Evaluation of a pre-operative staging protocol in the management of colorectal carcinoma.

OBJECTIVE: The optimum strategy for pre-operative staging of colorectal carcinoma (CRC) has yet to be defined. A protocol for staging CRC patients was set up in this hospital in 1998. The protocol included complete colonic visualization together with assessment of the liver and lung for potential metastatic disease. Pelvic imaging was required to assess the local spread of rectal tumours. Our aim was to evaluate prospectively this protocol. PATIENTS AND METHODS: Data from all patients diagnosed with primary CRC between January 1999 and December 2002 were prospectively collected and analysed. RESULTS: There were 295 patients; 56 (19%) patients presented as an emergency and were excluded. The study group consisted of 239 patients (206 had elective surgery and 33 had no resectional surgery). In the patients who presented electively; 88% had complete colonic imaging; 87% chest imaging; 90% had liver imaging; 91% of rectal tumours had pelvic imaging. Overall 75% of the elective patients completed the staging protocol. Reasons for incomplete staging were numerous and most were justifiable. Findings which influenced clinical management included alteration in surgical approach (14), lung metastases (7), primary lung cancers (2), definite liver metastases (25), possible liver metastases (8), neo-adjuvant radiotherapy required (27), advanced local disease (9) and other incidental findings (12). CONCLUSION: Our protocol influenced further management decisions in 39% of patients. Better stratification of patient care is possible, with the ultimate aim to avoid unnecessary surgery. However, complete staging is not always possible to perform.

Rectal retroflexion: an essential part of lower gastrointestinal endoscopic examination.

PURPOSE: Retroflexion of the endoscope during rectal examination may increase diagnostic yield but is not routinely performed because of concerns about safety and a lack of appreciation of its importance. The purpose of this study was to examine the yield, safety, and tolerance of endoscopic rectal retroflexion. METHODS: Prospective cohorts of subjects undergoing unsedated screening flexible sigmoidoscopy were examined with and without routine retroflexion. Pain scores were recorded. RESULTS: A total of 526 subjects (mean age 60 (range, 55-66) years) underwent flexible sigmoidoscopy in the first period when the endoscope was not routinely retroflexed. Of these, 480 (mean age 60 (range, 55-66) years) were subsequently examined with routine retroflexion. Retroflexion was impossible in 17 subjects (3.5 percent) because of discomfort. In the second group, 12 subjects (2.5 percent) had polyps in the lower rectum seen only on retroflexion. Of these, eight had metaplastic and four had adenomatous polyps (3 tubular <5 mm, 1 tubulovillous 15 mm). There was no difference in mean pain scores between the groups (no retroflexion = 2.13, retroflexion = 2.18). CONCLUSION: With an adenoma pick-up rate of 8 to 12 percent for screening flexible sigmoidoscopy, retroflexion increases adenoma detection by approximately 1 percent without adverse effects and should be an integral part of flexible sigmoidoscopy.

Use of the monoclonal antibody DAKO-ERbeta (8D5-1) to measure oestrogen receptor beta in breast cancer cells.

BACKGROUND: Oestrogen receptor beta (ERbeta) is highly homologous with the classical ER (known now as ERalpha). The exact role of ERbeta in breast cancer and its contribution in influencing patient response to endocrine therapy remains unclear. The aim of this study was to develop and evaluate a flow cytometric method for the detection of ERbeta in breast cancer cells using the DAKO monoclonal anti-ERbeta 8D5-1 antibody. METHODS: MCF7 cells were used as a positive control and U937 as a negative control for titration of the antibody. Cell lines and tumour samples were fixed with 1% paraformaldehyde and permeabilised with 0.5% saponin prior to flow cytometric analysis. RESULTS: A ten fold difference in expression of ERbeta within the different breast cell lines studied was found. Confirmation of antibody specificity against ERbeta protein by Western blot analysis detected a single band at approximately 65kDa. ERbeta immunopositive nuclei were identified in MCF7 cells by immunohistochemistry. CONCLUSIONS: DAKO ERbeta 8D5-1 antibody is specific for ERbeta protein and does not cross react with ERalpha protein. Using this antibody, ERbeta can be detected and accurately quantified in cell lines and solid breast tumours by flow cytometry.

