Women diagnosed with interval breast cancers — those detected between routine screenings — had a higher risk for aggressive disease and death, a restrospective study in Canada indicated.
Among women participating in a national screening program, and with a median follow-up of 7 years, breast cancer-specific mortality was more than threefold higher for women diagnosed with interval breast cancers compared to those whose cancers were found on screening, which included a sojourn time of 2 years to account for lead-time bias (hazard ratio [HR] 3.55, 95% CI 2.01-6.28, P<0.001), reported Saroj Niraula, MD, MSc, of the University of Manitoba in Canada, and colleagues.
The findings highlight the differences in the natural history of these cancers “and highlights inadequacies in current breast cancer screening practice,” the group wrote in JAMA Network Open. “Many of the aggressive and lethal forms of breast cancers either go unnoticed on mammogram or develop in the interval between mammograms.”
Compared to those found on screening, interval cancers were more likely to be high-grade (odds ratio [OR] 6.33 for grade III vs grade I, 95% CI 3.73-10.75, P<0.001) and estrogen-receptor negative (OR 2.88, 95% CI 2.01-4.13, P<0.001), according to the findings.
Adjustments for patients’ age and income did not change the overall findings, nor did sensitivity analyses that extended the sojourn time up to 4 years, though the effect size diminished.
It’s well established that interval breast cancers are more likely to be aggressive subtypes (triple-negative, HER2-positive) that require chemotherapy, Sarah Cate, MD, of the Icahn School of Medicine at Mount Sinai in New York City, told MedPage Today.
“Screening breast cancers are more likely to be indolent and usually require less adjuvant treatment, meaning chemotherapy. Thus, this article reinforces these two known facts,” said Cate, who was not involved in the study. “When I see a patient who had a normal mammogram and then developed a breast cancer 6 months later, these are usually more aggressive subtypes and will need chemotherapy.”
Cate also noted some limitations of the study, including that no information on patients’ breast density and family or genetic history was included.
“We know that dense breasts can lead to a missed cancer diagnosis, as cancer appears white on a mammogram, and so do dense breast tissue. This is known as a ‘masking effect,'” said Cate. In the U.S., women with dense breasts will typically also receive ultrasound to supplement their mammography, she noted.
“Patients with genetic mutations and those with a strong family history of breast cancer should undergo annual breast MRI, in addition to mammography,” she added. “In these patients, we already know that mammography is not enough.”
For their study, Niraula’s group used the Manitoba BreastCheck registry, the Manitoba Cancer Registry, and Statistics Canada census data to examine outcomes in 69,025 women (ages 50 to 64) who from 2004 to 2010 underwent a total of 212,579 mammograms, resulting in 1,687 cases of invasive breast cancer.
In all, 705 of the cancers were detected by screening (within 6 months of an abnormal finding on mammography), 206 were interval cancers (within 2 years of a normal mammogram), 275 were diagnosed in non-compliant patients (over 2 years since their last mammogram), while 501 were diagnosed among women outside the screening program.
There were 170 deaths from breast cancer: 20 in the screen-detected group, 29 in the interval group, 27 in the non-compliant group, and 94 among women not participating in the program. Patients who were non-compliant with the screening program had a higher risk of death compared to those with screen-detected cancers (HR 2.18, 95% CI 1.21-3.95, P=0.002), as did those who didn’t take part in the screening program to begin with (HR 6.14, 95% CI 3.73-10.11, P<0.001), again with a sojourn time of 2 years.
The authors noted, however, that “differences owing to selection bias could exist among those who participated in screening vs those who did not, as well as with those who were diagnosed outside the screening program and/or were noncompliant.”
There were also 55 deaths from other causes, but no significant difference was seen between the interval-detected and screening-detected groups (HR 1.33, 95% CI 0.43-4.15).
Patients had an average age of 58 at the time of their diagnosis. Stage I cancers were more often diagnosed in the screen-detected compared with the interval-cancer group (63% vs 25%), while higher-stage cancers were more common in the interval-cancer group (stage II: 45% vs 29%; stage III: 24% vs 6%; stage IV: 6% vs 1%).
Niraula had no disclosures. One co-author received a Manitoba Medical Services Foundation Allen Rouse Basic Science Career Development Research Award.