In 2009, The New England Journal of Medicine reported the case of an otherwise healthy 2-year-old boy in Canada who died after surgery. He had received a codeine dose in the recommended range, but an autopsy revealed that morphine (a product of codeine metabolism) had built up to toxic levels in his blood and likely depressed his breathing. Genetic profiling revealed him to be an “ultrarapid codeine metabolizer,” due to a genetic variation in an enzyme known as CYP2D6, part of the cytochrome P-450 family.
While codeine is no longer used at Boston Children’s Hospital, it’s this kind of genetic profiling that Shannon Manzi, PharmD, would someday like to offer to all patients—before a drug is prescribed.
Not all people respond the same way to drugs. The results of randomized clinical trials—considered the gold standard for drug testing—often produce a dose range that worked for the majority of the patients in the study. They don’t take people’s individuality into account, and that individuality can dramatically affect drug efficacy and toxicity.
Adverse reactions are more common than you might think. Nationwide, serious adverse drug reactions result in more than 700,000 Emergency Department visits and 120,000 hospitalizations annually. According to the Institute of Medicine’s influential 2000 report, To Err is Human, adverse drug events incur a cost of about $2 billion annually for hospital inpatients alone. (The Institute now believes this figure is an underestimate.)
“If we are aware of the genetic information at the time of medication prescribing, we can often substitute another drug, or in other cases reduce the dosage to avoid the adverse reaction,” says Manzi, who directs the Clinical Pharmacogenomic Service at Boston Children’s and chaired the hospital’s Adverse Events Committee for a decade.
Over the past two years, Manzi and colleagues have initiated preemptive testing for two drug/gene pairs that tend to cause severe toxicity in some patients. One test identifies patients likely to have toxic reactions to thiopurine drugs, used to treat inflammatory bowel disease (IBD) and sometimes to suppress the immune system after transplant. Another test picks up poor metabolizers of warfarin—a serious concern for patients requiring anti-coagulation. An electronic warning pops up when a prescriber tries to order a drug that doesn’t mix well with a patient’s genetic variation.
In a new research study, called InforMED Kids, Manzi and colleagues are taking pharmacogenomics to the hospital’s epilepsy, end-stage renal and IBD programs, where patients typically require multiple medications. Currently, they’ve enrolled more than 90 patients, along with their health care providers, with a goal of studying 1,000 patients. Several other departments at Boston Children’s, including Psychiatry and Cardiology, hope to eventually participate.
As the study gathers data on the effects of different genetic variants on drug responses, Manzi and colleagues hope to build a repository and database in order to develop and validate prescribing guidelines that tailor treatments to patients’ genetic makeup. Few such guidelines now exist.
In the long term, Manzi hopes to not only make pharmacogenomics data portable, traveling with patients as they change providers, but also to scale up the system by including local pharmacies. Also, while testing right now is limited to panels of genes known to be involved in the metabolism of medications, Manzi foresees this being replaced by next-generation sequencing—testing a patient’s entire portfolio of protein-coding genes—as sequencing costs continue to diminish.
“Running an expanded platform, such as a whole-exome or whole-genome sequencing, will soon become cheaper than the single-gene-based test we run today,” Manzi says.
Manzi will be giving a presentation Nov. 13 at the Advances in Pediatric Health Care continuing medical education (CME) meeting in Wakefield, Mass. For a PDF about the event, please click here.