"I would urge you to cultivate a keenly skeptical attitude toward the pharmacopeia as a whole…" Sir William Osler, 1907
Indeed, even a seemingly harmless topical ointment may pose unanticipated lethal risks which are undisclosed on the FDA-approved label.
A concerned dermatologist, a former FDA medical officer, brought to our attention the finding of a 6-year randomized clinical trial conducted by the Veterans Administration involving 1,131 elderly veterans. 566 were randomized to test tretinoin cream (0.1%), and 565 to placebo. Tretinoin, a member of the retinoid class of medications is commonly used for the topical treatment of acne and fine lines and wrinkles. It is marketed as Retin-A and Renova, as well as other branded products.
The purpose of trial was to establish the effectiveness of Retin-A as a chemoprevention intervention for nonmelanoma skin cancer. The trial failed to demonstrate effectiveness as a cancer prevention treatment. Instead, there were statistically significant, unanticipated extra deaths among those applying tretinoin, compared with those given a placebo: the trial was terminated 6 months early (in 2004).
In 2005, the authors published preliminary results in abstract form in the Journal of Investigative Dermatology, stating that there was NOT a statistically significant difference in overall mortality between tretinoin and placebo groups: "We conclude tretinoin did not cause the mortality difference between groups, and in retrospect the termination of the intervention was unnecessary."
Four years elapsed before the study results were finally published in January in Archives of Dermatology http://archderm.ama-assn.org/cgi/content/full/145/1/18
According to published report, there were significant differences in mortality between retinoid users and those on placebo:
Table 1. Cumulative Deaths among 1131 Study Participants,shows that the retinoid treated group consistently had an excess number of deaths compared to the placebo group: by April 2004, there were 135 deaths: 82 (14%) deaths in retinoid group compared to 53 (9%) placebo. Before the end of intervention in 2004 the deaths climbed to 184: 108 (19%) deaths in retinoid group compared to 76 (14%) placebo. And by the end of study (follow-up phase) there were 212 deaths: 122 (22%) deaths in retinoid group compared to 90 (16%) placebo.
Table 3 Cause of Deaths reveals that greatest disparity in deaths between the retinoid group and placebo involved:
lung cancer (15 deaths vs 8), respiratory thoractic disorders (15 vs 7), and vascular disorders (12 vs 3)–when deaths from Arteriosclerosis and Atherosclerosis are added, the vascular risk increases to 22 vs 5.
The study investigators attribute the difference to chance.
However, the accompanying editorial (below) suggests caution:
"While debate will continue regarding whether the association between topical tretinoin and death found in the VATTCresulted from chance or a real biological effect, until additional data from other studies emerge, practitioners should view the results of the VATTC with discretion. Public health ideally uses the precautionary principle—that possible harm should be avoided before harmful effects are unquestionably proven.8 At a minimum, this principle should cause prescribing physicians to discuss the results of the VATTC with elderly men using topical tretinoin."
And a published letter (below) cites earlier controlled retinoid trials with similar results:
"A causal link between tretinoin and mortality due to lung cancer or other lung diseases is consistent with previous RCT data. Specifically, the Alpha-Tocopherol, Beta Carotene Cancer Prevention Trial 7 and the Beta-Carotene and Retinol Efficacy Trial8 both linked vitamin A–related compounds to lung cancer. Ironically, both trials were intended to demonstrate that these compounds could prevent lung cancer. In both studies, however, lung cancer rates in subjects taking vitamin A–related substances were, unexpectedly, significantly higher than in subjects taking placebo, leading to early discontinuation of the vitamin A–related interventions in both trials."
The FDA has shown utter indifference: taking no action whatsoever to alert prescribing physicians to the potential lethal risk. The tretinoin labeling fails to reflect the results of the VA study.
The principle investigator, Dr. Martin A. Weinstock, MD, PhD, discloses receiving financial support from: received support from Galderma Laboratories, Johnson & Johnson, and Ligand Pharmaceuticals. The manufacturer of tretinoin, Ortho-McNeil–Janssen Pharmaceutical is a division of Johnson & Johnson.
