Adèle C. Green, MB, BS, PhD
Cancer and Population Studies Group
Queensland Institute of Medical Research
Brisbane, Queensland, Australia,
and University of Manchester, Manchester, UK
Gail M. Williams, PhD
School of Population Health
University of Queensland
For the last two decades, the question of whether or not sunscreen use can protect against melanoma has been a topic of hot debate. While the notion is appealing and biologically plausible, various reasons why sunscreen use might not protect people from melanoma and might even cause melanoma have been put forward. Most clinicians and researchers on both sides have agreed on one thing, however: that a very large, well-conducted, randomized controlled trial is the only way to provide an answer.
Last year, our research team in Queensland, Australia, published the first clinical trial results regarding sunscreen and melanoma prevention, in the Journal of Clinical Oncology.1 We showed in a Queensland trial that regular use of sunscreen could halve people’s risk of developing melanoma. While the trial was not very large, it was conducted over a considerable period. Its results go a long way towards resolving, if not clinching, the debate, and we hope it will assist clinicians in making recommendations about appropriate preventive care.
Historical Stumbling Blocks
For many years, researchers trying to answer the question of sunscreen’s protective potential against melanoma faced a number of stumbling blocks in trying to design a valid, convincing study. The best evidence available until recently was the data from some less than ideal animal and human studies.
Previous support for sunscreen use in melanoma prevention
Apart from studies specifically focusing on sunscreen and melanoma prevention, basic science strongly suggested that sunscreen could help. We have long known that the majority of cutaneous melanomas are caused by excessive exposure to solar ultraviolet (UV) radiation.2-4 As early as 1957,3 Lancaster and Nelson attributed causation of many melanomas to solar UV. Therefore, shielding the skin from UV with effective sunscreens along with other sun safety precautions would logically appear to be beneficial in long-term reduction of melanoma formation.
Of course, research has shown that people tend to apply sunscreen too thinly and infrequently,5 and for decades, even sunscreens that provided excellent protection against the sun’s shorter, UVB wave- lengths often protected inadequately against the longer, UVA wavelengths. However, by fundamental principles, if proper amounts of effective broad-spectrum UVA/UVB sunscreen were used,2 it could be considered a significant adjunct to clothing, shade, and other physical means of solar UV protection.6 It goes almost without saying that this would hold true only if sunscreen were used in the manner intended — to protect skin from the effects of heightened sun exposure, rather than to encourage and facilitate even greater exposure.7,8
Beyond basic science, supporting evidence for the benefits of sunscreen against melanoma development were scarce, since reviews of case-control and cohort studies of sunscreen use and melanoma risk were uninformative.1,6 Some support, however, did come from a randomized trial of sun- screen application in Canadian children.9 Conducted from 1993 to 1996, it showed a small reduction in new melanocytic nevi, the strongest predictors of melanoma, in children allocated to sunscreen, especially if they had freckles.9 Other affirming evidence of the long-term benefit of regular sunscreen use against melanoma was indirect; several trials had established that regular sunscreen use could reduce the occurrence of other sun-induced skin tumors, specifically actinic keratoses and squamous cell carcinomas (SCC).6,10
Previous arguments against sunscreen use in melanoma prevention
The case against sunscreen use has been made many times in the past.7,11,12 The criticism was based largely on poor non-trial evidence that mainly reflected the lack of reliability in people’s self-reports of past sunscreen use (frequency and/or duration). It also suffered from the profound problem of incorrectly inferring causation from association. In linking an increase in melanoma development with sunscreen use, investigators could not properly account for certain confounding factors – e.g., that patients at high melanoma risk (such as patients with many nevi) might be advised to use sunscreen regularly because they are known to be at high risk, thereby making sunscreen use appear to be associated with heightened melanoma risk.