Glutaraldehyde as a possible cause of diarrhoea after sigmoidoscopy.

BACKGROUND: An outbreak of proctitis at the start of a colorectal cancer screening programme utilizing flexible sigmoidoscopy prompted scrutiny of the incidence of this complication and the role of glutaraldehyde in its aetiology. METHODS: Questionnaires completed 1 day and 3 months after sigmoidoscopy were reviewed for 388 patients, and glutaraldehyde levels in the recycled rinse water of the endoscope washing machine were measured. The incidence of symptoms in the subsequent 612 patients after installation of a washer that does not recycle rinse water was examined. RESULTS: Five patients (1.3 per cent) presented to hospital with bloody diarrhoea occurring immediately after a normal flexible sigmoidoscopy. Repeat examination confirmed the presence of proctitis. Symptoms subsided rapidly with either no treatment or steroid enemas. Eight additional patients (2.1 per cent) recorded similar problems but received no treatment and the symptoms settled spontaneously. Glutaraldehyde levels rose progressively in the rinse water after each wash cycle with 2 per cent glutaraldehyde solution. Only one possible case of proctitis (0.2 per cent) was identified from the questionnaires completed by 612 patients after changing to a washer that did not recycle the rinse water. CONCLUSION: These observations should prompt the careful assessment of cleaning techniques. The use of washing machines that do not recycle rinse water may avoid this complication.

Cytokeratin expression in breast cancer: phenotypic changes associated with disease progression.

Transition from a normal to a cancerous state is marked by alterations in the cytoskeletal structure of those cells involved. We have examined such changes to determine if these transitions are markers of disease progression. Cytokeratin (CK) protein and messenger RNA (mRNA) expression were examined in malignant and benign breast tissues. Flow cytometric results demonstrated a significant correlation between cytokeratin protein expression detected by 5D3 antibody, specific for cytokeratins 8, 18, and 19 and axillary node metastasis (P = 0.01). A threshold of positivity of 338,000 molecules/cell was determined and reflected the wide range in cytokeratin levels expressed by normal or benign tissues. Examination of cytokeratins 8, 18, and 19 revealed a consistent pattern of expression with respect to tumor grade. Only cytokeratin 19 showed significant correlation with increasing tumor size (P = 0.006). mRNA expression for cytokeratin 8 was significantly higher in node-positive compared with node-negative disease (P = 0.02). Cytokeratin 18 mRNA levels were significantly lower in both node-negative (P = 0.03) and node-positive (P = 0.02) patients when compared with benign samples. Increased levels of cytokeratin 18 mRNA showed an inverse relationship with protein expression (P = 0.05). The results indicate that cytokeratin expression in breast cancer may be associated with tumor progression. Furthermore, the alteration in the expression of individual cytokeratins deserves further investigation to determine the consequences of these changes with respect to cellular function.

The effect of 3-week tamoxifen treatment on oestrogen receptor levels in primary breast tumours: a flow cytometric study.

