Methods: In this study, samples of bone marrow and venous blood, collected pre and postoperatively for 40 cases, were screened for disseminated tumour cells using immuncytochemistry with a pan cytokeratin antibody and E48 RT-PCR, to establish the most appropriate samples and assay to incorporate into prospective studies. The molecular approach was also applied to intraoperative central venous blood.
Results: The concordance between the molecular and immuncytochemical tests applied to preoperative samples was moderate (BM 0.62, CVB 0.12), likely reflecting sample errors. Testing samples of bone marrow and central venous blood preoperatively with either approach gave results that predicted disease-free survival and distant- metastases-free survival. Testing the intraoperative samples was also a sensitive predictor of distant metastases. However, testing bone marrow or blood post-surgery was not helpful
Conclusions: The preliminary data revealed that detection of disseminated tumour cells pre and intraoperatively can identify individuals at high risk of local and distant recurrence. The discordance between the results obtained with the different tests postoperatively may reflect the presence of epithelial cell debris in the haematopoietic cell compartment post-radiotherapy. These findings suggest that detection of disseminated tumour cells pre- or intraoperatively signals for a high risk of local and distant recurrence and reduced survival in head and neck cancer.