Antigenic heterogeneity and individuality in adenocarcinomas of the rectum and their secondaries

Histopathological and Prognostic Evaluation of Immunohistochemical Findings in Colorectal Cancer

Many immunohistochemical studies have investigated the relationship between immunohistochemical characteristics and histopathological findings in colorectal tumors. One of the most extensively studied markers has been tissue CEA, although the prognostic significance of this and other antigens is still uncertain. The authors report results relative to three tumoral antigens (carcinoembryonic antigen, CEA; tissue polypeptide antigen, TPA, and carbohydrate antigen 19–9, CA 19–9) determined by immunohistochemical methods in tissue samples of 52 colorectal carcinomas. The relationship between the immunohistochemical characteristics of the neoplasms and the clinicopathologic parameters, as well as their influence on the prognosis of the patients, were examined. Positive CEA reaction has a significant relationship with grade of differentiation of the tumor while diffuse cellular expression of this antigen often indicates neoplasms extending beyond the intestinal wall and invading the lymph vessels. The number of tissue antigens expressed is significantly related to the extent of tumor spread through the intestinal wall. A greater incidence of recurrence and shorter disease-free interval and survival were observed in neoplasms that expressed tissue TPA antigen or more than one tissue antigens. In the present study the latter parameter has demonstrated to have independent prognostic significance for the disease-free interval. Immunohistochemical evaluation of antigens in colorectal carcinoma tissue shows a possible independent prognostic value of the antigenic heterogeneity of tumors, which could be related to their different biological behavior.

Molecular testing in oncology: problems, pitfalls and progress

Recent advances in the understanding of the molecular basis of cancer and the development of molecular diagnostics based on this knowledge have done much to progress the fields of oncology and pathology. Technological developments such as Next Generation Sequencing (NGS) and multiplex assays have made feasible the widespread adoption of molecular diagnostics for clinical use. While these developments and advances carry much promise, there are pitfalls to implementing this testing. Choosing appropriate biomarkers is a vital first step for clinical use and being able to understand the complex relationship between predictive and prognostic biomarkers is a crucial component of this. Testing for standard of care biomarkers is not straightforward, one must choose carefully between clinical trial assays, assays that analyse the same biological phenomenon or surrogate biomarkers. Sample heterogeneity and population specific difference is assays may skew results and must be controlled for at the assay design stage. At a technical level, NGS has the potential to revolutionise laboratory practice and approaches to cancer treatment. However, use of this technology requires careful planning and implementation if one is to avoid technical and ethical quagmires. Finally, with FDA regulation of companion diagnostics one may be limited to therapy specific assays.

Shared tumor antigens in colorectal carcinoma and neuroendocrine tumors

ND4 monoclonal antibody recognizes a tumor marker found on poorly differentiated colorectal cancer. We demonstrate its expression in 25% of gastrointestinal neuroendocrine tumors, which also express CEA in 37% of cases. As in colorectal cancer the ND4 marker is predominantly membrane bound in a colonic neuroendocrine tumor cell line, LCC-18 (p < 0.05). The ND4 marker is absent in a poorly differentiated colorectal cancer cell line that does not express CEA or other tumor antigens. Shed antigen in the serum of patients with neuroendocrine tumors is detected in only five of seven patients with the carcinoid syndrome and two of four of those without evidence of the syndrome. However, the reactivity was less in the patients with localized disease, and this test is unlikely to be of diagnostic utility in this group of patients. The sharing of this antigen in colorectal cancer and neuroendocrine tumors is not universal, but does support the common-cell progenitor theory for the origin of these tumors.

An immunohistochemical study of the distribution of blood group substances and related antigens in primary colorectal carcinomas and metastatic lymph node and liver lesions, using monoclonal antibodies against A, B, H type 2, Le(a), and Le(x) antigens

