New model helped predict absolute risk for breast cancer in Italian population

Petracci E. J Natl Cancer Inst. 2011;doi:10.1093/jnci/djr172.

Issue: August 10, 2011

Researchers from Italy have developed a model to determine the absolute risk for breast cancer and which lifestyle interventions could aid in disease prevention.

The researchers collected data for 2,569 patients and 2,588 controls who were part of an Italian study from June 1, 1991, to April 1, 1994; and data on incidence and mortality from Florence Registries. Using this information, they designed an absolute risk prediction model for breast cancer that included five standard non-modifiable risk factors (reproductive characteristics, education, occupational activity, family history and biopsy history) and three modifiable risk factors (BMI, alcohol consumption and physical activity). Independent data from an Italian cohort was used to validate the model, and researchers calculated the percent risk reduction in high-risk subgroups identified by the Lorenz curve. According to the researchers, the model was reasonably well-calibrated, but the discriminatory accuracy was modest (ratio of expected-to-be observed cancers=1.10; 95% CI, 0.96-1.26). In addition, the model predicted that changing modifiable risk factors reduced the absolute risk for developing cancer by quantifiable amounts. Among women aged 65 years, the average 20-year reduction in absolute risk was 1.6%; among those with a positive family history of disease, the absolute risk was 3.2%; and it was 4.1% among those with the most non-modifiable risk factors.

This model may aid in the creation of programs designed to encourage healthy lifestyle changes, according to a press release. In a general population of women, a 1.6% absolute risk reduction amounts to 16,000 fewer cases of cancer. However, according to the model, a 3.2% reduction in a high-risk group, such as postmenopausal women with a family history, amounts to only 2,560 fewer cases.

“[The study] provides extremely important information relevant to counseling women on how much risk reduction they can expect by changing behaviors, and also highlights the basic public health concept that small changes in individual risk can translate into a meaningful reduction in disease in a large population,” Kathy J. Helzlsouer, MD,Mercy Medical Center in Baltimore, wrote in an accompanying editorial.

These estimates, Helzlsouer said, are based on optimistic assumptions of changes in behavior, and the results do not apply to those in other countries, such as the United States, where breast cancer rates among older women are much higher than in Italy.

“These differences, however, serve to emphasize the need for models such as these to be created on the basis of local data to provide the best estimates of risk to aid decision-making, whether for the individual or the population in public health settings,” she said.

Follow HemOncToday.com on Twitter.