Search

In many organs the so-called blood group antigens A, B and O(H) are constituents of the surface of normal epithelial cells. Deletion of these antigens in cases of carcinoma may serve as a clinically useful tumor marker, especially in cases of carcinoma of the bladder. Further progress is hampered by uncertainty concerning the detectability of A, B and H tissue isoantigens on normal urothelium in tissue sections. We examined th expression of these tissue isoantigens on normal ureteral urothelium obtained at autopsy from 68 adults using the specific red cell adherence test and found that they were not detectable in normal urothelium from a significant fraction of individuals in blood group O and that results in blood groups A, B and AB also were somewhat variable. The variability did not appear related to the secretor status or postmortem interval. Our results suggest that the current specific red cell adherence test may not be suitable for detecting deletion of tissue isoantigens in bladder tumor patients in blood group O.

Abnormal cellular DNA content, a hallmark of malignancy, is known to occur in squamous cell carcinoma of the head and neck. The significance of this finding has yet to be elucidated. Paraffin-embedded blocks from 46 squamous cell carcinomas of the head and neck were sectioned, dewaxed, and made into a suspension of bare nuclei by mechanical mincing and enzymatic digestion. The nuclei were stained with propidium iodide, and quantitative DNA measurements were obtained with flow cytometry. The results were compared with existing prognostic factors: DNA aneuploidy was identified in 41% of the specimens and was significantly correlated with tumor size and surgical stage; the relationship between DNA aneuploidy and anatomic site approaches statistical significance. There was no correlation with histologic grade or presence or absence of cervical metastasis.

A Bladder Cancer Flow Cytometry Network study has been carried out aimed at identification of the sources of inter- and intralaboratory variability. Replicate "cocktail" samples containing a mixture of peripheral blood lymphocytes and an aneuploid cell line and samples of peripheral blood lymphocytes serving as a DNA reference standard were distributed to five network laboratories. The samples were stained for DNA using propidium iodide, with each laboratory using its own staining protocol. Sets of these samples were analyzed by flow cytometry to obtain cellular DNA distributions. DNA index and hyperdiploid fraction were calculated for each histogram using an automated technique. Results were evaluated by analysis of variance to identify sources of variability. Three important sources of variation were found that affect flow cytometry in general and- the transportability of flow cytometry results to routine clinical use in particular. The significant variation among laboratories that is constant across time most probably represents stable differences in instrumentation, instrument set-up, and laboratory techniques. This variation can be compensated for, if it is known and stable, to develop transportable classification criteria. The second type of variation, termed the interaction component, represents differences among laboratories that are not constant across time. Sources of this variation include inconsistency in sample preparation, staining, and analysis. The elimination of this type of variation is required for meaningful comparison of data within and among laboratories and the creation of interlaboratory data-bases. The third type of variation represents pure measurement variability and affects the sensitivity of the technique.

Sixty-eight colonic polyps of various histologic types were studied using retrospective flow cytometric (FCM) DNA analysis to determine the prevalence of aneuploid cell populations and whether they were associated with any particular histologic features. Overall, 13 of 68 polyps (19%) contained DNA-aneuploid cells, including 3 of 9 villous and 4 of 10 villoglandular polyps with histologic features of carcinoma in situ (CIS), 4 of 11 villous and 1 of 18 villoglandular polyps without CIS, and 1 of 12 adenomatous polyps. Eight hyperplastic polyps were diploid. These results show retrospective FCM analysis can detect DNA aneuploidy in polyps; DNA aneuploidy may occur before histologic evidence of invasive carcinoma; and this change is more frequent in types of polyps thought to have increased malignant potential (i.e., those with villous morphology and/or CIS).

Flow cytometric DNA analysis of human urinary bladder specimens may be clinically useful for prognosis in transitional cell (urothelial) carcinoma and for detecting recurrence after treatment. However, many important methodological differences exist among institutions which have described this technique, and it has not previously been shown that data from different institutions are comparable. The National Cancer Institute has created a Flow Cytometry Network to address the need for technology assessment of flow cytometry. This report describes the independent flow cytometric analysis and interpretation of "unknown" paraffin-embedded bladder tumor specimens by the five Network institutions. Although important differences in method existed among the institutions, substantial agreement was achieved in actual data generated and their interpretation. This suggests that a consensus regarding acceptable laboratory performance of this technique could be reached, which should facilitate its more widespread clinical implementation.

