FAST MRI Research Programme

(Diagnostic Yield study for Average MammOgraphic screeNing Density): A multicentre study offering women a FAST MRI scan in addition to their screening mammogram, to see if FAST MRI can find cancers missed by mammography

Aim:

To see if FAST MRI, a new imaging test, can detect cancers that have been missed by mammograms for women with average breast density at their first screening mammogram (age 50-52).

Background:

Finding breast cancers early saves lives. The NHS breast screening programme uses mammograms tofind breast cancers early but sometimes a mammogram misses a cancer which keeps growing until the woman finds it herself. MRI (Magnetic Resonance Imaging) detects aggressive cancers better than mammograms but is expensive and the NHS uses it to screen only women at very high risk. FAST MRI is a shorter MRI test which might benefit more women by finding more breast cancers earlier and saving more lives.

Every woman’s breasts are different. Breast composition affects how they look on mammograms. Women with denser breasts are more likely to have their cancers missed on mammograms because dense breast tissue can hide a cancer. Of four density categories (A,B,C and D), A is the least dense and D is the densest. At age 50-52, about 8 of 10 women will be in one of the middle categories, B or C, about half in each. We already know from previous research that FAST MRI can find cancers missed by mammograms in women with denser breasts (C and D). But women with average breast density (B) also have cancers missed, because mammograms are better at finding some types of cancer than others. We want to find out if FAST MRI can detect additional cancers for this group of women (category B). Women in category A are least likely to benefit as their mammograms show any cancers present more clearly and easily.

Design and methods used:

We will invite women aged 50-52, the age when women at average risk of breast cancer are first invited for an NHS screening mammogram. We will use computer software to measure women’s breast density from their mammogram and invite women with breast density in category B and a normal mammogram result to have a FAST MRI scan. 1,000 women will be scanned at 4 NHS sites, chosen for the ethnic diversity of the screened population and the experience the site has in working with FAST MRI. This choice of NHS sites will ensure our sample is representative of the UK. NHS professionals who have completed FAST MRI reader training will interpret the FAST MRI scans. We will count the total number of cancers detected by FAST MRI and record the types, aggressiveness and size of cancer found. We will count how many women need further tests but turn out not have cancer. We will ask women to share with us their experience of having a FAST MRI. The results will help us decide which women should be included in a future FAST MRI trial to measure if FAST MRI is clinically and cost effective for the NHS by finding breast cancers earlier and saving lives.

Patient and public involvement (PPI):

Patients and the public are integral in our work and we have ongoing support from:

1. Breast Cancer Unit Support Trust (BUST)
2. National Cancer Research Institute (NCRI) Breast Group
3. Independent Cancer Patients’ Voice (ICPV)

Our two lay research team members with personal experience of breast cancer and of breast screening will work with the wider PPI group to write public-facing documents so that participants and the public understand the research. They will also work to increase diversity, equality and inclusion in the PPI group.

Dissemination:

We will publish the results in academic journals and present at international meetings. Our PPI network will help us to share the results with charities and support groups locally and nationally.

Funded by the NIHR and MRC via the EME funding stream (NIHR 150502).

Project Details

Chief Investigator: Dr Lyn Jones

Co-Lead: Dr Rebecca Geach

Planned End Date: 01/11/2026

Local Ref: 5273

(OPtimisation of the FAST MRI protocol: an Evaluation of what makes a good breast MRI through detailed Analysis of scans from multiple NHS sites)

Background:

Finding breast cancer early saves lives. The NHS Breast Screening Programme (NHSBSP) uses mammograms to detect early breast cancers. However, not all cancers show on a mammogram so a cancer can be missed and continue to grow until the woman finds it herself. Magnetic Resonance Imaging (MRI) scans are better at detecting cancers than mammograms. However, MRI is expensive, and the NHS only uses it to screen women classed as high risk of developing breast cancer.

Recent studies have shown that using only part of the full breast MRI scan detects cancer equally well as the full scan, but is a much quicker scan with lower costs. This technique is called FAST MRI and has the potential to save more women’s lives by finding breast cancers earlier than a mammogram and providing value for money for the NHS. A group of research studies led by North Bristol NHS Trust aim to develop a better breast screening programme using FAST MRI for women who currently have mammograms, to screen for breast cancer.

How easy it is to see a breast cancer on an MRI scan depends on the scan quality and the technical details of the scan, known as the protocol. Quality control is therefore crucial for breast screening to optimise the detection of cancers.

This pilot study will develop a standardised and optimised protocol to be used in a separate multicentre trial of FAST MRI for women having their first screening mammogram (DYAMOND).

Aims:

1. Define the parameters that make a good quality FAST MRI scan.
2. Standardise the scan protocol across NHS sites to enable quality control during FAST MRI research trials and future clinical practice.

Design and Methods:

Our study will analyse breast MRI scans from different scanners across NHS sites within the FAST MRI Research Programme. Anonymised scans will be sent electronically to a panel of Breast Radiologists who will each score the scans for multiple aspects of scan quality. A team of Medical Physicists will also extract the numerically measurable aspects within each of the sites’ scan protocol and images. These two information sets, radiologists’ visual assessments and objectively measured values, will then be analysed to discover which settings make the optimal FAST MRI scan for each type of MRI scanner used. Site specific recommendations will be made to improve scan quality.

Results:

Study in progress.

Funded by Southmead Hospital Charity Research Fund.

Project Details

Principal Investigator: Dr Katherine Klimczak

Planned End Date: 31/12/2024

Local Ref: 5268

(EvaluatioN of an Artificial Intelligence (AI) Tool developed within and owned by the NHS to accurately measure mammographic breast Density): Selection for personalised screening with FAST MRI.

Background:

Finding breast cancer early saves lives. The NHS uses mammograms to try and detect early breast cancers. However, as mammograms do not show some cancers very well, a cancer can be missed and continue to grow until the woman finds it for herself. MRI (Magnetic Resonance Imaging) is a test that can find cancers better than mammograms, but it is expensive and so the NHS only uses it to screen women at very high risk of breast cancer. A quicker, shorter MRI test is now available called FAST MRI. Not only might this test benefit more women, it may also provide better value for money for the NHS to find breast cancers early and save lives. 

Every woman’s breasts are different. One way they differ is in a characteristic known as mammographic or breast density, which affects how they look on mammograms. Women with denser breasts can have their cancers missed on mammograms, as the dense normal tissue can hide the cancer. FAST MRI is better at finding these cancers.

To find out which women have dense breasts and could benefit from a FAST MRI, the mammograms need to be studied and measured. Currently, breast density is looked at and estimated by the radiologist but as each radiologist might view images slightly differently, results for breast density might not always be correct. There are now better systems to do this using (expensive) technology.

Aims:

To evaluate the accuracy and reliability of a breast density measurement tool. This will provide the National Breast Screening Programme (NHSBSP) with the ability to describe a woman’s breast density. If this tool is successful, it will further enable the North Bristol NHS Trust led FAST MRI research programme to develop a better breast screening programme.

Methods:

Our study uses an NHS developed and owned artificial intelligence (AI) software tool to automatically categorise the breast density. The cost to other NHS organisations will therefore be much lower when compared to commercial software. In this study we have tested our tool on anonymised mammograms, held in a research database called OPTIMAM, and compared the results to measurements made by commercial technology.

Our study looked at women aged 50-55, the age when women at average risk of breast cancer in the population are first invited to attend for an NHS screening mammogram.

Results:

The results of this study will be presented at the Symposium Mammographicum Conference, June 2023

Funded by Southmead Hospital Charity Research Fund.

Project Details
Principal Investigator: Dr Katherine Klimczak
Planned End Date: 01/04/2023
Local Ref: 5086

A study to improve FAST MRI interpretation training and to enhance Breast Clinician and Advanced Practitioner Radiographer understanding of breast MRI at multidisciplinary team (MDT) discussions.

