Several years ago, women's health vendor Hologic introduced the concept of 3D digital breast tomosynthesis, a new technology thought to revolutionize the breast imaging specialty by increasing the performance of standard 2D digital mammography. Interest from the health care community immediately sparked, with many experts indicating breast tomosynthesis could ultimately supplant digital mammography as the modality of choice for the screening and diagnosis breast cancer in any setting. Despite its promise, the years came and went, but still no tomosynthesis product. The lack of clinical data, a very complicated FDA reviewal process, unclear indications for use, and myriad other factors were, and still are, to blame. Now, it appears one very important and cumbersome barrier to tomosynethsis development is beginning to budge.
Last Friday, the FDA's Medical Devices Advisory Committee - a group consisting of clinicians, statisticians, patient advocates, and industry representatives - convened to advise the FDA on whether or not to proceed with approval for Hologic's Selenia Dimensions 3D tomosynthesis system. In a 12-0 vote, the panel overwhelmingly, and somewhat surprisingly, came out in favor of the FDA approving the device and moving forward in the pre-market approval (PMA) process. While many radiologists, technologists, administrators, and members of industry anxiously await the final decions from the FDA, last week's panel recommendation to proceed in the formal FDA review process marks a historic first step towards a definitive answer.
Recap of the FDA's Review of Digital Breast Tomosynthesis
My colleague Rachel Klein attended the FDA meeting, hearing expert testimony from numerous stakeholders in the breast imaging terrain. Over the course of the meeting, both Hologic and the FDA made presentations to the panel reviewing two Hologic-sponsored studies and one independent study on the efficacy of tomosynthesis compared to FFDM. At this point in time, available data supporting, or refuting, the value of digital breast tomosynthesis has been scarce, which served as a major impediment to a PMA review process first initiated in 2008, and then tabled in light of the need to accumulate new and more comprehensive data. As expected during the meeting last Friday, the Hologic-sponsored studies were highly criticized for their selection bias in the study pools. Patients included in both studies were selected based on their initial FFDM study and had a higher than normal recall rate (55% vs. 10%). While this would likely select against tomosynthesis rather than for it, panel members were concerned that these cases may have been "home runs" for tomosynthesis and artificially increased its sensitivity. In addition, such a high recall rate shows a group of patients quite different from the general screening population--tomosynthesis' target audience. Despite these concerns, panel members noted the significant clinical advantages of tomosynthesis over FFDM alone and lauded its ability to provide more and new information to radiologists.
Another cause of contention to date, panel members also expressed concern over radiologist training requirements for tomosynthesis. Two reader studies rather than one were conducted at the FDA's urging, as the results suggested the radiologists participating in the first study did not accurately adhere to the training provided. While concerns regarding appropriate training and use of the technology were valid for this meeting and particular technology, the concept resonates throughout the field of radiology, as deliberated by several panel members. Many difficult and somewhat philosophical questions were presented, including, "How does this community ensure quality in its reads?", "Whose responsibility is it to train radiologists?", and "What role should the manufacturer have in ensuring quality outcomes?". While no consensus was reached, this debate will likely reappear if and when tomosynthesis finally reaches market and is placed in clinical practice. The extent of additional training required for optimal use of the technology will likely be a significant factor inflecting future adoption.
Finally, and as hot a topic as any in the field of diagnostic imaging, radiation dose concerns resonated throughout the discussion as tomosynthesis delivers a higher dose (nearly two times) than that of FFDM based upon available data. While both exams are well below the levels required of the Mammography Quality Standards Act (MQSA), risk-benefit was discussed to determine if the added risk of cancer due to radiation dose would be outweighed by the identification of novel and invasive cancers. Various models were presented to demonstrate the relatively small difference in radiation dose was overcome by the reduced recall rate and subsequently fewer exams for the patient. In this regard, and in light of a time when radiation dose management will weigh heavily upon the regulation of new imaging technologies moving forward, this viewpoint from the panel represents a significant step forward for radiation-emitting medical imaging products.
Lastly, considerable debate existed as to tomosynthesis' role as a first-line screening modality or second-line diagnostic tool or both, and the final decision on this will not likely be known until it is ushered further along in the FDA's review process and the technology is implemented in clinical practice. In either case, tomosynthesis was shown to have made vast improvements in detecting mammography-occult cancers as documented by both the vendor and the FDA.
What The Future Holds for Digital Breast Tomosynthesis
Given the possible outcomes from this early meeting in the FDA review process, it is safe to say 3D digital breast tomosynthesis passed a significant test and is moving in the right direction for approval. When available, it can be expected hospitals and imaging centers of all types will be lining up to make sizeable investments in the technology in light of the many benefits afforded by the novel imaging system. However, the extent of hospital and imaging center adoption of tomosynthesis still remains uncertain as its premium price (upwards of $400,000 for a tomosynthesis-capable digital mammography system) and unknown reimbursement may make this a hard sell for many institutions, especially those in smaller markets or hospitals without an active research program. Nonetheless, estimates for full FDA approval are still in the late 2011 range though the unanimous panel vote may accelerate this process. For organizations considering new mammography technology or tomosynthesis, here are 5 thoughts to consider before making the investment:
- At this time, data supporting tomosynthesis use is still in its infancy. Is your organization prepared and willing to accept some risk in purchasing premium, tomosynthesis-capable mammography systems in the absence of robust clinical evidence supporting its use?
In the current health care environment, securing premium reimbursement for tomosynthesis from Medicare and private payers may be challenging and several years away. What other factors does your organization need to consider in making a business case for tomosynthesis in the absence of incremental reimbursement?
- Tomosynthesis has the potential to dramatically change the perception of the breast imaging experience for patients. How might an early adopter best be able to leverage this technology to strategically grow its breast imaging program, or alternatively, how can an organization without tomosynthesis better compete in a market with tomosynthesis products?
- To fully benefit from tomosynthesis, radiologists will need to be fully supportive of its benefits at the expense of likely decreased productivity during the early phases of adoption. As tomosynthesis is a novel breast imaging device, will your radiologists be amenable to having longer interpretation times without an guarantee of incremental professional reimbursement?
- Early results indicate tomosynthesis has the capability to significantly reduce recall rates, suggesting less downstream imaging opportunity. As downstream imaging and interventional procedures are business drivers for breast imaging, how do you balance the clinical benefits of the system versus the potential business implications of decreased downstream opportunity, especially in light of the new imperatives set forth in the post-health care reform era?