The prognostic significance of DNA aneuploidy was studied restrospectively in 177 Stage I cutaneous melanomas. DNA content was determined by flow cytometry of propidium iodide-stained nuclei recovered from formalin-fixed, paraffin-embedded material. Of 162 evaluable histograms, 124 were diploid, 35 aneuploid, and 3 tetraploid. Aneuploidy strongly correlated with established predictors of unfavorable prognosis, namely, thickness p less than .005, level p less than 0.005, ulceration p less than 0.005, and presence of vertical growth phase p less than 0.02. Overall, aneuploidy was strongly correlated with recurrence (p less than 0.005) and shorter disease-free survival (p less than 0.0001). Aneuploidy was an independent predictor of recurrence for tumors less than 1.5 mm thick (p less than 0.0001) and greater than or equal to 3 mm thick (p = 0.031). For melanomas 1.5-2.9 mm thick, aneuploid tumors had a 27% higher recurrence rate than diploid tumors (63% vs. 36%). This was not statistically significant (p = 0.247). In a multivariate analysis of common predictors stratified by thickness, DNA aneuploidy was the most significant independent parameter (p less than 0.002). DNA content appears to be an important stratification parameter for Stage I cutaneous melanoma.

Neoplastic and non-neoplastic tissue specimens from ten patients with primary adenocarcinoma of the urinary bladder were examined. Most of these tumors were associated with either foci of transitional cell carcinoma and/or with glandular metaplasia of the bladder epithelium. The mucin produced by the neoplastic cells was PAS, alcian blue, mucicarmine, PB/KOH/PAS, and RPB/KOH/PAS-positive. ABH isoantigens of these tumors were not always deleted. Ultrastructurally, the neoplastic cells resembled goblet cells. Their plasma membrane had numerous microvilli with prominent glycocalyx. Proliferation and attenuation of tight junctions were noted. The gap junctions were few and small. Two types of desmosomes were found. The ultrastructural features of the neoplastic cells were attributed in part to the malignant transformation and in part to the direction of their differentiation. We have not observed any distinctive morphologic, histochemical, immunologic or ultrastructural features that might be diagnostic for these adenocarcinomas.

Although current methods of histologic substaging for incidental prostatic carcinoma are useful, they offer only general indications of potential tumor behavior. To further define the biologic tendencies of stage A cancers, an examination was made of the role of DNA ploidy combined with histologic staging in archival material selected to achieve both a representative sample and long-term follow-up. With histology alone, 36% of stage A2 cancers and 9% of A1 neoplasms were progressive. Adding DNA flow cytometry to histology resulted in a significant improvement in the capacity of pathologic evaluation to predict outcome. Progression occurred in 67% of aneuploid stage A2 prostate cancers and in none of the nonaneuploid stage A1 tumors. Despite current limitations in the interpretation of DNA histograms from archival tissue, flow cytometry has significant potential in the pathologic evaluation of incidental prostatic carcinomas.

DNA ploidy of melanocytic skin tumors from 87 patients (53 primary melanomas, 34 nevi) was determined by flow cytometry from routinely prepared paraffin blocks. Ploidy data correlated strongly with conventional morphological parameters. Only 1 of 34 nevi, but 13 of 53 melanomas were aneuploid. Among the melanomas, none of 21 levels I-III melanomas was aneuploid, but 13 of 32 levels IV and V melanomas were aneuploid. There was also a significant correlation between increasing Breslow thickness and the presence of DNA aneuploidy. For 33 melanoma patients with over 2 yr of follow-up (average, 7.1 yr), only 4 of 23 diploid tumors have recurred, but 9 of 10 aneuploid tumors have recurred. We conclude that the biological potential of melanocytic skin tumors is strongly linked to DNA aneuploidy. Since this parameter can be conveniently determined from paraffin blocks, determination of ploidy abnormalities in these tumors may be clinically useful.

