NEW YORK – A team of academics, pharmaceutical firms, patient organizations, and life sciences companies in Europe has embarked on a new project to improve the diagnosis of patients with rare diseases by pairing genetic newborn screening and machine learning tools.
The effort, called Screen4Care, commenced last October with a budget of €25 million ($28 million), which will sustain it through September 2026. The Innovative Medicines Initiative, a joint undertaking of the EU and the European Federation of Pharmaceutical Industries and Associations (EFPIA), is funding Screen4Care, and the project involves numerous biopharmaceutical firms, including Pfizer, Roche, Takeda Pharmaceuticals, Sanofi, Novartis, ProQR Therapeutics, Lysogene, and Novo Nordisk.
Alessandra Ferlini, head of the medical genetics unit at the University of Ferrara in Italy, is the scientific coordinator of the project, which involves 21 academic partners. Nicolas Garnier, senior director and patient advocacy lead for rare disease global product development at Pfizer, is project lead on behalf of the EFPIA. Other participating industry partners include San Diego's Illumina and Genoox, a Palo Alto, California-based genetic analysis tools company.
According to Garnier, Screen4Care has sought to include a diverse group of contributors, including experts in genetics, bioinformatics, data management and data standards, imaging, ethics, decision-analytic modeling, and data repository management. He noted the consortium also involves the European Rare Disease Patient Advocacy Organization (EURORDIS). By including EURORDIS, Screen4Care will be able to "promote robust dialogue and ensure the needs and preferences of the rare disease community guide progression of the project," Garnier said.
At its core, Screen4Care is about reducing the time it takes for a patient to be diagnosed with a rare disease. Accomplishing this will require a "multi-pronged strategy" based around genetic newborn screening and AI-based tools. In Garnier's words, the use of "advanced genomic technologies," such as targeted next-generation sequencing panels and whole-genome sequencing, should provide an effective tool for early diagnosis of rare diseases in newborns, especially since "approximately 72 percent of rare diseases have a genetic cause and 70 percent of rare disease patients are children."
Here, Illumina will play a role in designing various panels to be used in the project. According to Ferlini, there will be a panel designed to screen newborns for treatable diseases, called Treat, and another one for actionable diseases, called Act. The project will rely on whole-genome sequencing for infants with early-onset symptoms. Altogether, around 18,000 patients are expected to be screened using one of these approaches over the course of the project.
"We are still refining the criteria that we will use to select which rare diseases and mutations will be included in the Treat and Act panels," she said. As part of the design process for the Treat panel, some feasibility considerations include "amenability to next-generation sequencing, penetrance, age of onset, severity, specificity, sensitivity, cost, stakeholder and community feedback, and availability of approved therapy," she said.
For the Act panel, the Screen4Care consortium is also refining the concept of actionability, mainly through community engagement carried out in collaboration with EURORDIS. The alliance last year published 11 key principles to support a harmonized approach to screening.
On the machine learning side, Screen4Care aims to develop new algorithms to flag patients at early disease onset based on their electronic health records, as well as to develop a repository of AI symptom checkers to assist patients as they undertake their diagnostic journeys.
There is also a patient-facing aspect to Screen4Care, Garnier pointed out, which includes establishing digital infrastructure for engaging patients, parents of newborns, and caregivers as "equal decision makers" in the process of receiving a diagnosis.
"The ecosystem will provide an open innovation platform, which allows for continuous data collection and information exchange, aiding the development of next-generation diagnostics and enabling physicians, patients, and relatives to make informed decisions at an earlier stage," said Garnier. The overarching goal is that the effort will "contribute to minimizing disease progression, improved patient health and quality of life, enable R&D, and an optimized use of healthcare resources," he said.
According to Garnier, since Screen4Care kicked off, the consortium behind it has started the foundational work of the project, related to "horizon scanning, community engagement, designing the gene panels, strategic mapping of data sources," and the establishment of stakeholder forums and boards, such as its scientific advisory board and patient advisory board.
"An ongoing effort is to forge bilateral communication lines with other initiatives and organizations with similar or overlapping goals, to find synergies and avoid duplication of effort," he noted.