This has always been the insurmountable stumbling block for non-randomized studies linking melanoma to sunscreen use. They have been unable to distinguish the main drivers of sunscreen use from those of melanoma causation, since they are one and the same: susceptibility to sunburn, high occupational or recreational sun exposure, and family history of melanoma.6 Detractors of these studies have also cited the short latency period allowed in case-control studies between reported sunscreen use and the development of melanoma, with the elapsed time insufficient to allow any valid scientific conclusions about causation or protection.11,12
Another key argument against sunscreen use as a melanoma preventive has been the belief that sunscreen use increases users’ time in the sun, thereby potentially increasing their risk of melanoma. This hypothesis was generated by study groups of dedicated European sunbathers who may have used sunscreen to heighten their sun exposure and consequent skin tanning.
Such groups are not using sunscreen the way it is intended, as an adjunct to other physical means of sun protection. Thus, any increase in melanoma risk among them may come from the misuse of sunscreens rather than the properties of sunscreens per se.6,8
Other potential safety concerns about sunscreen (e.g., photoactivation of certain ingredients, endocrine disruption, and vitamin D deficiency) have also been raised, but most sunscreens now on the market have been evaluated for safety and efficacy, and the safety concerns have been unsupported by direct human evidence of harm.6
The Nambour Skin Cancer Prevention Trial
The Queensland trial results regarding sunscreen and melanoma came as the culmination of 10 years of follow-up after a 4.5-year randomized controlled trial of 1,621 adults aged 25 to 75 who had been randomly selected from all residents of Nambour, a subtropical Australian township. Trial participants were centrally randomized either to apply a freely supplied broad-spectrum sunscreen with an SPF of 16 daily to the head and arms for 4.5 years, or to continue their usual level of sunscreen use or non-use. We followed study subjects for a decade after the end of the active intervention period by sending questionnaires every year asking about new skin cancers, ongoing sunscreen habits, and average time outdoors, and by monitoring regional pathology laboratories as well as the Queensland Cancer Registry for newly reported melanomas. Sun exposure patterns and history of skin cancer were identical in both treatment groups at baseline owing to successful randomization, and proved to be the same in both groups in the decade-long follow-up.1
Ten years after cessation of the sunscreen intervention, the group of 812 people randomized to daily sunscreen use had experienced 11 new melanomas, half the number identified in the control group of 809, who experienced 22 new melanomas. This result was of borderline statistical significance (p=0.051). Eleven of the 22 melanomas in the control group were invasive melanomas, compared with only three invasive melanomas among the daily sunscreen users (73 percent fewer) (p=0.045).1 We are uncertain about why the dedicated sunscreen group had a greater apparent decrease in invasive melanomas than in situ melanomas. Possibly, a proportion of in situ melanomas have a different developmental pathway and do not progress to invasive melanoma; or possibly simple chance influenced the findings, given the relatively small number of melanomas diagnosed in our study over all.
Although participants who had been randomized to use sunscreen daily were more likely to continue to do so after the trial than those in the control group (25 percent versus 18 percent), most of the effectiveness could be attributed to the regular sunscreen application during the trial,1 when 75 percent of participants assigned to daily sunscreen use applied sunscreen at least 3 or 4 days per week, compared with 25 percent of the control group.
It is difficult to say how the use of a higher SPF (about SPF 30 or higher) sunscreen in the Nambour Trial might have affected the results, since the study sunscreen (SPF 16) could filter out around 95 percent of the UVB (compared with 97 percent or higher filtered by SPF 30+ sunscreen). (Separately, being a broad-spectrum sunscreen, it also filtered a proportion of UVA wavelengths in sunlight.) Based on previous research, we know that lower than recommended amounts of sunscreen were undoubtedly applied by Nambour Trial participants on average;5 thus, increasing skin coverage or average thickness of application might have had an increased protective effect. In addition, longer use than the 4.5 years of the trial, or randomizing a much larger number of people to the sunscreen intervention group at the outset, might well have produced stronger results.
