[music]
Deborah Levine, MD Hi I'm Debbie Levine. I'm the Senior Deputy Editor for Radiology and I'm here doing a podcast
with Dr. Glen Lo who is a radiologist at Sir Charles Gairdner Hospital in Western Australia.
The title of his paper is "Evaluation of the Utility of Screening Mammography for High-Risk
Women Undergoing Screening Breast MRI."
This is going to be published in the October issue of Radiology and I'd like to mention
that the study was actually performed when Dr. Lo was in Toronto, Canada at the University
Health Network Mount Sinai Hospital and Women's College Hospital.
Sitting next to me I also have Dr. David Ballard who is our Olmsted Editorial Fellow.
He's here with us this week and he'll be asking some questions as well.
Dr. Lo, Dr. Ballard, welcome to the podcast.
Glen Lo, MBBS Hi.
D.L.
So Dr. Lo can you tell us a bit about your project and what you did and what you found?
G.L. Yeah sure.
So as a fellow in Canada we reviewed our high risk breast MRI screening program.
I think it was 2012/2013 two years consecutive and these were a mixed population of women
at high risk for breast cancer who were being screened with screening MRI as well as screening
mammography.
Our question was did the screening mammography add anything to the screening MRI.
So basically we retrospectively reviewed the data from our prospective screening program
which included a mixture of women who had the BRCA mutation, high risk family history
with the highest amount of lifetime risks, previous history of breast or ovarian cancer,
or a history or (inaudible) radiation.
And basically of all those studies that were done and the cancers that were diagnosed in
the group we found that there were 45 cancers, 43 of them were seen on MRI and 14 were seen
with both MRI and mammography, but mammography didn't see or detect the cancer that MRI
hadn't seen.
So the two cancers that were sort of missed were missed by both modalities.
So if we were thinking of having two screening tests mammography and MRI, the mammography
didn't add anything to just having the MRI.
D.L.
Excellent.
I have to ask, with only 45 cancers total, even though you had a very large screening
population, but was your study adequately powered to make strong policy statements like
you might not need mammography in these high risk women if you're having MRI?
G.L. Yeah I think that's a pretty bold conclusion that we did come to and we did suggest that
further studies and other institutions need to order their data as well to confirm if
that this is something that before a policy decision is made; as a policy change is a
big decision, so no I don't think it's going to be powered – we only have I think
1249 women included in the end and yes that wasn't a huge number of cancers, but I think
as a preliminary result it's something that warrants further investigation.
David Ballard, MD Can you comment on the economics of using MR as a primary imaging screening
modality in these high risk women such as the implemental cost, cost benefit ratio?
G.L. Obviously screening MRI is a very expensive test in terms of the cost of the examination
but also the time it takes the woman has to lie flat for an extended period of time.
It's obviously more expensive.
We didn't do a cost benefit analysis.
That's something to do next, but for the protocol that we used for the screening MRI
it's a full diagnostic protocol.
That is an expensive test.
I think to use MRI only in the future abbreviated MRI protocol so shorter screening protocols
would be needed to make a better cost benefit.
D.L.
Great yeah I think that brings up a good point because some of these limited sequence breast
MRIs have gadolinium some of them don't and one of the concerns that has been raised
recently is when we use contrast, gadolinium in particular, you get gadolinium deposition
in the brain we now know and so what about that concern using contrast repeatedly in
these relatively young women who are undergoing screening?
G.L.
That when they did the study hadn't really made it as a big concern in the general newspapers
and of public concern and it's not something that we actually investigated, but since having
performed this study and definitely since moving home to Australia, it's something
that the (inaudible) women that we definitely screened in Australia are concerned about
and we do experience women dropping out of MRI screening, first of all by a request can
we do the screening without gadolinium and we feel that's an incomplete examination
(inaudible) diffusion trying to develop non IV gadolinium and MRI protocols that they're
not available yet.
But yeah it is a consumer concern and it's something that we'll have to address.
D.B.
In your study there were no MR occult lesions.
Can you comment on what types of lesions MR is picking up and mammography is not?
G.L. Yeah so two of the cancers were MRI occult and so MRI only detected 43 of the 45 cancers.
One of the cancers was a 5 mm, I think I'll check the number, cancer that was incidentally
detected at a prophylactic mastectomy after having had the negative MR and then the other
occult cancer was a serendipitously biopsied cancer that was adjacent to a benign calcifying
process and the mammography had recorded the calcifications the lesion of calcification
wasn't cancer that in the biopsy sample a small cancer was found next to it.
In terms of what type of cancers the MRI had detected that the mammography hadn't, it
was a both DCIS and invasive so I think that's particularly interesting.
People usually consider that mammography detect calcifications representing DCIS that might
be MRI occult but of our DCIS 12 cases at mammography only detected one of those 12
and that was calcifications.
MRI detected all the other ones and that included up to 7 cm of DCIS so it's not like it's
a small number (inaudible) DCIS that wasn't seen in mammography that MRI did detect.
So it was both invasive and DCIS.
But in terms of the numbers MRI detected, compared to mammography and smaller cancers,
more node negative cancers, and I forget what the grades were it was getting a better screening
test because it was detecting cancers(inaudible.)
