August 01, 2019

ACR’s Pisano Lays Out Rationale for Joining TMIST Trial at AHRA Meeting

Speaking on July 24 at the AHRA annual meeting in Denver, Etta D. Pisano, MD, FACR, presented its members compelling reasons why radiology needs the Tomosynthesis Mammographic Imaging Screening Trial (TMIST) randomized clinical trial and how financial benefits await breast imaging services that participate in the mammoth research effort.

By measuring the relative clinical performances of digital breast tomosynthesis (DBT) and digital mammography (DM), TMIST will determine which of these two imaging technologies reduces advanced cancers in the population screened, an endpoint that correlates with breast cancer mortality. To generate enough statistical power to answer its primary questions, TMIST will involve 164,946 women, ages 45 to 75, who will randomly be assigned to breast cancer screening, typically in one-year increments during the four and a half year-long trial.

Pisano, who was the lead research in a similar randomized trial that tested the efficacy of DM in the 1980s, is again taking the lead on TMIST to generate answers to the relative performance of DBT and DM and which women might most benefit from the newer technology. Health care policy demands answers, Pisano noted, because of the multibillion-dollar costs of mandating annual breast cancer screening for all women beginning at age 40 and the opportunity to save lives through early breast cancer detection and treatment.

The study is funded by the National Cancer Institute through the ECOG-ACRIN clinical trials group.

Key questions to be answered by TMIST include:

  • How do DBT and DM compare in reducing advanced breast cancer in the screened population? (Advanced cancers are defined as metastatic cancer, cancers with positive nodes, invasive tumors that are equal to or greater than 2 cm, and all invasive cancer that are equal to or greater than 1 cm with prognostic markers that suggest aggressive behavior.)
  • How does DBT affect breast cancer mortality compared with DM?
  • What is the rate of overdiagnosis with the two technologies?
  • Do all makes and models of DBT perform with the same diagnosis accuracy, sensitivity and specificity for specified North American women?
  • How can genetic, demographic and other information be used to tailor individual screening strategies?

“The most important outcome from TMIST will be the information it will provide to us about how to individualize breast cancer screening,” Pisano said. “We have been using personalized therapies for breast cancer for several decades. We need personalized approaches for screening as well, and this study will give us lots of information on how we can achieve that.”

Pisano stressed that she and her fellow TMIST researchers are seeking assistance from AHRA members to assure the trial’s successful completion. As of July 8, 12,144 participants were enrolled at 71 sites in 27 states, Argentina, Canada and the U.S. territory of Puerto Rico.

“Beyond the professional satisfaction from direct involvement in historically significant research, participation also generates financial rewards,” Pisano said. Qualified sites can receive a $30,000 advance toward accrual to hire a full-time radiology assistant. They are paid $500 per subject for initial recruitment, randomization and data submission. They receive $150 for each additional screening and another fee for the submission of path/blood/buccal smears, and a subsidy for mammograms for uninsured women who otherwise cannot afford to pay for their own mammograms and are subject to the charity care policies of that institution.

A research agreement is needed so funds from TMIST can flow to participating institutions. There is no separate contacting process specifically for TMIST, Pisano noted. Master research agreements covering participation in all National Cancer Institute trials run through its National Clinical Trial Network (NCTN). They would have to be established between participating facilities and each NCTN group it is affiliated with. Most contracts are between oncology divisions or cancer centers.

If your radiology department is expected to receive funds services, such as personnel salary support, it will need an interdepartmental agreement with the department named in the master NCTN research agreement within your institution.

Pathways to TMIST participation may be found through the NCI Community Oncology Research Program (NCORP). Facilities can also take a non-NCORP pathway through a direct institutional membership in one of five NCI NCTNs. They are Alliance for Clinical Trials in Oncology, SWOG Cancer Research Network, NRG Oncology and the Canadian Cancer Trials Group (CCTG), only for Canadian sites.

Pisano urged U.S. and Non-Canadian international institutions that don’t have a NCTN affiliation to contact the ECOG-ACRIN membership office to learn about its application process. They should also apply for an institutional Cancer Therapy Evaluation Program ID by contacting ecuhelpdesk@mail.nih.gov or calling 703-738-9166.

At a minimum, your TMIST team will consists of a lead radiologist, a clinical research associate, mammography interpreters and mammography technologists. The facility must be equipped to perform DBT and DM. Because of the importance of patient recruitment, many facilities assign a full-time administrative assistant to the task.

At least one team member must complete TMIST-specific webinar training, which is now offered on demand on an ECOG-ACRIN Cancer Research webpage. Phantom imaging tests must be completed for each DBT and DM scanner in the trial. The credentialing images are to be submitted through Transmission of Imaging and Data for review by the Physics Quality Control Core.

Sample mammography reports must be submitted to Elodia Cole for de-identified sample radiology mammography reports covering the following: abnormal tomosynthesis screening mammography report (BIRADS 0), an abnormal tomosynthesis screening mammography report (BIRADS 0), an abnormal digital screening mammography reports (BIRADS 0), a tomosynthesis screening mammography report with no finding or with benign findings (BIRADS 1 or 2), a digital screening mammography report with no findings or with benign findings (BIRADS 1 or 2), and a diagnostic mammography report (with final BIRADS).

Completed physics and IT surveys should be submitted via Survey Monkey for each physical site. Most recent Mammography Quality Standards Act reports for systems to be used to screening TMIST participants are to be emailed to contact@tmistQC.org. An IRB-approved clearance is also necessary.

“The extra effort required to sign up to participate is definitely worthwhile, so your site can become part of a large group of committed researchers — the TMIST Investigators Group,” Pisano said. “You can become a part of this historic effort and be recognized as a coauthor of the papers and policies that will follow from participation in TMIST.”