By Linda Carroll

(Reuters Health) – Over the last three decades, improved screening and treatments may have kept hundreds of thousands of women from dying of breast cancer, computer simulations suggest.

Depending on the simulation, anywhere from 305,000 to more than 600,000 breast cancer deaths may have been averted, researchers reported in Cancer.

Because no one knows how many women might have died without advances in screening and treatment, the authors of the new study developed estimates of breast cancer trends based on information collected prior to 1990.

“The number of women per 100,000 who died each year from 1975 to 1990 – before the widespread implementation of screening – was increasing slightly from year to year,” said study coauthor Dr. Jay Baker, a professor of radiology at Duke University and chief of the breast imaging at the Duke University Hospital. “There are several assumptions you can make about the maximum mortality rate. For example, you could say it would have been flat after 1990 or you could say it kept increasing at the same rate it was between 1975 and 1990, or you could say it was increasing more than that.”

For the new study, Baker and his colleagues turned to data from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute to estimate how many women between the ages of 40 and 84 were dying from breast cancer each year before and after 1989. SEER collects data on cancer diagnoses, treatment and survival for approximately 30 percent of the US population.

First, they hypothesized four ways history could have gone had there not been advances in screening and treatments. (For example, the number of deaths from breast cancer remained constant after 1989, or death rates increased by 0.4 percent per year since 1989, etc.)

Then, they subtracted the annual number of deaths as calculated from the SEER data from the number of deaths predicted in their simulations. The resulting number, Baker said, was the number of deaths avoided.

Depending on which scenario they used, the number of breast cancer deaths averted between 1990 and 2018 ranged from 305,000 to 614,500.

The new findings need to be interpreted in light of the number of assumptions used in the researchers’ calculations, said Dr. Deanna Attai, an assistant clinical professor in the department of surgery at the David Geffen School of Medicine at the University of California, Los Angeles and UCLA Health Burbank Breast Care.

Another weakness of the study is that SEER data were only available through 2015, and the researchers used that information to estimate death rates going forward.

“I don’t think any of us would doubt the assumption that improvements in screening and improvements in treatment have led to better outcomes,” Attai said. “But you have to consider the fact that they don’t have any real data from 2016, 2017 and 2018. And you have to consider the fact that the SEER database doesn’t include every single woman who died of breast cancer.”

While the SEER database is very useful, it has changed over the years to be more inclusive of different parts of the country as well as minorities, Attai said. That means that the sampling from 20 years ago may differ from more recent years.

Beyond that, Attai said, there is no information in the SEER database on whether women got mammograms within a year of being diagnosed with breast cancer. So it’s impossible to say, based on this database, what impact screening may have had on declines in deaths.

“Studies like this are at best guestimates,” Attai said.

“I am most interested in the years 2017 through 2019 since that’s when there have been tremendous advances in targeted therapies,” she said.

SOURCE: Cancer, online February 11, 2019.