TORONTO - Canadian researchers have developed a software app that they hope will improve lung cancer screening around the world.
The risk-calculating software can predict with a high level of accuracy whether lesions, or nodules, in the lungs detected on a CT scan are likely to be malignant or benign.
That can help doctors determine whether a patient at high risk for lung cancer is likely to need an invasive biopsy or surgery, said co-principal investigator Dr. Stephen Lam, chair of the Provincial Lung Tumour Group at the B.C. Cancer Agency.
"We already know that CT screening saves lives," said Lam. "Now we have evidence that our model and risk calculator can accurately predict which abnormalities that show up on a first CT require further followup ... and which ones do not."
The software was developed by Martin Tammemagi, an epidemiologist at Brock University in St. Catharines, Ont., and a member of the cross-Canada research team whose findings from two studies are reported in this week's issue of the New England Journal of Medicine.
"One of the goals of the model is to help reduce the number of needless tests," Tammemagi said in a statement. "When we can identify with accuracy that a person's nodule has low probability for lung cancer, then they'll receive followups that include fewer CT screens, less biopsies and less surgery.
"That will help minimize needless risk to the individual and needless suffering."
To determine whether a spot seen on a CT of a current or former smoker is likely to be malignant, doctors feed information into the app. That data includes the size and density of the nodule and where in the lung it is located; the age and sex of the patient; and such clinical details as whether the person has emphysema or a family history of the disease.
"Then we get a risk index calculated within a couple of minutes," Lam said Wednesday from Vancouver. "It's accurate nine in 10 times." That's for small nodules; in larger lesions, the accuracy rate is even higher — 97 per cent, he said.
Their research showed that while nodule size is one predictor of lung cancer, the largest nodule appearing on a CT among study participants was not necessarily cancerous. The researchers found that nodules located in the upper lobes of the lung carry an increased probability of cancer.
"Essentially, in the first CT, over 20 per cent of the participants in a screening program would be found to have one or more abnormal spots on the CT scan," said Lam. "But only a few per cent would be cancerous."
The studies involved almost 3,000 current and former smokers aged 50 to 75, who had more than 12,000 lung nodules in total detected by CT scans. Of those, 144 were found to be cancerous. Participants were part of one of two studies, the first of which followed patients for three years, while the second averaged 8.5 years of followup.
One of those patients is Chris Douglas of Vancouver, who smoked for about 40 years and was only able to quit once he enrolled in the study six years ago.
When his third annual CT scan showed that a nodule in his lower right lung was getting larger and denser, doctors performed a biopsy and determined the lesion was cancerous, Douglas said.
He wasn't surprised.
"I knew I was engaging in risky behaviour," said Douglas, 65. "I had seen the result of smoking in my parents' generation and in my own. It can be very aggressive by the time it's diagnosed.
"I was almost more relieved that I'd been offered the opportunity for early detection."
He had surgery to remove the area in his lung that contained the tumour, and subsequent CT scans have turned up no evidence of new potentially cancerous nodules.
The retired health services consultant swims five days a week and said he's never felt better, in part because he's no longer smoking. "The lung study has saved my life."
There are no Canadian guidelines to tell doctors which patients should be screened for lung cancer or how often that screening should be done.
In July, the U.S. Preventive Services Task Force issued a draft recommendation that people at high risk for lung cancer receive annual low-dose computed tomography, or CT, scans.
In Canada, there are no formal lung screening programs funded by the provinces and territories, as there are for breast and colorectal cancer, even though CT scans have been shown to reduce lung cancer deaths by 20 per cent, Lam said.
"So this is a big finding, that we can actually do something to reduce lung cancer mortality significantly," he said. "That's why we hope that lung cancer screening will be implemented in Canada at the population level."
The studies were funded by the Terry Fox Research Institute, the Canadian Partnership Against Cancer and the U.S. Public Health Service National Cancer Institute.
Lung cancer kills more than 20,000 Canadians annually and is the leading cause of cancer death in Canada for both men and women, according to Canadian Cancer Society statistics.