Do cell kinetics have prognostic and/or predictive value in oesophageal cancer treated by surgery?

Abstract

Tumour cell kinetics could help in predicting the optimal duration of treatment for the individual patient. In order to assess the importance of cell kinetics in oesophageal cancer, 63 patients with cancer of the oesophagus and/or gastro-oesophageal junction were studied. Seven patients had T1 tumours, six T2, 47 T3, and three T4. Twenty patients had no pathological nodes, while 43 patients had node-positive disease. Thirty-one patients had squamous cell carcinoma, 31 patients suffered from adenocarcinoma and one patient had signet ring cell carcinoma. The primary treatment was surgery. 5-iodo-2'-deoxyuridine (IUdR) was injected 6 to 10 h before surgery, and five biopsies per tumour were taken. The labelling index (LI), S-phase duration (Ts) and potential doubling time (Tpot) on the 305 biopsies taken were measured using flow cytometry. Overall, 1-year disease-free survival (DFS) was 57%, with the 2-year DFS being 38%. T-stage, pathological node status and sex significantly influenced the DFS. The mean Tpot from our 63 patients was 5.6 days, with a standard deviation of 3.6 days. When DFS was studied as a function of Tpot, no statistically significant difference was found between fast- and slow-proliferating tumours (log-rank, P = 0.84). A trend developed with fast-proliferating tumours recurring earlier than slow-proliferating ones. From our results it was also clear that intratumour variability exists, and is a confounding factor using Tpot as a predictor for treatment outcome. When tumours were classified into two categories, 'fast' or 'slow', according to their mean Tpot value (with the median Tpot (4.6 days) as the cut-off value), the trend for fast tumours recurring earlier was less clear than when the classification of tumours was based on the confidence interval (CI) being situated entirely below or above the mean. This classification into 'fast' or 'slow' included 27 of 63 tumours. The other tumours (n = 36) had a CI which included the cut-off, and could thus not be classified in this way. DFS at 1 year was 55% for the fast-proliferating tumours (n = 32) vs 63% for the slow-proliferating ones (n = 31). DFS at 1 year on the basis of the CI, however, was 38% for the fast-proliferating tumours (n = 17) vs 45% for the slow-proliferating ones (n = 10). Tumour cell kinetics have the greatest chance of predicting outcome in relatively long treatment schedules, where proliferation will have more time to occur. The fact that oesophageal tumours are almost all fast-proliferating tumours means that for new treatment schedules overall treatment time should not be unduly prolonged, whichever combination of surgery, radiotherapy and chemotherapy is used.