Dave J.-

I think you are misinterpreting the 1 in 22 comment from R.A. Fisher. He was stating his convenient choice for liking P=0.05 for a cutoff was based on it coinciding with a differerce very close to 2 standard deviations from the mean, and that meant the expectation would be that at that low P value, 1 in 22 trials could be concluded a real effect. It was stated in 1925, when people had to look up statistics in tables, and so the correlation with 2 standard deviations was somewhat a matter of convenience. To him, though, it represented a reasonable risk. This 1 in 22 has absolutely nothing to do with translation to humans, as the P value calculation is simply a mathematical tool to compare two datasets to conclude if they are different.

A far as the other question considering history, the probability would be zero, or very low if you consider Rilutek to apply. But, even if the tool is blunt and has historically low translation success, I don't know if we can blame only the tool. I am optimistic that it is also related to the difficult nature of ALS, and us not testing the right therapy. Think of it this way: Should we also throw away human clinical trials of ALS because of the lack of success?


Wayne-

You caught me. I was just trying to make a point, that the evolved definition of statistical significance is somewhat arbitrary. Many think we should stop using 5% and simply state the confidence level for others to decide for themselves, if it is statistically significant. Especially so, when we now, 86 years later, have computers to calculate this quickly. The value should always be given, not a simple mention of which side of the line. How close was it??? (e.g. Apocynin discussions).

For us PALS, with our lives on the line, I am just saying that the line shouldn't be so absolute. I suspect some even forget the real meaning. P=0.06 and P=0.04 are really close, yet one happens to fall on the right side of a threshold. I don't know for sure that ALSTDI decided to abandon just because of the p value. But, just as you infer a lot regarding something like statements and not data, I am doing the same regarding the many instances of reading "was not statistically significant". And, when I know statistical significance is VERY effected by sample size, I would hope that those that are close to the cutoff can have a repeat (or enlarged) test to see if the difference is REAL. This gets back to whether the trials are even powered sufficiently to ever have small P values with a small difference. I would love to calculate this myself, and am only missing one input to do so, standard deviation of mouse survival. Certainly ALSTDI has this data and could quickly tell us the power to detect various size effects with their standard test protocol.

This statistical power and sample size concern is partially fueled by this:

From Supplementary Table 4 of the Nature Genetics Report: Additionally, all studies were undertaken using the criteria delineated in Scott et al. 2008 which indicates a group size of N=20 or 24 has over 90% power to detect a 15% effect, >85% power to detect a 10% effect.

To me, this implies they were expecting to see larger differences, using numbers like 15% effect and 10% effect. I realize their actual sample size may be larger (unless we segregate genders). Given the history, it would seem more relevant to discuss the power to detect 1.5%, 3% and 5% effects.

BTW- Can we call a truce, my friend? I'll concede that you can make it go higher. LOL

---

per·se·vere [pur-suh-veer] - to persist in anything undertaken; maintain a purpose in spite of difficulty, obstacles, or discouragement; continue steadfastly.

Wayne is more then just a pretty face. But you held your own, Pers.
He has longevity on his side, .
I believe they should immediately stop all single drug experiments , for they have proven to be failures. You just can't keep repeating failures. I'll say one thing . They are persistent . But at our expense. This ancient way of thinking has cost us dearly.
Multiple drug trials on mice are just as easy as single drug trials..who cares which drug, agent, or product caused the hit? As long as you get a hit . Isn't that what we are searching for?
As far as 846 ? I believe it's not worth bringing to trial and that's why alstdi hasn't done it. For them to fail would be catastrophic as far as confidence goes in the research and ALS community. But if they used it in a cocktail with something else? And maybe they are?
But who are we to suggest they change their ways? They will have another crop to work with soon . They will never run out of subjects that's for sure.

Sorry, Persevering, I wasn't very clear on what I meant there.

The historical application of the P criterion to ALS mouse trials is the one-drug non-wonder versus the placebo. To an ALS patient, "statistical significance" has an entirely different meaning, like "what's the chance this does something for me?"

So, let's suppose we have a hypothesis, that mouse trials of the kind customarily done produce more of significance for human ALS victims than sitting watching another episode of Survivor on TV. As it turns out, we already have a huge data set, lots more than the Magic 22, and the results are in: the results of the mouse trials versus the couch spud controls are indistinguishable.

And nobody should even be surprised at such overwhelming evidence of worthlessness to human ALS victims. After all, the mouse trials are never attempts to find out if it's possible to help the poor mouse! With nobody trying to help the mouse, nobody's succeeding in helping the mouse, and therefore the mouse trials produce nothing that might be a springboard for effective human ALS therapy.

And as long as there's no success to report, "the matter requires further research" and 10 years from now we'll be about where we were 10 years ago. The research establishment continues to have jobs, drug company investors who rely on the research establishment to exercise good judgment on how to make research produce something useful are being ripped off, and ALS patients are being ripped off.

And then there's the ALS patient trials....... just like the mouse trials. The medical establishment isn't trying to find out what can be done to help ALS patients, so they have nothing useful to report on that front. They spent a lot of research money feeding the establishment, and that's the only success they thought necessary.

--Dave J.

Thanks Rob..I thought you might have been testing 846 with other drugs.. I can only hope you continue to do this. You are bound to cause a effect sooner or later , as you mix more and more prospects..THANK. YOU...

Yes. Rob, thanks for monitoring this thread and commenting. We can be a critical group. I know many of us have offered more criticism than positive sugestions. I still believe in the tool and am still intrigued by the rare results of 846.

A friend gave me access to the Sean Scott et al paper, Design, power, and interpretation of studies in the standard murine model of ALS. Having only begun to delve into it, it is obvious that it addresses many of the questions raised throughout this thread.

I hope to share some highlights of new revalations to me, later.

Thanks again!

---

per·se·vere [pur-suh-veer] - to persist in anything undertaken; maintain a purpose in spite of difficulty, obstacles, or discouragement; continue steadfastly.

One quick interjection about Apocynin - the paper disussing mouse studies at ALS TDI offered that mice typically arrive to them at 35 to 45 days old. This correlates with the dosing beginning about 2.5 weeks later than the 14 days at University of Iowa, if on the early end of typical.

And, while discussing days when dosing began, the Ceftriaxone studies mentioned in the paper had dosing beginning at 84 and 90 days in 2 separate studies. It was also completed at 200 mg/kg. Current phase III human dose is 4 g, irreregardless of body mass. But, this would be 88 mg/kg to 44 mg/kg for people weighing 100 to 200 lb (45 to 90 kg), so a very different dosage between those preclinical tests and the current human clinical test. (But it appears ALS TDI may have retested, since the median survivals in their Research Archive do not match.)

Rob and the other fine folks at ALS TDI - Will you please consider adding just a little more detail to available study results, such as dosage and time at first dose? I think it is natural to assume the results posted have dosages the same as we'd see in clinical studies, and it seems important to know if true. I think we all assume all results were from the same study parameters, so it would be informative to know when there are exceptions. Thanks!

---

per·se·vere [pur-suh-veer] - to persist in anything undertaken; maintain a purpose in spite of difficulty, obstacles, or discouragement; continue steadfastly.

This is a good thread with a lot of good questions:

First, to address specifics about study designs and data generated in specific mouse studies at ALS TDI

Apocynin
Harraz et al. tested apocynin in B6-SjL SOD1-G93A mice. These are the same strain of mice in which apocynin was tested at ALS TDI.
Harraz et al reported a 113 day life extension in these mice using 300 mg/kg/day of apocynin in drinking water starting at age 14 days. They also reported an 81 day life extension using 150 mg/kg/day of apocynin in drinking water starting at age 14 days, a 65% life extension. This was the dose that was repeated at ALS TDI. The only difference between our 150 mg/kg/day study and their 150 mg/kg/day study is that we started our study dosing at approximately 30 days of age rather than 14 days of age. I cannot speak to the weening practices by Harraz et al., but at 14 days old, mice are not yet consuming “drinking water”. They are nursing. Typically mice will nurse to 18-21 days old. Unless the mice were being treated with apocynin via maternal milk, then the actual difference in duration of treatment may have been as little as 9 days between the two studies. It is not outside the realm of possibility that 9 days difference in treatment onset could produce approximately an 80 day difference in survival effect, but I find that unlikely.
Before initiating our apocynin studies, we contacted Harraz et al. directly to try to confirm that formulations techniques were identical and we confirmed stability of the molecule in our formulation by LC/MS.
We saw no statisitically significant effects on survival in our study of apocynin.
We saw no statistically signifiant effects on
We did observe a statistically significant effect on onset of paresis in male mice. This is the small effect referenced in the press release about the our presentation at the Chicago neuroscience meeting.