Another limitation of the Nambour Trial findings is that they flow from a single trial only: their replication would give further weight to our relatively slim amount of evidence. Furthermore, the trial was carried out only in adults: it would be very reassuring to have more randomized trial evidence of long-term benefit against melanoma for children using sunscreen regularly, even if, like the Canadian study in 2000,9 the study focused on prevention of melanocytic nevi as an intermediate surrogate for melanoma. Third, our trial was carried out in Queensland, Australia, a relatively sunny part of the world where year-round caution in the sun is highly warranted. Thus, we can only presume that the results would also apply to Northern Hemisphere populations in the summer months or during travel to sunny countries (when sun exposure and risk of melanoma development rise).
Recommendations for Patients
What advice should we give melanoma patients based on these results? In the Nambour Trial, participants were asked to apply sunscreen to the head (face, ears, scalp if exposed), neck, arms, and hands every morning, and advised to reapply after heavy sweating, bathing, or long sun exposure. The daily sunscreen users also tended to apply sunscreen more regularly to the trunk and lower limbs than the discretionary (control) group did, and we assume this was the reason for the decrease in melanoma across all body sites, not only on the prescribed application sites. Thus, medical professionals can advise that regular application of a broad-spectrum (UVA/UVB) sunscreen with an SPF of 15+ or higher in this fashion is likely to be beneficial for melanoma protection. Clothing should cover those body parts to which sunscreen hasn’t been applied. Other supplementary techniques including sun avoidance and/or seeking proper shade in high sun exposure settings should also be recommended. We hope that more behavioral research will be carried out to give us deeper insights into facilitators and barriers to regular sunscreen use and other sun protection behaviors, especially in high sun exposure settings among persons highly susceptible to melanoma.
- Green AC, Williams GM, Logan V, Strutton GM. Reduced melanoma after regular sunscreen use: randomized trial follow-up. J Clin Oncol 2011; 29(3):257-263.
- El Ghissassi F, Baan R, Straif K, et al. A review of human carcinogens--part D: radiation. Lancet Oncol 2009; 10(8):751-752.
- Lancaster HO and Nelson J. Sunlight as a cause of melanoma: a clinical survey. School of Public Health and Tropical Medicine, Sydney. Med J Australia 1957; 44(14):452-56.
- Parkin DM, Mesher D, Sasieni P. Cancers attribut- able to solar (ultraviolet) radiation exposure in the UK in 2010. Brit J Cancer 2011; 105:(S66-S69). doi:10.1038/bjc.2011.486.
- Neale R, Williams G, Green A. Application patterns among participants randomized to daily sunscreen use in a skin cancer prevention trial. Arch Dermatol 2002;138(10):1319-1325
- Green AC, Williams GM. Point: sunscreen use is a safe and effective approach to skin cancer prevention. Cancer Epidem Biomar Prev 2007; 16(10):1921-1922.
- Autier P, Boniol M, Doré J-F. Is sunscreen use for melanoma prevention valid for all sun exposure cir- cumstances? J Clin Oncol 2011; 29(14):e425-e426.
- Green AC, Williams GM. Reply to P. Autier et al. J Clin Oncol 2011; 29(14):e427.
- Gallagher RP, Rivers JK, Lee TK, et al. Broad-spec- trum sunscreen use and the development of new nevi in white children: a randomized controlled trial. JAMA 2000; 283(22):2955-2960.
- Green A, Williams G, Neale R, et al. Daily sunscreen application and betacarotene supplementation in prevention of basal-cell and squamous-cell carci- nomas of the skin: a randomised controlled trial. Lancet 1999; 354(9180):723-729.
- Berwick M. Counterpoint: sunscreen use is a safe and effective approach to skin cancer prevention. Cancer Epidem Biomar Prev 2007; 16(10):1923- 1924.
- Berwick M. The good, the bad, and the ugly of sunscreens. Clin Pharmacol Ther 2011; 89(1):31-33.