Search tips
Search criteria 


Logo of jclinpathJournal of Clinical PathologyVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Clin Pathol. 2007 March; 60(3): 332–335.
Published online 2006 August 1. doi:  10.1136/jcp.2006.038802
PMCID: PMC1860555

Squamous differentiation in primary urothelial carcinoma of the urinary tract as seen by MAC387 immunohistochemistry


Squamous differentiation (SqD) is variably present in urinary tract tumours, but its significance remains unclear. In this study, SqD was assessed by immunohistochemistry using the monoclonal antibody Mac387 in 145 urothelial tumours (bladder, n = 115; renal pelvis, n = 30). Mac387 detects the myelomonocytic L1 antigen; a member of the calgranulin family shared by epithelial cells and keratinocytes. L1 antigen was shown in SqD in urothelial carcinomas of the bladder or the renal pelvis, including 11 cases with focal SqD unrecognised by conventional analysis. SqD is more frequent in renal pelvic tumours (p = 0.027) and increases with grade/stage mainly in bladder carcinoma (grade, p = 0.05; stage, p = 0.005). Stage Ta/T1 bladder carcinomas with SqD recurred more (p = 0.021). In conclusion, Mac387 efficiently shows SqD in urothelial tumours.

Urothelial tumours have a great capacity for divergent 1differentiation and, it is not uncommon to see areas of squamous differentiation (SqD).1,2,3,4,5 True incidence is unknown because no strict criteria were defined previously, but some reports suggest that SqD occurs in 16–25% of urothelial carcinomas of the bladder1,3,5 and varies according to grade and stage.2 The clinical significance of SqD remains uncertain, but seems to be an unfavourable prognostic feature predicting local recurrence after cystectomy.6 Also, it might be an indicator of a poor response to radiation therapy or systemic chemotherapy.1,7 Information concerning its effect on the recurrence capacity in low‐stage bladder urothelial carcinomas is lacking.

Immunohistochemistry might be helpful in the assessment of SqD, and some reports suggest that cytokeratin 14, cytokeratin 19, cytokeratin 10, caveolin 1 or the myelomonocytic L1 antigen could be immunohistochemical markers of SqD.8,9,10,11,12

Our purpose was to evaluate SqD in urothelial tumours as seen by the immunohistochemical detection of L1 antigen in a series of 145 patients with bladder (n = 115) or renal pelvis (n = 30) tumours.

Materials and methods

The retrospective study included specimens from 145 randomly selected patients having primary renal pelvis (n = 30) or bladder (n = 115) urothelial carcinoma retrieved from the files of the Reina Sofia University Hospital, Cordoba, Spain. Treatment included complete transurethral resection of the bladder or cystectomy. Clinical information was obtained from the patient's record. Patient follow‐up (mean (SD)), calculated as the number of months from the date of surgery to the date of the most recent cystoscopy (or the last visit or death) was 65 (28) (range 6–120) months. Tumour recurrence in stage Ta/T1 bladder tumours was defined as reappearance of tumour after treatment with at least one tumour‐free cystoscopy interval.

Specimens were fixed in formalin and embedded in paraffin. Available H&E stained slides of tumours were reassessed without knowledge of the clinical status. The resulting grade and stage were in accordance with the criteria of the 1973 World Health Organization and the 2002 Tumour, Node and Metastasis, respectively.2,13

The presence of intercellular bridges or keratinisation was indicative of SqD.1,2,3,4,5 For immunohistochemistry, 4‐μm‐thick sections were deparaffinised in xylene and rehydrated in graded alcohol. Immunohistochemical analysis was carried out using streptavidin–biotin peroxidase. Briefly, sections were incubated overnight with the primary mouse monoclonal antibody detecting myelomonocytic L1 antigen at 4°C (Mac387, Dakopatts, Glostrup, Denmark) diluted at 1:400. Biotinylated secondary antibody and the streptavidin–biotin peroxidase complex were sequentially applied (LSAB2 kit, Dako, Glostrup, Denmark). Diaminobenzidine was used as chromogen (Dakopatts). Endogenous peroxidase activity was blocked by incubating the slides in 1% hydrogen peroxide in methanol for 10 min. For epitope retieval, the slides were boiled in 10 mM citrate buffer (pH 6.0). Negative and positive controls were included in every procedure. A control group for the immunoreactivity included normal urothelium (n = 5), flat urothelial hyperplasia (n = 5) and normal trigone mucosa (n = 5).