Using indirect immunoperoxidase staining, we examined the distribution of blood group substances and related antigens, including Le(a), A, B, H type 2, and Le(x), in 87 carcinomas and 42 normal mucosal specimens of colon and rectum, as well as in metastatic lesions of 32 lymph nodes and 9 liver specimens. The compatible expression rate of A/B/H type 2 antigens was 33.3% (3/9) in proximal normal colon, but only 6.1% (2/33) in distal normal colon. Compatible expression was observed in 14 of 25 carcinomas in the proximal colon (56.0%), but these antigens in distal colon cancers reappeared with a high positive rate, 62.9% (39/62). The rate of H type 2 accumulation with the deletion of A and/or B antigens was 6.9% (6/87). Incompatible expression was not observed in colorectal cancer. Le(a) and Le(x) antigens were expressed in normal mucosa and primary carcinoma throughout the colon. Le(a) and Le(x) in primary carcinoma that showed A/B deletion with/without H type 2 accumulation was expressed more than in carcinoma with A/B/H type 2 compatible expression (88.2% vs 71.7% for Le(a); 94.1% vs 88.7% for Le(x), respectively). In 16 of 32 patients (50.0%), A/B/H type 2 antigen expression in metastatic lesions had disappeared or was decreased compared with that in primary carcinoma, followed by metastasis to regional lymph nodes. These results suggested that: (a) A/B/H type 2 blood group substances in the distal colon behave as tumor-associated antigens in colorectal carcinoma. (b) Most frequently, A/B/H type 2 antigens expressed in primary carcinoma were weakened or had disappeared in metastatic lymph nodes. Further investigation of the biological function of carbohydrate chains, such as those of blood group substances and related antigens on cancer cell surfaces may lead to a solution of the problem of the metastatic behavior of tumor cells.

Immunohistochemical expression of sialyl Lex antigen in relation to survival of patients with colorectal carcinoma

BACKGROUND

The alterations of carbohydrate chains in cancer cell membrane can be related, not only to the formation of tumor-associated antigens, but also to cell biologic significance. There is, furthermore, a possibility of their relationship to tumor metastatic behavior and subsequent survival of patients with cancer. Recent clinical studies elucidated that a carbohydrate antigen, sialyl Lex, is a useful tumor marker in colorectal cancer. However, the sialyl Lex antigen immunoreactivity in colorectal carcinoma in relation to patient survival is unknown. The aim of the study was to elucidate whether sialyl Lex expression in tumors was correlated with patient survival.

METHODS

Immunohistochemical expression of sialyl Lex antigen, as detected by monoclonal antibody CSLEX1, was studied in 175 specimens of primary colorectal carcinoma from 120 patients who received radical surgery.

RESULTS

The positive expression of sialyl Lex was observed in 87 of 120 patients (72.5%). There was no statistically significant association between the negative or positive of sialyl Lex and clinicopathologic factors, excluding Dukes stage and histologic type. The difference between overall 5-year survival of patients with sialyl Lex-negative tumors and that of patients with sialyl Lex-positive tumors (81.2% versus 60.6%) was statistically significant (P = 0.0263). In proportion to staining intensity that was expressed as a score based on the percentage of the total field stained positive with CSLEX1, 5-year survival of patients indicated a worse outcome (P = 0.0113). The prognostic value was then studied in a Cox regression model. Dukes stage had the strongest association with patient survival, whereas sialyl Lex expression was found to be the second-ranking parameter.

CONCLUSIONS

When examining the expression of sialyl Lex antigen, as detected by monoclonal antibody CSLEX1, clinically useful information for patient survival after radical resection of colorectal cancer is given.

Immunotargeting with monoclonal cytokeratin 8 antibodies of human urothelial cancer transplanted to nude mice

The possibility of using cytokeratin antibodies for the radioimmunolocalization of urinary bladder cancer was studied. A monoclonal murine IgG antibody was raised against cytokeratin 8 and labelled with iodine-125; normal murine IgG was used for control purposes. The urothelial cancer cell line RT4 was transplanted into immunodeficient nude mice. The anti-cytokeratin 8 antibody was administered intraperitoneally and its uptake in the tumour and other organs was analyzed with a computerized gamma camera. Optimal scintigraphic visualization occurred 11 days after antibody administration. The tumour/blood ratio of the specific antibody was 5.64 (+/- 5.01 SD) on day 11, compared with 0.73 (+/- 0.35 SD) in the control. Autoradiography demonstrated antibody uptake preferentially in viable sections of the tumour. The antibody uptake is presumed to be the result mainly of binding to the released cytokeratin in and around cells lysed during natural cellular death. The monoclonal murine anti-cytokeratin antibody is of potential interest in studies aimed at improving the clinical staging of urinary bladder cancer.

