Malignant melanoma has attained epidemic status throughout the world. Its incidence has increased more rapidly than that oí any other cancer, and the lifetime risk has risen exponentially from 1 per 1,500 population in 1935, to 1 per 600 by 1960, to 1 per 105 by 1991. Projections are that 1 in 75 children born in 2000 will eventually have malignant melanoma. Australia has the highest rate at 1 in 15.1,2
WHY IS MELANOMA SO IMPORTANT IN PEDIATRICS?
Although non-melanoma skin cancers (basal and squamous cell carcinomas) are also increasing and are 16 times as prevalent as malignant melanoma in the United States, melanoma is much more important in pediatrics. It causes more than 3 times as many deaths as the other two skin cancers combined and mortality rates for non-melanoma skin cancer are dropping instead of rising, as for melanoma.1,2 Second, non-melanoma skin cancers are primarily diseases of the elderly. In contrast, 2% to 3% of melanomas occur in children and adolescents. Among U.S. women, melanoma is now the most frequent cancer in those between 25 and 29 years old, and is second only to breast cancer in those between 30 and 35 years old.1,2 Finally, malignant melanoma is most important in pediatrics because of the unique way that exposure to ultraviolet light (UV) during childhood causes melanoma throughout life.
INTERMITTENT, NOT TOTAL, UV EXPOSURE IS THE DOMINANT MODIFIABLE RISK FOR MELANOMA
Both malignant melanoma and non-melanoma skin cancers are more common in those with light skin (blondes, redheads, or light brunettes), who sunburn easily, suntan poorly, or freckle with sun exposure.1-3 Malignant melanoma and non-melanoma skin cancers are less common in Asians and Hispanics and least common in blacks, and things that increase UV exposure (eg, living at a latitude closer to the equator) elevate the risk for both. However, non-melanoma skin cancer and melanoma differ in the way UV causes them, and this influences prevention.
Non-melanoma skin cancers, especially squamous cell carcinoma, are most strongly related to lifelong, chronic UV exposure and are more common in those living in rural, farm settings. These cancers occur most often on areas with the greatest total sun exposure - the head, the neck, the back of the hands, and especially the nose.1"4 The relative tumor density for squamous cell carcinoma is 200 times greater on the nose than on the trunk.
In contrast, malignant melanoma correlates best with episodic exposure to UV. There is no direct relationship between total sun exposure and malignant melanoma; rather, this cancer is most strongly related to intermittent exposures and blistering sunburns. Long-term, low-level exposure may be protective, perhaps because tanning reduces the risk for UV damage and burns. Malignant melanoma, as opposed to non-melanoma skin cancer, does not correlate with farm life. It is more common in office workers or in those who work inside but engage in outdoor play or vacations in the sun, where UV damage to unprepared (untanned or congenitally pale) skin is most likely.1"6 Also, malignant melanoma is less common in areas of the body with long-term sun exposure (the nose and the face) and most prevalent where UV is intermittent (the trunk in men, the legs in women).13 These observations have led many to believe that a tan conferred by regular, moderate exposure provides some protection for melanoma.13
To carry this a step further, intermittent UV exposure of unprepared skin during childhood and adolescence is most important in predisposing to melanoma because this is when the most sun exposure and burns occur.1,2,4,7 This makes prevention during the pediatric years so important.
DOES SUNSCREEN WORK?
After reviewing the literature, I found that I needed to modify anticipatory guidance about sun protection. I placed too much reliance on sunscreen. The effectiveness of sunscreens in preventing melanoma has recently become a hot topic in the world literature.6,7
It is clear that the way sunscreen has been used does not work well. For example, greater sunscreen use has been associated with more, not fewer, childhood sunburns6 and case-control studies, contrasting groups with and without melanoma, found either that there was no difference in history of sunscreen use between the melanoma group and the control group8 or that sunscreen use had been more prevalent in those who had melanoma.910 Interestingly, one of these studies found that a history of sunbaths was protective, as long as patients did not report sunburn.10
Finally, the number of nevi during childhood is a strong predictor of future melanoma. Autier et al. took a history of sun exposure and sunscreen use from 6-year-old and 7-year-old children.11 They then counted nevi at the same time and again 3 years later. Children who used more sunscreen had more new nevi. Once more, sunscreen use was associated with an increased risk for melanoma instead of being protective.
