First Global Survey of Cancer Survival Released

The first global survey of cancer survival finds very wide variations across five continents. Five-year survival for the four types of cancer included in the study were generally higher in North America, Australia, Japan, and the southern, northern, and western sections of Europe. It was lower in Algeria, Brazil, and eastern Europe.

“The idea for CONCORD germinated in 1999,” says Michel Coleman, lead author and professor of epidemiology and vital statistics at the London School of Hygiene and Tropical Medicine in England. “No population-based studies of transatlantic survival differences had been done in over 40 years, and a number of one-off comparisons had shown higher survival in the United States when compared to Europe. In addition, no one had done a fully rigorous and quality controlled central analysis of data using all patients with a given form of cancer.”

The study gathered data from almost 2 million cancer patients from 101 population-based cancer registries across 31 countries. They compared 5-year survival for cancers of the breast (women), colon, rectum, and prostate. The registries covered a total population approaching 300 million. Patients were diagnosed between 1990 and 1994, and then followed until death or the end of 1999.

“What started off years ago as a comparison between the United States and Europe grew into this concept of pulling together data from as many places as possible,” says John L. Young, Jr., PhD, professor of epidemiology in the Rollins School of Public Health at Emory University in Atlanta and one of the US participants. “This is exciting as a first attempt to look at differences in cancer survival on a much wider basis than in the past.”

Possibly the hardest part of the project was correcting for background mortality and life expectancy across many different databases, countries, languages, and healthcare systems. Nearly 2,800 life tables were constructed to compensate for variations related to sex, year of age, calendar year, country or geographic region, and race in the US.

“When a patient with cancer dies, it is not always the cancer that killed them,” noted Dr. Young. “You don’t want to include all of those other causes of death, so we looked at survival for the cancer cohort regardless of what they actually died from and compared it to a cohort from the general population. As it worked out, measuring actual survival was easy; expected survival was much more difficult.”

The highest survival rates recorded in the 31 participating nations were seen for breast and prostate cancer in the US, colon and rectal cancers in men in Japan, and France had the highest survival for colon and rectal cancers in women. Algeria had by far the lowest survival rates for all cancers in both genders.

Within the US, 16 states and metropolitan areas were included using data from both the Surveillance, Epidemiology, and End Results (SEER) program and the National Program of Cancer Registries (NPCR). These databases accounted for over 42% of the population.

“US survival overall was much better than was seen in the rest of the world,” says Young. “There were some differences within the US, but there was much less variation in outcomes. The striking finding is really an old one as we found a wide gulf in cancer survival between African American and white populations.”

The difference was largest in breast cancers (14%) and smallest in prostate cancer (7%). The authors suggested this was most likely due to racial difference in stage at diagnosis, access to care, and treatment compliance issues.

“What we hope happens as a result of this study is that participating countries will see how they stack up with others across the continents,” says Young. “Instead of saying ‘that’s interesting’ and going back to business as usual, we encourage the countries involved to look at where they were and how they can improve.”

One major concern is the age of the data, since most of it is approaching 10 years old. Both Drs. Young and Coleman noted that one of the reasons to go back so far was to include information from developing nations that may run 3 or 4 years behind in compiling their statistics.

“We are hoping that these results may be a catalyst for updating the study,” says Young. “This should also be a motivation for the individual health care systems to see if the differences still exist today.”

- The CONCORD Working Group. Cancer survival in five continents: a worldwide population-based study (CONCORD). Lancet Oncol. 2008 Aug;9:730-56.

Kurt Ullman is a veteran health care writer and reporter.