Bivariate and multivariate statistical analyses included the χ2 test with the SPSS 8.0 for Windows Software. All were two sided tests. A p value [less-than-or-eq, slant]0.05 was considered significant.


The morphological patterns of keratinising and non‐keratinising SqD were: (i) wide nests and cords showing diskeratosis, intercellular bridges and corneous pearls; (ii) small foci often overcome under conventional evaluation showing intercellular bridges, but not diskeratosis or corneous pearls; (iii) additional features suggestive of koilocytosis; (iv) occasional but focally prominent intracytoplasmic lumina in nests of SqD and (v) rare cases had rather clear cells forming nests of variable size well demarcated from concomitant urothelial carcinoma (fig 11).

figure cp38802.f1
Figure 1 Trigone mucosa (A) with suprabasal immunostaining using the L1 antigen (B; A, H&E ×200; B, Mac387 monoclonal antibody ×200). Squamous differentiation (SqD) in urothelial carcinoma showing extensive corneous pearls ...

By immunohistochemistry, the suprabasal cells were labelled in normal trigone mucosa, whereas normal or hyperplastic urothelium and all urothelial carcinomas were negative. Keratinising SqD and the non‐keratinising type of SqD, both in tumours of the bladder and the renal pelvis, showed strong homogeneous labelling for the L1 antigen (fig 11)) and the intensity of Mac387 staining was dependent on the degree of SqD rather than pathological grade. Mac387 positively stained macrophages, with granulocytes acting as internal control. SqD was more frequent in renal pelvic than in bladder tumours (p = 0.027). Table 11 shows the demographic features of the series; SqD was seen only in grade 2 and 3 renal pelvic tumours (p = 0.439), but increased from grade 1 to 3 in bladder tumours (p = 0.05). SqD also increased according to stage in the renal pelvis (p = 0.106) and bladder tumours (p = 0.005). Survival status was not related to SqD in both renal pelvis and bladder tumours; patients having stage Ta/T1 bladder tumours with SqD had more chances of recurrence (p = 0.021; table 11).). Sensitivity and specificity for SqD using Mac387 reached 100%, while conventional evaluation reached a sensitivity of 54% and 100% specificity. Concerning the clinical significance of SqD in stage T1 bladder cancer, 10 cases with SqD recurred (55.5% of 18 stage T1 recurrent cases) and 2 additional cases with SqD had no recurrences (3.6% of 55 stage T1 non‐recurrent cases). In seven stage‐T1 cases Mac387 immunohistochemistry helped to delineate SqD that was not already evident with conventional evaluation.

Table thumbnail
Table 1 Clinical and pathological variables in 145 cases of urothelial carcinoma of the urinary tract according to squamous differentiation


Clinical significance of SqD in urothelial tumours remains unsettled with evidence suggesting that it might be an indicator of poor response after radical surgery, radiation or systemic chemotherapy.1,2,3,4,5,6,7 Most reports to date have concentrated on bladder urothelial tumours showing SqD, but evidence concerning renal pelvis tumours is lacking. Also, the significance of SqD in low‐stage bladder tumours in terms of tumour recurrence is unknown. This is in part owing to the fact that SqD in low‐stage tumours is often non‐keratinising and focal; these foci are often missed during conventional evaluation. The use of immunohistochemistry in assessing SqD is helpful in detecting foci of non‐keratinising SqD thus providing an objective identification of the lesion. Our results showed that SqD increased with grade and stage both in the bladder and renal pelvis specimens. Similar results were obtained in earlier studies by Lopez‐Beltran et al3 and more recently by Sanchez‐Carbayo et al12 who also found a significant association between SqD and tumour grade and stage. The absence of differences in renal pelvic tumours might be due to the limited number included since all cases that had SqD were grade 2 or 3 showing similar a incidence. There was no association between SqD and survival status in both renal pelvis and bladder cancer suggesting that its effect on survival is probably related to the association with higher grade and stage. An original finding in our study is the association with high recurrent tumours in patients with bladder Ta/T1 carcinomas having SqD, which favours SqD as a marker of aggressiveness. In this study, we chose myelomonocytic L1 antigen (a member of the cytoplasmic calcium binding proteins known as calgranulins) expression identified by the monoclonal antibody Mac387, which consistently labelled SqD of all types and allowed more objective identification of SqD than that obtained with routine investigations in some cases.8,9 Our results indicate that Mac387 is also seen in non‐keratinised squamous cells, a fact which is corroborated by staining patterns of the non‐keratinised mucosa of the bladder trigone.9 Our study suggests that Mac387 may be a reliable marker for studying the development of SqD in the urothelium and its tumours where many of the early or partial lesions can be difficult to detect morphologically. This may be due to some forms of SqD being focal or extensive and involving a subjective element in their identification, unlike the typical keratinising types which are easily recognised.