Posted by Vera Hassner Sharav
Arch Dermatol. 2009;145(1):18-24Topical Tretinoin Therapy and All-Cause Mortality
Martin A. Weinstock, MD, PhD; Stephen F. Bingham, PhD; Robert A. Lew, PhD; Russell Hall, MD; David Eilers, MD; Robert Kirsner, MD, PhD; Mark Naylor, MD; James Kalivas, MD; Gary Cole, MD; Kimberly Marcolivio, MEd; Joseph Collins, ScD; John J. DiGiovanna, MD; Julia E. Vertrees, PharmD; for the Veterans Affairs Topical Tretinoin Chemoprevention (VATTC) Trial Group
Objective: To evaluate the relation of topical tretinoin, a commonly used retinoid cream, with all-cause mortality in the Veterans Affairs Topical Tretinoin Chemoprevention Trial (VATTC). The planned outcome of this trial was risk of keratinocyte carcinoma, and systemic administration of certain retinoid compounds has been shown to reduce risk of this cancer but has also been associated with increased mortality risk among smokers.
Design: The VATTC Trial was a blinded randomized chemoprevention trial, with 2- to 6-year follow-up. Oversight was provided by multiple independent committees.
Setting: US Department of Veterans Affairs medical centers.
Patients: A total of 1131 veterans were randomized. Their mean age was 71 years. Patients with a very high estimated short-term risk of death were excluded.
Interventions: Application of tretinoin, 0.1%, or vehicle control cream twice daily to the face and ears.
Main Outcome Measures: Death, which was not contemplated as an end point in the original study design.
Results: The interventionwas terminated 6months early because of an excessive number of deaths in the tretinointreated group. Post hoc analysis of this difference revealed minor imbalances in age, comorbidity, and smoking status, all of which were important predictors of death. After adjusting for these imbalances, the difference in mortality between the randomized groups remained statistically significant.
Conclusions: We observed an association of topical tretinoin therapy with death, but we do not infer a causal association that current evidence suggests is unlikely.
Trial Registration: clinicaltrials.gov Identifier: NCT00007631
~~~~~~~~~~
Archives of Dermatology. 2009;145(1):18-24
EDITORIAL: Dealing With Unanticipated Mortality in a Large Randomized Clinical Trial of Topical Tretinoin
. . . I would urge you to cultivate a keenly skeptical attitude toward the pharmacopeia as a whole. . . . Sir William Osler, 19071
Millions of people have used topical tretinoin to treat acne and photoaging. [2-4] In this issue, Weinstock et al 5 report an unexpected result that led to the premature halt of their randomized, placebo-controlled Department of Veterans Affairs Topical Tretinoin Chemoprevention(VATTC)trial designed to determine whether topical tretinoin, 0.1%, cream prevents basal and squamous cell skin cancer in US veterans.
Limited topical tretinoin use (up to twice daily to the face and ears) was associated with increased death (end-ofintervention hazard ratio, 1.54; P=.01).
While some argue that topical tretinoin might increase pulmonary-related mortality,6 Weinstock et al5 ultimately reject this explanation and chalk their results up as a chance finding based on the following: (1) minimal systemic absorption of topical tretinoin; (2) the lack of a dose-response association; (3) the lack of specificity of causes of death; and (4) the lack of a statistical interaction between tretinoin use and smoking with mortality.
(On this last point, asbestos provides the classic epidemiological example of an exposure producing a statistical interaction with smoking: smoking and asbestos exposure together result in greater risk of lung cancer than the sum of the risks from the individual exposures.)7 While debate will continue regarding whether the association between topical tretinoin and death found in the VATTCresulted from chance or a real biological effect, until additional data from other studies emerge, practitioners should view the results of the VATTC with discretion.
Public health ideally uses the precautionary principle—that possibleharmshould be avoided before harmful effects are unquestionablyproven.8 At a minimum, this principle should cause prescribing physicians to discuss the results of the VATTC with elderly men using topical tretinoin. More circumspect practitioners may wish to discuss the results of the VATTC with all patients using topical tretinoin. This dialogue should include that the results of theVATTC may have beendueto chance, but also that the outcome of death wasnot initially anticipated, and owing to the adhoc analysis, various important risk factors, such as smoking status, might not have been completely ascertained. These discussions provide an opportunity for all health care providers prescribing tretinoin to emphasize tobacco prevention and cessation with their patients.9
Clearly, future trials of topical retinoids, especially in elderly patients and in current and former smokers, should mortality with validated, sound methods. With increased mortality as a foreseeable outcome, rigorous, predetermined stopping rules and appropriate statistical methods will ensure that trials are not halted prematurely owing to invalid (but essential) interim analyses.10-13 Finally, unlike other investigators who have not emphasized unexpected mortality data,14,15 we highly commend Weinstock et al 5 for reporting and highlighting these results.
Correspondence: Dr Dellavalle, Dermatology Service, Department of Veterans Affairs Medical Center, 1055 Clermont St, Box 165, Denver, CO 80220(robert n .dellavalle@uchsc.edu ).Financial Disclosure: None reported.