Background:

FAST MRI (a shortened form of breast MRI) has been developed to address limitations of full protocol breast MRI (fpMRI), by shortening the time needed to acquire and to report the scan. FAST MRI diagnostic accuracy is similar to fpMRI and therefore has potential for wider use in screening but to roll it out on a larger scale more readers skilled at interpreting FAST MRI would be needed. Initial evaluation showed NHSBSP mammogram readers could be trained to interpret FAST MRI with a single day of structured training, but novice MRI readers were still learning at the end of the final assessment task and therefore it is likely that further training could further improve their performance. The current study aims to see if further training can enable novice MRI readers to match the performance of experienced breast MRI readers at FAST MRI interpretation. The improvement in novice reader performance during the training will be monitored to enable evaluation of their "learning curve” so that we can find out how much training mammogram readers need to enable them to interpret FAST MRI scans in clinical practice.

Methods:

Mammogram readers from seven NHS sites in England will undertake the developed North Bristol FAST MRI interpretation training programme. The final assessment task of the training programme has been updated for this study (since its use in the previous Multi Centre Reader Training Study) so that feedback (the true results of each FAST MRI scan) is given to the readers immediately after they have recorded their opinion about the scan. This modification to the assessment task is known as "formative assessment” in educational theory and has been shown in education research to be an effective teaching tool. Because the FAST MRI scans of the final assessment task are presented to each reader in a different random order, we will be able to map the learning curve of each reader as they complete the final (formative) assessment task. 

Results:

The results of this study will be presented at the Symposium Mammographicum Conference, June 2023

Funded through the National Breast Imaging Academy by Health Education England.

Project Details
Principal Investigator: Dr Liz O’Flynn
Planned End Date: 30/12/2022
Local Ref: 5041

Background:

To be able to get high quality FAST MRI images it is important that the correct protocol (information/sequences programmed into the scanner) is optimised. To do this test (phantom) breast need to be developed to trial the protocols to put into the MRI machine.  

Aims:

To develop breast test objects to ensure an optimum FAST MRI protocol (the information put into the MRI scanner to run the scan) and develop a quality assurance (QA) programme that can be used across different sites and MR scanner vendors.

Methods:

This phantom development work includes the design and construction of 2 magnetic resonance imaging (MRI) test objects that will be used for quality assurance (QA) tests of MR scanners at centres participating in the FAST MRI project.

One of these test objects will be used to assess the dynamic range of the dynamic contrast enhanced (DCE) sequence and the other the resolution in 3 dimensions. The MRI test objects will be trialled at NBT initially in order to finalise the design. 

Results:

The results of this study will be presented at the Symposium Mammographicum Conference, June 2023

Funded by the National Institute of Health Research (NIHR) Research for Patient Benefit funding stream.

Project Details
Principal Investigator: Dr Lyn Jones
Planned End Date: 30/10/2022
Local Ref: 4543

Background:

After the promising results of the single centre study, the FAST MRI training programme was awarded an NIHR grant to expand and develop the FAST MRI training further.

Aims:

The aim of this study was to refine and pilot a training programme for FAST MRI interpretation within the NHS Breast Screening Programme (NHSBSP) workforce, to support the delivery of a future multicentre study of FAST MRI versus mammogram for breast cancer screening.

• Produce an electronic version of a standard teaching tool and data collection tool working in close partnership with NHS based SciCom team (Workstream 1).
• Pilot the standard teaching tool across six NHS sites within the South West region, UK and collect data on the accuracy and speed of FAST MRI interpretation following training (Workstream 2).

Methods:

The aim was achieved in two work streams:

• Workstream 1: Production of electronic versions of a standard teaching tool and of a data collection tool. 
• Workstream 2: Pilot the standard teaching tool across six NHS sites (NHSBSP screening units) within the South West region of England (Truro, Plymouth, Taunton, Avon, Cheltenham and Swindon) and collect data on the accuracy and speed of FAST MRI interpretation following training. In addition, a Budget Impact Analysis to assess potential cost savings and affordability to the NHS to be assessed by the research team’s Health Economist as part of this workstream.
• Training: The teaching tool was used to train at least 12 readers from six centres to read FAST MRI. One day group training replaced the one-to-one training (used in the previous single centre study). This was made possible by the electronic teaching device.
• Follow-up interview: Study participants who gave consent to be contacted for future studies were invited to take part in a follow up interview to find out more information about how the readers felt about the study.

