoma of the urinary bladder was 33% (15/46 cases), among which 13 cases involved C228T (13/15, 87%) and 2 involved C250T (2/15, 13%) mutations. The incidence of TERT promoter mutations in histological variants of primary adenocarcinoma was 29% in enteric, 55% in mucinous, and 22% in mixed adenocarcinoma (Fig. 1; Table 1). The TERT mutation incidence among the histological variants was not statistically significant. There was no significant correlation of TERT promoter mutation with any clinicopathologic characteristics when considering all specimens or cystectomy specimens alone.

Fig. 1: TERT promoter mutation in primary adenocarcinoma of urinary bladder.
figure 1

Hematoxylin and eosin-stained sections of enteric adenocarcinoma A, mucinous variant B, and mixed variant C show C228T TERT mutation D, E, F as indicated by a distinct melting temperature demonstrated by the red vertical bar. The TERT wildtype G, H, I corresponding to D, E, F, respectively, is demonstrated by the green vertical bar. The mutant and wildtype melting points have a 4 °C difference. In the cohort, 33% of the primary adenocarcinomas of bladder were positive for TERT promoter mutations.

Full size image

We investigated KRAS mutations in a previous study [13]. TERT promoter mutations were absent in three KRAS-positive adenocarcinomas of the urinary bladder.

TERT promoter mutations in bladder urothelial carcinoma with glandular differentiation

TERT promoter mutations were detected in 20 (67%) of each paired urothelial carcinoma and glandular differentiation components, including 19 C228T and one C250T (Fig. 2). Identical TERT promoter mutation status was shared by the urothelial carcinoma and the glandular components in all cases, whether positive or negative for C228T or C250T mutations (Fig. 2).

Fig. 2: Concordant TERT promoter mutation in urothelial carcinoma with glandular differentiation.
figure 2

Hematoxylin and eosin-stained sections from two urothelial carcinoma components A, C and coexisting glandular components B, D showed concordant TERT promoter mutations in either C228T E, F or C250T G, H as demonstrated by the distinct melting temperature indicated by the red vertical bar. The corresponding wildtype I, J, K, L is indicated by the green vertical bar. The mutant and wildtype melting points have a 2.5–4 °C difference. In the cohort, 67% of urothelial carcinomas with glandular differentiatin were positive for TERT promoter mutations. M Correlation between TERT promoter mutations in urothelial carcinoma and glandular components. The TERT mutation status in 30 urothelial carcinomas (upper row) with glandular components (lower row) was lined up and color coded. Each column represents a case. The TERT promoter mutation status was highly concordant whether mutated or not.

Full size image

Compared to primary adenocarcinomas of bladder, urothelial carcinomas with glandular differentiation had a higher incidence of TERT mutation (33% vs 67%; P = 0.0048).

TERT promoter mutations in nephrogenic adenoma, villous adenoma, florid cystitis glandularis, and intestinal metaplasia of the bladder

Additionally, we analyzed TERT promoter mutations in nephrogenic adenoma (n = 24), villous adenoma (n = 8), florid cystitis glandularis (n = 31), and intestinal metaplasia (n = 20) of the bladder (n = 20). None of these lesions had TERT promoter mutations.

Discussion

Primary adenocarcinoma of the urinary bladder is a rare malignancy, accounting for 0.5–2.0% of all malignant vesical tumors found in patients in their sixth and seventh decades of life, and characterized by an aggressive behavior [1, 2]. One-third of the primary adenocarcinomas of the urinary bladder have advanced cancer with or without lymph node metastasis at the time of diagnosis [14, 15]. These cancers also occur in association with schistosomiasis, endometriosis, bladder augmentation, and other irrigative conditions of the urinary bladder. Primary bladder adenocarcinoma has a variety of growth patterns, pure or mixed, including enteric, mucinous, and mixed patterns [3]. The morphological features of these cancers are identical to those from other organs. Urothelial carcinomas also have diverse morphological manifestations, including glandular differentiation [16]. Although studies have identified common molecular pathways in conventional urothelial carcinoma, molecular data are lacking for adenocarcinomas of the urinary bladder [7, 8, 17, 18].

