When I first got exposed to ALS.net’s Precision Medicine Program (PMP) in the fall of 2014, what intrigued both Beth and I the most was the aspect that the lab would have Beth’s motor neuron cells in a dish and be able to see how they responded to a wide variety of drugs and biologics. We thought: “What a wonderful way to find a likely path to a cure for her.” Too good to be true? It turns out that this program, and others like it, are likely to be important in more ways than one; not only may it help find a cure for Beth, but for all patients who are living with this horrific and complex disease.
The Food & Drug Administration (FDA) evaluates drugs and biologics in a manner that hasn’t changed in almost 100 years (Here is a great video from my fellow ALS.net Ambassador, Matt Bellina on this). By and large the current FDA approval process works very well for diseases where we have a thorough understanding of the disease’s biologic process. The process is pretty straight-forward; do a randomized double blind placebo controlled clinical trial and then compare the results of the treated group to the placebo group looking to see if there is greater than a 5% chance that the treated group result could be replicated in a placebo group. This probability is referred to as a "p value." When you have a disease that creates the same observed result in all afflicted, and an objective result measure (a biomarker), this technique works very well, especially in large size trials.
However, with ALS no two patients progress in identical fashion. ALS is neither a homogeneous disease nor is there an accepted biomarker of ALS disease progression. That creates a real problem in using the so-called FDA “gold standard” when measuring a drug’s effect on disease progression. Researchers and doctors have tried to address this problem by expanding trial size hoping the law of large numbers will solve the problem of heterogeneity. However this just leads to larger, more costly and longer trials. Since Big Pharma, which are the only players who can possibly afford this process, largely view ALS as the graveyard for biopharma investment, we are left with a variety of smaller, less well capitalized companies who create promising offerings that make it through tiny Phase 1 safety trials and Phase 2a trials and then nothing happens while they try to raise capital for the larger trial. There are countless examples of this in ALS now, several during just the time that Beth has been diagnosed!
Not a pretty picture – hence the name “graveyard of biopharma.” But let me digress for a moment. Being an almost 60 year old male and hopefully without one foot in the grave, I tend to wake up sometimes in the middle of the night – let’s call it the "Hebron p value." The odd thing I often realize when I wake up this way is that a seemingly unsolvable problem that has pre-occupied me the previous day is in fact solvable. This usually causes me to stay up the rest of the night, mind speeding up by the moment, pondering the solution.
This has happened a few times since the fall of 2014 with respect to the PMP that we fund through ALS.net. I know that the purpose of the PMP is to find treatments for people like Beth, but I realized that there may be other applications of the ALS.net funded PMP and certainly others being developed by groups like Answer ALS, Biogen, etc too.
The first thing I realized was that with so much data on so many patients, we should be able to change the way clinical trials are done. With so much data, predictive progression algorithms, like the one that won the Prize4Life contest, could be developed or refined even further and be made “FDA ready.” Once that is accomplished, all future trials could have what I call “virtual placebo arms” where each patient’s progress on the drug would be compared to their predicted progression without the drug and no patient gets a placebo. (Let’s say that again: No patient gets a placebo.) The positive result: no more statistical noise from patients progressing in unique ways; in other words a homogeneous relationship between the treated group and the placebo group because they are comprised of the “same people”. How do you improve the algorithms? Why not run them on the PMP patients’ historical baseline data and then compare their results to the PMP data at a later point in time? Then focus study on the patients for whom the progression algorithm doesn’t work and look for patterns to identify these patients in advance. That knowledge can also translate into trial exclusion data or be used to stratify trial results to exclude the results of those who cause the remaining statistical noise.
I’ve seen the power of predictive algorithms in my own career working at the New York Life Insurance Company. But guess what, you have too, all the time in your Search Engine results. Remember back in the early days when there was no predictive technology to return results when you searched for something on the internet? The list was in alphabetical order or worse! That is what we are doing in ALS clinical trials today in a way, we are starting with the letter A when the treatment we need is way down the list. Predictive algorithms may help speed up the process to find the right letter or groups of letters to hone in on, rather than just going through them in a regular fashion order.
Predictive algorithms aren’t going to solve everything. But, I realized that those running the programs like the PMP at the Institute need to start thinking about their opportunity to advance ALS research and clinical trials by applying predictive algorithms to the massive data sets they are compiling. To do this best it is obvious that data sharing and data unification are things that Answer ALS, Biogen and the ALS Therapy Development Institute, and others, should commit to now. I know that the ALS Therapy Development Institute has already committed to sharing data; heck, they are sharing lots of it in real time with my daughter as part of the program.
The best news I can give is that some of these ideas are capturing the attention of the experts. They are creeping into ALSA Clinical Trial Guidance drafts and biopharma companies are showing interest. And I now sleep with a bottle of water bedside hoping for more sleepless nights and finding yet more value in the P’s of PMP.
And by the way this is why Beth and I and our entire family wholeheartedly support ALS.net. It is the first organization to actually get a PMP program for ALS off the drawing board and start collecting data. It is imperative that the ALS community support this effort and hopefully expand it.
I leave you with one last thought: “Reality is for people who lack the imagination to just enjoy the trip.” Not sure who said this – maybe Timothy Leary…
Here is some of what I've written here in Beth's words: