Limiting breast cancer screening to women at higher levels of risk for the disease — and not offering it to women at lower levels — could improve cost-effectiveness, reduce overdiagnosis, and maintain the benefits of screening, conclude researchers from the United Kingdom.
The results are published online July 5 in JAMA Oncology.
“The age-based or ‘one-size-fits-all’ breast screening approach does not take into account the individual variation in risk,” write the authors of the new study, led by Nora Pashayan, MD, PhD, from the University College London, United Kingdom.
The researchers simulated three hypothetical cohorts of 50-year-old women who were free of breast cancer at enrollment and then followed up for 35 years. Each cohort had 360,000-plus women.
The first cohort received no screening. The second cohort was offered breast screening mammography at age 50 years and every 3 years thereafter until age 69 years (ie, the current standard in the United Kingdom).
In the third cohort, risk estimation with genomic testing was simulated. The women were then offered screening per the current standard in the United Kingdom — but only if their risk scores were above a threshold risk (which varied in the study).
The researchers report that, in general, as the risk threshold for screening was lowered, the incremental cost of the program increased linearly compared with no screening. In other words, the more women who are screened, the higher the cost of the screening. Furthermore, no additional QALYs were gained below the 35th percentile risk threshold. Thus, after screening the top 65th percentile, there was no gain in QALYs to be had among the remaining, lower risk women.
The team also uncovered a number of sweet spots where limiting the number of women who are screened resulted in a variety of acceptable outcomes.
For example, in the risk-stratified cohort, the screening threshold at the 70th percentile (which would exclude the 30th percentile and lower from screening) had the highest net monetary benefit, at a willingness to pay of £20,000 (US $26,800) per QALY gained, with a 72% probability of being cost-effective.
They also report that, compared with age-based screening (ie, the current UK standard), risk-stratified screening at the 70th percentile risk threshold or less would cost £537,985 (US $720,900) less over the study period and have 71.4% fewer overdiagnoses but would avert 9.6% fewer breast cancer deaths.
As a comparison, risk-stratified screening at the 32nd percentile risk threshold or less would cost £20,066 (US $26,888) less, have 26.7% fewer overdiagnoses, and avert 2.9% fewer breast cancer deaths.
In other words, there is still a bit of a price to be paid — fewer breast cancer deaths prevented — for the cost savings and the reduction of the harm of overdiagnosis at various threshold cutoff points.
Genomic Testing a Stumbling Block?
Writing in the accompanying editorial, Megan Roberts, PhD, from the National Cancer Institute, Bethesda, Maryland, comments that the study provides evidence for “a promising approach to precision screening in breast cancer.”
“Multilevel barriers hinder genomic testing adoption and uptake,” she writes.
The barriers include costs; problems with collecting genomic information (especially with suboptimal electronic health records); low clinician and patient knowledge; and ethical, legal, and social implications associated with genetic testing, such as consent, confidentiality, and privacy and reporting incidental findings.
In comments to Medscape Medical News, lead study author Pashayan responded to these concerns one by one.
About cost: “Although genotyping adds cost, there are saving from less screening, less treatment of overdiagnosed cancers, and less investigation (biopsy) of false-positive screens,” she said.
About collecting and storing results: “Neither the saliva or blood samples nor the genotyping profiles need to be stored; rather, what is needed is the risk score,” she said.
About low levels of knowledge among patients/doctors: “This is not a barrier. Rather, it is work that needs to be done on how to communicate risk and screening recommendation,” she said, adding that genomics are now in the medical curriculum.
About ethical, legal, and social implications (ELSI): “A large body of research is being done on ELSI to make sure insurance and employability won’t be affected. To address these issues, we are working with international groups like the PHG Foundation of Cambridge; Professor Bartha Maria Knoppers and her group at McGill University in Canada; and Professor Inez de Beaufort and her group at Erasmus Medical Center in the Netherlands.”
In an online podcast accompanying the study, Pashayan said that she and her team “expect to see similar patterns” in terms of cost-effectiveness and benefit-harm ratios in studies from other countries.
But she also acknowledged that the addition of genomic testing complicates the screening process.
“A risk-based approach is much more complex,” she commented. Among other things, the risk threshold for a population must be established. And the public and health professionals must be educated, she said.
A risk-based approach is much more complex.
Pashayan said major studies are underway globally to evaluate a risk-based approach that exceeds only one variable — age.
“Before we move from ‘one-size-fits-all’ to risk-based screening, we need evidence that the risk-based approach will do more good than harm, would be value for money…and that it would be acceptable to the users and the providers, accessible to all, and feasible to run,” she said.
With international collaboration, studies are underway to address those issues, such as the Canadian PERSPECTIVE I & I (Personalized Risk Assessment for Prevention and Early Detection of Breast Cancer: Integration and Implementation) study.
Editorialist Roberts points out that recommendations in the United States to change mammography screening standards were “controversial among clinicians and patients alike” and led to “confusion” among the public.
If the model proposed by the British research team were adopted in the United States, screening would have to be “de-implemented” among low-risk women, she contends. More research is needed about how this would all be managed by healthcare systems, clinicians, and women, she notes.
Roberts concludes with a call for more research and careful deliberation: “As prevention becomes increasingly more precise, we must address these issues such that the promise of precision screening can be achieved.”