Funding/Support: This study was supported in part by Colorado Health Informatics Collaboration Academic Enrichment Funds from the University of Colorado Denver School of Medicine (Drs Schilling and Dellavalle) and NCI K07 Cancer Prevention and Control Career Development Award grant K-07CA92550 (Dr Dellavalle).
Additional Contributions: Martin Weinstock, MD, PhD, answered additional questions via e-mail regarding the manuscript.
REFERENCES
1. Osler W. The reserves of life. St Marys Hosp Gaz. 1907;13:95-98.
2. Weiss JS, Ellis CN, Headington JT, Tincoff T, Hamilton TA, Voorhees JJ. Topical tretinoin improves photoaged skin. JAMA. 1988;259(4):527-532.
3. Swinyer LJ, Swinyer TA, Britt MR. Topical agents alone in acne: a blind assessment study. JAMA. 1980;243(16):1640-1643.
4. Goldfarb MT, Ellis CN, Voorhees JJ. Topical tretinoin: its use in daily practice to reverse photoageing. Br J Dermatol. 1990;122(suppl 35):87-91.
5. Weinstock MA, Bingham SF, Lew RA, et al; Veterans Affairs Topical Tretinoin Chemoprevention (VATTC) Trial Group. Topical tretinoin therapy and all-cause mortality. Arch Dermatol. 2009;145(1):18-24.
6. Katz KA. Topical tretinoin, lung cancer, and lung-related mortality. Arch Dermatol. 2008;144(7):945-946.
7. Liddell FDK. The interaction of asbestos and smoking in lung cancer. Ann Occup Hyg. 2001;45(5):341-356.
8. Goldstein BD. The precautionary principle also applies to public health actions. Am J Public Health. 2001;91(9):1358-1361.
9. Dellavalle RP, Johnson KR. Time for all dermatology societies to adopt smoke free policies. J Am Acad Dermatol. 2006;54(1):149-150.
10. Goodman SN. Stopping at nothing? some dilemmas of data monitoring in clinical trials. Ann Intern Med. 2007;146(12):882-887.
11. DeMets DL, Pocock SJ, Julian DG. The agonizing negative trend in monitoring of clinical trials. Lancet. 1999;354(9194):1983-1988.
12. DeMets DL. Futility approaches to interim monitoring by data monitoring committees. Clin Trials. 2006;3(6):522-529.
13. Nissen SE. ADAPT: The wrong way to stop a clinical trial. PloS Clin Trials. 2006; 1(7):e35.
14. Bombardier C, Laine L, Reicin A, et al; VIGOR Study Group. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343(21):1520-1528.
15. Curfman GD, Morrissey S, Drazen JM. Expression of concern reaffirmed. N Engl J Med. 2006;354(11):1193.
~~~~~~~~~~~~~
COMMENTS AND OPINIONS
Topical Tretinoin, Lung Cancer, and Lung-Related Mortality
By Kenneth A. Katz, MD, MSc, MSCE
Amid continuing controversies over drug safety,1,2 results of a trial of topical tretinoin—a commonly used medication for acne3 and skin wrinkles, hyperpigmentation, and roughness4—raise serious concerns for the public health. The Veterans Affairs Topical Tretinoin Chemoprevention (VATTC) trial5,6 was a vehicle-controlled randomized controlled trial (RCT) that studied whether topical tretinoin, 0.1%, cream applied to the face and ears could prevent nonmelanoma skin cancer.
As reported in an abstract published in 2005,6 the study observed 1131 subjects for at least 2 years. After 6 years, and about 6 months prior to the study’s scheduled conclusion, a safety monitoring committee stopped the study because of excess mortality among subjects who applied tretinoin (n=82 deaths [14%]) compared with those who applied vehicle (n=53 [9%]) (P=.01). Differences in mortality from pulmonary disease (12 vs 4) and non–small cell lung cancer (NSCLC) (11 vs 4) were reported.5
A causal link between tretinoin and mortality due to lung cancer or other lung diseases is consistent with previous RCT data. Specifically, the Alpha-Tocopherol, Beta Carotene Cancer Prevention Trial 7 and the Beta-Carotene and Retinol Efficacy Trial 8 both linked vitamin A–related compounds to lung cancer. Ironically, both trials were intended to demonstrate that these compounds could prevent lung cancer. In both studies, however, lung cancer rates in subjects taking vitamin A–related substances were, unexpectedly, significantly higher than in subjects taking placebo, leading to early discontinuation of the vitamin A–related interventions in both trials.