The expression of BGR-Ags is often aberrant in human carcinomas. The observation that BGR-Ag expression in human bladder carcinomas correlates with prognosis for patients with these tumors is especially interesting in light of the numerous reports of correlations between cell surface glycosylation and malignant phenotype in experimental animal tumors. Many observations suggest how this relation might be mediated. It seems reasonable to anticipate that the study of the BGR-Ags and their expression in carcinoma may emerge from its current predominantly descriptive phase and become an important part of the investigation of human tumor biology.

The expression of the ABH blood group antigens (BGAg) by carcinomas of the urinary bladder at the time of initial biopsy heralds a relatively benign clinical course, whereas antigen deletion in histologically similar tumors is often predictive of tumor recurrences and invasion. The use of BGAg as a tumor marker in the clinical setting has been limited by difficulties in demonstrating BGAg in tissue sections with the erythrocyte adherence test, especially for antigen H. In this study, immunoperoxidase methods were used to detect BGAg in deparaffinized tissue sections of normal and malignant human urothelium, and the results were compared with those of the erythrocyte adherence test performed on adjacent sections. Results with the two methods were similar for antigens A and B; however, a novel lectin-antilectin immunoperoxidase method was clearly superior for detecting antigen H. The immunoperoxidase method significantly reduced false-positive test results for antigen deletion in normal urothelium and more accurately predicted the subsequent clinical course in patients having urinary bladder cancer.

Ethanol preservation of voided and catheterized urine has long been the standard for urinary cytologic study. In this report ethanol preservation of bladder irrigation specimens was evaluated for flow cytometric analysis in a multiinstitutional study requiring specimen transport. Specimens from ten patients obtained at one center were preserved for varying periods of time in 50% ethanol, then distributed by express mail to four other participating centers, up to 3000 miles distant. On receipt, from 1 to 3 weeks after collection, the samples were processed and examined by flow cytometric study using propidium iodide as the fluorochrome. Forty-six of the 50 (92%) analyzed specimens gave satisfactory histograms. In 41 of the 46 adequate samples (89%), an aneuploid peak in the alcohol preserved (propidium iodide stained) specimen correlated well with the fresh (acridine orange stained) specimen. However, there was variable loss of DNA stainability with a broadening of the coefficient of variation in some alcohol-preserved specimens and an increase in cellular debris so that measurements of DNA index and percent hyperdiploid cells were considered unreliable. The authors conclude that ethanol preservation of bladder irrigation specimens for short periods of time may be a feasible alternative when flow cytometric analysis cannot be carried out on fresh specimens, but that this is not optimal fixation for specimens that must be transported and further studies of other fixatives are recommended.

Both cytogenetic analysis and flow cytometry (FCM) have prognostic efficacy in the analysis of transitional cell carcinoma (TCC) of urinary bladder. To correlate results of the two methods, we studied a unique group of patients whose tumors had undergone prospective conventional cytogenetic analysis. Paraffin blocks of these tumors were processed for FMC, then analyzed for nuclear DNA content. Of the 34 tumors processed, 30 (88%) yielded interpretable DNA histograms. In 26 (87%) of these, there was good correlation between the two methods with respect to the presence or absence of a hyperdiploid cell line. Discrepancies may have resulted from sampling error or from interpretation of a tetraploid peak as a prominent G2M region. Retrospective FCM analysis of paraffinized TCC tissue correlates well with conventional, prospective cytogenetic analysis and is applicable to the majority of urinary bladder transitional cell carcinomas.

