SAINT TMS

[nexus73]

SAINT

Saw this posted over on reddit, new variation of TMS with "significant reduction of depression symptoms" in 79% of participants. Uses fMRI targeting, figure 8 coil TMS, 10 treatments per day x 5 days.

SAINT Depression Treatment: About Stanford Accelerated Intelligent Neuromodulation Therapy

This noninvasive type of nerve therapy is helpful for treatment-resistant depression.

psychcentral.com

I'm hoping the smart psychiatrists here can tell me is this is new and better, or vaporware

---

Members don't see this ad.

 

[Penguin10]

SAINT

Saw this posted over on reddit, new variation of TMS with "significant reduction of depression symptoms" in 79% of participants. Uses fMRI targeting, figure 8 coil TMS, 10 treatments per day x 5 days.

SAINT Depression Treatment: About Stanford Accelerated Intelligent Neuromodulation Therapy

This noninvasive type of nerve therapy is helpful for treatment-resistant depression.

psychcentral.com

I'm hoping the smart psychiatrists here can tell me is this is new and better, or vaporware

Chris Aiken did a brief podcast on the studies that led to approval for The Carlat Psychiatry Podcast here. He raised the question of whether similar results might be achievable by incorporating aspects of the SAINT strategy with current systems which was interesting.

 

[calvnandhobbs68]

I can't see this protocol becoming mainstream enough to make a dent the way it's currently structured but that's just me.
1) Insurances are gonna have to actually pay for fMRI targeting and 10 sessions a day or you'll have to rely on people paying OOP
2) The big kicker, who is gonna be up for 10 hour x 5 day treatment sessions? Patients are going to have to take vacation or FMLA or something to sit around in a clinic for a week straight.

I mean I guess the people who have real deal TRD might be up for it, but not the same volume you have for regular TMS right now where people who fail 1-2 SSRIs will go try TMS. Theta burst stimulation is already so brief on its own that when you tell someone they can either come in for 3 minutes a day for a few weeks or have to sit in a clinic for 5 days straight for 10 hours a day....also why wouldn't you just try MRI targeting a regular run of theta burst TMS first? All this being said I don't actually do TMS, so maybe someone who does may have more insight.

It's kind of like the Zulresso thing....oh here's this rapid treatment for post-partum depression but guess what it's super expensive and you have to sit in an infusion center for 2.5 days. Big fanfare but barely anyone actually does it in real life and Sage is now trying to get zuranolone to work so it'll be an oral option they can market that people will actually take.

 

[WisNeuro]

Anyone have the SAINT validation articles? All I see is the N-14 treatment group article that measures efficacy at 4 weeks. I have teh same concerns as others. One, this needs some direct comparison studies with real world patients as opposed to super clean samples that are usually included in proof of concept studies. Because, this would need to show clear superiority to other methods, or work very well in those that traditional TMS doesn't get, given the cost and time associated with such a protocol.

 

[G Sheb]

Anyone have the SAINT validation articles? All I see is the N-14 treatment group article that measures efficacy at 4 weeks. I have teh same concerns as others. One, this needs some direct comparison studies with real world patients as opposed to super clean samples that are usually included in proof of concept studies. Because, this would need to show clear superiority to other methods, or work very well in those that traditional TMS doesn't get, given the cost and time associated with such a protocol.

I think you're talking about this article Psychiatry Online
It's still early days. There are only two studies out there, one is a feasibility and safety study and the other is an RCT.
I believe it's being advertised as a potential inpatient treatment that can replace ECT.

 

[WisNeuro]

I think you're talking about this article Psychiatry Online
It's still early days. There are only two studies out there, one is a feasibility and safety study and the other is an RCT.
I believe it's being advertised as a potential inpatient treatment that can replace ECT.

Yes, this is the one that I've seen. And, while interesting, is far from groundbreaking. Also, cost-wise, this still looks more expensive than ECT. Especially if a facility has to essentially keep an MRI set aside for 1 or 2 patients for an entire day.

