This article is more than 5 years old. Information may no longer be current.
Drugs approved without supporting trials need more safety-related label changes
Click Here to Manage Email Alerts
Cancer drugs with indications not supported by randomized controlled trials required more adverse event-related modifications to their safety labels, study data showed.
“Driven by an urgency to bring effective therapies to patients with cancer, some drugs with potentially important anticancer activity have been approved by the U.S. FDA without supporting phase 3 randomized controlled trials,” Daniel Shepshelovich, MD, medical oncology fellow at Princess Margaret Cancer Centre and University of Toronto, and colleagues wrote. “The extent of modifications to cancer drug indications, dosing and related toxicities after FDA approval without a supporting randomized controlled trial has not been reported extensively.”
Researchers searched the Drugs@FDA website for new drug indications for solid tumors approved between January 2006 and December 2016. They identified 109 new solid tumor indications for 59 individual drugs.
Shepshelovich and colleagues obtained study characteristics, regulatory pathways and label modifications from drug labels beginning at approval and continuing through October 2017.
Researchers used logistic regression to compare indications approved with and without randomized controlled trials, and they used the Benjamini-Hochberg false discovery rate method to adjust for multiplicity.
Investigators considered modifications “major” if they were described as such on warning labels.
Seventeen indications (15.6%) were not supported by randomized controlled trials.
Unsupported indications appeared more likely than those with randomized controlled trial support to need companion diagnostic tests (OR = 3.9; P = .02).
Unsupported indications also appeared more likely to include surrogate endpoints as primary outcomes (OR = 7.88; P < .001), be designated as breakthrough therapies (OR = 7.62; P = .006) or receive accelerated approval (OR = 17.67; P < .001).
Indications not supported by randomized controlled trials appeared significantly more likely to require modifications in common adverse events after approval (71% vs. 29%; OR = 5.78; P = .002).
Indications not supported by randomized controlled trials also appeared more likely to require major modifications to warnings and precautions after approval (88% vs. 62%; OR = 4.61); however, the difference did not reach statistical significance.
Supported and unsupported indications required comparable postapproval modifications to indications and usage, dosing and administration, boxed warnings and contraindications.
“Cancer drugs approved by the FDA without supporting randomized controlled trials were associated with more postmarketing safety-related label modifications relating to adverse events,” the researchers wrote. “Health care professionals should be aware of this and should practice increased vigilance when using such drugs in the early approval setting.” – by Andy Polhamus
Reference:
Shepshelovich D, et al. J Clin Oncol. 2018;doi:10.1200/JCO.2017.77.5593.
Disclosures: Shepshelovich reports no relevant financial disclosures. One author reports honoraria from Apobiologix.