Mammogram Study: Findings And Reaction

By Chris Earl, Reporter

CEDAR RAPIDS, Iowa - Amid the release of a controversial 25-year study of nearly 90,000 Canadian women and the effectiveness of mammograms, what is the correct answer for 2014?

This depends on who is doing the talking.

On Tuesday, the British Medical Journal released the findings of a study from the Canadian National Breast Screening Study.

The synopsis: Starting in 1980 and ending in 1985, the study tracks 89,935 Canadian women from six provinces between the ages of 40 and 59. About half were given five annual mammography screens. The remaining women received a single examination and what is called "usual care".

The findings:

44,925 With Annual Mammography Screenings: 666 invasive breast cancers diagnosed from 1980-85.

44,910 With The Single Examination/Usual Care: 524 invasive breast cancers diagnosed from 1980-85.

During the initial five-year period and the 25-year follow-up period from 1985 to 2010:

Mammography Screening Group: 3,250 diagnosed with breast cancer.

Single Examination/Usual Care (Control) Group: 3,133 diagnosed with breast cancer.

The study also contends 106 cancer cases were observed in the mammography screening group that were attributable to over-diagnosis.

Conclusion: The CNBSS claims that annual mammograms in women between ages 40-59 did not reduce mortality from breast cancer beyond that of a physical examination/usual care.

Here is the full report:
http://www.bmj.com/content/348/bmj.g366

* * *

Behind all of the numbers are people.

Liz Decious, 44, of Lisbon, was in a doctor's office nine years ago when her physician made a suggestion.

"She always had her patients get a baseline at (age) 35 so they would have something to compare to in the future," said Decious.

A mother of two young children at the time, Decious was startled to learn she needed a biopsy. The procedure revealed a cancer at Stage 1 in her breast. She endured radiation and was able to avoid chemotherapy.

"I was lucky that I didn't have to go through all of that," said Decious.

Yet, when she learned of this report, Decious admitted she has a difficult time with it.

"When I read about the study, it frustrated me because there are so many women out there who are afraid to go get a mammogram and a study like this would probably deter them," said Decious.

* * *

We reached out to medical experts here in Eastern Iowa on this study. Their extended statements are below, whether by television interview or by e-mail correspondence:

Dr. Arnold Honick, a Cedar Rapids radiologist with UnityPoint Health St. Luke's Hospital:

"This particular study has been discounted in the past and I don't recall exactly the time as such. The concerns of the study are several fold. One is quality. Quality of the mammography that was performed and there's a variety of findings that point towards that. If you can't perform a good exam, you're not going to find the breast cancers.

"The design of the study of what's called a randomized control study. This really did not randomize patients appropriately – you're not supposed to know patients that go into each arm exactly, whether they have a breast mass or not. These patients had breast exams prior and were sent into the screening exam arm, which, obviously, dilutes the findings.

"Taken as a whole, the strong statements by the author that you basically need to look at things and not do mammography and we can get rid of some screening tests that are out there and I think that is totally false.

"Whether somebody wants to initiate a look on how to improve what you're doing for screening, that makes a lot of sense. There are quality measures going on for the technology and how it's interpreted that will allow us to detect breast cancers earlier, sooner and save additional lives.

"The other thing we talked about was the continued improvement and treatment. That is something that will improve patients morbidity, how they do with their disease and their mortality. That's going to continue to improve and may diminish the improvement, somewhat, in early detection.

"Initially, mammography was done on a table top like you take an x-ray of your arm. They've made some changes to get that. If you look at those mammograms, they're very difficult to interpret akin to what's proposed as a deficit for this study and in the 1980's, the mammograms that were performed look much different than they do now. There's a marked improvement of us having the ability to see through denser tissue, to see things when they're much smaller, being able to evaluate those exams. That has been a rapid improvement.

"There's continued improvement in the way mammograms are done. There's lower dose where the exposure rate is different. Down the line is tomosynthesis, which will allow us to see through denser breast tissue and be able to detect breast cancers early so there's a continued effort to improve this screening test which, in no way, shape or form, should be abandoned.

"There are enough difficulties with this study to discount the majority of the findings. There are plenty of studies and information that show an improved mortality rate in the early detection of breast cancer with mammography. Now the degree may differ among studies but, overall, it's the best tool we have for screening.

"What may or may not be coming down the line is some alteration in when you should start and how often but, right now, at age 40 and on, at least an annual mammogram is recommended by people who do that sort of thing. Whether it's decided upon that it's every other year or not, it's still not going to discount the utility of doing the screening mammogram."

* * *

Dr. Carol Scott-Conner, endocrine and breast oncology surgeon with the University of Iowa Hospitals and Clinics:

"As our treatment of breast cancer has become better and better, the advantage of "catching it early" has become, perhaps, less and less important. Furthermore, in our zeal to "catch it early" we may be picking up some women with cancers that were not destined to cause them any problems. This is what this study suggests. It's an extremely well-designed study with very long follow up and a large number of women. It adds support to the growing sense that screening mammography should be a individual decision in "young" women (certainly the 40-50 year old group, and maybe even, as this study suggests, some 50-59 year olds as well). Women in this age group should definitely discuss this with their physician, and individual risk factors (such as a strong family history) should be taken into account - the study did not address any high-risk subsets.

Women age 60 and above should continue with annual mammograms as per the current guidelines.

Other screening modalities are becoming available - for example, 3-D screening ultrasound. These may prove to be attractive alternatives in younger women.

Too many women still die from breast cancer even with early detection. More research in basic mechanisms of disease will, we all hope, provide better and easier cures for more women.

* * *

Dr. Laurie Fajardo, radiologist with the University of Iowa Hospitals and Clinics:

"This follow-up report, from the Canadian Breast Cancer Screening Trial – conducted 25 yrs ago – will refuel the original criticisms and issues pertaining to the trial and the quality of the mammography performed.

"The original trial had concerns that patients were not randomly entered into the mammography versus clinical breast exam only arms. Screening mammography is performed on healthy women without any breast exam abnormalities or symptoms. This trial was shown to include women with breast lumps – who are not candidates for screening. More women with abnormal exams and lumps were entered into the mammography arm of this trial. Women with these types of cancers have a higher probably of dying versus those who are asymptomatic and have a normal breast exam.

"External experts who reviewed the quality of the mammograms performed in this trial (which were film mammograms, not modern digital mammograms), cited many concerns regarding poor quality.

"The sensitivity statistic calculated for mammography in this study was only ~ 32% - which is extremely low. A minimum sensitivity in any program should be at least twice as high as this (and even higher). Any screening test with a sensitivity at this level should not be employed as this is unacceptable. So – not surprising the longer term follow-up data are not favorable for mammography.

We know that overdiagnosis of cancer by mammography is a component of screening programs, particularly for DCIS. However, the technology for breast imaging continues to improve and new screening technology, such as digital breast tomosynthesis has had very positive evaluations in the last 3 years. Fewer normal women (without cancer) are "recalled" by tomosynthesis, yet the technology detects more cancers. The additional cancers detected are invasive and higher pathologic grade – which are important cancers to detect. Tomosynthesis has not been shown to detect more DCIS than conventional digital mammography, which may reduce the overdiagnosis problem."
facebook twitter email alerts you tube hooplanow