Take‐home messages

  • Squamous differentiation (SqD) is more frequent in renal pelvic urothelial tumours and increases with grade/stage.
  • The clinical significance of SqD remains uncertain, but seems to be an unfavourable prognostic feature.
  • Patients having stage Ta/T1 bladder tumours with SqD had more chance of recurrence.
  • Mac387 detects the myelomonocytic L1 antigen; a member of the calgranulin family shared by epithelial cells and keratinocytes.
  • Sensitivity and specificity for SqD using Mac387 immunohistochemistry is high.

Other valid alternative markers for SqD at the immunohistochemical level include the detection of cytokeratin 14 or cytokeratin 19.10,12 Recent molecular genetic analyses suggest a characteristic signature of urothelial tumours having SqD; caveolin 1 and keratin 10 have been found to be overexpressed in urothelial carcinomas and cell lines, but more frequently in invasive T2–T4 carcinomas; data on sensitivity and specificity of these markers are not readily available for comparing with Mac387, which seems to be superior to conventional evaluations in our series. In conclusion, our study emphasises the value of immunohistochemistry in assessing SqD in urothelial tumours; in particular, detecting the L1 antigen might result useful in assessing the clinical significance of SqD.


This work is supported by the Grant FIS 03/0952 (Madrid, Spain).


SqD - squamous differentiation


Competing interests: None.


1. Martin J E, Jenkins B J, Zuk R J. et al Clinical importance of squamous metaplasia in invasive transitional cell carcinoma of the bladder. J Clin Pathol 1989. 42250–253.253. [PMC free article] [PubMed]
2. Mostofi K F. ed. In: Histological typing of urinary bladder tumours. WHO international histological classification of tumours Geneva: WHO, 1973.
3. Lopez‐Beltran A, Martin J, Garcia J. et al Squamous and glandular differentiation in urothelial bladder carcinomas. Histopathology, histochemistry and immunohistochemical expression of carcionoembryonic antigen. Histol Histopathol 1988. 363–68.68. [PubMed]
4. Sakamoto N, Tsuneyoshi M, Enjoji M. Urinary bladder carcinoma with a neoplastic squamous component: a mapping study of 31 cases. Histopathology 1992. 21135–141.141. [PubMed]
5. Lopez‐Beltran A, Sauter G, Gasser T. et al Urothelial tumors: infiltrating urothelial carcinoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn I , eds. World Health Organization classification of tumors. Pathology and genetics of tumors of the urinary system and male genital organs Lyon: IARC press, 2004.
6. Honma I, Masumori N, Sato E. et al Local recurrence after radical cystectomy for invasive bladder cancer: an analysis of predictive factors. Urology 2004. 64744–748.748. [PubMed]
7. Logothetis C J, Dexeus F H, Chong C. et al Cisplatin, cyclophosphamide and doxorubicin chemotherapy for unresectable urothelial tumors: the MD Anderson experience. J Urol 1989. 14133–37.37. [PubMed]
8. Clark B R, Kelly S E, Fleming S. Calgranulin expression and association with the keratinocyte cytoskeleton. J Pathol 1990. 16025–30.30. [PubMed]
9. Tungekar M F, Heryet A, Gatter K C. The L1 antigen and squamous metaplasia in the bladder. Histopathology 1991. 19245–250.250. [PubMed]
10. Harnden P, Southgate J. Cytokeratin 14 as marker of squamous differentition in transitional cell carcinomas. J Clin Pathol 1997. 501032–1033.1033. [PMC free article] [PubMed]
11. Sanchez‐Carbayo M, Socci N D, Charytonowicz E. et al Molecular profiling of bladder cancer using cDNA microarrays: defining histogenesis and biological phenotypes. Cancer Res 2002. 626973–6980.6980. [PubMed]
12. Celis J E, Celis P, Osergaard M. et al Proteomics and immunohistochemistry define some of the steps involved in the squamous differentiation of the bladder transitional epithelium: a novel strategy for identifying metaplastic lesions. Cancer Res 1999. 593003–3009.3009. [PubMed]
13. Greene F L, Page D, Fleming I D. et alMorrow M: AJCC cancer staging manual. New York: Springer‐Verlag, 2002.

Articles from Journal of Clinical Pathology are provided here courtesy of BMJ Group