D.L.
Great.
One of the questions we always have when breast MRI is used for screening is the recall rate
and the concern of increasing the anxiety of women and bringing a lot of women back
for more diagnostic tests, you actually had a recall rate of 23% and in a screening population
that would be quite high.
What do you recommend we do in the future to lower that rate?
G.L. I think what's important to stress is that this is a retrospective study from
2012 to 2014, so this is when the screening MRI and reading was still relatively junior
in the department I think and they definitely changed the reading practices since I was
there.
We included BI-RADS 3 as a positive screening test.
So part of that really high recall rate was the fact that we had I think 10 set of MRI
studies where BI-RADS 3 read.
So the radiologist thought it was probably benign but with some early interval follow-up
imaging whether that was another MRI or a second ultrasound and then if that was negative
to go back to screening.
With experience this included both prevalent and incident rounds with experience and actually
doing this retrospective study we recognize cases that we can actually probably just say
these are actually probably benign, call them negative from the get go and improve that
so we do need to actually follow-up and see how practice has changed.
D.B.
In your study the exams weren't blinded to the reader, in other words if a patient
had a mammogram earlier they had access to that when interpreting the MR.
Do you have any data suggesting which test was most often first and how that could affect
the performance of the interpretation of the other test?
G.L.
We didn't actually collect that data unfortunately so I can't give you the exact numbers, but
what I can explain is that if the cancer was visible at mammography and MRI then both tests
would have been positive.
Consequently also if the MRI was performed first and the mammogram was read out after
the MRI was being reported or if it was by the same radiologist that the mammogram probably
would have – a subtle mammographic finding would have been read as positive given the
MRI finding, yet we didn't really see that in a number of positive mammograms.
So I think that this is a clinical practice study and it reflects real life screening,
it's how it happens.
Women might be scheduled for an MRI on the weekend on a Saturday and then come in on
Monday for their mammogram.
The studies might not be read by the same radiologist.
It's usually a chase to get to the MRIs on a Monday morning so different people might
have picked them up, but it's a potential confounder and it's a limitation in a retrospective
study, but I think it would probably have equally helped the sensitivity of both studies.
D.L.
One thing I was thinking about and this gets back to something we were earlier speaking
about the cost of doing MRI screening and would you be able to drop the mammography,
the BRCA women were obviously the highest risk group.
They comprised 8.4% of your population, but they also had over half of the cancers.
And so one of the things I was thinking is that since they're known to be such high
risk women should we be considering them completely separately from the other high risk groups
in your study when we're talking about potential policy changes of switching MR screening to
replace some aspects of mammography screening?
G.L. Yeah, I think that would probably be the group to start given that mammography
is less sensitive in that group already and our population is a heterogeneous group.
It would be useful to pick the group of women that would benefit most from MRI and benefit
least from mammography to drop the mammography because obviously it's the safer group to
go with.
My personal thought would be to go to biannual screening or perhaps do single view and that's
because I've returned to Australia where our screening mammography for the general
population is biannually anyway so I'm actually used to having a two-year mammography screening
in a general population.
Our high risk women have the same they do annual MRI and mammography.
I think it would be useful to pick your population and maybe be driven by what they want.
They're usually young women.
They usually don't want to have too much radiation if they can avoid it and if given
the choice the consumer might try that.
They can see what they want.
D.L.
You mention some differences in Australia compared to North America.
You have this interesting experience spanning the different continents, how is breast imaging
handled in Australia that's different from Canada and the United States?
G.L.
So our general population breast cancer screening is age 50 to 74 and every two years and it's
free.
The government pays for it.
Women can self-present from age 40 onwards but we only screen every two years.
So the mammography that does help us to detect changes because obviously it is a longer time
to see change for mammograms.
For the high risk population we do have similar screening protocols of annual mammography
with MRI, however our government funding only pays from age 30 to age 50.
It then becomes up to the woman to choose whether she privately wants to pay for that
to continue with MRI screening beyond age 50.
Currently where I work, I'm in the tertiary hospital centers we do offer tomosynthesis
and then sometimes adjunctive screening ultrasound that's actually evidence based given that
we might have more false-positives that might be less sensitive screening testing in that
population.
But it is certainly different in North America, in Toronto, it's annual screening both modalities
and they continue that.
D.L.
Terrific.
Well thank you so much for taking the time to speak with us today.
I really appreciate your time and I love your study.
G.L. Thank you very much.
For more infomation >> Used starcraft buses for sale | 2008 Ford E350 Non-CDL Wheelchair Shuttle Bus - Duration: 3:04. 
For more infomation >> Great .8K In Maple Grove Raising Money For Lupus Awareness - Duration: 2:57. 
For more infomation >> Twin Cities Live TCL at Mall of America for 25th birthday 8/11/17 - Duration: 58:06. 
For more infomation >> Journalists Say They Are Not to Blame for Police's Reputation - Duration: 0:53. 
For more infomation >> Official Photo ID Makeup (for bad quality pictures) - Duration: 5:58. 
Không có nhận xét nào:
Đăng nhận xét