Bend Memorial Clinic introduces 3-D mammography

But some experts wonder if technology is necessary

By Markian Hawryluk / The Bulletin / @markianhawryluk


Published Sep 2, 2014 at 12:01AM / Updated Sep 2, 2014 at 09:14AM

Bend Memorial Clinic is bringing 3-D mammography to Central Oregon, offering women access to a technology that many believe will significantly improve breast cancer screening. Initial studies suggest the new imaging technique could identify more invasive cancers while reducing the rate of false positives. But many experts warn that the question is far from resolved.

Where traditional mammography produces two views of each breast from a stationary X-ray tube, 3-D mammography, formally known as tomosynthesis, takes a series of X-ray images as the tube arcs around the breast. A computer then reconstructs those images into a series of 1-millimeter-thick slices. The radiologist can then scroll through anywhere from 100 to 300 cross-sections to get a better idea of the internal anatomy of the breast.

That helps to reduce some of the major weaknesses of standard mammography, where overlapping tissue can mimic cancer and dense breast tissue can mask problems.

Mammography recommendations have been controversial in recent years as experts debate how to best balance the benefits of finding cancers before they have a chance to spread against the harm of finding and treating anomalies in the breast that will never pose a risk to a woman’s life. Tests are evaluated based on their likelihood to catch cancers when they exist, known as sensitivity, and the likelihood that something identified as cancer actually is cancer, called specificity.

“Usually there’s a trade-off between sensitivity and specificity. There’s no free lunch,” said Dr. Kimberly Ray, assistant professor of clinical radiology at the University of California, San Francisco. “But this technology seems to be improving both measures, so that’s really very promising.”

A study of 3-D mammography, published in the Journal of the American Medical Association in June, examined more then 450,000 screening records involving some 140 radiologists at both large academic medical centers and smaller community clinics throughout the U.S. They compared results from before and after clinics added 3-D machines.

Where 2-D mammography found nearly three invasive cancers for every 1,000 women screened, doing both 2-D and 3-D screening found four cancers. Moreover, there was no difference in the amount of ductal carcinoma in situ, or DCIS, found. DCIS is a precursor to invasive cancer that many critics say leads to overdiagnosis and overtreatment.

Meanwhile, the number of women called back for follow-up dropped from 107 per 1,000 women with 2-D mammography alone to 91 women with 3-D scans.

Finding more cancers with fewer call-backs suggests tomosynthesis is better at predicting who has cancer than traditional mammography. With 2-D mammograms, 4.3 percent of women called back turned out to have cancer, while with tomosynthesis, 6.4 percent did.

“It goes up by a chunk,” said Dr. Jamie Ockner, a radiologist with Medford Radiology Associates working at BMC. “It’s not just incremental — a 40 percent increase in detection of invasive cancers — and nobody is arguing whether those need to be treated.”

Questions remain

While the study has many excited about 3-D mammography, some experts are pumping the brakes.

“Unfortunately, it was a before-after study that cannot reliably attribute a small effect, like the small reduction in false alarms — to the addition of 3-D mammography,” said Dr. Gil Welch, professor of medicine at Dartmouth Institute for Health Policy & Clinical Practice. “Most importantly, the study can’t tell women what they really want to know: Is diagnosing 40 percent more women with breast cancer good or bad for their health?”

Welch, who authored the book “Overdiagnosed,” has been a frequent critic of cancer-screening programs, which he says are overstating their benefit in protecting patients while minimizing the potential harm.

Without long-term data, doctors can’t know whether the technology will ultimately save lives. Radiologists say one of the strengths of 3-D mammography is that it allows them to rule out cancers on the screen, rather than bringing women back for more testing. But there is no way to know whether 3-D mammography in the JAMA study overlooked some cancers until those women come back for the next round or two of screening.

“The sad thing is the study is likely to be overhyped,” Welch said, “leading every hospital in the country to feel pressured to buy a new piece of expensive equipment and, to recoup their investment, pressure women to use it.”

Numerous press reports have cited the price for the 3-D mammography machine made by Bedford, Massachusetts-based Hologic at about $500,000. That’s two to three times the price of a standard 2-D mammography machine, but not close to the multmillion-dollar price tags for CT scanners or surgical robots.

Still, BMC had to make expensive upgrades in its computer systems and storage capacity to be able to handle the larger files associated with 3-D mammograms.

“The machine is just-getting-warmed-up money,” Ockner said.

It’s still unclear whether 3-D mammography will increase costs to the health care system. Once cost analyses are done, they may find that the higher price is recouped by calling fewer women back for screening and catching more invasive cancers earlier when they can be treated at a lower cost.

The bigger immediate financial issue for clinics adding the machines is that most public and private insurance companies aren’t paying extra for tomosynthesis.

Regence Blue Cross/Blue Shield of Oregon, for example, indicated in July it would not pay for 3-D mammograms, and the Medicare program last year ruled that tomosynthesis would be considered part of digital mammography and not eligible for additional payment. The Centers for Medicare & Medicaid Services is expected to issue another decision in October whether to reimburse clinics for tomosynthesis and whether to pay radiologists more for the extra work involved in reading 3-D mammograms.

