Diagnostic accuracy, clinical effectiveness and budget impact of screening BRCA1/2 mutation carriers by MRI. A health technology assessment
Health technology assessment
|Published
The Commission forum for the Regional Health Authorities in the National System for Managed Introduction of New Health Technologies within the Specialist Health Service, commissioned a health technology assessment (HTA) of the diagnostic accuracy, clinical effectiveness and budget impact of breast cancer screening using magnetic resonance imaging (MRI) in combination with mammography, for women with BRCA1/2 genetic mutations.
Key message
BRCA1 and BRCA2 genetic mutations are important risk factors for breast and ovarian cancer etiology. Women carrying one of these mutations have a high life-time risk of developing breast or ovarian cancer. The current screening strategy in Norway for women with BRCA 1/2 mutations is annual magnetic resonance imaging (MRI) and mammography from the age of 25 to 75.
The key messages in this report can be summarized as follows:
- More true positive breast cancers will be identified when MRI is used in addition to mammography (higher sensitivity), but at a cost of more false positives (lower specificity)
- We were not able to detect a decrease in breast cancer mortality when annual MRI was added to an annual mammography-screening program. The certainty of the evidence was considered very low, due to imprecision and very wide confidence interval.
- Annual savings would be approximately 6.2 million NOK if annual MRI screening of BRCA1 and BRCA2 mutation carriers was removed from the current practice and replace with mammography alone
- Annual savings would be approximately 2.5 million NOK if annual MRI screening was only offered to for BRCA1 and BRCA2 mutation carriers between 25 to 50 years of age, followed by annual mammography alone up to the age of 70
Summary
Background
The Commission forum for the Regional Health Authorities in the National System for Managed Introduction of New Health Technologies within the Specialist Health Service, commissioned a health technology assessment (HTA) of the diagnostic accuracy, clinical effectiveness and budget impact of breast cancer screening using magnetic resonance imaging (MRI) in combination with mammography, for women with BRCA1/2 genetic mutations.
BRCA1 and BRCA2 genetic mutations are important risk factors for breast and ovarian cancer. These mutations are not very common in the general population, but women carrying one of these mutations have a high life-time risk of developing breast or ovarian cancer.
Current strategies for early detection and risk reduction of breast cancer are screening using mammography and/or magnetic resonance imaging (MRI), prophylactic mastectomy and oophorectomy. While Norwegian national clinical guidelines describe prophylactic mastectomy as the best option to reduce breast cancer risk, many women in Norway prefer to have annual breast cancer screening using both mammography and MRI. It is important to assess which preventive measure is the most effective and cost-effective. This information would help determine appropriate methods for preventing and treating breast cancer among women with high breast cancer risk, and would help these women make personal choices.
Objective
The objective of this health technology assessment is to examine the diagnostic accuracy, clinical effectiveness and budget impact of breast cancer screening using magnetic resonance imaging (MRI) in combination with mammography versus mammography alone in women with BRCA1 or BRCA2 genetic mutations.
Method
We conducted systematic literature searches for systematic reviews and for primary studies. Individual search strategies were designed for each database. Search strategies were based on a combination of subject headings and text words for BRCA, MRI and breast cancer. Two reviewers independently screened all identified records and critically appraised the selected publications. The outcomes of interest were cancer mortality and breast cancer mortality.
Quantitative data for the included studies were combined for meta-analysis using Review Manager. We report the diagnostic accuracy and used a random effects model to estimate odds ratios or risk ratios and corresponding 95 % confidence intervals. We used the GRADE tool (Grading of Recommendations Assessment Development and Evaluations) to assess the certainty of the evidence.
Health economic evaluation
In current practice, women are screened annually using MRI in combination with mammography from 25 to 75 years old. In this HTA, we compare this practice with two alternative strategies:
1) An annual screening with mammography only from age 25 to 70
2) A combination of annual MRI and mammography from age 25 to 50, followed by annual mammography alone up to age 70.
Results
The literature search was completed in December 2016, and resulted in five included references: one systematic review and four clinical studies.
Diagnostic accuracy
The combination of MRI and mammography was associated with higher sensitivity and lower specificity than mammography only. This means that more true positives will be identified (13 and 12 more per 1000 per year for BRCA1 and BRCA2, respectively) at the cost of more false positives (140 and 118 more per 1000 per year for BRCA1 and BRCA2 carriers, respectively). The certainty of the evidence was considered high.
Clinical effectiveness
We were not able to detect a reduction in breast cancer mortality when adding MRI to an annual mammography screening program compared to only mammography (RR 0.64; 95% CI 0.16-2.54). The certainty of the evidence was considered very low, due to imprecision and very wide confidence interval. The mortality of women in the non-screening group was significantly higher than for women who attended a screening program with either mammography alone or a combination of MRI and mammography.
Economical outcomes
The current breast screening strategy for BRCA1 and BRCA2 carriers is annual MRI and mammography from the age of 25 to 75. An alternative strategy examined in this report involves annual screening with mammography as currently prescribed, but MRI only from age 25 to 50, thus saving approximately 1.4 million NOK for BRCA 1 carriers and 1.1 million NOK for BRCA2 carriers each year. A further reduction in cost will be achieved by introducing a screening program involving only annual mammography compared to the current practice resulting in 6,2 million NOK annual savings for both BRCA1 and BRCA2 mutation carriers.
Discussion
The ideal way to investigate the effect of screening interventions is prospective studies starting follow-up when women are identified as mutation carriers and to follow them until potential breast cancer development. In our evaluation, only one study fulfilled this criterion. In the other studies, women were enrolled at the time of diagnosis and were retrospectively divided in groups depending on which screening regime they had been following.
Potential risk of radiation-induced breast cancer is highly relevant when choosing a screening modality for identifying breast cancers, in particular in young women carrying a mutation. However, we have not considered this in the present report.
Conclusion
Higher sensitivity but lower specificity are obtained when MRI and mammography are used in combination compared to mammography only for detection of breast cancers in BRCA1 and BRCA2 mutation carriers. Therefore, by the combined screening, more true positives will be found, but also more false positives. Adding MRI to an annual mammography-screening program has not shown to statistically significant reduce breast cancer mortality among women with hereditary breast and ovary cancer generally, or BRCA1 and BRCA2 mutations specifically, compared to mammography screening alone. The results suggests that if MRI is removed from the current Norwegian screening strategy, the consequence would be a reduction in MRI screening-related costs. Future studies should have longer follow-up and report the association between detected breast cancer, stage distribution at diagnosis and treatment costs.
