Cancer survival estimates vary by calculation method, study finds

A European study of nearly 1.9 million cancer cases found that the method used to calculate survival can significantly affect the results. For head and neck cancers, 10-year survival estimates differed by as much as 11 percentage points, ERR reported.

Cancer survival is a key measure of treatment outcomes and how they change over time. One calculation method counts deaths caused by a specific cancer; another compares cancer patients’ survival with that of people of the same age and sex in the general population.

The international study, published in the journal Cancer Medicine, examined registry data from 13 European countries, including Estonia. Researchers compared cancer-specific and other-cause mortality for 11 cancer types.

Kaire Innos, research director at the National Institute for Health Development and a member of the study team, said relative survival calculations generally assume that cancer patients face the same risk of dying from other causes as people of the same age in the general population. The findings showed that this assumption did not hold for several cancer types, she said, adding that patients’ other health risks should be considered when interpreting survival figures.

The largest difference between the two measures was for head and neck cancers: 10-year cancer-specific survival was 7 to 11 percentage points higher than relative survival. Differences were also found for bladder, stomach, lung, cervical, ovarian and kidney cancers.

For breast and prostate cancers, the pattern was reversed: cancer-specific survival was lower than relative survival. The difference was particularly large among older patients.

The study found that the risk of dying from other causes varied according to cancer type. Compared with the general population, men with head and neck cancer had 2.08 times the risk of dying from other causes, while the risk for women with the same cancers was 2.21 times higher. For women with lung cancer, it was 2.6 times higher.

By contrast, women with breast cancer and men with prostate cancer had a lower risk of dying from other causes than the general population. Innos said the differences may be linked to health behaviour, other illnesses and treatment. Smoking, for example, is a risk factor for both head and neck cancers and several other chronic diseases. For breast and prostate cancers, health characteristics of people who attend screening and health checks may also affect the results, she said.