Over the past two decades, the field of medical genomics underwent nothing less than a revolution in terms of both technological advancement and accumulated knowledge. This revolution holds the promise of changing the entire medical practice, and while the industrycontinues to improve genome sequencing technologies and decrease the price of sequencing a genome, other challenges are lurking that hinder the prospects of this revolution and undermine the efforts of wide-scale integration of genomics into mainstream medicine.

To emphasize this point further, even though the technologies to help diagnose patients with rare genetic diseases exist, the rate of underdiagnoses and misdiagnoses is still alarmingly high, and patients who receive diagnoses end up waiting too long for them, sometimes years. These extensive diagnostic journeys directly impact the ability to recruit patients for clinical trials, and thus the ability to develop more treatments for rare diseases. To date, only 5% of more than 7,000 known rare genetic diseases have FDA-approved treatments.

At my company, a leading digital health company, our mission is to end the diagnostic odyssey for undiagnosed pediatric patients with rare diseases. I've seen that the main contributors to this state of affairs are the excruciatingly long wait times for genetics appointments, coupled with the significant workforce shortage of experts in the field.

To reach more than 400 million patients globally (50% of whom are estimated to be young kids) with earlier intervention to improve outcomes and help many of them live relatively healthy and productive lives, the diagnosis must shift from the geneticists clinic to the primary point of care, or at least it must be initiated much earlier by primary care physicians.

Without adopting technological solutions that will support the integration of genomics into mainstream medicine, genomics will never live up to its promise and become a standard of care. In my opinion, realizing that vision will be a balancing act between the affordability for payers, accessibility for providers and actionability for patients, and it will depend on technological solutions combining AI-based phenotyping, as well as connecting front-line providers with human experts in genetics, alongside the most advanced genome sequencing technologies.

High Throughput Genetic Testing

As noted, genome sequencing technologies have made huge strides over the last two decades. The affordability of genomics is now increasingly dependent on the ability to sift through and interpret vast amounts of data produced from a genome, and to determine which data is pertinent for a medical diagnosis and for disease treatment a task fitting for AI.

Indeed, in the last few years, we have witnessed many AI-driven solutions sprouting to address this problem. Some of these solutions are home-grown, in leading laboratories such as Invitae, GeneDx and PerkinElmer Genomics. (Full disclosure: PerkinElmer Genomics uses FDNA's technology in its genetic analysis.) Others are developed as software platforms by vendors such as Sophia Genetics, Fabric Genomics, Congenica and Emedgene.

Harnessing AI to perform data analysis challenges has proven to be very successful and is a direct contributor to the affordability of genetic testing today, as well as the gradually increasing rate of reimbursement by payers. I believe AI will continue to play a key role in driving down prices to the $100 range, which will make genomics extremely affordable, both for health systems and for individuals paying out-of-pocket.

Phenotyping Driven By AI

AIs impact goes far beyond applying machine learning algorithms that sift through genetic variations and proprietary knowledge bases of pathogenicity. As more OMICS technologies stack up with genomics, and more AI modalities like natural language processing and computer vision image analysis are integrated directly into the genome analysis pipeline, we will see an increasing standardization of data across disparate data silos and a closing of the genotype-phenotype gap between the clinic and the lab. This trend will drive genomic data to become more actionable for patients and allow them to make informed decisions about their health.

Much of todays phenotyping is performed by humans and is inherently subject to biases such as age, gender and ethnicity. If we approach this problem with legal and ethical rigor, care and are cautious of patient privacy, and with respect to the providers and their workflow, AI could enhance human skills and capabilities. I think that helping primary care providers collect, structure and analyze phenotypic information of patients with rare diseases is an area worth prioritizing.

Connecting The Expert Community

Finally, technology is more than AI. Technology is also an enabler for fostering connections and interactions between humans. Some tasks in practicing medicine must be left to humans, but even then, technology can assist. An alternative abbreviation of AI (augmented intelligence) is my preferred one. It implies a symbiotic relationship between people and machines, making each other stronger, rather than threatening to replace each other.

Tailoring a solution combining all three components (genomics, AI-based phenotyping and community connection) like the one described above is not an easy task, and it depends on the ability of stakeholders from many disciplines to work together, share data and collaborate on research and development.

To achieve this, a best-of-breed approach should be taken, and not only should data be shared, but a global collaboration between commercial companies, academic research institutions and caregivers should occur. The integrity of the data, ethical and privacy policies, and trust in workflow should be established. This requires an open dialogue between all parties involved, as well as a fast-pace framework to allow developers to move quickly in building these tools.

Certainly, working with different stakeholders with sometimes conflicting interests is challenging, but the one common goal we all have is helping patients, especially kids with rare and undiagnosed genetic diseases.

Read more here:
Access And Actionability Are Key For Genetic Testing And Precision Medicine - Forbes

Comments are closed.