Read on Twitter
Bupropion and Naltrexone in Methamphetamine Use Disorder | NEJM https://www.nejm.org/doi/full/10.1056/NEJMoa2020214#.YACoDTr9HmI.twitter
Results of the multisite ADAPT-2 trial just published in @NEJM - 11% effect, NNT 9
Here's my thread on first takes..
Results of the multisite ADAPT-2 trial just published in @NEJM - 11% effect, NNT 9
Here's my thread on first takes..
1. Wow. Amazing seeing such a well-conducted #addiction trial published in @NEJM.
Pre-screened 2763 participants, screened 932, randomized 403. This is *hard* work for the trial team, who appear to have done an incredible job recruiting.
Pre-screened 2763 participants, screened 932, randomized 403. This is *hard* work for the trial team, who appear to have done an incredible job recruiting.
2. Very low attrition rate and high adherence rate (75-80%).
This is really quite unusual. Staying in clinical trials is difficult- multiple urine screens/wk, lots of long research visits- so this retention rate is spectacular. But this limits the generalisability of results.
This is really quite unusual. Staying in clinical trials is difficult- multiple urine screens/wk, lots of long research visits- so this retention rate is spectacular. But this limits the generalisability of results.
3. Participants look quite naturalistic/ real-world.
70% male, 38% employed, on average using 27/30 days prior to the trial (nearly daily) with mod-severe MUD, 19% mainly injecting (versus smoking).
This sounds fairly similar to other MA trials, including in Australia.
70% male, 38% employed, on average using 27/30 days prior to the trial (nearly daily) with mod-severe MUD, 19% mainly injecting (versus smoking).
This sounds fairly similar to other MA trials, including in Australia.
4. The 2 stage sequential: parallel trial design is statistically complex (for me, anyway!). Non-responders in stage 1 were randomized to stage 2- in a way that enriched the stage 2 sample with the placebo non-responders. This inc. chance of detecting an effect if it existed.
6. The way the trial was conducted- the next step would be to do a 'naturalistic' study, with a straightforward placebo/active RCT- but this would prob be *less* likely to detect an effect. So this is as good as it gets.
7. So my take-home:
Really inspiring to see such rigorous work in this field.
Would I prescribe this? No, not routinely- firstly, no NTX-XR in Australia; secondly, overall effect of 11% is not staggering, and in non-responders in phase 1 (after 6 wks), only 3% responded.
Really inspiring to see such rigorous work in this field.
Would I prescribe this? No, not routinely- firstly, no NTX-XR in Australia; secondly, overall effect of 11% is not staggering, and in non-responders in phase 1 (after 6 wks), only 3% responded.
8. This would also not work for people with comorbid opioid + MA use- not uncommon- because NTX is an opioid blocker. And high rate of adverse events (not severe.. but nausea, GI problems, insomnia, irritability).
9. MA use is a really significant problem, both here in Australia as in the US. We need to keep looking for effective medication treatment options.
Funding #clinicalresearch and well conducted trials like this one will eventually get us there.
Funding #clinicalresearch and well conducted trials like this one will eventually get us there.
@GillBedi what do you reckon?
