The Relationship of ABH(0) Blood Group Antigen Expression in Intraepithehl Dysplastic Lesions to Clinicopathologic Properties of Associated Transitional Cell Carcinoma of the Bladder

7

Takumi Yamada, MD,* Iwao Fukui, MD,* Tsuyoshi Kobayashi, MD,* Hideaki Sekine, MD,* Masayuki Yokogawa, MD,* Takashi Yamada, MD,t and Hiroyuki Oshima, MD*

Paraffin-embedded, giant-step sections of 13 bladders with transitional cell carcinomas were stained with monoclonal anti-A or anti-B antibodies to investigate whether intraepithelial dysplastic lesions are related to obvious tumors. Normal and/or hyperplastic lesions were retained in only eight bladders; severe dysplasia and/or carcinoma in situ were found in all bladders except two. AB-antigen expression was retained in intraepithelial lesions of bladders with invasive carcinoma. Most intraepithelial lesions were AB-antigen negative in bladders with frequently recurrent tumors. In bladders with initially multiple tumors, AB-antigen expression was negative in almost one half of the intraepithelial lesions. Therefore, it appears likely that most multiple or recurrent bladder carcinomas arise from dysplastic cells in intraepithelial lesions which have acquired malignant potential; initially invasive tumors quickly develop from a limited lesion acquiring a high malignant potential without changes of cell phenotype in most intraepithelial lesions. Cancer 67:1661-1666,1991.

T IS NOT UNCOMMON that various histocytologic I changes such as hyperplasia, dysplasia, and carcinoma in situ associate with papillary and/or sessile tumors in apparently normal transitional epithelium of the bladder. Among these intraepithelial lesions, carcinoma in situ may develop into invasive bladder carcinoma. ' - I Expression of ABH antigens (ABH-Ag) in apparently normal epithe- lium in the bladder with transitional cell carcinomas has been investigated by immunohistological methods. Dele- tion of ABH-Ag occurs in associated dysplastic lesions and obvious turn or^.'^-'^ It is, however, unknown whether such intraepithelial dysplastic changes cause obvious transitional cell carcinoma or not. The current report in- vestigates the distribution of ABH-Ag in whole epithelium of the bladder using giant-step sections of the bladder to

From the *Department of Urology, Tokyo Medical and Dental Uni- versity School of Medicine, Tokyo, and the ?Department of Pathology, Dokkyo Medical College, Tochigi, Japan.

Address for reprints: Takumi Yamada, MD, Department of Urology, Kasukabe Municipal Hospital, 7-2- I , Chyuo, Kasukabe City, Saitama, Japan.

Accepted for publication September 1, 1990.

explore the relationship of phenotypic expression of cells composing intraepithelial lesions to clinicopathologic features of overt tumors.

Materials and Methods

Materials

Thirteen bladders removed for transitional cell carci- noma of the bladder from 1982 to 1985 were divided into three groups on the basis of the clinicopathologic prop- erties of the tumors: Group 1, cystectomy done as the initial therapeutic procedure for invasive cancer; Group 2, cystectomy done as the initial therapeutic procedure for multiple superficial bladder tumors; and Group 3, cystectomy done because of multiple and frequent recur- rences after prior treatment of papillary tumors. The clin- icopathologic profiles of the tumors and patient data are summarized in Table 1. Tumors in Group 1 were deeply invasive (pT2 and pT,) and had grade 3 anaplasia in con- trast to those in the others which were superficially in- vasive (pTI) and had grade 2 anaplasia. Group 1 included eight patients with invasive solitary nonpapillary carci-

1661

1662 CANCER March 15 199 1 Vol. 67

TABLE 1. Patient Profiles

Group 2 Group 3 Group 1 initially frequently

invasive f91* multiole (2) recurrent (2)

Median age (yr) (Range)

Sex Male/female

Histologic grade 2 3

Stages TI T2 T3

62 (32-69)

81 1

0 9

0 3 6

61 (55-67)

210

2 0

2 0 0

52 (40-64)

210

2 0

2 0 0

* Number of cases.

nomas and one with invasive multiple papillary carci- nomas.