Other initiatives are indeed underway across Europe. In the UK, the Welsh branch of the National Health Service began whole-genome sequencing of critically ill newborns and those with unexplained conditions in 2020. Also underway in the UK is the Newborn Genomes Program, which is exploring the idea of offering whole-genome sequencing to all infants to identify rare diseases early on.
Ferlini noted that the Screen4Care initiative has just started its activities and is "willing to widely collaborate with other ongoing newborn screening initiatives."
Common objectives
Companies and organizations taking part in Screen4Care are numerous. In addition to the heavy pharma presence, researchers are participating from universities and hospitals in the Czech Republic, Denmark, Germany, Ireland, Italy, Spain, Sweden, and Switzerland. Tools providers, such as Illumina and Genoox, are also central to the effort. The latter is perhaps best known for its Insights platform, which combines AI with a large, real-word genomic evidence dataset to produce, for example, clinical reports or to simplify genomic data as part of research efforts.
Moshe Einhorn, chief technology officer at Genoox, said the firm will support the genetic data analysis and bioinformatics needs of Screen4Care. As a first step, the company will support the design of the newborn screening panels with experts, pharma firms, and Illumina, he noted. Once sequencing of patients starts, the company will provide data interpretation and analysis from raw data through generation of clinical reports.
Additionally, Genoox will provide a means for sharing and connecting insights from the data through Franklin, its genetic community platform, to all the stakeholders in the project, Einhorn added. On a more technical level, the company will develop a bioinformatics pipeline to support the interpretation of "challenging regions and variants," such as the SMN1 and SMN2 genes and copy number variants, which Einhorn said are important for newborn screening. The company's AI Prioritization Engine will also be used for interpreting genetic information, prioritizing data, and classifying variants.
A representative for Illumina confirmed the vendor would be taking part in Screen4Care as a sequencing and diagnostic partner. "Illumina will share its genomics expertise and insights to the project to enable early intervention and shorten the difficult diagnostic journey for rare disease patients and their families," the representative said.
The Centre Nacional d'Anàlisi Genòmica (CNAG), based in Barcelona, Spain, is also contributing bioinformatics and data transfer and sharing capabilities to Screen4Care. According to Ivo Gut, director of CNAG's Center of Genomic Regulation, his institution's responsibility will also be to provide whole-genome sequencing for patients with unexplained symptoms, which he said CNAG already has experience with.
According to Gut, the long-term goal is to test gene panels and genomics methodologies for newborn screening in order to understand which is the more appropriate approach to be translated in newborn screening across Europe. "Some rare diseases are currently treatable also by fairly simple solutions, such as vitamin B12 supplementation," observed Gut. "It is so important to achieve their early diagnosis and identification."
Pfizer's Garnier noted that "Screen4Care is a research project aiming to develop a diagnosis framework even when a treatment is not available because it will still be better for patients, and will ultimately facilitate targeted R&D."
He added that his firm will contribute to Screen4Care "both [at] the scientific and leadership level as well as the operational level, with contributions anticipated from multiple members of the Pfizer team." He added that Pfizer's participation is "consistent with its goal to improve healthcare around the world" and of "advancing the diagnosis of rare disease patients everywhere."
It may take time, though. Gut acknowledged that the EU has been seeding innovation with large projects such as Screen4Care in recent years under Horizon 2020 and Horizon Europe, as well as via initiatives like the IMI, but that the benefits from these investments could take decades to materialize. He said, however, that such projects were worth it, likening them to the development of genomics research as a whole.
"If you think about genomics 15 years ago, we had the sequence of the human genome and that was about it," said Gut. "Now we can generate the human genome in two days. Technologically, we are on a different planet."
Similarly, Screen4Care could take 10 or 15 years to fully reach its benefit. "I do believe [this project] has to be done," said Gut. "If we don't do it now, then we'll have to do it in 10 years, and it will take 25 years to make a difference."
For now, Ferlini said that the next immediate steps for Screen4Care include concluding panel design. Disease prioritization will also be finalized shortly, she said.
"This preparatory activity is the prerequisite to start our planned clinical trials and enroll infants," Ferlini said. "We will have five intense working years with the common objective to transform the current diagnostic odyssey rare disease patients often experience into a standardized, fully accessible, and robust diagnostic flowchart by integrating digital health and genetic newborn screening strategies," she added.