Having now read the E & P Neuren update report (dated 03/02/23) and the pre-print manuscript discussed in that report, I’ll offer my opinion. Please note, it's just my opinion, and that opinion is certainly not expert.
In the new E & P report, the analysts have downgraded trofinetide’s probability of NDA approval (from 95% to 85%) “mostly due to concerns around the validity of the RSBQ co-primary endpoint for assessing clinical severity “.
While the analysts acknowledge that criticisms of the reliability of the RSBQ are not new, and point out that they have raised the issue in earlier reports, a new manuscript draft has recently come to their attention. This manuscript draft analyses whether the RSBQ is a valid measure for use in both adult and paediatric Rett patients and considers whether a novel measuring system might be more reliable. The manuscript has been authored by numerous highly respected Rett experts and is yet to be published in
American Journal on Intellectual and Developmental Disabilities.It’s not just medical experts who question whether a more sensitive and reliable measuring system could be developed. Rett parents have also expressed frustration with current trial measurement systems.
Mel Lancaster is quoted in the Voice of the Patient Report as saying“The scoring methodologies didn’t really capture the true improvement that she (Mel’s daughter)
had in so many different areas.… So the measuring system that’s being used, yes, we need one, but there has to be another way to capture the true improvements.”The RSBQ, which was used in the Lavender trial, measures 45 items, grouped into 8 subscales (General Mood, Breathing Problems, Hand Behaviours, Repetitive Face Movements, Body Rocking and Expressionless Face, Night-time Behaviours, Fear/Anxiety, and Walking/Standing) whose ratings reflect severity and frequency of Rett symptoms. A total, representing the sum of the 45 items (maximum score 90), and 8 subscale scores are obtained.
The authors of the draft manuscript referenced in the E & P report propose a novel, revised RSBQ measure, in which just 6 subscales are used for paediatric patients and 7 for adult patients. They argue that this revised measure has a higher clinical relevance and a better fit than the current 8 subscale RSBQ measure and conclude that their proposed novel measure is “suitable for use in research and clinical practice.” In other words, this draft manuscript is essentially an argument for adoption of a new, revised RSBQ measure devised by the authors and proposed as a suitable measure for use in Rett clinical trials.
It’s worth noting that the E & P analysts stress that, irrespective of this argument for a revised RSBQ scale, all the Rett experts they consulted remained optimistic that trofinetide will be cleared, given the FDA's familiarity with this metric and the lack of an objective outcome measure.
Focusing for a minute on trofinetide : -
- Trofinetide met not just the RSBQ in its pivotal Phase 3 trial, which was agreed with the FDA prior to the trial, but also the clinician-rated CGI-I that the FDA insisted on.
- Efficacy in the Lavender trial was seen across all 8 RSBQ subscores (see below).
- The RSBQ endpoint used in the trial was the Total score. The manuscript authors state that the RSBQ Total scale has "good psychometric properties".
- As Jon Pilcher points out, the manuscript recommends some reorganisation of the RSBQ subscales, but that doesn't impact the Total scale.
- As Jon Pilcher also points out, neither the manuscript nor the E&P report says anything about Acadia submitting or having to submit any new analysis.
- As the E & P analysts state, all experts they consulted remained optimistic that trofinetide will be cleared, given the FDA's familiarity with this metric and the lack of an objective outcome measure.
- The manuscript puts forward an argument for a proposed, new, revised RSBQ measure. This measure has not been used to date prospectively in a clinical trial and may or may not be acceptable to the FDA as a pivotal clinical trial endpoint measure.
View attachment 5048638As has been noted here already, the lead author of this draft manuscript is Dr. Walter Kaufmann, a well-known expert on Rett syndrome. Dr. Kaufmann was promoted to the position of Anavex’s Chief Scientific officer 12 months ago. Just three weeks after his appointment, Anavex’s somewhat controversial Phase 3 results in Rett syndrome were released.
Anavex’s 33 patient adult Rett Phase 3 trial was originally planned as a Phase 2 trial and was changed midway. In addition, the original primary endpoint of the trial was shifted from the change from baseline in RSBQ (as was used as a co-primary endpoint in the trofinetide Phase 3 trial) to AUC responder analysis of the RSBQ score. This change was only advised on the FDA’s clinical trial database two weeks prior to the announcement of the trial results. Anavex originally gave the impression that they might file an NDA on the strength of this trial but now indicate that they are waiting for data from their ongoing Phase 2/3 paediatric trial.
I’ve already observed that Anavex has used the RSBQ as the
only efficacy primary endpoint for both their Phase 3 adult Rett trial and their ongoing Phase 2/3 paediatric Rett trial. That’s notable because the FDA insisted that Acadia use not only the RSBQ, but also a co-primary, physician-measured, CGI-I endpoint in the Phase 3 Lavender trial.
So why the E & P analysts’ statement -
The experts we consulted are not aware of a revised subscale analysis of RSBQ utlilised in the FDA submission?
I think it’s just a poorly worded statement (I make them all the time!). What I think the analysts were trying to say was that if Acadia had chosen to include a revised subscale analysis of the RSBQ in their NDA submission, it was not something that any of the experts knew anything about.
The other question that comes to mind is why, suddenly, the appearance of this draft manuscript?
Its appearance, during the FDA’s trofinetide NDA review, might just be random timing. Perhaps its significance is that Anavex is preparing the ground for this proposed, revised RSBQ to be somehow used in their NDA submission for Anavex2-73 in Rett. Or perhaps, if you tend towards conspiracy theories, someone is desperately hoping that trofinetide stumbles on the home straight, leaving Anvex2-73 to race through first to the finish line.
Whatever the answer to either question….I’m not concerned.
(20min delay)