Duke University researchers have figured out why some human papilloma virus (HPV) infections go away while other become cancer.

For people infected with the HPV, the likelihood of clearing the infection and avoiding HPV-related cancer may depend less on the body's disease-fighting arsenal than has been generally assumed.

Researchers found that the body's ability to defeat the virus may be largely due to unpredictable division patterns in HPV-infected stem cells, rather than the strength of the person's immune response.

More than six million people in the U.S. become infected with HPV every year. Most people clear the virus on their own in one to two years with little or no symptoms. But in some people the infection persists. The longer HPV persists the more likely it is to lead to cancer, including cancers of the cervix, penis, anus, mouth and throat.

To better understand why some HPV infections go away and others progress, researchers developed a model of HPV infection at the level of the infected tissue. By combining their model with data from a population of 313 teenage girls who were tested for HPV every six months for four years, the researchers were able to measure the influence of symmetric and asymmetric cell divisions on the time it takes to get rid of the virus, and compare it to the body's ability to kill virus-infected cells via immune cells called killer T-cells.

Surprisingly, random division patterns in infected cells were found to play a critical role in eradicating the virus.

In particular, the researchers calculated that as much as 83 percent of the body's ability to clear the infection can be explained by the pattern of divisions in HPV-infected cells.

"There is no doubt that the immune system plays a role in HPV clearance," Marc Ryser, co-author of the study, said in a statement. "However, the contribution from the stem cell division patterns may play a non-negligible role in the process, something that has not been acknowledged in previous studies."

The team's next step is to use their model to try to understand why some people who take oral contraceptives for a long time have an increased risk of developing cervical cancer.

The findings are detailed in the journal PLOS Computational Biology.