Present status of boron neutron capture therapy

The neutron capture reaction 10B(1n,4He)7Li produces two energetic particles, 4He2+ and 7Li3+ that are strongly cell toxic. Due to the short range of these nuclear fragments (5-9 microns) mainly those cells that have bound or internalized a 10B-containing substance are growth-inactivated. The most critical and difficult step in an efficient boron neutron capture therapy (BNCT) is the tumour targeting. It is today possible to synthesize a large number of boron compounds and conjugate them to tumour-seeking macromolecules, such as monoclonal antibodies or different polypeptides. The boron-containing substances presently considered for therapy are sulfhydryl boron hydride (BSH) and boron-phenylalanine, (BPA) for the treatment of gliomas and malignant melanomas respectively. Other boronated compounds considered are ligands for receptor-amplified tumour cells, antibodies for tumour cells with specific antigens and thioureas for treatment of melanotic melanomas. The required boron concentration is given by the relative dose due to neutron capture in 10B and that of the competing capture reactions in nitrogen and hydrogen. Capture in nitrogen produces protons with a range of about 10-11 microns and this gives a radiation dose to all cells in the neutron activated area. Calculations show that the local concentration of 10B near the critical radiation target, DNA, must be higher than 10 ppm (10 micrograms/g). Increased emphasis will be put on the development of combinations of treatments that fulfil the requirements for attacking the microscopic spread of the tumour.

Heterogeneity in antigenic expression and radiosensitivity in human colon carcinoma cell lines

A panel of human colon carcinoma cell lines were characterized regarding both antigenic heterogeneity and variations in radiosensitivity. Monoclonal antibodies were used to study the expression of carcinoembryonic antigen (CEA), gastrointestinal cancer antigen (GICA or CA 19-9) and carcinoma-associated antigen (CA-50). Radiosensitivity was studied with the clonogenic survival technique. Three cell lines, LS 174T, HCTC, and SW 1116 stained positive for all three antigens. HT-29 was positive for CA 19-9 and CA-50 whereas Caco-2 was positive for CEA and CA 19-9. The cell lines SW 620 and LIM 1215 only stained positive for one of the antigens, CA-50 and CEA, respectively. In nearly all positive cases the stainings were very heterogeneous with mixtures of positive and negative cells. One exception was the HCTC cells which stained homogeneously for the CA 19-9 and CA-50 antigens. The neuroendocrinelike COLO 320 cells were negative in all cases. The radiosensitivity varied strongly between the cell lines with Dq-values between 0.8 and 1.9, extrapolation numbers between 2.0 and 4.7, Do-values between 1.1 and 2.8. The surviving fraction at 2 Gy varied between 0.3 and 0.7 with HCTC as the most radiosensitive and HT-29 as the most radioresistant cell line. Thus, there were differences in antigenic expression and intrinsic radiosensitivity between the cell-lines and antigenic heterogeneities within each cell line. The analyzed panel of cell lines will be valuable in studies of dose-effect relations for monoclonal antibodies labeled with toxic radionuclides simulating both antigenic heterogeneity and variations in radiosensitivity.

Expression of blood group antigens H-2, Le(y), and sialylated-Le(a) in human colorectal carcinoma. An immunohistochemical study using double-labeling techniques

In this study, double-labeling immunohistochemistry was used to gain insight into the coexpression or interrelationship between blood group antigens (BGA) that are differentiation antigens in the normal colon, and BGA that are sequential moieties in the same synthetic pathway. Paired-wise Sialylated-Le(a)/Le(y) and H-2/Le(y) was studied. The Sialylated-Le(a) and Le(y) are synthesized from type 1 and type 2 backbones, respectively. In the normal colon, the Le(y) and Sialylated-Le(a) are expressed by cells at the base and surface of the crypt, respectively, representing undifferentiated and differentiated enterocytes. The H-2 is considered oncofetal in nature, and is considered to be the immediate precursor in the synthesis of Le(y). In individual cancers. Sialylated-Lea and Le(y) were detected in different cancer cells within the same malignant glands, separately in different glands, and in different subcellular compartments of the same cell. Both H-2 and Le(y) were coexpressed in the same individual cells in 92% of cancers expressing both these BGA. In 50% of the cancers, the H-2 and Le(y) also were expressed separately in different malignant glands within individual tumors. These findings indicate that, in colorectal cancers, differentiation antigens (Sialylated Le(a) and Le(y)) are expressed by different individual cells within the same malignant gland somewhat, recapitulating the normal colon crypt. Antigens of different backbones occasionally may be expressed in the same cells but within different subcellular compartments. Precursor accumulation is common in cancers, and antigens in the same synthetic pathway are coexpressed in the same cell. The expression of H-2 and Le(y) in different glands (lack of coexpression) may be explained possibly by aberrant synthesis of Le(y) by an alternate pathway.

