The Pathway to Discovery and Hope for ALS Patients and Researchers: A massive biological genetic and clinical resource for ALS and neurodegenerative diseases.

Every year, 5,000 Americans are diagnosed with amyotrophic lateral sclerosis (ALS), also known as Lou Gehrig’s disease, that robs them of their ability to walk, talk, swallow, and even breathe. Although scientists have identified more than 30 genes linked to ALS, 90% of those diagnosed have no known cause for their condition. With a wide heterogeneity in the age of onset and clinical course, scientists have struggled to identify the molecular causes of ALS that can lead to impactful therapies.

Five years ago, the Robert Packard Center for ALS Research at Johns Hopkins assembled a group of researchers and patient advocates to launch the Answer ALS initiative to fix this. The goal was simple, and ambitious: to deeply study the molecular characteristics of a patient’s nervous system cells and link these findings to detailed clinical data to identify subtypes of ALS. These subtypes and other insights gleaned from the data, coupled with advanced artificial intelligence methods, will help to develop targeted treatments for everyone affected with the disease, says Jeffrey Rothstein MD, PhD, (the founder of the Answer ALS research program) Director of the Answer ALS Research Program and Founder and Director of the Robert Packard Center for ALS Research. This could yield a truly personalized approach to ALS, he says.

“With these answers, I believe we can begin to identify new therapies and the individual patients they are most likely to benefit,” says Rothstein.

Today, Rothstein and a group of more than 100 other scientists published the massive Answer ALS resource paper in the journal Nature Neuroscience. The article outlines essential details of the large patient population, their aggregate clinical and genetic characteristics, and the diverse biological data sets generated from the patients. Uniquely, the biological data was generated not from traditional and historically relatively low yield sources like serum, but from actual spinal cord cells, known as induced pluripotent stem cells (iPS) individually derived from each of the enrolled ALS patients. More than 1000 iPS cell lines formed the basis of this biological phenotyping.

As the program was massive, industrial in size, the paper outlines the careful attention to iPS cell generation and the data quality and reproducibility critical data that other researchers will need in order to utilize this information in their own work. Notably, all the data and the iPS cell lines are to be freely usable and shared amoung academic and commercial researchers worldwide. For Brian Wallach, who was diagnosed with ALS in 2017 and founded the patient advocacy group I AM ALS, this openness and sharing is the point.

“I hope that when people look at Answer ALS, they see the opportunity to use the data and that they seize that opportunity. It’s not an endpoint, it’s the beginning,” he says.

The idea for Answer ALS first arose in 2013 in a conversation with Steve Gleason, a former safety for the New Orleans Saints, who was diagnosed with ALS. He believed that a large scientific collaboration would enable experts to begin to understand more about the molecular basis to ALS. Rothstein immediately saw the potential. With some quick, back of the envelope calculations, Rothstein estimated that if he could gather around 1000 patients, he might have enough data to begin to group patients into different subsets based on underlying biological and clinical similarities.

Gleason and Rothstein were especially excited about the possibility of created iPS cell lines for each patient. For conditions like cancer, doctors can take a small piece of affected tissue and do a biopsy, thereby learning more details about a patient’s disease. But biopsies of the brain and spinal cord are too dangerous, meaning that Rothstein and other scientists studying neurodegenerative diseases like ALS have to rely on animal models to understand the disease. The use of iPS cells, Rothstein says, solves that problem.

“The motor neurons and other spinal cells grown from these cells are like a biopsy. We can study ALS in a patient’s own cells, without the limitations of animal models,” Rothstein says. This is especially important for those with sporadic ALS, which has no animal models.

Wallach is one of more than 1000 individuals with ALS who participated in Answer ALS. The 41-year-old donated blood cells (which were genetically reprogrammed to create iPS cells) and downloaded a special app to his phone. The app allowed him to log data on his breathing, speech, and cognition that would accompany more formal measurements at clinic visits.

The blood cells from Wallach and others were shipped to Clive Svendsen, PhD of Cedars-Sinai Medical Center in Los Angeles, California to be reprogrammed into iPS cells and then matured into motor neurons. Differences both large and small can occur between different batches of iPS cells and any mature cells that result, so creating multiple internal controls was crucial to allow Svendsen’s team to ensure the results were identical no matter when they had processed the cells. Scientists ordering stem cell lines will receive not only the cells they ordered but also batch of technical and differentiation controls to account for variability. “The optimization process was lengthy and arduous,” Svendsen says, “but well worth it.”

“To be honest, nothing has ever been done at this scale. I think we’re probably one of the first to take this on. Most papers only have maybe three or four lines, so to have 1000 along with all the matching biological and clinical data is a pretty big deal,” Svendsen says.

Svendsen’s team then shipped the differentiated motor neurons across town to the lab of Leslie Thompson, PhD at the University of California, Irvine, where she directed a team to carry out a range of transcriptomics, proteomics, and epigenomics on these cells. These analyses will enable scientists to dive more deeply into the molecular basis of ALS. To Thompson, -omics data will play a key role in identifying subtypes of ALS and developing better therapies.

“We’re getting close to finding variables hidden in the data that we intend to publish soon,” Thompson says. “There are some disease subtypes that are emerging.”

With each Answer ALS participant creating more than 1 terabyte of data, information management is a crucial part of the process. Over 2.6 trillion data points have already been generated, and the team estimates this number will reach 20 trillion by the end of the project. With so much information, the team of Ernest Fraenkel, PhD at MIT has turned to machine learning and artificial intelligence to wring coherence from the data tsunami.

Fraenkel’s team also spent months determining the best way to label samples that would enable the ready transfer of important clinical and metadata with each sample but do so in a HIPAA-complaint fashion. The Answer ALS Data Portal connects all this information and was designed to be an easy-to-use resource freely available to scientists, both in academia and industry, around the world.

“We want people to freely use this data,” Rothstein says. “This paper is us introducing it to the world.”

From Wallach’s perspective, this paper is a major breakthrough in ALS research. “This work showed it was possible to enroll a large number of patients and gather data from them over a long period of time. That had never been done before in ALS,” he says. “It shows the possibility of what we can do in the field.”

Unit leading co-authors include: Emily Baxi, PhD, Answer ALS Program Director; Nicholas Maragakis, MD at Johns Hopkins Medicine and James Berry, MD, MPH at Massachusetts General Hospital, who co-coordinated the clinical aspects of the program; Clive Svendsen, PhD of Cedars-Sinai Medical Center in Los Angeles, who coordinated the stem cell research; Leslie Thompson, PhD of the University of California, Irvine, who oversaw the -omics analyses; Jennifer Van Eyk, PhD who has carried out the proteomics; Ernest Fraenkel, PhD at MIT and Steve Finkbeiner, MD, PhD at the Gladstone Institutes who analyzed and managed the volumes of data; Raquel Norel at IBM Watson, who developed machine learning algorithms to analyze data collected from a smartphone app; and Terri Thompson, PhD, data and portal director for Answer ALS.