Main

Primary squamous cell carcinoma of the urinary bladder is an uncommon malignant neoplasm, the etiology of which is not well understood. It accounts for 3–5% of bladder cancer in Western populations.1, 2, 3, 4 In areas of endemic Schistosoma haematobium (bilharziasis) infections, primarily in Middle Eastern countries, the contribution of squamous cell carcinoma to overall bladder cancer incidence is reportedly as high as 75%.5, 6 In Western populations, the development of urinary bladder squamous cell carcinoma has been attributed to chronic irritation from such things as urinary calculi, catheters, diverticula, or chronic nonbilharzial infections. In comparing reported cases of muscle-invasive bladder cancer, urinary bladder squamous cell carcinoma consistently presents at a higher stage and carries a worse-stage matched survival than urothelial carcinoma of the bladder.

While the association between human papillomavirus (HPV) and uterine cervical and head/neck squamous cell carcinoma is well established, the role of HPV in the pathogenesis of urinary bladder squamous cell carcinoma is not defined clearly. In urothelial carcinoma of the bladder, HPV has been suggested as a causative agent by some,7, 8, 9 while others disagree with this concept.10, 11, 12 p16 expression is often used as a surrogate marker for HPV infection in cervical neoplasia and squamous cell carcinoma at other organ sites.13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33 However, whether p16 expression can be used as a surrogate marker for HPV infection in urinary bladder squamous cell carcinoma is also poorly understood. The purposes of this study were to investigate the possible role of HPV in the development of squamous cell carcinoma of the urinary bladder and to determine if p16 expression could serve as a surrogate marker for HPV in this malignancy.

Materials and methods

A total of 69 cases of interest were retrieved from the surgical pathology files of participating institutions, between the years of 1992 and 2011. Of these cases, 52 were from cystectomy specimens, 14 from transurethral resections, and the remaining 3 cases were from biopsy material. The study included 42 cases of pure squamous cell carcinoma urinary bladder and 27 cases of urinary bladder urothelial carcinoma with squamous differentiation. The latter subset cases were defined by tumors that exhibited keratin pearls, intercellular bridging, or intercellular keratin in the areas of squamous differentiation with at least some areas of classic urothelial carcinoma present. All the cases were reviewed and the diagnoses confirmed. None of our cases in this study were known or suspected of having a history of infection with S. haematobium. One case of ‘urinary bladder squamous cell carcinoma’ from a 57-year-old woman was positive for HPV. However, after reviewing the clinical chart, the patient had a history of uterine cervical squamous cell carcinoma. Therefore, the correct diagnosis for this case was uterine cervical squamous cell carcinoma involving the urinary bladder, and the case was excluded from the study.

Immunohistochemical studies were conducted on 5-μm formalin-fixed, paraffin-embedded tissue sections using an anti-HPV mouse monoclonal antibody (Clone K1H8, 1:50; Dako, Carpinteria, CA, USA) and an anti-p16 mouse monoclonal antibody using the CINtec Histology kit (CINtec Histology, Westborough, MA, USA). The HPV antibody can detect HPV subtypes 6, 11, 16, 18, 31, 33, 42, 51, 52, 56 and 58. Nuclear staining was considered positive for HPV. HPV in situ hybridization was performed using Inform HPV II Family 6 Probe (detecting HPV genotypes 6 and 11) and HPV III Family 16 Probe (detecting HPV genotypes 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 66) (Ventana Medical Systems, Tucson, AZ, USA).

The CINtec Histology kit involves a two-step antibody method for the detection of the human p16 antibody. The first antibody is a monoclonal mouse antibody (E6H4) directed towards the human p16 protein. A secondary antibody, for visualization purposes, involves a polyclonal goat anti-mouse antibody conjugated with horseradish peroxidase. These steps were performed and have only been validated on formalin-fixed, paraffin-embedded tissue after heat-induced epitope retrieval at 95–99°C in the included epitope retrieval solution. For p16 immunohistochemistry, any nuclear immunoreactivity of p16 was deemed positive. Only nuclear staining was considered as true immunoreactivity, but the presence of cytoplasmic staining was recorded. Appropriate negative and positive controls were used.

Results

The clinicopathological information is summarized in Table 1. Most cystectomy cases were found to be of high stage (pT3–pT4) for both squamous cell carcinoma (23 of 31, 74%) and urothelial carcinoma with squamous differentiation (15 of 21, 71%). Of the 17 noncystectomy specimens, 14 were pT2. There was a 1:1 male-to-female ratio in these patients and they had a mean age of 63 years (range of 37–84 years). Staging characteristics were similar to that seen in the overall study population.

Table 1 Clinicopathological information for urinary bladder squamous cell carcinoma and urothelial carcinoma with squamous cell differentiation
Full size table

The results of immunohistochemical staining for HPV and p16 and in situ hybridization for HPV are summarized in Table 2. HPV DNA and protein were not detected in any of the 42 cases of squamous cell carcinoma of the urinary bladder (0%, 0/42) or in any of the 27 cases of urothelial carcinoma with squamous differentiation of the urinary bladder (0%, 0/27). p16 expression was detected in 13 cases (31%, 13/42) of squamous cell carcinoma and 9 cases (33%, 9/27) of urothelial carcinoma with squamous differentiation (Figures 1,2,3,4,5,6 and Table 2). The staining extent is summarized in Table 3. While nuclear staining was the only parameter used in scoring, only two cases demonstrated pure nuclear staining. The majority showed focal to diffuse cytoplasmic staining coupled with nuclear staining. Pure cytoplasmic staining was not identified in any examined case. Concordant staining between the urothelial and squamous components of the urothelial carcinoma with squamous differentiation was seen. The p16 expression in all cases occurred in the absence of demonstrable HPV DNA or HPV protein expression. p16 immunoreactivity did not correlate with any clinicopathological characteristics.