Liver fatty acid-binding protein: a marker for studying cellular differentiation in gut epithelial neoplasms

Human liver fatty acid binding protein is a 127 residue cytoplasmic protein synthesized in liver and in the intestinal epithelium. Previous studies of normal and transgenic mice indicated that the liver fatty acid-binding protein gene is a sensitive marker of enterocytic differentiation. This study shows the use of immunohistochemical methods to examine liver fatty acid-binding protein gene expression in normal human colonic epithelium, colonic villoglandular adenomas, nonmucinous and mucinous adenocarcinomas, and several types of noncolonic epithelial neoplasms. Cells containing liver fatty acid-binding protein were found in normal colonic epithelium, in two thirds of colorectal villoglandular adenomas and nonmucinous adenocarcinomas, and in one third of mucinous adenocarcinomas but not in noncolonic, nonhepatic carcinomas. All liver fatty acid-binding protein-positive colonic adenomas and adenocarcinomas contained patches of immunoreactive cells distributed among histologically identical patches of cells without liver fatty acid-binding protein immunoreactivity. This "mosaicism" was also found in metastases from liver fatty acid-binding protein-positive colonic adenocarcinomas. Immunostaining of these liver fatty acid-binding protein-positive tissues for carcinoembryonic antigen did not show a mosaic cellular pattern in its expression. These data suggest that within a given neoplasm, differences exist in the differentiation programs of monoclonally-derived, malignant colonic epithelial cells and that liver fatty acid-binding protein is a useful marker for operationally defining these subpopulations. Liver fatty acid-binding protein is also a potentially useful diagnostic marker for colorectal and hepatic carcinomas.

Immunohistochemical assessment of growth fractions in colorectal adenocarcinomas with monoclonal antibody Ki-67. Relation to clinical and pathological variables

Immunostaining with monoclonal antibody Ki-67 (MAb Ki-67) has been employed to determine the growth fractions in a series of 139 primary adenocarcinomas of the large bowel. A wide range (18.9-71.4%; mean 39.4%; median 37.2%) in the percentage of Ki-67 reacting cells (Ki-67 index) was observed. The Ki-67 index was found to be unrelated to tumour stage, lymph node involvement, and presence of synchronous distant metastases. Mucinous carcinomas showed higher levels of Ki-67 reactivity than non-mucinous adenocarcinomas (P = 0.0003). Among non-mucinous adenocarcinomas a significant inverse correlation was demonstrated between the percentage of Ki-67 stained cells and the degree of differentiation (P = 0.002), and preservation of nuclear polarity (P less than 0.001). Moreover, tumours of patients younger than 45 years were generally characterized by high numbers of proliferating cells. There was no correlation between Ki-67 index and the other clinical and pathological variables examined. In most cases small differences in Ki-67 reactivity were observed in different samples from the same tumour. These results demonstrate that immunohistochemical assay with MAb Ki-67 represents a simple and reliable method for the assessment of proliferative activity in colorectal adenocarcinomas and suggest that Ki-67 labeling may provide information of clinical relevance in the management of patients with large bowel cancer.

Can mortality from rectal and rectosigmoid carcinoma be predicted from histopathological variables in the diagnostic biopsy?

