Molecular detection of micro-metastic dissemination in colorectal carcinoma using novel tumour specific markers.

Principal Investigators: Dr Jenny Hardingham

Colorectal cancer (CRC) is a leading cause of morbidity and mortality in developed countries. In Australia, there are about 11,000 new cases diagnosed annually. The development of disseminated (metastatic) disease accounts for the majority of deaths in CRC. Presently, patients with early stage (A or B) disease are considered 'cured' by removal of the primary colon tumour, while patients with stage C (spread to the local lymph nodes) are offered additional chemotherapy, or radiotherapy in the case of rectal cancer. However, up to 30% of patients who have been diagnosed as stage A or B primary tumours, relapse and die within 5 years of surgery as a result of the metastatic spread of cells from the primary tumour to other organs. This suggests that tumour cells with metastatic potential had been shed from the primary tumour prior to surgery. Alternatively, tumour cells had been dispersed during surgery, either into the peritoneal cavity causing local recurrences, or into the blood circulation or lymph nodes causing metastases in distant organs, particularly the liver.

The ability to detect rare disseminated tumour cells may lead to the reclassification of the tumour stage and therefore identify additional patients at risk of developing metastatic disease. These patients would then be offered adjuvant chemotherapy or radiotherapy. A novel method has been established in this laboratory (immunobead RT-PCR) that uses magnetic beads, coated with an epithelial-specific monoclonal antibody to isolate epithelial (colon) cells. The isolated cells are identified as tumour cells using a molecular technique, RT-PCR for specific gene expression markers. A minimum 3-year post surgery follow-up of the clinical outcome for 42 patients (all stages) showed that the patients who were positive for disseminated tumour cells have a significantly shorter disease-free survival. In A and B stage patients, there was a trend for a poorer outcome in patients who were positive compared to patients who were negative, but further patients need to be included in the analysis to obtain statistically significant results. We have shown that the presence of marker-positive cells in blood samples or intra-abdominal wash samples means that these patients are 5 times more likely to suffer disease recurrence than patients who were negative.

Detection of disseminated tumour cells at presentation may have important prognostic implications, altering the staging of the disease and resulting in the early identification of patients at high risk of disease recurrence. It is now recognised that undetected micro-metastatic disease can contribute to the failure of primary treatment. Establishment of a panel of molecular expression markers specific for tumour cells will enable specific and sensitive detection of rare disseminated tumour cells amongst a population of other cells. Long-term follow-up of patients post surgery will determine if detection of positive cells either in the peritoneal cavity, blood samples or lymph nodes is an independent prognostic marker of disease-free and overall survival. If so, then such 'high risk' patients identified at the time of surgery may be offered adjuvant therapy, which at present is only offered to more advanced stage patients.

Detection of disseminated tumour cells in blood and abdominal wash samples by Immunobead RT-PCR.
Results to date show that 34/171 patients were positive for one or more markers in pre-operative samples, while 46/171 were positive in post-operative samples, suggesting that tumour cells were shed during surgery. The positive patients included 14 stage A, and 16 stage B. Long term follow-up will be done to determine if the presence of disseminated tumour cells correlates with a poorer outcome in these early stage patients.

Detection of disseminated tumour cells in lymph nodes
The reliability of conventional microscopic examination of lymph node sections is hampered by several limitations. It can be difficult to identify single, or even small clumps, of tumour cells by routine histopathology. Also, examination of only a few sections from lymph nodes means that most of the specimen is not reviewed. More sensitive techniques are thus needed to detect micrometastatic disease.

We have used expression of CEA, CK20 or guanylyl cyclase C (GCC) by RT-PCR assay to detect the presence of tumour cells in lymph nodes from CRC patients and from controls (patients undergoing bowel resection for benign conditions). CEA and CK20 are established markers of colon tumour cells. GCC has been found to be a selective marker for colon tumour cells in extra-intestinal tissues (Carrithers et al ., 1996). We have analysed 213 lymph nodes that were reported histologically negative for tumour cells, from 40 colorectal cancer (CRC) patients (8 stage A, 12 stage B, 15 stage C and 5 stage D). As controls, 49 lymph nodes from 14 benign control cases were harvested from the tissue removed at operation, and 27 'no operation' control blood samples were collected to test for expression of the markers in white blood cells (WBC). Overall there were 9/40 cases positive for all 3 markers, and 35/40 positive for at least one marker. In patients with stage A or B disease, 4 stage A and 10 stage B patients had lymph nodes that were positive for one or more markers. These results indicate that RT-PCR was able to detect the presence of tumour cells in an additional 14 patients who were considered lymph node metastasis-free by conventional pathology. This would result in a staging shift from stage A or B to stage C, which has important implications for treatment decisions.

Targeting over-expressed genes for therapeutic intervention

We are developing a technique to reduce the activity of genes that are over-expressed in colorectal cancer by preventing the process of translation of protein from those genes. This is achieved by harnessing a naturally occurring process called RNA interference. Short interfering (si) RNA sequences are made in the laboratory specifically for each of the target genes. Transfection of specific siRNAs into the colon cancer cells in culture should result in abrogation of expression of that gene. The effect of gene silencing will be measured using an invasion chamber in which the rate of invasion through a basement membrane of the transfected cells will be compared to non-transfected cells. These experiments will allow us to select the most appropriate genes to target for therapeutic purposes.