The patients had blood types of either A (seven pa- tients), B (three), or AB (three). Patients with blood group 0 were excluded from the current study because of the weak staining of the H-antigen, which may lead to false- negative staining.

Removed bladders were processed according to the method of Austen and Friedell.'53'6 Briefly, the bladder was inflated like a balloon immediately after removal with buffered 20% formalin solution and then immersed in the formalin solution for 1 week. The fixed bladder was cut into 1-cm transverse ring pieces, and a 6-pm giant section was prepared from each paraffin-embedded ring piece.

Staining Technique

Staining of ABH blood group isoantigen (ABH-Ag) fol- lowed the avidin-biotin-peroxidase-complex method,"

using the Vectastain ABH kit (Vector Laboratories, Bur- lingame, CA). All procedures were done at the room tem- perature. For staining of the AB-Ag, anti-A or anti-B monoclonal antibody (Dako, Carpinteria, CA) was used as the primary antibody according to the blood type of the patient, and biotinylated anti-mouse immunoglobulin G was used as the second antibody. Erythrocytes, vessel endothelium, and normal urothelium served as positive controls for AB-Ag, and muscle and connective tissue were used as negative controls. Negative-control staining was done by either using different anti-blood type antibody or omission of the primary or secondary antibody, which eliminated staining of normal epithelium, vessel endo- thelium, and erythrocytes. AB-Ag expression in a lesion was defined as follows; negative, a lesion where positively stained epithelial cells occupied less than 25% and positive, a lesion where more than 25% of epithelial cells were pos- itively stained.

Results

Distribution of Intraepithelial Histocytologic Lesions in Apparently Normal Epithelium

Histocytologic findings of epithelial cells in apparently normal epithelium were divided into three categories: normal and hyperplasia, mild and moderate dysplasia, and severe dysplasia and carcinoma in situ. The classifi- cation of dysplasia followed the criteria previously re- ported?-I2 Representative features of each type are shown in Figure 1. The relative distribution of each category in apparently normal epithelium was measured using at least three giant sections of each bladder specimen, averaged and summarized in Figure 2. Normal and/or hyperplastic epithelium was present in only eight of 13 bladders; severe

FIGS. IA-1C. Representative intraepi- thelial changes (original magnification, X 100). (A) Hyperplasia, (B) moderate dys- plasia, and (C) carcinoma in situ.

No. 6 AB-AG EXPRESSION IN BLADDER ABNORMALITIES - Yamada et al. 1663

Papillary der tumors have been noted in apparently normal epi- thelium, but the distribution of these changes has not been investigated.

Deletion of AB-Ag expression occurs more frequently in lesions of severe dysplasia and carcinoma in situ than in those of mild and moderate dysplasia. Since AB-Ag expression is retained in low-grade bladder tumors and deletion occurs in many high-grade turn or^,'^-*^ deletion of AB-Ag expression could be used as a sign of cellular phenotypic changes expressing malignant potential.

Our results provide evidence that AB-Ag expression in severe dysplasia and carcinoma in situ is retained in Group 1 with invasive tumors, but AB-Ag expression is lost in the invasive tumors themselves. The finding contradicts

lnvasive group l n i W Frequently the widely accepted hypothesis'-' that high-grade and high-stage carcinomas of the bladder originate from car- cinoma in situ. However. it has been suggested that most

recurrent group group -

FIG. 2. Relations between intraepithelial histocytologic changes and type of tumors. Open bar: normal epithelium and hyperplasia; dotted bar: mild and moderate dysdasia: shaded bar: severe dvsulasia and car-

invasive carcinomas primarily occur in the bladdeI-25 and are detected Soon after their 0ccurrence-25-27 _ _ _ _

cinoma in situ. Therefore, it appears likely that intraepithelial lesions with a high malignant potential occur in a limited area and quickly develop into an invasive carcinoma and that AB- Ag-positive intraepithelial lesions occupying most of the area of the epithelium are not the source of invasive car- cinomas because of their distinct phenotype which retains AB-Ag expression.