The effect of 3-week, preoperative tamoxifen treatment on oestrogen receptor (ER) levels, expressed by primary breast tumours, was examined. Patients (age-matched) with breast cancer, confirmed by fine-needle aspiration, were either treated with 20 mg ml(-1) oral tamoxifen per day or received no medication in the 3-week interval between assessment and surgery. Quantification of ER using flow cytometry was performed on the surgically removed tumour samples from tamoxifen-treated (n = 40) and control (n = 38, untreated) patient groups. The tumours were mechanically disaggregated, and saponin treatment rendered these cells permeable to antibodies. Using dual-parameter labelling with a FITC-conjugated antibody (NCL-5D3) directed against cytokeratin 8/18/19 and a biotinylated antibody (DAKO-ER 1D5) directed against the oestrogen receptor, ER quantification was determined on a number of receptors per cell basis. Using QC quantum bead standards, ER levels in the epithelial cell population, the non-epithelial cell population and the whole-cell population (ER+) were calculated. ER levels were significantly lower in the total cell population than tamoxifen-treated patients (P = 0.002) when compared with the control (untreated) group. By using a gating procedure using 5D3 antibody positivity, a significantly lower level was detected on examining the cytokeratin-positive population alone (P = 0.006). Using a complementary gating technique, ER levels were quantified in the cytokeratin-negative cell population. Examination of this group of cells showed no significant difference between the levels of oestrogen receptor found in the tamoxifen-treated and untreated groups (P = 0.4). We have demonstrated that ER levels can be monitored by flow cytometry. ER levels in patients treated with tamoxifen 3 weeks before operation are significantly lower than in a comparative group of patients who received no drug. Furthermore, the most significant difference in receptor levels is seen by quantification of total ER levels expressed by all the tissue.

Examination of multidrug resistance in cell lines and primary breast tumours by flow cytometry.

The aim of this study was to measure multidrug resistance (MDR) by flow cytometry and quantify the expression of P-glycoprotein (using antibody) glutathione transferase (using alpha-GSTpi antibody) in alpha-JSB-1 and alpha-GSTpi of a series of cell lines and primary breast cancers, and to assess the relationship between these MDR proteins and a selection of oncogene and prognostic markers in breast cancer. Flow cytometry was performed using permeabilised cells stained with fluorescent antibodies using well-established methods. Antibody staining was confirmed for JSB1, but not GSTpi by use of known positive and negative controls. No correlation was seen when comparing the number of molecules of alpha-JSB-1 with alpha-GSTpi (P = 0.1, r2 = 0.4, n = 14) using a selection of cell lines. Examination of 45 breast tumours for expression of JSB-1 and GSTpi revealed a significant association between these two antibodies (P < 0.00001, r2 = 0.5, n = 45). On examining the breast tumours, alpha-JSB-1 showed a positive association with c-erbB-2 (P = 0.003), c-myc (P = 0.0004) and c-jun (P = 0.02) but not ER or EGF-R expression. alpha-GSTpi showed a positive association with c-erbB-2 (P = 0.03) and c-myc (P = 0.0004) but not ER, EGF-R or c-jun. Flow cytometric MDR levels were not related to tumour grade or axillary node status. In solid tumours, a relationship between the two antibodies used has been clearly demonstrated, however, specificity of alpha-GSTpi is questioned. Both antibodies show an association with c-erbB-2, which is associated with poor prognosis and with c-myc which is involved in cell cycling and differentiation. Monitoring MDR markers (Pgp) using this methodology may be useful for evaluation of prognosis in breast cancer.

Inhibition of endothelial adhesion and invasion by breast carcinoma cells may contribute towards the anti-metastatic effects of tamoxifen.

The anti-metastatic actions of tamoxifen on the oestrogen receptor-(ER-) positive cell line, MCF-7 and Hs578T, which is ER-negative, were investigated by measuring changes in the tumour cell adherence to endothelium and invasion of Matrigel. The endothelial hybridoma EA.hy926 was grown to confluence on the bases of 96-well plates. Either tamoxifen, the pure ER antagonist ICI 182,780 or the control, phosphate-buffered saline (PBS), was added to each well in varying concentrations. Adhesion of tumour cells to the endothelium was then measured using an isotopic adhesion assay. Invasion was determined by measuring the number of cells passing across a Matrigel-coated filter with 8 microm diameter pores. After 24-h incubation, the number of cells which had invaded was determined by an XTT colorimetric assay. Tamoxifen and ICI 182,780 inhibited both adhesion to the model endothelium and Matrigel invasion of the ER-positive cell line at therapeutic concentrations (P<0.005). Neither compound, however, had an effect on the ER-negative cell line. This action of the ER antagonists may play a role in prolonging the disease-free survival seen in women with breast cancer who are treated with adjuvant tamoxifen.