Results:

Jones LI, et al., Evaluating the effectiveness of abbreviated breast MRI (abMRI) interpretation training for mammogram readers: a multi-centre study assessing diagnostic performance, using an enriched dataset. Breast Cancer Research. 2022 Jul; 24(1):55.

Conclusions:

• NHSBSP mammogram readers who completed the FAST MRI training programme achieved diagnostic performance at FAST MRI interpretation (in the final assessment task of the training programme) within international benchmarks (standards) published for full protocol breast MRI.
• The single day of training was not enough to enable mammogram readers who were new to breast MRI to interpret FAST MRI quite as well as mammogram readers who were experienced in full protocol breast MRI interpretation.
• Mammogram readers who were new to breast MRI interpretation continued to improve at how well they could detect breast cancers from FAST MRI scans during the final assessment task of the training programme. This showed that they were still learning during the final task and means that their diagnostic performance might improve further if given further training.

Funded by the National Institute of Health Research (NIHR) Research for Patient Benefit funding stream.

Project Details
Principal Investigator: Dr Lyn Jones
Study Completion: 07/09/2020
Local Ref: 4543

Background:

Mammographic screening programmes result in both over diagnosis and under diagnosis of breast cancer. Under diagnosis leads to cancers presenting symptomatically between screening visits (interval cancers), and to cancers being detected by screening only once they have already reached more complex and life-threatening stages.Although MRI is the most sensitive method to detect breast cancer, currently only women classified as high risk (>30% lifetime risk) are offered screening MRI in the UK. However, in the future, personalised screening could enable larger numbers of women to be offered different screening regimes, each incorporating different imaging modalities, according to their level of risk.

Finding breast cancer early saves lives, and there is therefore a need to develop cost-effective imaging tests that will benefit women at risk of breast cancer by finding significant disease early. First post-contrast Acquisition SubtracTed (FAST) MRI is a type of abbreviated (shortened) breast MRI. FAST MRI is essentially as accurate at breast cancer detection as full protocol breast MRI, but is much faster to acquire and report. This technique might benefit more women than are currently offered screening with full protocol breast MRI. FAST MRI may be especially useful for women with dense breasts, since cancers obscured by dense tissue on mammograms are often visible on MRI

Aims:

The aim of this study was to explore whether NHS breast screening programme (NHSBSP) mammogram readers can learn to effectively interpret FAST MRI with less than one day’s additional training and to match the capabilities of expert breast MRI readers at this task in terms of accuracy and speed of interpretation.

Methods:

FAST MRI images were created by using previously acquired full protocol breast MRI scans. The anonymised images were reformatted and simplified into a form equivalent in its display to a FAST MRI. Using these FAST MRI images, training was offered to colleagues at Bristol Breast Care Centre.

Four consultant radiologists who were qualified to report full protocol breast MRI and four screening mammogram film readers who had not previously been trained to report MRI were trained to read FAST MRI. They were shown a set of training images with answers in a one to one session with the Chief Investigator, and the length of time taken to train each person was recorded.

Results:

The findings showed that the brief structured training carried out in the study enabled multi-professional mammogram readers to achieve similar accuracy at FAST MRI interpretation to that of the consultant radiologists experienced at breast MRI interpretation.

For more information about this study, published results can be found on the British Institute of Radiology website (main results), European Journal of Radiology website (training methodology) and the Pub Med website (review of published literature).

Funded by North Bristol NHS Trusts’ Research Capability Fund.

Project Details
Principal Investigator: Dr Lyn Jones
Study Completion: 26/02/2019
Local Ref: 4002