In the meantime, many clinics that add the 3-D technology may have to do traditional digital mammograms as well to get at least some payment.

“If they don’t do the traditional mammogram in conjunction with that, they’re going to be doing everything for free,” Ockner said. “And they’re not going to take that hit.”

That also means patients will be exposed to double the radiation. The combined dose is still lower than the FDA limit for mammograms and lower than what women experienced with older film mammography.

Radiologists still need the 2-D view of a traditional mammograms, but now that view can be created from the 3-D scans using computer software. If the benefits of the initial studies hold up and the billing issues are resolved, that could mean women would get better screening with little trade-off.

Some clinics have been charging patients an extra $50 to $100 to offset the cost of 3-D mammograms, although Ockner said BMC will not do so.

“There is going to be no charge to the patient for this,” he said. “This is BMC on a mission and to me, this is good patient care and it’s the right thing to do.”

BMC will have the first 3-D machine in Central Oregon. There are some 1,100 units in place nationwide, and 10 others at sites in Oregon, all in the northwest corner of the state. Legacy Health has three of those units, paid for by the hospital’s charitable foundation. That has allowed Legacy to provide 3-D mammograms since 2012 without charging patients extra.

“It’s asking a lot to do what we’re doing, potentially doing a lot of it for free with the expectation that the insurance companies will eventually pay for it,” said Dr. Sam Gruner, a radiologist at Legacy Breast Health Center–Good Samaritan in Portland. “The insurance companies … are happy to let you do that and would like to define that as the way it’s going to be.”

Gruner recently compared Legacy’s last eight months of 2-D mammograms with eight months of 3-D, finding improvements in cancer detection similar to those found in the JAMA study with only slightly higher call-back rates.

“It’s a meaningful improvement,” he said. “It is a lot bigger improvement than going from (film) to digital.”

Shifting landscape

It’s still unclear whether 3-D mammography might benefit some groups of women more than others. Many have speculated that tomosynthesis could eliminate the need for follow-up testing for women with dense breasts. According to one study, 76 percent of all missed cancers occur in dense breasts.

Where in an ideal mammogram, the background of the breast is black, in dense breast tissue, it can appear white.

“Then you’re trying to find a white blob on a white background,” Gruner said.

Many states, including Oregon, have passed laws requiring doctors to notify women with dense breasts with specific language that suggests the possible need for follow-up imaging, such as MRI or ultrasound. Following up with ultrasounds finds cancers that mammograms miss, about three per 1,000 women, but with a high number of false positives.

“For all but the densest of dense breasts, you could make a case for 3-D being all you need to do,” Gruner said. “Whereas ultrasound buys you those three extra cancers with a lot of additional false alarms, 3-D buys you additional cancers without a lot of false alarms.”

BMC was still installing its machine last week, and it expected to get all its certifications in place by Oct. 1. Officials from Cascade Medical Imaging, a joint venture between St. Charles Health System and Central Oregon Radiology Associates, indicated they, too, have been reviewing the evidence for 3-D mammography and that it is in the group’s strategic plan for the near future.

“Our plan is to serve the region by bringing this technology to both our Bend and Redmond Women’s Imaging Centers,” Dave Magness, the group’s director of imaging wrote in an email to The Bulletin.

That does raise the concern of whether tomosynthesis will spur a similar arms race for mammography as for CT scanners or MRIs. Over the past two decades, competing hospitals and health centers have tried to one-up the competition by adding increasingly powerful scanners and marketing them heavily to consumers. But studies showed the more imaging equipment was added, the more scans patients received as institutions tried to offset their costs.

Moreover, new technology often proves irresistible for patients, who equate modern technology with better care.

“The latest and greatest may not always be the best,” said Dr. Linyee Chang, a radiation oncologist and medical director of the St. Charles Cancer Center in Bend. “In my field, we’ve jumped on the bandwagon a little bit too soon just to find out five years down the line it really didn’t make things any better.”

It’s too early to tell whether 3-D mammography could improve cancer screening to the point of changing screening recommendations. The U.S. Preventive Health Services Task Force is reviewing its guidelines and in an email to The Bulletin confirmed that the studies of 3-D mammography would be part of its review. The task force in 2012 recommended against screening for women in their 40s, concluding the risk of false positives and the resulting harms from unnecessary treatment outweighed the few cancers that would be detected in younger women. If it pans out, 3-D mammography could cut the risk of harms but not at the expense of missing some cancers. And that could change the screening calculus.

“I think it’s going to make a difference but there’s no proof of that, except for the current standards that if you find it earlier, it’s better for the patient,” Ockner said. “This is the future of mammography. This will make every single radiologist a better radiologist.”

—Reporter: 541-617-7814, mhawryluk@bendbulletin.com

Editor’s note: This article has been corrected. In an earlier version of the story, Ockner’s description of the danger of radiation was unclear. The Bulletin regrets the error.