New Method Speeds Diagnosis of Rare Genetic Disease

A new laboratory method developed by researchers at Columbia University Vagelos College of Physicians and Surgeons may now help physicians more quickly diagnose patients with suspected genetic disorders of the immune system, many who have been trapped in diagnostic limbo for years.  

The researchers, who published their findings June 20 in Cell, applied the method to one rare inborn error of immunity called activated-PI3Kδ syndrome (APDS) and found dozens of additional genetic variations that could cause the syndrome. As a result, they also found that the disease was substantially more common than previously assumed.  

“Our findings give physicians a resource that can help them rapidly diagnose and treat patients and avoid cumbersome assays and long diagnostic odysseys that delay treatment,” says study co-leader Benjamin Izar, the Vivian and Seymour Milstein Family Associate Professor of Medicine. 

“Many more people with diseases caused by their immune system could now benefit because there are genetic causes that can be detected, and effectively treated,” adds study co-leader, Joshua Milner, a professor of pediatrics and director of the Division of Pediatric Allergy, Immunology, and Rheumatology. 

“For APDS patients, rapid diagnosis is particularly critical because there is an effective, FDA-approved precision therapy available,” says Zachary Walsh, an MD/PhD student in the Izar lab who conducted much of the research. 

The findings have a real-time impact on patients’ lives: one patient has already received a diagnosis of APDS, which causes a wide range of health problems, including infections, autoimmune disease, and increased risk for certain cancers at a young age. This patient is now receiving a precision therapy for APDS, a drug called leniolisib that targets the aberrantly functioning protein. 

“There are more patients to find,” Izar says. 

Based on the success with APDS, the Columbia researchers, together with their colleague Dusan Bogunovic, also from the Department of Pediatrics at Columbia, are now looking to apply their method to other genetic diseases that affect the immune system. 

“There are so many diseases we could do this for, and hopefully it's just the tip of the iceberg,” Walsh says.   

The problem of interpreting genetic test results in medicine 

The method developed by the Columbia team was designed to test the functional impact of genetic variants in the two known genes that cause APDS. 

Patients are diagnosed with APDS only when genetic testing reveals a variant that is already known to cause the syndrome, which makes patients eligible for leniolisib, the only targeted treatment for APDS.  

But genetic testing isn’t always clear-cut. For every variant that's known to cause APDS, there are hundreds of variants of uncertain significance, or VUSs, that have not been classified because their functional impact is unknown. 

“The problem is we don’t know whether a VUS is relevant to the person’s condition or just reflective of normal differences from one person to another,” Milner says. “VUSs pose a major challenge, since it’s possible that they could be the cause of the patient’s symptoms and point to a therapy. But until we know that they cause cells to have the abnormality seen in APDS, the patient is in limbo.” 

New method helps lift genetic uncertainty 

To speed the functional evaluation of VUSs in APDS, the Columbia researchers used a CRISPR base editor to make thousands of mutations in the APDS genes and then measured the impact of each of those genetic changes on healthy human T cells in the lab. Variants that caused too much of a certain function in the T cells were classified as gain-of-function, which is the type of consequence seen in APDS. Further clinical observation can enable their classification as pathogenic.  

“What made our study so powerful was our ability to create thousands of variants in the genes, whether they had previously been encountered in patients or not,” says Walsh. “By proactively classifying variants, even before they're found in patients, we hope we can get out ahead of the VUS problem.” 

“Beyond rare disorders, this method could usher in an era of the Human Genome Project Version 2, where we not only describe whether or not a variant exists, but begin to understand whether such genetic variation, either alone or in combination, has an impact on a given phenotype,” Izar says.  

APDS may be more common than we thought 

Only a few hundred Americans are thought to have APDS, but based on a search of hundreds of thousands of genomes, the new study suggests that APDS could be magnitudes of order more common than previously estimated, possibly affecting one in every 10,000 Americans. 

The researchers found potential gain-of-function variants in about one of every 5,000 Americans in the All of Us precision medicine program, which has sequenced the genomes of more than 250,000 Americans. Some people who carried the variants had signs and symptoms consistent with APDS recorded in their health records but had not been diagnosed with the syndrome. 

“These people may have had milder symptoms but because of this finding, they could potentially benefit directly from the targeted treatment,” Milner says. “Physicians need to be more aware of the signs and symptoms seen in those carrying these types of variants, so that far more patients can undergo genetic testing and be diagnosed.” 

“These findings also hint that a lot of ultrarare or rare genetic diseases may not be as rare as we think,” says Walsh. “The framework we developed for APDS could be applied to many other diseases, both rare and more common, to identify more pathogenic variants and more patients and get a better sense of the true prevalence of these diseases.” This point is critical, adds Milner, because “we are entering an era where the genetics point directly to precision treatments, even when symptoms vary in type or severity.”