Minocycline
Numerous groups have reported efficacy by minocycline in SOD1 mice when treatment begins before symptom onset. I’ll have to review each of the published pdf’s, but I know that at least one of these reports was based on sorely underpowered study designs that did not account for gender or litter. These studies suffered from the early years using the G93A-SOD1 mouse when researchers did not yet know the variability inherent in the model and had not reached a consensus that minimum standards were necessary for interpretable survival results.
ALS TDI did test minocycline with a dosing regimen that commenced prior to symptom onset. In our hands, minocycline was not efficacious and, like the subsequent clinical trial, actually demonstrated untoward effects. Because I know how we ran the studies and I have the utmost confidence in our methodologies with the SOD1 mouse (the ALS worldwide community has largely acknowledged our methodologies as the gold standard now), I am reasonably confident that minocycline can be deleterious in SOD1 mice whether delivered before or after onset of overt symptoms.

---

Next, I’ll begin to address the question of dosing before or after disease onset in the SOD1 mouse for pre-clinical testing.


Imagine for a moment a scenario where a person is diagnosed immediately with ALS following the very first symptoms. These symptoms might include something as seemingly trivial as hyperreflexia. In the 40 days immediately following this aknowledgement of hyperreflexia, each and every motor unit (motor neuron connected to muscle fiber bundle) is completely ravaged with disease. In the first week, a 180 lbs patient loses 20 lbs of muscle mass and is very fatigued. Three weeks later, this 180 lbs patient is down to 130 lbs and is in a wheel chair. One week after that, the patient is 120 lbs and no longer able to breathe on their own.

In the scenario above, there is virtually no grey area between “pre-symptomatic” and diseased. There are no healthy motor neurons to be protected by a therapeutic. There is not opportunity to slow down a disease that is ravaging each and every motor neuron and muscle fiber virtually simultaneously during a 40 day period. This is the fate of the SOD1 mouse.

However, even in the most aggressive cases of ALS in humans that I have encountered, none have rivaled what I describe above. In most cases, patients enter the clinic with symptoms affecting one or two limbs or they present with bulbar symptoms. The truth is that often times, as much as 90% of an ALS patient’s motor units are completely functional and “pre-symptomatic”. Many of you have lived or are living the relentless degeneration caused in ALS and you know that it creeps along robbing motor unit after motor unit of its ability. I am interested in protecting those functional “pre-symptomatic” motor units. By beginning treatment presymptomatically in the SOD1 mice, we have an opportunity to identify drugs that might slow or stop the relentless march of ALS against each functional unit. To limit initiation of treatment to the first signs of overt symptoms in these mice does not provide the best opportunity to identify treatments that can help many patients living with ALS right now.

Of course, if a drug shows efficacy when we “give it our best shot” by treating prior to symptom onset, we will try to test the limits of the therapeutic by testing it at onset, like we did with 00846. Additionally, if we have a certain understanding or hypothesis about the molecular biological sequence of events in the SOD1 mice and our therapeutic is hypothesized to impact a 10 day window in the disease course of the mouse, we also design experiments accordingly. However, in order to test a given therapeutic in multiple paradigms, we sacrifice the ability to test other equally promising therapeutics. We are resource constrained, but do our very best.

Finally, I would like to take a moment to commend my friend and colleague, Dr. Albert Ludolph and his close associate Caterina Bendotti along with the whole European Consortium of ALS/MND researchers for taking an active role in implementing guidelines that were in large part defined by the work of Sean Scott, James Heywood, and others here at ALS TDI during our early years. I have been at both European Consortium meetings organized by Dr. Ludolph. He and his European colleagues are very receptive to data driven dialogue about best practices for pre-clinical testing in the ALS research space.

---

Fernando G. Vieira, M.D.
Director of In Vivo Operations
ALS Therapy Development Institute

Dr. Vieria-

Thanks for the great explanation of the SOD1 mouse model and the analogy to an equivalent human living 130 days.

Regarding Apocynin, I hope you may also give us the median survival change and the actual P value from each the two early dosing studies, rather than simply state "no statistically significant effects on survival".

Thank you!

---

per·se·vere [pur-suh-veer] - to persist in anything undertaken; maintain a purpose in spite of difficulty, obstacles, or discouragement; continue steadfastly.

Hi Dr Vieira,

Thanks so much for your great reply. Could you tell us please which other drugs or compounds you have trialled which have accelerated disease progression. I was very interested to hear that you saw accelerated disease progression with minocycline. Were you expecting that and what reasons did you conclude were behind this? Would you consider a trial with 000846 in a model with lower copy number? It's amazing that you only have a 10 day window to hit the jackpot. Does CD40L activity just keep increasing after this time in mice and is that reflected in humans throughout the disease process?

Mary

Hey Mary!

To answer your question, I have to say that I have no idea! Maybe an e-mail to Dr. Vieira would help?

Wayne