In September 2015, the General Assembly adopted the 2030 Agenda for Sustainable Development that includes 17 Sustainable Development Goals (SDGs). Building on the principle of “leaving no one behind”, the new Agenda emphasizes a holistic approach to achieving sustainable development for all. Perhaps it is time to add #18 to the list – No Disease Orphan – 2030.

In India, within the frameworks of both the National Policy for Rare Diseases published in 2020 by the Government of India and the recently announced National Digital Health Mission on August 15th 2020, it is clear that the focus of the government is on technologies for better diagnoses and care delivery by leveraging digital platforms at population scale. The question of how all this focus on technologies can help build solutions for rare and orphaned diseases remains to be worked out. Let me suggest two examples of how progress can be made.

1. For example, the efforts at the University of Toronto through the spinoff company Deep Genomics is a case study that needs to be examined closely. Deep Genomics uses deep learning platform Spidex, that provides “a comprehensive set of genetic variants and their predicted effects on human splicing across the entire genome.” Its first success story appears to be a treatment for Wilson disease in patients who possess Met645Arg, a mutation shown to lead to the loss of function of the ATP7B copper-binding protein. DG12P1 is the oligonucleotide therapy that caused approximately 70% skipping of exon 6. That exon skipping in turn resulted in frameshift and stop gain, which is expected to cause loss of ATP7B function. It is important to note that the approach taken by DART and its sister concern Hanugen can now be scaled to have a much larger impact using these methods.
2. Syndromic screening of orphan diseases can be facilitated by use of software like Phenomizer (Kohler et al.,) to carry out differential diagnosis and generate a gene list for a patient case. Phenomizer accepts as input a collection of phenotypes that a clinician has collected from a physical/clinical examination of the patient and uses the rich disease-phenotype knowledge in the Human Phenotype Ontology62 (HPO) to generate a ranked list of diseases and genes. Or we could use DeepGestalt (Gurovitz et al.) or Pedia (Krawiz et al.) based on deep learning methods, which are exciting recent developments in the field of phenotyping that can possibly supplement the physical examination step. The facial analysis frameworks are powered by a model trained on a large collection of images representing more than 200 rare syndromes. The model can then be used to analyze a patient’s photograph and generate a candidate list of diseases and genes, thereby adding efficiency to the manual phenotyping step in many cases.
3. To make real progress towards this goal #18 “No Disease Orphan -2030” we need to promote an active dialogue among stakeholders in orphan diseases with a view to create a better world for persons with these disabilities. Fortunately, there has been a growing awareness around orphan diseases and many stakeholders representing global, national and local interests in orphan diseases are active today with many information resources in web portals and foundations supporting diagnostic and therapeutic advances. Most important are the extraordinary heroic efforts by patients and their caregivers, such as the group at DART, that give us hope that the audacious goal of no diseases orphan is attainable.