 |
5-Year survival rates under scrutiny
A comparative analysis of how changes in 5-year survival rates relate to changes in mortality and incidence indicates a tenuous relationship between these measures of cancer burden, according to researchers from the Department of Veterans Affairs Medical Center in Vermont, USA. While increased 5-year survival is commonly inferred to correlate with improved treatment and decreased mortality, Dr Gilbert Welch and co-authors of the study [1], claim that 5-year survival is an unreliable indicator of progress and may be extremely misleading.
The authors used population-based statistics reported by the National Cancer Institute's (NCI) Surveillance, Epidemiology and End Results (SEER) programme - the US Federal Governments primary effort to collect and report cancer incidence, initial treatment and survival - to calculate the change in 5-year survival from 1950 to 1995 for the 20 most common solid tumours. They then correlated changes in 5-year survival with changes in mortality and incidence.
Welch and colleagues found an increase in 5-year survival for each of the 20 tumour types, ranging from 3% for pancreatic cancer to 50% for prostate cancer. Mortality rates fell for 12 cancer types and increased for the remaining eight during the same period. They found little correlation between the change in 5-year survival for a specific tumour and the change in tumour-related mortality. Conversely, the change in 5-year survival was positively correlated with the change in tumour incidence rate.
The authors conclude that their data suggest that inferences can not be made about the effectiveness of early diagnosis or treatment from temporal changes in 5-year survival, and improved 5-year survival may instead reflect more cases diagnosed and unchanged mortality. They stress that they are not implying there has been no progress in cancer treatment but point to population-based mortality rates as a less ambiguous measure of progress.
Commenting on the study, Roger Black, Head of the Scottish Cancer Intelligence Unit in Edinburgh, UK, said "Few would disagree with the assertion of this study that increasing survival rates cannot be interpreted as unequivocal evidence of progress in combating cancer. However, we should also be aware of the limitations of mortality data, recognising in particular that trends in mortality result from underlying trends in incidence and survival". He added, "Another limitation is that, for most types of cancer, deaths observed in a particular period include patients diagnosed and treated many years before. Surely, our understanding of the occurrence and outcome of cancer is best served by studying incidence, survival and mortality. This is feasible in the many countries and regions of Europe which have complete population-based cancer registration and death registration systems".
1Welch HG, Schwartz LM, Woloshin S. Are increasing 5-year survival rates evidence of success against cancer? JAMA 2000, 283, pp2975-2978.
European Journal of Cancer
International Cancer News
Compiled by: Brad Timms (Acting News Editor)
European Journal of Cancer: 36 (2000) pp1725-1729
Copyright 2004 Published by Elsevier Ltd
Back . . .
 |
 |
 |
 |
 |
 |
 |
 |