Mutations in the TERT promoter have recently been implicated in urothelial carcinogenesis and are present in 60–80% of conventional and variants of urothelial carcinomas [4,5,6, 9, 19]. TERT promoter mutations are proposed as the earliest genetic alterations in bladder carcinogenesis [20]. TERT promoter mutations contribute to tumorigenesis by promoting immortalization and genomic instability in two phases [21]. The TERT promoter mutation analysis may be a useful adjunct test in difficult cases [4, 7, 9, 10, 12, 22,23,24,25,26,27,28,29,30,31,32,33]. To investigate the degree to which TERT promoter mutations are involved in adenocarcinoma of the bladder, and to gain insight into the potential of linked tumorigenesis to conventional urothelial carcinoma, we evaluated TERT promoter mutations in a series of 46 cases of primary nonurachal adenocarcinoma of urinary bladder and compared to a cohort containing 30 urothelial carcinomas with glandular differentiation. We found TERT promoter mutations in 33% (15/46) of primary adenocarcinomas of the urinary bladder, providing evidence that some adenocarcinoma of the urinary bladder followed a common tumorigenesis pathway. However, the incidence of TERT mutations was far less than those observed in urothelial carcinoma with glandular differentiation (33% vs 67% in current study).

There were conflicting data on incidence of TERT promoter mutation in primary adenocarcinoma of the urinary bladder. In a study by Vail and colleagues, TERT promoter mutation was absent in an analysis of 10 cases of primary adenocarcinoma [32]. TERT promoter mutation is also rare (4%) in urachal adenocarcinoma [34]. In contrast, Cowan et al. investigated 14 cases of primary adenocarcinomas of urinary bladder and found 29% had TERT mutations [29]. These discrepancies may reflect differences in patient population, tumor location (urachal vs nonurachal), sample size, methodology divergence, tumorigenesis, and biology. Together, these data suggested that TERT promoter mutation may not be the only driver activation event in the carcinogenesis of primary adenocarcinoma. We also found TERT promoter mutations from 67% of urothelial carcinomas with glandular differentiation, in which the concordant TERT promoter mutation status was seen in all cases (Fig. 2). Our findings support that both glandular and urothelial carcinoma components were derived from a common clonal origin.

Glandular lesions of the urinary bladder are frequently encountered. The major benign glandular lesions in the differential diagnosis include nephrogenic adenoma, villous adenoma, florid cystitis glandularis, and intestinal metaplasia of the bladder [1, 2, 35]. None of these benign glandular lesions harbored TERT promoter mutations in the current study. The distinction of benign glandular lesions from malignant tumors is usually straightforward, but it could be diagnostically challenging in some cases. Likewise, some malignant urothelial tumors may also mimic adenocarcinoma of the bladder. One example is nested variant of urothelial carcinoma. High incidence of TERT promoter mutation was found in nested variant urothelial carcinoma in recent studies [9, 31, 33]. These data suggest that TERT promoter mutation analysis could potentially be helpful in the differential diagnosis of glandular lesions of the urinary bladder (Fig. 3).

Fig. 3: TERT promoter mutation status in benign lesions of urinary bladder.
figure 3

Hematoxylin and eosin-stained sections of nephrogenic adenoma A, villous adenoma C, florid cystitis glandularis E, and intestinal metaplasia G showed TERT wildtype, B, D, F, H, corresponding to A, C, E, G, respectively, as indicated by a distinct 58 °C melting temperature B, D, F, H.

Full size image

TERT promoter mutations exhibited high organ specificity. Metastatic adenocarcinoma of bladder from cancers originating from prostate, colorectal, breast, and lung rarely harbor TERT mutations [11, 32, 34, 36,37,38,39]. Wu et al. screened 302 patients with a variety of urogenital cancers and determined the clinical relevance of TERT promoter mutations in urogenital cancer [40]. Urothelial carcinomas showed the highest mutation incidence, but nonurothelial cancers showed very low or no mutation [40]. TERT genotypes have been shown to be conserved across spatially, temporally, and morphologically distinct components of a single urothelial tumor, further supporting its use as a relatively stable and reliable molecular biomarker [41]. Urachal adenocarcinoma can have a variety of morphological appearances and can secondarily involve the bladder. Thiem et al. investigated 23 urachal adenocarcinomas and found TERT promoter mutation only in one case (4%) [34]. In contrast, a high percentage of primary bladder adenocarcinomas harbor TERT promoter mutation. These findings further support the utility of TERT promoter mutation test as a differential diagnosis tool.

In summary, TERT promoter mutations were observed in 33% of primary adenocarcinomas of the urinary bladder and in 67% of urothelial carcinomas with glandular differentiation. Identical TERT promoter mutation status in both urothelial and glandular components from the same tumor indicate a common clonal origin. TERT promoter mutation analysis may be a useful ancillary study in the differential diagnosis.