 

[G Sheb]

Yes, this is the one that I've seen. And, while interesting, is far from groundbreaking. Also, cost-wise, this still looks more expensive than ECT. Especially if a facility has to essentially keep an MRI set aside for 1 or 2 patients for an entire day.

They just just do one 8 minute session resting-state fMRI to localize the area you need to target and one structural scan. Patients are not scanned during treatment delivery. Right now rs-fMRI is still only a research tool.
I don't know how it works out expense-wise, but absence of significant cognitive side effects would be a big advantage over ECT, so is the no need for anesthesia. This has potential imo, but yes, it's too early. There were other fads in the past couple of decades that were touted as possible replacement for ECT but they didn't go anywhere.

 

[WisNeuro]

They just just do one 8 minute session resting-state fMRI to localize the area you need to target and one structural scan. Patients are not scanned during treatment delivery. Right now rs-fMRI is still only a research tool.
I don't know how it works out expense-wise, but absence of significant cognitive side effects would be a big advantage over ECT, so is the no need for anesthesia. This has potential imo, but yes, it's too early. There were other fads in the past couple of decades that were touted as possible replacement for ECT but they didn't go anywhere.

Ah,, that makes it better. As for ECT, the cognitive side effects are temporary. Anything longer term appears to be largely due to expectancy effects, somatoform/cogniform disorder, or compensation seeking. Most of these cases would likely pose the same issue in any neurologically based treatment protocol.

 

[FlowRate]

I think it's promising but, as they point out in their initial trials, they now need to find a way to decompose some of the alterations to standard TMS in order to figure out what it is about SAINT that led to apparent improvement. Keeping depressed people out of the house for a week straight? Frequent contact with caring others? fMRI targeting? Increased number of treatments in shorter timespan? Every one of those things is a potential causative factor.

They are still studying SAINT. I had a patient apply to be a study participant recently but he was rejected largely because his amotivation is so bad that he's unlikely to actually fly out of state and be able to keep up with the lodging/transportation/attendance requirements.

 

[G Sheb]

Ah,, that makes it better. As for ECT, the cognitive side effects are temporary. Anything longer term appears to be largely due to expectancy effects, somatoform/cogniform disorder, or compensation seeking. Most of these cases would likely pose the same issue in any neurologically based treatment protocol.

Yes, ECT is a high bar to reach and I am skeptical it can be matched in a real world setting. Until this question is answered, this is still a fad. How useful it is outside this context is doubtful.
The challenge about finding a replacement for ECT is that no one really knows why ECT is so effective. Likely it is precisely because of the widespread seizure that is involving disparate areas in the brain.
Years ago there was FEAST, which was supposed to trigger a more focal seizure and hence avoid the side effects. And yes there were less side effects, but it was also 50% less effective.

 

[clozareal]

Yes, this is the one that I've seen. And, while interesting, is far from groundbreaking. Also, cost-wise, this still looks more expensive than ECT. Especially if a facility has to essentially keep an MRI set aside for 1 or 2 patients for an entire day.

I disagree that it's not groundbreaking. I think that the active group having a 79% remission rate compared to sham of 13% is huge, twice the efficacy of ECT for TRD, and better than any treatment we have in psychiatry for TRD or overall. They also used a population sample with SI as well. Name me one other RCT that had this wide of a margin of difference between active vs placebo. It will be even more groundbreaking once it's replicated.

I do agree that it will be difficult to implement since it's 10 sessions a day x 5 days, although they did use theta burst which cuts down actual treatment duration to a few minutes. It will probably need to be like a PHP or inpatient type of setting. The other difficulty is in the fMRI and neuronavigation (also we do have Nexstim), but the Stanford group is claiming they have their own proprietary AI to localize the treatment target.

One thing I wonder is what if you just do 50 sessions of regular TMS rather than 30-36 and if that dose would actually be better for TRD.