How can this be? One possibility is that sunscreen is not used properly. Another is that earlier sunscreens provided UV-B but not much UV-A protection. But at least part of the answer is that these largely retrospective studies also found that subjects who used more sunscreen were at higher risk for sun damage because of greater sun exposure. It appeared that sunscreen was associated with increased risk because it was used to spend more time in the sun.
One study went beyond this confounder by randomizing 458 Canadian children and parents into a group that received sunscreen plus education about its use or one that did not.12 Nevi, counted at the outset and 3 years later, increased more in the control group than in the sunscreen group (28 vs 24, P = .05). This suggests that sunscreen can be effective if used properly.
BEST PRACTICE FOR MELANOMA PROTECTION
We lack sufficient evidence to finalize methods for optimal protection now. However, considering the above, perhaps anticipatory guidance should teach parents to focus on preventing sunburns in their children. They should understand that sunscreen is important, but that overreliance on sunscreens can increase the risk for melanoma. They should limit their child's UV exposure by decreasing time in the sun during peak intensity (9 AM to 3 PM). Children should use shade and wear a brimmed hat, shirts and pants, and eyeglasses with UV protective coating, as well as use sunscreens. Parents should individualize interventions based on their child's skin color and susceptibility to burns, family history (including diseases that predispose to melanoma), and presence of moles. Sunscreen should not be used as an excuse to stay in the sun longer.7
How about a tan? Should parents individualize sun exposure for their children to allow or encourage tanning without burning? This is more controversial, but I would say yes. However, we need evidence-based research that tells us whether parents can balance all of these factors to achieve tanning without burning, and what approaches work best to reduce sunburns, the development of childhood nevi, and malignant melanoma.
1. Gloster HM Jr, Brodland DG. The epidemiology of skin cancer. Dermatol Surg. 1996;22:217-226.
2. Williams ML, Pennella R. Melanoma, melanocytic nevi, and other melanoma risk factors in children. J Pediatr. 1994;124:833-845.
3. Coggon D, Inskip H. Is there an epidemic of cancer? BMJ. 1994308:705708.
4. Vacchino MN. Poisson regression in mapping cancer mortality. Environ Res. 1999;81:1-17.
5. Aase A, Bentham G. Gender, geography and socio-economic status in the diffusion of malignant melanoma risk. Soc Sd Med. 1996;42:1621-1637.
6. Robinson JK, Rigel DS, Amonette RA. Summertime sun protection used by adults for their children. / Am Acad Dermatol 2000;42:746-753.
7. Etzel RA, Balk SJ, Bearer CF, et al., eds. Pediatric Environmental Health. Elk Grove Village, IL: American Academy of Pediatrics; 1999.
8. Naldi L, Gallus S, Imberti GL, Camelli T, Negri E, La Vecchia C. Sunscreens and cutaneous malignant melanoma: an Italian case-control study, lnt J Cancer. 2000;86:879-882.
9. Westerdahl J, Ingvar C, Masback A, Olsson H. Sunscreen use and malignant melanoma, lnt J Cancer. 2000;87: 145-150.
10. Wolf P, Quehenberger F, Mullegger R, Stranz B, Keri H. Phenotypic markers, sunlight-related factors and sunscreen use in patients with cutaneous melanoma: an Austrian casecontrol study. Melanoma Res. 1998; 8:370-378.
11. Autier P, Dore JF, Cattaruzza MS, et al. Sunscreen use, wearing clothes, and number of nevi in 6- to 7-yearold European children. / Natl Cancer Inst. 1998;90:1873-1880.
12. Gallagher RP, Rivers JK, Lee TK, Bajdik CD, McLean DI, Coldman AJ. Broad-spectrum sunscreen use and the development of new nevi in white children: a randomized controlled trial. JAMA. 2000;283:2955-2960.