Psychedelics And Treatment-Resistant Depression: COMPASS Pathways Replies To Comments

(Part four of a four-part series)

See previous stories: 

1. Psychedelics And Treatment-Resistant Depression: An Overview Of COMPASS' Latest Trial Outcomes
2. Psychedelics And Treatment-Resistant Depression: Reactions To COMPASS' Published Results
3. Psychedelics And Treatment-Resistant Depression: Companies Weigh In On The Important Issue

In order to get all sides of the issue on the table and a clearer understanding of what the results of COMPASS Pathways CMPS’ latest Phase 2b trials mean, Benzinga interviewed Dr. Steve Levine, SVP of patient access and medical affairs, who worked closely with PI Dr. Guy Goodwin on the study.

One of the most resounding concerns over these trials focused on seemingly “uneven proportions of severely depressed people” within the three dose groups, pointing out that the 25mg arm would have included “significantly fewer” participants of this population.

Levine recalled that, as for the interpretation, the primary endpoint in the study was the change measured from baseline to three weeks in the MADRS depression score and, taking this into account, topline results showed there were no meaningful differences in the mean score across all three 25mg, 10mg and 1mg groups.

“Specifically, supplemental table 4 shows the baseline MADRS severity did not affect efficacy on that primary outcome. So I question the interpretation of what the critic is looking at, because within a study you may see some differences in the raw numbers, but that differs from what is statistically significant, what actually can be shown to be due other than the chance and actually affect the outcome,” Levine told Benzinga.

While the study had some numerical differences in mean baseline MADRS scores, table 4 evidence shows those were not meaningful across the groups, and the baseline severity of the MADRS score was uncorrelated with psilocybin’s clinical efficacy, he explained.

A second review stood out in the form of an accompanying editorial in the medical journal where COMPASS’ trial results were published. This one pointed at the potential difficulty of scaling up clinical trial protocols to broader patient populations.

Levine gave an extensive reply:

Firstly, the COMPASS Phase 2 study was “quite robust, unusually so, in some ways it was even fairly Phase 3-like,” referring to the number of participants (233) and the fact that it was conducted in 22 sites across 10 countries in 7 languages. 

“And that’s not typical. One of the common criticisms in Phase 2 studies, at least in terms of the generalizability to larger populations and the likelihood of being able to confirm findings thereafter, is that they may have been done in just one country, or in a limited number of sites, or smaller population.”

Levine also extended his reply to what he understands as a “kind of speculation” as to how psilocybin therapy might scale in the real world.

“This is something that we have been looking very closely at, and have been working on for four years now, because our goal does not just stop at regulatory approval. If we only have FDA approval or regulatory approval in other markets, that would not be success. Our goal is broad and equitable access for all the patients that may benefit as shown they benefit in our trials.”

And that means that it actually needs to get to patients, therefore broad coverage by payers, uptake into health systems, and safe and effective delivery.

To that end, Levine told Benzinga, COMPASS has already done significant work with payers and providers of care. 

In fact, current specialty Treatment-Resistant Depression (TRD) sites in the U.S. are already taking care of patients with existing antidepressant and interventional treatments. “And they already have in place the physical place, the room types and the workforce that would still need to be trained in our therapy, but they are ready to do so,” he said.

As for the poignant commentary regarding potential expectancy or selection biases in psilocybin non-naive participants, Levine stated that 94% of the study participants had no previous experience with psilocybin and did not know what to expect, and this is reflective of the general population.

All participants were told that they would be given some dose of psilocybin. “So that helps to mitigate the effect of the expectancy bias and to some degree the unblinding,” COMPASS Dr. Levine explained.

On the other hand, the trial’s design was very intentional as a comparison was not made between a 25mg and a placebo group, but rather between 1mg, 10mg and 25mg groups.

“The 10mg dose very much had a psychedelic effect and so, especially to someone naïve to psilocybin, it would have been difficult knowing which dose they got. Interestingly, even some people who got 1mg still had a subjective psychedelic effect,” Levine said.

After three weeks "and certainly if you follow it out even further,” the 10mg group does not look any different statistically from the 1 mg. But 25mg did have quite a robust effect: 6.6 point difference after three weeks and after a single administration at that primary endpoint, compared to 1mg.

In view of these results, Levine is certain that “there’s clearly a drug effect.”

What About Psychotherapy?

The COMPASS Phase 2 trials actually did not entail psychotherapy, but rather what Levine calls “psychological support.” That is, the primary safeguarding of patients across all three groups, which includes preparation sessions before, support during the drug administration, and integration following. 

“But that preparation and integration is not psychotherapy. The distinction I’m drawing there is that a psychotherapy would stand on its own as therapeutic, and is typically a more directive model of care. In this case, with psychological support, the aims are to prepare people naive to psilocybin so that they have some reasonable idea of what they might go through, to build trust and rapport, and to equip them with tools to be able to stay with the experience should it be challenging on that administration day,” Levine explained.

He further detailed that the administration session itself was largely silent in many cases, with the therapist standing as a continued supportive presence to ensure psychological and physical safety yet without the aim to proactively engage participants in therapy. And the integration session constituted a non-directive opportunity for participants to talk about their experience and insights they might have arrived to after administration.

But, Levine insists, it’s different from psychotherapy. 