The role of beta 1 integrins in adhesion of two breast carcinoma cell lines to a model endothelium.

Interactions between tumour cells and the endothelium are vital to the formation of haematogenous metastases. Binding to model endothelium of one oestrogen receptor positive breast carcinoma cell line (MCF-7) and one receptor negative line (HS578T) was examined in vitro together with endothelial retraction induced by these tumour cells. Adhesion was inhibited by monoclonal antibodies specific for the VLA integrins and by peptides containing the RGD motif which is commonly recognised as a ligand by the VLA adhesion molecules. However, binding of the two tumour cell lines was inhibited by monoclonal antibodies specific for different VLA molecules; anti-alpha 6 beta 1 inhibited MCF-7 adhesion but anti-alpha 5 beta 1 inhibited Hs578T. These results were consistent with flow cytometric quantification of the expression of these VLA integrins on the surfaces of the two tumour cell lines. Enzyme-linked immunosorbent assays (ELISA) demonstrated that laminin was present on the endothelial cell surface but collagen IV was absent. ELISA failed to detect increased exposure of the subendothelial matrix during the first hour after addition of either cancer cell type. This was supported by assays which demonstrated maintenance of the endothelial permeability barrier during this period. Slight endothelial retraction was detected within 2 hours of the addition of tumour cells. It is concluded that binding between tumour cells and confluent endothelium is inhibited by the blockade of adhesion molecules which are normally associated with interactions between the cell and the subendothelial matrix. Tumour cell to matrix interactions rather than direct tumour to endothelial cell adhesion may be the limiting step in tumour cell binding to the endothelium.

Renal allograft rejection: expression and function of VCAM-1 on tubular epithelial cells.

The interaction between vascular cell adhesion molecule-1 (VCAM-1) and very late antigen-4 (VLA-4) is known to play an important role in stabilizing the adhesion of lymphocytes to endothelial cells. Such cellular adhesion is crucial to many immunological processes including lymphocyte-mediated cell lysis. In this study the expression of VCAM-1 on renal tubular epithelial cells is demonstrated on biopsy sections recovered during acute renal allograft rejection. Experiments performed using epithelial cells cultured from renal tubules show that VCAM-1 is up-regulated by addition of the inflammatory cytokines tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma). Combination of TNF-alpha and IFN-gamma synergized to induce high levels of VCAM-1 expression. Further experiments demonstrated that the cytokines produced by activated lymphocytes in mixed leucocyte culture also up-regulate expression of VCAM-1. Assays of the adhesion of lymphoid cells to cultured renal epithelial cells showed that cytokine pretreatment of the renal cells enhanced the binding of lymphoid cells. The proportion of bound lymphoid cells was significantly reduced by addition of an antibody capable of blocking the interaction of VCAM-1 with VLA-4. This result indicated that the VCAM-1 induced on renal epithelial cells by inflammatory cytokines is functionally capable of binding VLA-4, thereby enhancing the adhesion of potentially graft-damaging lymphoid cells.

Immunologic status of the cancer patient and the effects of blood transfusion on antitumor responses.

For patients undergoing potentially curative surgery for cancer, the perioperative period is when they appear to be most vulnerable. The antitumor immune response is subjected to a number of iatrogenic insults at this time, not the least of which is the surgery itself. Any improvements that can be made by a further understanding of the pathophysiology of the events in the perioperative period, and the therapeutic interventions to control them, would obviously benefit the patient. Over the past 10 years, the desire for such benefits has led to the intense investigation of "the blood transfusion effect" in cancer surgery, which this article reviews.