In contrast to Group 1, AB-Ag expression is lost in most intraepithelial lesions in Groups 2 and 3 . There are

ficial papillary carcinomas and associated intraepithelial lesions at least from the viewpoint of AB-Ag expression. Furthermore, deletion of AB-Ag expression has been re-

dysplasia or carcinoma in situ were found in all bladders except two.

AB-Antigen Expression in Tumors and Intraepithelial Lesions by Immunostaining

Figure 3 illustrates typical positive and negative staining

of each group were found to be AB-Ag negative. As shown in Figure 4, however, AB-Ag expression was retained in most intraepithelial lesions in Group 1 with invasive tu- mors. Conversely, AB-Ag expression was lost in most ap- parently normal epithelium including normal, hyperpla- sia, and other dysplastic lesions in Group 3 with frequent recurrence (Fig. 5). In Group 2 undergoing cystectomy as initial therapy for multiple tumors, AB-Ag positively stained in approximately one half of intraepithelial changes (Fig. 6). These findings are summarized in Figure 7. The incidence of deleted AB-Ag expression in nontu- mor epithelium increased from Groups 1 to 3. Deletion of AB-Ag expression also increased with the progression of histocytologic anaplasia in intraepithelial lesions.

OfAB-Ag in each type ofintraepithelial lesion. ,411 tumors similar cellular properties of recurrent or multiple super-

Discussion

The current study demonstrates that areas of normal and hyperplastic epithelial cells without atypia in appar- ently normal epithelium are very limited in most of re- moved bladders with invasive, multiple, or frequently re- current tumors and that large portions of nontumor ep- ithelium are occupied by dysplastic epithelial cells in most cases. Various intraepithelial lesions associated with blad-

FIGS: 3A AND 3B. Representative staining expression of AB-antigens in intraepithelial changes (original magnification, X25). (A) positive staining and (B) negative staining.

1664 CANCER March I5 199 1 Vol. 67

FIGS. 4A AND 4B. Expression of AB-Ag in intraepithelial changes of a bladder with an invasive tumor. (A) H & E staining of a giant-step section. (B) Schematic presentation of AB-Ag expression in the section of panel A.

ported to occur before malignant progression after re- peated recurrence of bladder tumors.24 Therefore, intra- epithelial dysplastic lesions of the bladder could obtain

malignant potential with loss of AB-Ag expression, and thus they may be the origin of recurrent or multiple su- perficial tumors.

FIGS. 5A AND 5B. Expression of AB-Ag in intraepithelial changes of a bladder with frequently recurrent tumors. (A) H & E staining of a giant-step section. (B) Schematic presentation of AB-Ag expression in the section of panel A.

No. 6 AB-AG EXPRESSION IN BLADDER ABNORMALITIES - Yamada et al. 1665

FIGS. 6A AND 6B. Expression of AB-Ag in intraepithelial changes of a bladder with initially multiple tu- mors. (A) A & E staining of a giant- step section. (B) Schematic presen- tation of AB-Ag expression in the section of panel A.

REFERENCES

I . Melicow MM, Hollowel JW. Intra-urothelial cancer: Carcinoma in sitzr, Bowen’s disease of the urinary system: Discussion of thirty cases. J Urol 1952; 68:763-772.

2. Schade ROK. Some observation on the pathology and natural his- tory of urothelial neoplasms. Beitr Path Bd 1972: 145:325-335.

3. Koss LG, Tiamson EM, Robbins MA. Mappings cancerous and precancerous bladder changes: A study of the urothelium in ten surgically removed bladders. JAMA 1974; 227:281-286.

4. Koss LG, Nakanishi I, Freed S Z . Nonpapillary carcinoma in .&i and atypical hyperplasia in cancerous bladders: Further studies of sur- gically removed bladders by mapping. Urolo%qy 1977; 9:442-455.

5. Farrow GM, Utz DC, Rife CC. Morphological and clinical obser- vations of patients with early bladder cancer treated with total cystectomy. Cancer Res 1976; 36:2495-2501.

6. Wallace DMA, Hindmarsh JR, Webb JN ct al. The role of multiple biopsies in the management of patients with bladder cancer. Br J Urol 1979; 51:535-540.

