Survival from oral cancer in Mumbai (Bombay), India

OBJECTIVE: Cancer survival information is available for only few populations in developing countries. Data on survival of oral cancer patients registered by the Bombay population-based cancer registry, India, during 1992-1994 is presented in this paper. METHODS: Vital status of the patients was established by matching against death certificates, reply paid postal enquiries, telephone enquiries, scrutiny of hospital records and house visits. Of the 1808 eligible cases for analysis, 1204 (66.6%) were dead; and 604 (33.4%) were alive at last follow-up, of whom 136 were lost to follow-up during the first 5-years after diagnosis. Observed and relative survival rates are presented and hazard ratios and corresponding 95% confidence intervals were estimated using Cox proportional hazard model. RESULTS: The overall 5-year observed and relative survival rates were 30.5% and 39.7%, respectively. Survival steadily declined with advancing age and advanced clinical stages. Five-year observed survival was 59.1% for localised cancer, 15.7% for cancers with regional extension and 1.6% for those with distant metastasis. Those with tongue, buccal mucosa and retromolar trigone cancers had poor survival rates. CONCLUSIONS: Our study shows that detecting oral cancer in early stages, when these are amenable to single modality therapies, offers the best chance of long-term survival. However, primary prevention by tobacco/alcohol control measures is important in the long-term perspective.