 

[clozareal]

Yes, this is the one that I've seen. And, while interesting, is far from groundbreaking. Also, cost-wise, this still looks more expensive than ECT. Especially if a facility has to essentially keep an MRI set aside for 1 or 2 patients for an entire day.

Patients would have the MRI once prior to starting treatment, not every day.

 

[WisNeuro]

I disagree that it's not groundbreaking. I think that the active group having a 79% remission rate compared to sham of 13% is huge, twice the efficacy of ECT for TRD, and better than any treatment we have in psychiatry for TRD or overall. They also used a population sample with SI as well. Name me one other RCT that had this wide of a margin of difference between active vs placebo. It will be even more groundbreaking once it's replicated.

I do agree that it will be difficult to implement since it's 10 sessions a day x 5 days, although they did use theta burst which cuts down actual treatment duration to a few minutes. It will probably need to be like a PHP or inpatient type of setting. The other difficulty is in the fMRI and neuronavigation (also we do have Nexstim), but the Stanford group is claiming they have their own proprietary AI to localize the treatment target.

One thing I wonder is what if you just do 50 sessions of regular TMS rather than 30-36 and if that dose would actually be better for TRD.

Depends on the definition of remission, as this can be as simple as not diagnostic at the time of measurement. Which, is why I asked for any other studies, as the one I found was underwhelming. Not all RCTs are equal. This could be, I just haven't seen the data yet. If you have the papers, I'd love to see them.

 

[mistafab]

I didnt read their saint method but an 8 minute FMRI for targeting sounds like balogna.

Most likely they just theta bursted the same target we all use anyway (DLPFC). I bet theta bursting anyone 10 times a day for a week would have the same effect SAINT Mapped vs unmapped lol. Just seems like an excuse to use expensive technology that has found no real purpose in medicine right now.

 

[clozareal]

Depends on the definition of remission, as this can be as simple as not diagnostic at the time of measurement. Which, is why I asked for any other studies, as the one I found was underwhelming. Not all RCTs are equal. This could be, I just haven't seen the data yet. If you have the papers, I'd love to see them.

They used MADRS 4 weeks after treatment. Response was ≥50% reduction and remission was ≤10.

 

[clozareal]

I didnt read their saint method but an 8 minute FMRI for targeting sounds like balogna.

Most likely they just theta bursted the same target we all use anyway (DLPFC). I bet theta bursting anyone 10 times a day for a week would have the same effect SAINT Mapped vs unmapped lol. Just seems like an excuse to use expensive technology that has found no real purpose in medicine right now.

It's because using DLPFC using traditional methods missed the target 30% of the time. Neuronavigation would make sure that the stimulation target is actually where your 8-figure coil is directed at.

It's not an 8 minute fMRI. The MRI takes as long as it usually takes to do a regular MRI. It's similar to the technology they use for neurosurgical planning for procedures like DBS.

 

[G Sheb]

It's because using DLPFC using traditional methods missed the target 30% of the time. Neuronavigation would make sure that the stimulation target is actually where your 8-figure coil is directed at.

It's not an 8 minute fMRI. The MRI takes as long as it usually takes to do a regular MRI. It's similar to the technology they use for neurosurgical planning for procedures like DBS.

I don't think that's the point of the rs-fMRI.
It is an 8 minute resting state fMRI to capture the area of the dlPFC that is most anti-correlated with the sgACC. The idea is that if you target this specific sublocation of the dlPFC you get better results. It is a personalized targeting so two subjects may not have the same area targeted.
And as they discuss in the paper, they simply don't know how much this targeting contributes to the effect. There are many things that they've tried for the first time, including 10 sessions a day for 5 straight days.

 

[G Sheb]

I didnt read their saint method but an 8 minute FMRI for targeting sounds like balogna.

Most likely they just theta bursted the same target we all use anyway (DLPFC). I bet theta bursting anyone 10 times a day for a week would have the same effect SAINT Mapped vs unmapped lol. Just seems like an excuse to use expensive technology that has found no real purpose in medicine right now.

That's a usual rs-fMRI session.