The prognostic information provided by histopathological variables in the diagnostic biopsy was evaluated retrospectively in 276 consecutive patients "potentially curable" by surgery using criteria recently proposed by Jass et al. concerning the main specimen. In 252 (91%) cases, the biopsy material was appropriate for the intended assessment. The majority of tumours were tubular (67%), tubular configuration mostly irregular (76%) and the tumour cells usually had small nuclei (81%). Remnants of adenomatous structures were identified in 21%. Most tumours were moderately differentiated (57%), and poorly differentiated tumours were somewhat more common (27%) than highly differentiated ones (16%). There was no real correlation between these histopathological variables, as expressed in biopsy, and Dukes' stage. Patients with a well-differentiated tumour had somewhat better survival than those with a moderately or poorly differentiated one (p = 0.04). Tumour type tended to influence cancer-specific mortality in that patients with papillary tumour had a slightly better prognosis than those with a different type. An attempt was also made to classify the patients into two groups on the basis of the observers' subjective impression of the tumour pattern and the expected outcome. Among the 143 patients in whom the over-all subjective impression was an "aggressive tumour pattern", 43% died (median follow-up 47 months). The cancer-specific death rate in patients in whom the tumour pattern was "non-aggressive" was 27% (p = 0.04). In summary, the extent to which these variables predicted mortality was far from clinically useful.

Homogeneous penetration but heterogeneous binding of antibodies to carcinoembryonic antigen in human colon carcinoma HT-29 spheroids

The monoclonal antibodies 38S1, directed against the carcinoembryonic antigen (CEA), were tested for penetration and binding in human colon carcinoma HT-29 spheroids. Penetration was studied with a method which has not previously been used in immunological investigations. The method, which allows unbound substances to be visualized, is based on freeze drying, vapour fixation, dry sectioning and dry autoradiography. The antibodies penetrated easily and all parts of the HT-29 spheroids seemed to be reached within 15 min. The penetration was even faster than in control glioma U-118MG spheroids that did not express CEA. Binding of the 38S1 antibodies was demonstrated after processing with conventional histology and autoradiography. The binding in the HT-29 spheroids was, after a 1-h incubation period, extremely heterogeneous and occurred mainly in the peripheral parts. More cells were binding the antibodies after 8-h and 32-h incubations and these cells were arranged in peripheral clusters. No binding at all was seen in the CEA-negative glioma spheroids. The distribution of CEA antigens in monolayers and in frozen sections of spheroids of HT-29 cells was analysed with immunohistochemical staining using polyclonal CEA antibodies. The CEA antigens were heterogeneously distributed in both spheroids and monolayers and were as heterogenous as the binding of the monoclonal antibodies in the living spheroids. Thus, the heterogeneous binding in the living spheroids was not due to penetration barriers, but instead to the heterogeneity in the CEA antigen expression.

The spatial distribution of parenterally administered monoclonal antibodies against CEA in a human colorectal tumour xenograft

A recently developed experimental model consisting of athymic rats carrying human colonic tumours from the cell line LS 174 T in both hind legs was used. 125I-labelled anti-carcinoembryonic (anti-CEA) monoclonal antibodies were injected either intra-arterially after a bolus injection of mannitol, or intra-peritoneally with or without mannitol. On the fourth day the rats were killed and pieces from the tumours and various organs were measured in a well scintillation counter. Tumour pieces were then submitted to autoradiography and immunohistochemistry for examination of the antibody distribution at the cellular level. In all examined tumours injected with anti-CEA antibodies, most of the antibodies were located in the periphery close to fibrovascular septa. It appears, in addition to the specificity of the antibody for the CEA, that the tumour vascular permeability and anatomy are of utmost importance for tumour targeting in this experimental model with the particular antibody used.

The expression of carcinoma-associated antigens and blood-group-related antigens in rectal carcinoids

The immunohistochemical expression of blood-group substances and carcinoma-associated antigens were compared in rectal carcinoids and adenocarcinomas. Rectal carcinoid tumors, in contrast to rectal carcinomas, were consistently negative for CEA, gastrointestinal cancer antigen GICA (or CA 19-9), and carcinoma-associated antigen CA-50 (except in one case where less than 10 percent of the cells expressed CA-50). The carcinoids and rectal carcinomas extensively expressed blood-group substance A, B, and H, Lewis B antigen, and difucosylated carbohydrate antigens (DFCA). Thus, rectal carcinoids and adenocarcinomas possess both similar and different tumor antigen profiles. The occurrence of discrepant antigen determinants may be used in the differential diagnosis of these two types of tumors. The coexpression of blood-group substance A, B, and H, Lewis B antigen, and DFCA is consistent with the opinion that both rectal carcinomas and carcinoids have a common entoderm origin, but carcinoids are considered to rise from the endocrine-differentiated and the adenocarcinomas from the nonendocrine-differentiated enterocytes.