Cationized ferritin (CF) was used as an ultrastructural marker to study differences in the distribution and density of surface anionic charges between normal and neoplastic cells. In anchorage-independent cell systems, CF induces a redistribution of cell surface receptor-ligand complexes into clusters, patches, and caps on the surfaces of transformed cells, but not on the surfaces of normal cells. In the present report, the authors have, for the first time, extended the CF labeling studies to a system of anchorage-dependent, rat bladder epithelial cell lines: normal RBTC cells and RBTCC-8 carcinoma cells. Cells were grown to confluency and labeled with 500 micrograms/ml CF for 3 minutes. After completion of CF labeling, cells were incubated for 0, 15, 60, 120 minutes, or overnight and were fixed then with buffered glutaraldehyde for electron microscopy. The results show that (a) CF induces grouping of surface anionic charge sites into patches and clusters on RBTCC-8 carcinoma cells (CF covers 53.86 +/- 2.15% of the plasma membrane), but not on normal RBTC cells (CF covers the entire plasma membrane); (b) CF densities are well correlated to the cell surface sialic acid contents of RBTC (2.42 +/- 0.30 microM/10(9) cells) and RBTCC-8 cells (1.26 +/- 0.25 microM/10(9) cells); and (c) CF-labeled membrane is internalized at equal rates by RBTC and RBTCC-8 cells (30% in 15 minutes; 60% in 60 minutes; 70% in 120 minutes; 100% overnight). These data indicate that CF-induced patching of anionic sites is a surface characteristic that is common to anchorage-dependent and independent tumor cells. Mechanisms of CF patching in neoplastic cells are discussed.

Conventional prognostic parameters for colon carcinoma are predominantly determined after resection and have limited predictive value. For an evaluation of the significance of flow cytometric (FCM) DNA analysis in colon carcinoma, 56 invasive colon carcinomas were prospectively studied to compare DNA ploidy with established prognostic parameters obtained from pathologic examination of resected specimens. Objective parameters, i.e., depth of invasion and node status, were strongly linked to DNA ploidy; diploid tumors tended to be Astler-Coller stage A or B and nondiploid tumors stage C or D. Diploid and nondiploid tumors did not differ according to subjective criteria such as histologic grade and microscopic invasion of vessels and nerves. These results suggest that FCM DNA analysis may be a valuable tool in managing patients with colon carcinoma, since analysis of biopsies could indicate the likelihood of tumor spread before surgery.

Murine squamous carcinoma cells (KLN205) grown in a medium supplemented with the retinoid, 13-cis retinoic acid (RA), had dose-dependent, selective increases in the expression of certain lectin receptors, which correlated with a dramatic decrease in the ability to form pulmonary colonies (P = .0003) (Couch MJ, Pauli BU, Weinstein RS, Coon JS: JNCI, 78:971-977, 1987). These findings suggest a possible relationship between the RA-induced glycoconjugate alterations and the decreased experimental metastatic behavior. We further define the mechanism of RA's action. The finding that RA treatment (5 X 10(-6) M, 5 X 10(-7) M) did not perturb the cell cycle of KLN205 cells provides further proof that the decreased metastatic behavior is not attributable to any inhibition in the rate of growth or to alterations in the cell cycle. Furthermore, since stable subpopulations with variable lectin binding could not be detected, the mechanism of RA's action does not appear to be due to selection of variant tumor-cell subpopulations. Finally, in a series of experiments designed to determine the reversibility of the RA treatment, the RA-induced decrease in metastatic behavior reverted back to a more metastatic state in the same time frame (3 days) as the reversion of the RA-induced changes in cell-surface glycoconjugate expression. This reversion provides further evidence for a close relationship between the RA-induced modulation of tumor cell-surface glycoconjugate expression and the decreased metastatic behavior; it suggests that transient, reversible modulation of the tumor cell surface may play a role in determining metastatic behavior.

Several markers of initial bladder carcinomas described recently may be clinically significant predictors of biological behavior of future recurrences. Comparison of the marker systems and assessment of the value of using multiple markers have been difficult, because the various markers have been studied in different patients. In this study, we compared four markers [chromosome mode, marker chromosomes, and expression of the ABH "blood group" antigens and the Thomsen-Friedenreich antigen (using immunoperoxidase or lectin immunoperoxidase methods)] in 39 patients presenting initially with low-stage bladder carcinoma and followed 3 to 11 years or until deep muscle invasion occurred. Each of the markers was significantly related to subsequent recurrences with deep muscle invasion, and each marker system was able to identify those patients with a very low risk of subsequent invasion. For detection of a subpopulation of patients with Grade II carcinomas who were at high risk for development of subsequent invasion, combinations of markers, especially hyperdiploidy and abnormal expression of the Thomsen-Friedenreich antigen, were significantly more effective than any single marker system.