A link between tretinoin and lung-related mortality is biologically plausible, with the putative culprit not tretinoin itself but harmful tretinoin metabolites. This line of association begins with the finding that topically applied tretinoin can be absorbed systemically9 and therefore can reach lung tissue. Once inside cells, tretinoin can induce its own metabolism; continuous dosing with tretinoin may lead not to higher levels of tretinoin but to higher levels of tretinoin metabolites.10 It is those tretinoin metabolites that may injure lung tissue, particularly in the presence of cigarette smoke. This was demonstrated in a study that exposed ferrets to beta carotene (a vitamin A precursor) or cigarette smoke or both or neither for 6 months; lungs of all ferrets exposed to beta carotene showed a strong proliferative response and squamous metaplasia that was enhanced by exposure to cigarette smoke.11 A hypothesis linking lung cancer to adverse effects of tretinoin metabolites is also supported by the finding that patients with certain types of NSCLC (squamous or large-cell carcinomas) metabolize tretinoin more rapidly than patients with another NSCLC type (adenocarcinoma) or patients without lung cancer.12 This study raises the possibility that rapid metabolizers of tretinoin may be more likely to develop lung cancer because, compared with normal metabolizers, they have a relative deficiency of tretinoin and a relative excess of injurious tretinoin metabolites present in their lung tissue.
Additionally, the link between tretinoin and lung disease may be multifactorial; others have proposed that tretinoin- mediated downregulation of defensins in lung tissue contributed to lung-related mortality in the VATTC trial.13
It is not clear whether tretinoin caused the excess lungrelated deaths in the VATTC trial. But concern is warranted, certainly, both because a causal link is plausible and because topical tretinoin is indicated for the treatment of relatively minor conditions.
Correspondence: Dr Katz, 1360 Mission St, Ste 401, San Francisco, CA 94103 ( n Kenneth.Katz@post.harvard.edu ).
Financial Disclosure: None reported.
REFERENCES:
1. Committee on the Assessment of the US Drug Safety System, Baciu A, Stratton K, Burke SP, eds. The Future of Drug Safety: Promoting and Protecting the Health of the Public. Washington, DC: National Academies Press; 2006.
2. Drazen JM, Morrissey S, Curfman GD. Rosiglitazone–continued uncertainty about safety. N Engl J Med. 2007;357(1):63-64.
3. US National Library of Medicine. Retin-A (tretinoin) Cream, Retin-A (tretinoin) Gel, Retin-A (tretinoin) Liquid [ORTHO DERMATOLOGICAL]. http:
//dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?id=3734. Accessed July 8, 2007.
4. Renova [package insert]. Raritan, NJ: Ortho Pharmaceutical Corporation. http://www.aboutrenova.com/RENOVA.05.pdf. Accessed July 8, 2007.
5. Weinstock MA, Marcolivio K, Bingham S, et al. Topical tretinoin and allcause mortality. J Invest Dermatol. 2005;124:A52.
6. Department of Veterans Affairs. Determine the efficacy of topical tretinoin cream for the prevention of nonmelanoma skin cancer.
http://www.clinicaltrials.gov/ct/show/NCT00007631?order=1. Accessed July 22, 2007.
7. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med. 1994;330(15):1029-1035.
8. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease.
N Engl J Med. 1996;334(18):1150-1155.
9. van Hoogdalem EJ. Transdermal absorption of topical anti-acne agents in man; review of clinical pharmacokinetic data. J Eur Acad Dermatol Venereol.
1998;11(suppl 1):S13-S19.
10. Rigas JR, Francis PA, Muindi JR, et al. Constitutive variability in the pharmacokinetics of the natural retinoid, all-trans-retinoic acid, and its modulation by ketoconazole. J Natl Cancer Inst. 1993;85(23):1921-1926.
11. Wang XD, Liu C, Bronson RT, Smith DE, Krinsky NI, Russell M. Retinoid signaling and activator protein-1 expression in ferrets given beta-carotene supplements and exposed to tobacco smoke. J Natl Cancer Inst. 1999;91(1):60-66.
12. Rigas JR, Miller VA, Zhang ZF, et al. Metabolic phenotypes of retinoic acid and the risk of lung cancer. Cancer Res. 1996;56(12):2692-2696.
13. Rosenberg EW, Skinner RB Jr. Topical retinoids: another piece for the retinoidcigarette-lung cancer puzzle? J Thorac Oncol. 2006;1(7):732.