Following are the comments that I submitted to FDAALSGuidance@alsa-national.org regarding their draft of the FDA ALS Draft Guidance document:I tried to look at this document with two standards in mind:
1. Would it be helpful to someone developing an ALS therapy?
2. Would it change anything from what the FDA is doing today?
I worry that it achieves neither.
Introduction
The commentary on the ice bucket challenge isn't valuable to drug developers or the FDA (but since you brought it up as the enabler of this document, how much was spent on this document and what was it spent on?).
The reference to the ALS Registry adds no value.
The explanations of the difference between the ALS "guidance" and "guidelines" are just silly. The guidelines meeting was fare more specific and focused than this document. I would hope that you could have simply dropped some of the advice from the guidelines group directly into this document.
"The goal of clinical trial guidelines is to lead to more effective and efficient trials but they do not directly impact the FDA regulatory process." What in this guidance draft will directly impact the FDA regulatory process? I would love for that to be the case, but I just don't see it in what is presented in this draft.
Background
I think that the references to statistics are not needed and the numbers presented are not helpful. Any drug developer seeking guidance from the FDA probably already knows more about the ALS market than those statistics reflect. The "1 of 800" citation is in a paper that uses it with citations to two other papers. Did the references to the demographics come from the CDC's Registry? In that case there are questions as to the completeness and validity of the data. On page 17 there is a reference to 25,000 Americans living with ALS with a citation to a paper that just says it without any further background.
At the bottom of page 14 it boasts of guidance that will produce a "clinical development strategy that will provide the best opportunity to demonstrate a treatment's effectiveness and safety..." I think that isn't the main goal. The main goal here is to save some lives.
Benefit Risk
Words matter. On page 15 we see that people with ALS are "vulnerable" to investigational new treatments supported by only minimal evidence of tolerability, safety, and efficacy. Vulnerable? The whole idea of risk-tolerance is the ability and right to make informed choices. Vulnerable? And then we read about the "rigorous" standards that must remain in place... so that informed decisions can be made that do not unnecessarily "compromise the health and safety of people with ALS." Oh, please. These people are dying a difficult death and don't need such paternalistic attitudes. If we put those words in the FDA's mouth, we're crazy.
On page 16 there are references to oncology patients versus ALS patients attitudes. Does the fact that many cancers have available therapies make this an apples-to-oranges comparison
There is a reference to PatientsLikeMe.com perhaps having an influence on lower percentages of people with ALS taking riluzole. Perhaps a more basic reason is the same underlying reason that people stop using drugs in general -- the cost-benefit proposition. Why blame social media?
At the bottom of page 16 there are comments on exclusion criteria that indicate that patients want them "less strict." Again, words matter. "Less strict" gives an impression that changing exclusion criteria would diminish the science. I don't think that is the case.
The subsequent comments about the difficulty of some with ALS to get to clinical trial sites was a wonderful opportunity to introduce guidance that would permit more telemedicine and remote monitoring of participants in clinical trials. Will the FDA guidance encourage that they will accept more remote data from participants?
The first guidance on page 18 left me scratching my head. What difference does that guidance make from the status quo?
There is a reference to DiPALS vs DPS to question the use of historical controls. My understanding is that there were procedural differences between DiPALS and the original DPS participants and that this isn't resolved science. It's probably not a great example to cite.
On page 19 there is a passing reference to Type II errors. This needs much more attention in my opinion. This is the crux of much of angst over the eteplirsen situation for DMD. If a drug has a reasonable safety profile and the disease is quickly fatal without an effective treatment, the risk of making a Type II error is a problem that needs to be addressed. The FDA should be willing to accept some uncertainty regarding efficacy rather than dismiss drugs that may work on some patients. If we don't discuss this in this guidance document, we will have missed a major opportunity to actually "impact the regulatory process."
Expanded Access and Accelerated Approval
Why in the world were those two things glumped together? The are vastly different. They need to be clarified.
Expanded Access programs have a mechanism that may or may not be attractive to a drug developer.
Accelerated Approval gets products to market faster, and that is very different from EAPS and is far superior from many aspects. Drug developers need to know that the FDA is willing and able to use the Accelerated Approval path aggressively, and that should be spelled out very clearly in this guidance document.
Trials
There is a lot of description of standard interventions affecting ALS prognosis, but are these interventions ever considered to be standard-of-care issues for clinical trials? Some trials exclude people on NIV or feeding tubes. Should FDA guidance be standing up for these standard interventions to be a patient right as standard-of-care?
On page 71 you actually used the "gold standard" term. Last year I got the message from a gentleman with ALS that the gold standard isn't even the "gold standard" any longer. Words matter. Let's drop the baggage that makes us think that there is only one way to do clinical trials well.
Things I Didn't See That I Wish Were Included
There was discussion at the guidelines meeting that perhaps biomarker trials should be separated from efficacy trials. Are we slowing down efficacy trials by adding companion biomarker components? Are we putting trial participants at risk by tracking biomarkers during washout periods as was done in the NP001 Ph 2 trial?
Is there any mention in this document on rejecting placebo procedures that pose risks to patients? Many people find placebo surgeries draconian. It would be refreshing for the FDA to say that they expect alternatives to placebo surgeries or other procedures that would put placebo participants at risk.
Is there anything in this document that would have changed the big dexpramipexole trial design?
Many ALS trials start measuring volunteers from the day they enter the trial -- as if their ALS started on that day. Others have short lead-in periods to gather some "history," and that lead-in actually delays access to the investigational drug. It seems to me that historical medical data from clinical trial participants could be very helpful in determining whether an investigational drug is actually doing something in certain patients. It would be helpful if the FDA guidance could specify acceptance of some lead-in data from trial participants medical histories.
I think that trial data using ALSFRS-R should always include the components of the score. Perhaps the more granular data might yield some insights that the summary score does not reflect.
The Last Word
There were 38 references to "subjects" in this document.
"Subjects" is a disrespectful word for the most important people in clinical research. It objectifies them. It conjures up images of royal underlings.
Many pharmaceutical companies eliminated the word from their vocabularies long ago.
We need to eliminate the word from ALS research, and it has no place in an FDA Guidance document. If anything, the FDA should tell drug developers that it expects that they will not use the offensive term "subjects" for people who volunteer for clinical studies.
Thank you for the opportunity to submit comments.
