Concern and skepticism
Other clinicians who regularly prescribe these drugs expressed concern about the observation as well as skepticism about the prevalence of patients who could potentially experience similar effects.
Catherine Hackett/MDedge News
Dr. Steven E. Nissen
“We don’t know how common are these genetic abnormalities. If this is extremely rare, then it doesn’t have many clinical implications, but if a large portion of the population has this [genetic] abnormality, it’s something we’d need to pay attention to,” Steven E. Nissen, MD, chair of cardiovascular medicine at the Cleveland Clinic Foundation, said in an interview. “It will be important to know the prevalence” of the genetic polymorphism carried by the reported patient, said Dr. Nissen, who has done research on lipid-lowering medications and drug safety.
“This could be important, or a very rare one-off. I can’t say which,” based on what’s currently known, he said. “There are many unanswered questions that make it hard to know how important this will be. It requires further investigation. There is a lot of uncertainty.”
Dr. Nissen particularly endorsed studies that approach this issue by looking at the prevalence rates of the implicated genetic polymorphism rather than pharmacovigilance studies that make epidemiologic assessments of adverse-effect prevalence. Studies that look for adverse-effect associations in large patient populations are “sloppy, and unless the interaction is incredibly intense they are not very sensitive,” he said.
But Dr. Juurlink, a pharmacoepidemiologist whose specialty includes studies of this sort, said that they could be useful if carefully designed. He suggested, for example, comparing in large patient databases the observed incidence of rhabdomyolysis among patients on an SGLT2 inhibitor and also on rosuvastatin with those on pravastatin, a statin with a different metabolic profile. Another approach to further examining the observation would be dosage studies with rosuvastatin and canagliflozin in healthy volunteers, he said.
Dr. Nissen noted that rosuvastatin is a key agent from the statin class because it’s the “most effective” for lowering low density lipoprotein cholesterol. “Rosuvastatin is a go-to drug,” he declared. On the other hand, canagliflozin is “a little less used” than other drugs in the SGLT2 inhibitor class, specifically dapagliflozin (Farxiga) and empagliflozin (Jardiance), he said.
One in a million?
“This was a freak accident. I don’t find it at all concerning. It was definitely one in a million,” Yehuda Handelsman, MD, an endocrinologist and diabetes specialist who is medical director of The Metabolic Institute of America in Tarzana, Calif., said in an interview. “None of us have seen it” in either the several cardiovascular outcome trials now run on multiple drugs in the SGLT2 inhibitor class that included many patients also taking a statin, or in routine practice, he said. Dr. Handelsman noted that in his practice he had never seen a similar case despite treating “hundreds if not thousands of patients” with type 2 diabetes, virtually all of whom were on a statin and were also treated with an SGLT2 inhibitor, including many with canagliflozin.
Dr. Yehuda Handelsman
Dr. Handelsman cited the notably low estimated glomerular filtration rate in the reported patient, who was described as having a serum creatinine level of 150 mcmol/L (1.7 mg/dL) prior to canagliflozin treatment that then rose to 194 mcmol/L (2.19 mg/dL) at the time of hospitalization, which corresponds to estimated glomerular filtration rates of 29-31 and 21-23 mL/min per 1.73 m2, respectively, depending on the calculator used, rates that were possibly below the labeled minimum rate of 30 mL/min per 1.73 m2 for patients starting canagliflozin treatment. The case report cited the patient as having stage 3B chronic kidney disease, which corresponds to a eGFR of 30-44* mL/min per 1.73 m2.
“I think the patient had acute kidney injury” on starting canagliflozin “that may have affected the [rosuvastatin] metabolism,” Dr. Handelsman suggested. “She had severe kidney dysfunction to start with that fell further with SGLT2 inhibitor treatment,” a well described and usually transient effect of starting drugs in this class because of changes the SGLT2 inhibitors cause in renal blood flow. He noted that the patient had not been receiving an angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker, which may have contributed to her acute problems with fluid balance. Most similar patients with type 2 diabetes, cardiovascular disease risk, and chronic kidney disease would be on stable treatment with a drug that inhibits the renin-angiotensin system before starting an SGLT2 inhibitor, and not already having a RAS inhibitor on board before starting canagliflozin may have somehow contributed to the observed adverse effects, Dr. Handelsman said.
Dr. Juurlink was skeptical that the kidneys played a major role. “An abrupt change in renal function can influence statin clearance, but this was a 15-fold increase. You can’t explain such a dramatic increase by a transient reduction in renal function,” he said.
Dr. Juurlink and coauthors had no disclosures. Dr. Nissen had no relevant disclosures. Dr. Handelsman has been a consultant to companies that market drugs in the SGLT2 inhibitor class.
SOURCE: Brailovski E et al. Ann Intern Med. 2020 Aug 3. doi: 10.7326/L20-0549.
*Correction: This value was missing from the original article.