Re: UBC CCSVI clinical trial results to be presented Wednesday, March 8, 2017 at SIR meeting in Washington DC: "You will notice the Abstract published on the SIR conference website does not include any of the research outcomes. Normally, an abstract must summarize a study in its entirety, including patient outcomes. However, when a study is first going to be presented at a conference, it is common that researchers had to submit an abstract long before the full study was written up or even completed. You will note from the Abstract that this presentation will include "...all data from the first 48 weeks in all patients after their first procedure only, comparing sham to venoplasty." Thus, I find it surprising that the Abstract does not give a full summary of the research outcomes because we will not be getting full results from this presentation. We will get comparisons between patients who had venoplasty for jugular occlusions and those who did not at many data points beginning 12 hours after the treatments....
Note regarding upcoming UBC CCSVI research design and results
You will notice the Abstract published on the SIR conference website does not include any of the research outcomes. Normally, an abstract must summarize a study in its entirety, including patient outcomes. However, when a study is first going to be presented at a conference, it is common that researchers had to submit an abstract long before the full study was written up or even completed. You will note from the Abstract that this presentation will include "...all data from the first 48 weeks in all patients after their first procedure only, comparing sham to venoplasty." Thus, I find it surprising that the Abstract does not give a full summary of the research outcomes because we will not be getting full results from this presentation. We will get comparisons between patients who had venoplasty for jugular occlusions and those who did not at many data points beginning 12 hours after the treatments.
This type of research is called a crossover study, where the research begins with one group that receives an intervention (venoplasty) and another group that receives a sham intervention (mock venoplasty). Patients are randomly assigned to each of the groups and the team doing the evaluations after the treatment are blinded to whether the patients received the actual venoplasty or a sham procedure. The crossover part of the study means those patients who received the real venoplasty the first time would undergo a sham procedure a year later and those who received the sham would undergo a real venoplasty.
There are a couple reasons why UBC probably chose to do a crossover study. First, as we all know, MS is a highly variable illness. A crossover study controls for differences between the two groups – treated and untreated patients. Because every patient will eventually receive the treatment and every patient will receive a sham treatment, the two groups are statistically comparable. The results the team will present next week are NOT able to benefit from this advantage because it is not yet a crossover study – the second treatment round had not been done when they compiled the data.
Crossover studies also allow for slightly smaller sample sizes to return significant results. This is a good thing because crossover designs place a lot of burden on the participants – lots of questionnaires, lots of follow-up, and then you have to do it all again a year later!
A limitation of this type of study is it could accentuate a placebo effect. Patients who made a judgement about whether they did or did not receive the "real" venoplasty the first time will also make a judgement about what treatment they will receive next. For instance, if you had no effect at all the first time around, you might assume you got the sham procedure. This would give you high expectations for the second venoplasty. This limitation should be acknowledged by the team when they write up the full study after they complete all elements of the crossover design.
Another limitation is that half the group had to wait a year to get the real venoplasty. Why does this matter? Well, the groups were randomly assigned in the beginning, which means a year later the group that initially got the sham procedure would be a-year-old or, their MLS may have advanced, and they may have had other comorbidities arise. Basically, even though only a year has gone by, results coming from their real venoplasty means they are not totally comparable to the first group.
The presentation next week, however, only includes results up to 48 weeks, so it does not include the second round of treatments. We should look to see if they present any between group differences, for instance one group may contain people with higher EDSS scores, people who have had MS longer, or people who are older. These differences – and many more – may mean their response to venoplasty will vary. Random assignment ideally would even out these differences, but with a smaller sample differences inevitably occur. Remember – these differences will all be evened out once the full crossover study is released!
Sandra