December 06, 2018

Landmark Breast DBT Trial Seeks More Study Sites

Once every decade or so, radiology needs a statistically valid, multi-center, randomized clinical trial to answer a critical scientific question about a potentially lifesaving screening procedure.

The Tomosynthesis Mammographic Imaging Screening Trial (TMIST) is the most recent randomized trial to attempt to make such a significant advancement to diagnostic imaging science. Under the direction of Etta Pisano, MD, FACR, Chief Research Officer of the American College of Radiology, TMIST will determine whether digital breast tomosynthesis (DBT) or digital mammography (DM) is the best imaging-based screening test for finding the most aggressive breast cancers that pose the greatest mortality threat to women.

Initiated in July 2017, TMIST is recruiting approximately 165,000 asymptomatic women, ages 45 to 75, to be randomized to receive either screening DBT or DM. Annual exams for five years will be performed on most women in the trial and all subjects with specific known risk factors. Biennial screening performed three times over a five-year period will be reserved for post-menopausal women without those known risk factors.

Researchers will also collect patient demographic and health information, blood and buccal (cheek) smears, and if the women undergo breast biopsies, pathology samples from those biopsies for individualized risk evaluations of all subjects. These data will be deposited into a bio-repository that will serve as the basis for creating individualized breast cancer screening models.

Pisano and her collaborators are seeking the necessary 130 North American study sites so that they can recruit the huge voluntary patient population for the research.

Fifty-three sites were open and accruing patients as of December 5. A total of 9,032 subjects, including 3,000 women from a Canadian pilot study, have been enrolled to date.

Pisano spoke with Advocacy in Action to describe how more medical imaging facilities can participate.

AIA: What kinds of medical facilities are now contributing data to TMIST?

Pisano: We actually have a combination of community sites and academic health centers. We are using the National Community Oncology Research Program (NCORP) network to get more community hospitals and freestanding imaging centers into the trial. One of our top recruiting sites in Savannah, GA, is a community cancer center, and we have another community cancer center in Urbana-Champaign, IL. There is a site in Spartanburg, SC, and though Montefiore Medical Center in the Bronx is an academic research hospital, it is also a minority-dominate cancer center for the surrounding community. Our top accruing site is at the University of North Carolina-Chapel Hill, with almost 1,000 women recruited to the study to date.

AIA: How can a medical imaging service qualify to participate?

Pisano: The site has to have both tomosynthesis and digital mammography equipment at the site where they recruit subjects. We can help facilities join the National Clinical Trials Network (NCTN) and NCORP, if they aren’t already registered with those programs. There is no separate contracting process. Master research agreements that cover participation in all National Cancer Institute trials run through the NCTN are established between the facility and each NCTN or NCORP group it is affiliated with. The imaging service or department can apply for ECOG-ACRIN membership if it has no affiliation with NCTN or NCORP.

AIA: How much time and resources must be committed?

Pisano: The site trial team typically consists of a lead radiologist, who provides one or two hours a week for oversight and to make sure the study protocol is followed properly. Staff radiologists themselves don’t have a gigantic amount of work to do because it is already part of their job to read screening mammograms. Essentially the way we set up the study is that the screening mammogram report results get entered into a computer by the research assistant (RA).

One full-time equivalent should be budgeted for the RA and perhaps another for a recruitment coordinator to handle recruitment, enrollment, consents, case report form data entry, image transmission and other duties. Reading radiologists will integrate study interpretation into their daily routine, but they need to be aware of the involvement of patients under their care. Mammography technologists only need to be cognizant of imaging modality randomization of the trial subjects, and the quality control mammography tech will need to set aside about 30 minutes per month to perform TMIST QC imaging.

AIA: How much does the study pay its facilities?

The study pays facilities $500 for each recruited subject to NCTN sites and $150 for each follow-up screening mammogram required under the study protocol, with additional payments for submission of information about biopsy specimens, biopsy samples, and blood and buccal smear materials. NCORP sites receive the equivalent in NCORP credits. The University of North Carolina at Chapel Hill, for example, has recruited an average of 20 new enrollees per week since its program started in September 2017. By November 26 of this year, the program had enrolled 955 subjects and had accrued fees of over $477,000, which easily covers the cost of their full-time research assistant, recruiting coordinator and other staff.

The cost of screening mammography or tomosynthesis is reimbursed through the usual mechanisms of private insurance, Medicare or Medicaid, and there is a special fund in our trial that pays $138.17 per study for uninsured subjects to receive screening mammography, so this payment can be applied every year for every time a mammogram is required in the study for uninsured women. Participating sites receive additional annual fees and payments every time a subject returns for an additional mammogram, additional biopsy or biological specimens are submitted.

AIA: Have steps been taken to streamline the trial’s administration for its participating facilities?

Pisano: Most of the streamlining was done during our Canadian pilot study before we launched the trial in the US. From that experience, we learned not to have radiologists enter data themselves. Other protocol changes are ongoing. For example, our central Institutional Review Board gave us a 20-page consent form. We have since cut it down to 16 pages, and we are trying to make it even shorter. We are also working on a protocol amendment that reduces the image submission requirements for the sites. We have worked closely with the sites to hear their concerns and reduce their work to get the study done to the minimum that allows us to meet study endpoints.

AIA: Are there value-added benefits?

Pisano: When radiologists join the study, they are joining a group of engaged investigators who are intensely interested in this topic. They will be considered one of our collaborators, not just individuals who are running the study at their sites. If they wish, they can be part of the publication team. I want the help of people who want to actually write the papers and be listed as co-authors of the peer-reviewed papers that are published because of this trial. This is an opportunity for networking and for getting to know other people who are interested in this field and for learning how to run a clinical trial in breast imaging.

AIA: How can radiologists or imaging facility managers learn more about joining TMIST?

For information about becoming a TMIST study site, contact TMIST Project Manager Jessica Howell at

For information about TMIST itself, visit National Cancer Institute TMIST trial website or the ECOG-ACRIN Cancer Research Group website.