What to Watch for in the ATTRACT Study Data Presentation
An interview with Principal Investigator Suresh Vedantham, MD.
On Monday, March 6, at the Society of Interventional Radiology's annual scientific sessions in Washington, DC, the final results from the landmark ATTRACT (Acute Venous Thrombosis: Thrombus Removal With Adjunctive Catheter-Directed Thrombolysis) study will be presented and discussed by an expert panel.
WHAT: ATTRACT study final 2-year data presentation and panel
WHEN: Monday, March 6, 2017 from 10:30 AM to 12:00 PM
WHERE: Marriott Marquis Ballroom, Meeting Level 2
Walter E. Washington Convention Center, Washington, DC
To get an idea of what we might expect from this large, multicenter randomized trial evaluating pharmacomechanical catheter-directed thrombolysis plus standard therapy versus standard therapy alone, Endovascular Today spoke with Principal Investigator Suresh Vedantham, MD, ahead of the full presentation.
In conjunction with Dr. Vedantham and the ATTRACT investigators, we are inviting questions from our readership to be answered in a subsequent publication after the data are presented.
If you would like to submit a question to the investigators, please send it via email to askATTRACT@bmctoday.com. Endovascular Today will work with the ATTRACT team to address as many as possible.
Without speaking to the specific details you'll be presenting at SIR 2017, what should we expect to see from the much anticipated first look at the final ATTRACT trial data?
The ATTRACT study was designed and statistically powered to determine whether pharmacomechanical catheter-directed thrombolysis (PCDT), added to standard care as part of first-line therapy for above-the-knee proximal deep vein thrombosis (DVT), prevents the development of the postthrombotic syndrome (PTS) over 2 years. Physicians are also interested in other outcomes we assessed such as PTS severity, quality of life (QOL), and resolution of presenting DVT symptoms. In addition, there is interest in the outcomes related to important subgroups such as iliofemoral versus femoropopliteal DVT. It is important to realize that although ATTRACT will provide the largest, most rigorously obtained dataset from which to gain such perspectives, it has less statistical power to address some secondary outcomes and subgroups.
What is the potential impact on the global community interested in the treatment of DVT if the data support pharmacomechanical thrombectomy?
If PCDT is shown to prevent PTS, reduce PTS severity, and improve QOL, this would definitively validate the open vein hypothesis and define an important shift in our understanding of the disease. Such an outcome should drive major changes in clinical practice aimed at facilitating early identification and referral of patients who may benefit. If not, then DVT/PTS researchers should seek alternative directions to better understand the factors that determine whether a patient develops PTS and how severe it becomes.
How might favorable results potentially affect societal guidelines and referrals?
A positive primary outcome, especially following the positive results of the CAVENT study (although it employed a different thrombolysis method), should motivate a change in guidelines to support routine use of endovascular DVT therapy. Of course, the risks associated with thrombolytic drugs—especially if applied to a larger patient population than we are treating now—must be considered. Any such recommendations would need to be properly conditioned to ensure quality patient selection.
Is there any potential for the trial not to be sufficiently powered to favorably meet its endpoints? What factors could contribute to this occurrence?
Statistical power defines a continuum of imprecision; we apply cut-off values like P < .05, which is actually fairly arbitrary. ATTRACT was designed to have 80% power to identify a 33% reduction in PTS rates if one truly exists, accepting a 5% risk of over-calling a difference if none exists (alpha error, 0.05), and assuming a 10% loss to follow-up over 2 years. We did enroll the full sample, and I will disclose that the overall PTS rate (both arms combined) was similar to previous studies and higher than our original projection. The proportion lost to follow-up was slightly greater than our early projections. Therefore, it is highly likely that the study does have sufficient power to address its primary endpoint, which is presence or absence of any PTS. Secondary outcomes and subgroups are another matter because there is a greater chance of being underpowered for those assessments. Any conclusions we draw about secondary outcomes and subgroups will need to be appropriately conditioned.
If the primary endpoint is not met, is there still potential for ATTRACT to be favorable for pharmacomechanical thrombectomy?
It depends on the secondary outcome and subgroup assessments, especially safety, PTS severity, and QOL. We always appreciated that patients with iliofemoral DVT may fare differently than those with femoropopliteal DVT. Both populations were included in the study because it is reasonable to evaluate if the open vein hypothesis might apply to both, and because some experts have publicly advocated to extend treatment to patients with femoropopliteal DVT. With either a positive or negative primary outcome assessment, it will be interesting to look at these groups. However, a negative primary outcome assessment will greatly increase skepticism among physicians about the value of these procedures and, depending on what we see, that might be well justified considering the risks and costs likely to be associated with PCDT.
What are some potential hurdles in interpreting a dataset as large as this one?
When you have such a wealth of data, it takes a certain kind of group discipline to do things in an organized way, giving preference to the questions that were specified before the study. ATTRACT has benefited tremendously from the leadership of outstanding scientists in clinical trial methodology, including Clive Kearon, MB-PhD, who leads the Data Coordinating Center (DCC) at McMaster University; Susan Kahn, MD, MSc (McGill University); and lead biostatistician Professor Jim Julian (McMaster University). I have learned so much from them, from the world-class staff at the DCC, and other ATTRACT leaders.
One significant challenge is gaining consensus on data interpretation and messaging among so many experts who have such different baseline perspectives on the disease and different levels of experience with high-end trial methodology. But, we have a group of people with incredible collaborative maturity, so I am confident we will continue to succeed in this respect.
If you were to begin designing the ATTRACT trial in 2017, what would you do the same, and what would you do differently?
Any way the trial turns out, I am going to receive a lot of feedback from the physician community on exactly what we should have done—so maybe I can be forgiven if I don't feel too inclined to self-flagellate at this time! Our strategic decision to take a broadly collaborative approach with input from diverse investigators with different preconceptions about DVT, PTS, and PCDT, has served us very well and enabled us to maintain a high level of credibility for the study across a range of constituencies. This will ultimately increase the value of the study to providers and patients.
Have additional questions for the ATTRACT investigators? Send them to
askATTRACT@bmctoday.com for possible inclusion in upcoming Endovascular Today coverage of the ATTRACT study.