“We believe that it’s the drug doing the work, and the support is there to create a safe condition for that effect to happen. Because this is what the data shows so far: if you look at the effect of the integration component within the study, participants who were responders or remitters did so in the day following the administration but before that first integration happened. And nobody who wasn’t a responder became one after having integration. So it would be hard to say that integration was necessary in some way for the efficacy of COMP360 in the study,” he further added.

Following this logic, upcoming Phase 3 trials will have a very similar amount of psychological support as used in Phase 2.

COMPASS Medical & Commercial Model

As stated by Levine, COMPASS’ ultimate goal is to bring a new option to those suffering with a difficult to treat depression in a broadly and equitably available manner. With this aim, the company is first expecting its synthetic psilocybin COMP360 to obtain approval by the FDA and regulatory agencies and markets around the world. 

“Part of the consideration of that is that the FDA regulates the approval of drugs, but not of psychotherapy or medical practice. What would be important to FDA is to know that there is a drug effect, and that there’s adequate quality of that drug, and that patients are safe regarding administration.”

Following the expected regulatory approval, the next step would be for the new drug to be broadly taken up by both government and commercial payers, so as to make it affordable for patients.

“Payers in most cases are not going to cover something that isn’t FDA-approved, so it has to be a model that is approvable. But then, once it is, and payers see the value in covering that, then this is how at the patient level you’re able to access care. Not that we have a perfect healthcare system, and not that every single person in the US is insured, but the vast majority of people in this country need to access their healthcare using their healthcare benefits. And that means it’s a small co-pay as opposed to paying out of pocket, which is always the case in any other type of model,” Levine explained.

The COMPASS official disclosed that the company still doesn’t know yet exactly what will be required within an FDA label, or what will be included in the REMS program, as that content “will be subject to ongoing conversations and negotiations.”

Interestingly, Dr. Levine holds extensive experience in treating patients with ketamine priorly to joining the COMPASS team. He explained that the available therapeutic data for ketamine and esketamine is also for the TRD population, and that some ketamine clinics do use it as first-line treatment in an off-label way.

“Ultimately, if we look at the FDA-approved version of ketamine, which is esketamine, that’s the one currently being used on-label and covered by payers. Our hope is that, relative to the esketamine data, ours will look quite strong and will be seen by payers as something that demonstrates value and a favorable thing to be covered,” he said.

In terms of what will generate revenue, COMPASS will produce its proprietary synthetic psilocybin compound COMP360 for distribution to specialty pharmacies, bought by both government and commercial payers, and sent to sites of care.

Levine explained the company won’t employ therapists or other healthcare providers nor deliver services. Nonetheless, it will offer therapist training in psychological support –as it is already doing for upcoming Phase 3 studies,- as well as digital solutions for both therapist and patient.

The digital solutions currently used at COMPASS’ trials include a proprietary natural language processing program that can listen to the sessions in real time –as recording is required by the FDA,- which Levine says can help ensure both safety and quality in the therapy delivered.

Hopefully, in the future these AI functions may be a way of making this model more personalized, predictive and preventive. 

“That will happen if we’re able to validate biomarkers, digital phenotypes, as well as find other predictive markers to what both patient and therapist say during sessions and even prior to them. But eventually, that might help predict who’s able to benefit from this treatment,” Levine shared.

A Concluding Comment On The Phase 2 Study’s Results

Benzinga explicitly asked Dr. Levine why he thought the results publication was the target of such diverse and numerous comments, spanning from full endorsement and enthusiasm all the way to negative concerns.

“I think that some people didn’t really understand that this was a study of Treatment-Resistant Depression (TRD) participants and not regular Major Depressive Disorder (MDD),” said Levine. “Certainly the interpretation of the robustness of the response that we saw after a single administration three weeks later, if you really understand that population and what is the likelihood of this population to get any benefit on any treatment based on any other study ever done, then I think you see it a bit differently than if you’re thinking about this as a study of MDD,” he said.

COMPASS’ official then referred to StarD, the largest depression study ever run which looked at a best-practice algorithm for MDD participants in relatively real-world conditions.

That study was made in four stages, the last two -equivalent to the population included in COMPASS’ Phase 2 study,- showing the likelihood of remission was 13.7% (step 3) and 13% (step 4.) 

“And that was achieved with daily oral treatment. We saw 30% of our participants in remission at week 3 following a single administration,” he stated.

Lastly, Levine told Benzinga he believes that while some people were “really keying in on some of the serious adverse events” such as the suicidal thinking or behavior, the focus should be set in the “greater context.” 

COMPASS’ official further disclosed that as many as two-thirds of the population with TRD had had some history already of suicidal thinking or behavior before, and that “every participant that had one such adverse event in the study had previously had it.”

In fact, in order to enroll in COMPASS’ trials, participants’ suicidality had to be really low. “So there was only one way to go, which was up.” Plus, they were all off antidepressants.

“So, you start to put all this together: TRD population moderately to severely depressed, most of them with a history of suicidal thinking or behavior in the past, off of antidepressants, getting just a single administration... And then, those events were pretty remote from the administration. They didn’t happen on the same day nor the day after but weeks after, and only in people who didn’t benefit from the treatment,” he explained.

“Of course we take suicidality incredibly seriously, continuing to in Phase 3 the same way we did in Phase 2, but I think in terms of the safety, the headline is that most of those adverse events were mild to moderate, the vast majority,” Dr. Levine concluded.

Photo courtesy of diamant24 and PopTika on Shutterstock.