Population-based survival from cancers having a poor prognosis in Mumbai (Bombay), India.

BACKGROUND

Oesophagus, stomach, pancreas and lung cancers contribute more than 35% of the total cancer incidence in Mumbai and survival rates for these cancers are very poor in most populations in the world. The authors here report and discuss the population-based survival from these cancers in Mumbai, India.

METHODS

Follow-up information on 5717 cancers patients having a low prognosis, registered in the Mumbai Population-Based Cancer Registry for the period 1987-1991, was obtained by a variety of methods, including matching with death certificates from the Mumbai vital statistics registration system, postal/telephone enquiries, home visits and scrutiny of medical records. The survival for each case was determined as the duration between the date of diagnosis and date of death, loss to follow-up or the closing date of the study at the end of 1996. Cumulative observed and relative survival rates were calculated by the Hakulinen Method. For comparison of results with other populations, age-standardized relative survival (ASRS) was calculated by directly standardizing age specific relative survival to the specific age distributions of the estimated global incidence of major cancers in 1985. The log rank test was used with univariate analysis to identify the potentially important prognostic variables. The variables showing statistical significance in univariate analysis were introduced stepwise into a Cox Regression model to identify the independent predictors of survival.

RESULTS

The 5-year relative survival rates were 11.8% for oesophagus, 10.1% for the stomach, 4.1% for the pancreas, and 7.0% for lung. Females had higher survival rates than males, except with lung cancer. Lower survival was observed for those younger than 35 years for all 4 sites. For each site, survival declined with advancing age. Single patients who remained unmarried had better survival, except with pancreatic cancer. For all sites Muslims had a better survival and Christians had a lower survival as compared to Hindus. Education did not show any pattern for any site. Survival decreased rapidly with advancing clinical extent of disease for all sites. Survival for localized cancer ranged from 12.5% to 31.3%, for regional spread 1.3% to 3.4% and with distant metastasis not a single site recorded more than 1%. On multivariate analysis, extent of disease emerged as an independent predictor of survival with all the sites. Also, age for oesophagus, stomach and lung, religion for oesophagus and stomach, and education for stomach and lung, emerged as independent predictors of survival.

CONCLUSION

All the sites included in the study demonstrated very low survival rates with significant variation. Comparison with other populations revealed lower survival rates than for Shanghai-China. In remaining populations, survival proportions did not show much variation for pancreas and lung cancers. For stomach cancer, European countries showed better survival rates. Early detection with treatment is clearly important to reduce the mortality from these cancers.