7. Eisenberg RB, Roth RB, Schweinsberg MH. Bladder tumor and associated proliferative mucosal lesions. J Urol 1960; 84:544-550.

8. Melamed MR, Voutsa NG, Grabstald H. Natural history and clin- ical behavior of in situ carcinoma of the human bladder. Cancer 1964: 17:1533-1545.

9. Cooper PH, Waisman J, Johnston WH, Skinner DG. Severe atypia of transitional epithelium and carcinoma of the urinary bladder. Cancer

10. Heney NM, Daly J, Prout GR, Nieh PT, Heaney JA, Trebeck NE. Biopsy of apparentry normal urothelium in patients with bladder carcinoma. J Urol 1978: 120:559-560.

1 1. Wolf H, Hojgaard K. Prognostic factors in local surgical treatment of invasive bladder cancer, with special reference to the presence of urothelial dysplasia. Cancer 1983; 51:1710-1715.

12. Stein BS, Reyes JM, Pethersen RO. Mcnenllis D, Kendall AR.

1973: 3 1:1055-1060.

FIG. 7. Expression of AB-Ag by patient groups and progression of intraepithelial changes.

1666 CANCER March 15 199 1 Vol. 61

Specific red cell adherence: Immunologic evaluation of random mucosal biopsies in carcinoma of the bladder. J Urol 198 I ; 126:37-40.

13. Huland H, Kloppel G, Otto FU, Droese M. The value of histologic and staging random biopsies, tumor and bladder mucosa blood group antigens, in predicting progression of superficial bladder cancer. Eur Urol

14. Coon JS, Mccaal A, Miller AW 111, Farrow GM, Weinstein RS. Expression of blood-group-related antigens in carcinoma in situ of the urinary bladder. Cancer 198.5; 56:197-804.

15. Austen G Jr, Friedell GH. Observations on local growth patterns of bladder cancer. J CJrol 1964; 56:38-41.

16. Sat0 EA, Friedell GH, Tiltman AJ. Bladder cancer as seen in giant histologic sections. Cancer 1977; 39:447-45.5.

17. Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase technique: A comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem

18. Bergman S, Javadpour N. The cell surface antigen A, B or O(H) as an indicator of malignant potential in stage A bladder carcinoma: Preliminary report. J Urol1978; I 19:49-5 I .

19. Emmott RC, Javadpour N, Bergman SM, Soares T. Correlation of the cell surface antigens with stage and grade in cancer of the bladder. J Urol 1919; 121:31-39.

20. Richie JP, Blute RD Jr, Waisman J. Immunologic indicators of

1984; 10~28-31.

1981; 29577-580.

prognosis in bladder cancer: The importance of cell surface antigens. J Urol 1980; 123:22-24.

21. Decenzo JM, Howard P, Irish CE. Antigenic deletion and prog- nosis of patients with stage A transitional cell bladder carcinoma. J Urol

22. Lange PH, Limas C, Fraley EE. Tissue blood-group antigens and prognosis in low stage transitional cell carcinoma of the bladder. J Urol

23. Newman AJ Jr, Carlton CE Jr. Johnson S . Cell surface A, B or O(H) blood group antigens as an indicator of malignant potential in stage A bladder carcinoma. J Urol 1980; 124:27-29.

24. Yamada T, Fukui I, Yokogawa M, Oshima H. Changing expres- sion of ABH blood group and cryptic T-antigens of noninvasive and superficially invasive papillary transitional cell carcinoma of the bladder from initial occurrence to malignant progression. Cancer 1988; 61:721- 726.

25. Kaye KW, Lange PH. Mode of presentation of invasive bladder cancer: Reassessment of the problem. J Urol 1982; 128:31-33.

26. Prout GR Jr, Griffin PP, Daly JJ, Heney NM. Carcinoma in situ of the urinary bladder with and without associated vesical neoplasms. Cancer 1983; .52:.524-532.

27. Newman LH, Tannenbaum M, Droller MJ. Muscle-invasive bladder cancer: Dose it arise de novo or from pre-existing superficial disease? Urology 1988; 3258-62.

197.5; 1141874-878.

1978; 119:.52-5.5.