1. Home>
2. Specialties>
3. Dermatology>
4. Summary and Comment

Long-Term Topical Tretinoin and Excess Mortality in Older Patients

Topical tretinoin therapy was associated with increased mortality, but current evidence suggests a causal association is unlikely.

The possibility that nonmelanoma skin cancers might be prevented by use of topical and systemic retinoids has long been of interest, although few data are available on use in patients with extensive actinic damage. Recently, investigators conducted a placebo-controlled, double-blind, multicenter clinical trial to evaluate the efficacy and safety of 0.1% tretinoin cream in subjects at risk for nonmelanoma skin cancers. Sponsored by the U.S. Department of Veterans Affairs, the study enrolled 1131 veterans at six sites throughout the United States (95% men; mean age, 71). Subjects were instructed to apply 0.1% tretinoin cream to the face and ears up to twice daily, depending on tolerance, for 2 to 6 years.

The study was terminated 6 months early at the recommendation of the V.A.’s human rights committee because of excess mortality in the treatment group. At the time the study was stopped, 108 deaths had occurred among the 565 subjects in the tretinoin group, compared with 76 deaths among the 566 in the control group (P=0.01). Given the limited area of application, the disparity in mortality rates was completely unexpected. No association was found between mortality and frequency of application or the total number of tubes used. Significant increases in mortality from vascular, respiratory, thoracic, and mediastinal disorders in subjects applying topical tretinoin were seen, but not from myocardial infarction, stroke, or chronic obstructive pulmonary disease. The authors did not establish a cause-and-effect relation between topical tretinoin and increased mortality.

Comment: Clinical trials of pharmacologic agents can have unexpected results that are difficult to explain but have broader treatment implications. Tretinoin cream has been approved by the FDA and other regulatory agencies worldwide for decades, with no excess mortality observed. However, long-term use in older individuals has never been examined. Given the limited application area, it is unlikely that sufficient tretinoin was absorbed to produce systemic effects. More than 95% of these subjects were male, but real-world use is predominantly by women, for whom the applicability of these findings is unclear. Although a cause-and-effect relation between topical retinoids and mortality seems unlikely, this report makes prescription of these medications precarious, especially for long-term, off-label use by older patients. Evidence-based data should support prescribing decisions, and patients should be informed of these results so that they can decide whether they wish to use tretinoin.

— Craig A. Elmets, MD

Published in Journal Watch Dermatology January 30, 2009

Citation(s):

Weinstock MA et al. Topical tretinoin therapy and all-cause mortality. Arch Dermatol 2009 Jan; 145:18.

• Original article (Subscription may be required)
• Medline abstract (Free)

Schilling LM and Dellavalle RP. Dealing with unanticipated mortality in a large randomized clinical trial of topical tretinoin. Arch Dermatol 2009 Jan; 145:76.

• Original article (Subscription may be required)
• Medline abstract (Free)