Table 2 Results of HPV in situ hybridization (ISH) and immunohistochemical (IHC) staining and p16 IHC staining in urinary bladder squamous cell carcinoma and urothelial carcinoma with squamous differentiation
Full size table
Figure 1
figure 1

Squamous cell carcinoma of the urinary bladder (a–c). p16 immunostaining of the same case demonstrating strong nuclear staining and less intense cytoplasmic staining (d).

Full size image
Figure 2
figure 2

Squamous cell carcinoma of the urinary bladder showing well-formed keratin pearls (a) and corresponding immunoreactivity with p16 (b).

Full size image
Figure 3
figure 3

Squamous cell carcinoma of the urinary bladder (a) with negative human papillomavirus (HPV) in situ hybridization (ISH) (b) and positive p16 staining (c). A case of confirmed HPV-infected uterine cervical squamous cell dysplasia was used as positive control (d) with HPV-ISH (e) and HPV immunostaining (f), both displaying strong nuclear positivity.

Full size image
Figure 4
figure 4

Urothelial carcinoma with squamous differentiation. The classic urothelial component is demonstrated in part (a) with its corresponding p16 immunostaining in (b). Well-defined squamous differentiation from the same case is seen in part (c) with its corresponding p16 immunostaining in part (d).

Full size image
Figure 5
figure 5

The mixed morphology comprising the classic urothelial and squamous differentiation components of urothelial carcinoma with squamous differentiation are demonstrated clearly in this case (a and b). p16 immunostaining is shown in both the squamous (c) and the urothelial (d) components from this case.

Full size image
Figure 6
figure 6

High-grade invasive urothelial carcinoma with strong p16 immunostaining (a and b).

Full size image
Table 3 Extent of p16 immunostaining in urinary bladder squamous cell carcinoma and urothelial carcinoma with squamous differentiation
Full size table

Discussion

Information regarding an association between HPV infection and urinary bladder squamous cell carcinoma is limited. In this study, we were unable to detect either HPV DNA by in situ hybridization or HPV protein expression by immunohistochemical staining in 42 cases of primary urinary bladder squamous cell carcinoma or in 27 cases of urothelial carcinoma with squamous differentiation. These findings indicate that HPV does not play a demonstrable role in the development of urinary bladder squamous cell carcinoma or urothelial carcinoma with squamous differentiation.

The role of HPV infection in the genesis of urothelial neoplasms has been a subject of controversy. Many investigators have found little or no correlation between HPV infection and urothelial carcinoma and have concluded that the virus does not play an etiological role.10, 11, 12 However, in a study involving Iranian patients, HPV DNA was detected in slightly over 35% of urothelial neoplasia cases, and 81% of the viral strains identified were HPV type 18, which is considered a high-risk strain for the development of uterine cervical neoplasia.34 More recently, Li et al7 have also shown the presence of HPV infection in bladder cancer. They implicated HPV type 16, which is also considered a high-risk strain for the development of uterine cervical neoplasia. Shigehara et al9 have also shown HPV infection in bladder cancer. Their study showed a predilection for the virus in low-grade tumors in young patients (<60 years of age). Interestingly, they also reported p16 expression in 94% of cases with HPV infection.9

The signal-transduction pathways in HPV-related squamous cell carcinoma have been well studied in the uterine cervix.28 The HPV E7 protein inactivates the retinoblastoma tumor suppressor gene, which normally inhibits p16. Cervical cancer associated with HPV, therefore, has increased p16 expression29, 30 and its expression is used by some as a surrogate marker for HPV infection in cervical lesions.13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 It has been hypothesized that HPV infection might similarly play a role in the development of urinary bladder squamous cell carcinoma. In contrast to our findings and the previously mentioned reports, others have shown a positive correlation between high-risk HPV infection and high-grade urothelial cancers of the bladder.7, 8, 9, 35, 36 Cai et al37 have proposed that HPV may have a causative role in the genesis of urothelial neoplasms, particularly in nonmuscle-invasive bladder cancer. In our study, we detected p16 expression in 31% (13/42) of cases of urinary bladder squamous cell carcinoma and in 33% (9/27) of cases of urinary bladder urothelial carcinoma with squamous differentiation. However, despite the presence of p16 expression, none of these tumors had evidence of HPV infection at either the DNA level or the protein level, as measured by in situ hybridization or immunohistochemical staining, respectively. In separate studies using in situ hybridization, Guo et al38 and Westenend et al39 found no HPV DNA in any of the 32 cases of urinary bladder squamous cell carcinoma that they evaluated. More recently, using three different polymerase chain reaction techniques, Ben Selma et al40 did not find HPV DNA in 125 cases of urinary bladder carcinoma, including five cases of squamous cell carcinoma. These findings, in concordance with our findings, further emphasize the notion that HPV infection does not play a role in the pathogenesis of urinary bladder squamous cell carcinoma.

In summary, our findings indicate that HPV infection does not play a role in the development of squamous cell carcinoma of the urinary bladder or urothelial carcinoma with squamous differentiation. These findings also emphasize that p16 expression, although demonstrable in about one-third of our cases, should not be used as a surrogate marker for evidence of HPV infection in either squamous cell carcinoma of the urinary bladder or urothelial carcinoma with squamous differentiation, as neither HPV DNA nor protein was detected in these neoplasms.