Detection algorithm predicts best drugs for genetic disorders and cancer

A computer model built by researchers from the Institute for Research in Biomedicine (IRB Barcelona) and the Center for Genomic Regulation (CRG) can predict which drugs will be most effective in treating diseases caused by mutations that can interrupt protein synthesis, resulting in unfinished proteins.

The results, published today in Genetics of naturemarks an important step in personalizing treatment by matching patients with specific mutations to the most promising drug candidate. The predictive model, a publicly available resource called RTDetective, can accelerate the design, development, and efficiency of clinical trials for many different types of genetic disorders and cancers.

Truncated proteins are the result of a sudden stop in protein synthesis. In our bodies, this is due to the appearance of “nonsense mutations” that act like a stop sign or roadblock, causing the cellular machinery to suddenly grind to a halt. In many cases, these unfinished proteins stop working and cause disease.

The presence of these stop signs causes one in five single-gene disorders, including some types of cystic fibrosis and Duchenne muscular dystrophy. They also often appear in tumor suppressor genes, which normally help control cell growth. Stop signs inactivate these genes and are a major cause of cancer.

Diseases resulting from truncated proteins can be targeted with nonsense suppression therapies, drugs that help cells ignore or “read” the stop signals that appear during protein production. Cells with higher readthrough rates will produce more full-length or nearly full-length proteins.

The study demonstrates that clinical trials of nonsense suppression therapies to date have likely used ineffective patient-drug combinations. Indeed, the effectiveness of drugs in promoting readthrough depends not only on the nonsense mutation, but also on the genetic code immediately surrounding it.

The researchers made the discovery after studying 5,800 disease-causing premature stop codons and testing the effectiveness of eight different drugs on each. The data came from patient reports submitted to open-access public repositories like ClinVar, as well as research projects like The Cancer Genome Atlas (TCGA), which collected and analyzed genetic information from thousands of patients with cancer and genetic diseases, including premature stop codons.

They found that a drug that works well for one premature stopping signal may not be effective for another, even within the same gene, because of the local sequence context around the premature stopping signal.

“Imagine a DNA sequence as a road, with a stop mutation appearing as a roadblock. We show that navigating through this obstacle depends heavily on the immediate environment. Some mutations are surrounded by well-marked detours while others are full of potholes or dead ends. This is what characterizes a drug’s ability to bypass obstacles and work effectively,” explains Ignasi Toledano, first author of the study and co-PhD student at IRB Barcelona and the Center for Genomic Regulation.

The researchers generated a massive amount of data by testing many different combinations of drugs on stop sign avoidance, resulting in a total of more than 140,000 individual measurements. The data was large enough to train accurate predictive models, which they used to create RTDetective.

The researchers used the algorithm to predict the effectiveness of different drugs for each of the 32.7 million possible stop signs that can be generated in RNA transcripts of the human genome. At least one of the six drugs tested had to achieve a readthrough of more than 1% in 87.3% of all possible stop signs and 2% in nearly 40% of cases.

The results are promising, as higher read rates are generally correlated with better treatment outcomes. For example, Hurler syndrome is a serious genetic disease caused by a nonsense mutation in the IDUA gene. Previous studies have shown that with just 0.5% read rates, individuals can partially mitigate disease severity by creating very small amounts of functional proteins. RTDetective predicted that a read rate above this threshold could be achieved by at least one of the drugs.

“Imagine a patient is diagnosed with a genetic disease. The exact mutation is identified through genetic testing, and then a computer model suggests which drug is most appropriate. This informed decision-making is the promise of personalized medicine that we hope to unlock in the future,” says Ben Lehner, ICREA Research Professor, one of the study’s lead authors and group leader at the Centre for Genomic Regulation in Barcelona and the Wellcome Sanger Institute in the UK.

The study also suggests how to quickly deliver new drugs to the right patients. “When a new drug is discovered, we can use this approach to quickly build a model and identify all the patients who are most likely to benefit from it,” adds Professor Lehner.

The researchers next plan to confirm the functionality of proteins produced by direct-readout drugs, a crucial step to validate their clinical applicability. The team also plans to explore other strategies that can be used in combination with nonsense suppression therapies to improve the efficacy of treatments, particularly in cancer.

“Our study not only opens new avenues for the treatment of inherited genetic diseases, for which direct reading agents have already been tested, but also, and importantly, for the treatment of tumors, since the majority of cancers have mutations that cause premature protein termination,” concludes Fran Supek, research professor at ICREA of the IRB of Barcelona, ​​one of the main authors of the study.