In September, an advisory board to the National Cancer Institute overwhelmingly voted to form a working group to review the continued funding and running of its largest cancer screening trial. But the lone abstaining board member suggested that the reevaluation was a smoke screen for ending the $100 million Tomosynthesis Mammography Imaging Screening Trial (TMIST).
“We’re going to have a working group to kill the trial,” Peter Adamson, MD, of Sanofi, commented in an interpretation of the NCI’s intent. The randomized trial, which began enrolling women in 2017, is comparing tomosynthesis, or three-dimensional (3-D) mammography, with older, less expensive 2-D technology.
The NCI’s move to reevaluate TMIST comes 6 months after Medscape Medical News reported that the projected 165,000-women study was lagging in enrollment of both patients and participating sites/physicians.
Enlisting sites is difficult, in part because radiologists – informed by their experience and previous research results – already believe that the 3-D technology is superior, commented two TMIST study investigators at that time.
Furthermore, experts who were quoted in the Medscape story said that radiology practice in the United States has substantially transitioned to 3-D and will continue to do so. This point was supported by a review of imaging market data.
At the September virtual meeting of the National Cancer Advisory Board, which was first reported by the Cancer Letter, Philip Castle, PhD, MPH, director of the NCI’s division of cancer prevention, said it was “time to reevaluate TMIST’s feasibility and relevance.”
There is also a need to “examine the utility of TMIST, given the unanticipated challenges due to the pandemic,” he said, referring to low accrual of patients in 2020.
TMIST is “a huge budget item,” and accrual has been a problem “for years – essentially since the trial began,” added Ned Sharpless, MD, director of the NCI.
Fourteen advisory board members voted in favor of creating the working group to review the trial. The lone abstention, as noted above, was from Dr. Adamson, who is global head of oncology development and pediatric innovation at Sanofi and is also professor emeritus, University of Pennsylvania, Philadelphia.
Dr. Adamson said that he was “not denying the impact” of COVID-19 but speculated that other clinical trials have had the same problem.
He also wondered whether COVID-19 was an “excuse” for bringing TMIST to a “public forum” and objected to the fact that TMIST investigators were not present and had not been invited to offer “an action plan” to remedy their trial’s woes.
Dr. Castle explained that COVID-19 had reduced TMIST’s monthly patient accrual by 50% from March to August, compared with recent preceding months. The underaccrual “will likely result in delayed outcomes and escalating costs.”
However, TMIST was staggeringly behind in accrual even before COVID-19. The trial has averaged less than 1,500 women a month over the past 2 years, yet it needed to average 5,500 a month to meet accrual goals, Dr. Castle noted.
Further, Dr. Castle suggested that TMIST’s contribution to mammography practice may be negligible, citing three factors: changes in the imaging market, earlier research results, and other ongoing major trials of 3-D mammography.
“[The] relevance of the [TMIST] data by the completion of the trial is uncertain because of 3-D market ... penetrance by the conclusion of the trial,” Dr. Castle said. Even with optimistic accrual projections, study completion and reporting would occur somewhere between 2029 and 2032 – not in 2025, as originally planned.
Currently, 68% of certified mammography facilities in the United States have one or more 3-D units, and 40% of all units are 3-D, he noted. (However, as previously reported, this latter statistic on total units is likely an undercount because all 3-D units have a built-in 2-D function, but the two components in a single machine are equally counted by the Food and Drug Administration – even when 3-D is exclusively used.)
Dr. Castle also summarized earlier observational and randomized trial data on tomosynthesis mammography: “Evidence is already available suggesting that 3-D is no worse and probably better than 2-D.”
Additionally, he reviewed three other European mammography trials involving 3-D technology (in the United Kingdom, Germany, and Norway) and said that they will contribute “additional informative data.”
Notably, Dr. Adamson did not object to these critical points but said the “approach” taken by the NCI advisory board at this meeting was “incredibly disturbing.”
Dr. Castle defended himself, saying “nowhere” in his presentation was there any mention that the trial would be “shut down.”
Dr. Sharpless then interjected: “There is no easy way to do this. TMIST is the most troubled of our trials [in prevention and screening] from an accrual point of view.”