© 2005 Oxford University Press
CORRESPONDENCE |
RESPONSE: Re: Sun Exposure and Mortality From Melanoma
Affiliations of authors: University of New Mexico, Albuquerque, NM (MB); University of Sydney, Sydney, Australia (BKA, AK); Memorial Sloan-Kettering Cancer Center, New York, NY (LB-P); University of Connecticut Health Center, Farmington, CT (JF); Albert Einstein College of Medicine, New York, NY (CE); University of Miami, Miami, FL (RLB)
Correspondence to: Marianne Berwick, PhD, MPH, University of New Mexico, Department of Internal Medicine, New Mexico Cancer Research Facility, MSC08 4630, Rm. 103A, 1 University of New Mexico, Albuquerque, NM 87131 (e-mail: mberwick{at}salud.unm.edu).
We welcome the three letters responding to our article "Sun Exposure and Mortality From Melanoma" (1). Kalish suggests that "...melanomas induced by intense sun exposure have a more benign behavior than those not induced by intense sun exposure." We agree that this is a possibility; and one of us suggested it, somewhat cautiously, over 20 years ago "...melanoma might be biologically more benign if it occurs in association with high ambient sun exposure" (2). It is not the only possibility, though, for the point in the development of melanoma at which sun exposure might exert its effect on survival is still not known. For example, at present we cannot distinguish between the effects of sun exposure at the time melanoma is initiated, during the course of its preclinical development, or after it has been diagnosed and treated. The last is, perhaps, unlikely because our sun exposure measures were directed to past rather than present exposure; but it cannot be excluded entirely.
Dellavalle and Johnson advance the hypothesis that melanoma largely due to sun exposure is less aggressive than melanoma largely due to genetic predisposition. This is certainly a possibility and is consistent with the dual pathway hypothesis for melanoma; one pathway is characterized by the presence of p53 staining and indicators of high sun exposure, and the other is characterized by the absence of p53 staining, high nevus density, and a higher prevalence of nevus remnants in the lesions (3). We have found evidence supporting this hypothesis in a subset of subjects in a different study (4).
Dellavalle and Johnson make the interesting suggestion that family history might be useful in distinguishing more aggressive from less aggressive melanomas. However, in what is apparently the only relevant study published to date, there was no statistically significant difference in survival time between 26 patients with metastatic melanoma who were members of families with multiple melanomas (median survival = 57.4 months) and 78 closely matched patients with metastatic melanoma from families with no previous melanoma history (median survival = 50.0 months; P = .99) (5).
Autier and colleagues have written that our results suggest that "...tumors that arise more as a consequence of chronic sun exposure have a less aggressive phenotype than tumors that arise more as a consequence of intermittent sun exposure." Our results do not exclude a protective effect from some level of intermittent sun exposure. In Table 1 of our article (1), univariate analyses showed inverse associations between ever sunburned and high levels of recreational sun exposure and death from melanoma (hazard ratio [HR] = 0.5, 95% confidence interval [CI] = 0.3 to 0.9 and HR = 0.6, 95% CI = 0.3 to 1.0, respectively) that were similar to that between solar elastosis and death from melanoma (HR = 0.5, 95% CI = 0.3 to 0.9). It may be that, of the sun exposure variables, only solar elastosis remained statistically significant after adjusting for other important variables [Table 2 of (1)]. Because one dermatopathologist measured solar elastosis, it is likely to be less misclassified than the other variables. In addition, multivariable analyses of sunburn alone, excluding recreational sun exposure or solar elastosis, suggested an inverse association with death from melanoma, although it was not statistically significant (HR = 0.6, 95% CI = 0.4 to 1.1; P = .08), as did recreational sun exposure alone, excluding sunburn and solar elastosis (HR = 0.6, 95% CI = 0.3 to 1.2; P = .18).
|
We did describe the body site distribution for melanomas with and without solar elastosis in our article (1). The age distribution of solar elastosis (Table 1) certainly indicates that older individuals have more solar elastosis, as expected. Solar elastosis is not associated with indicators of intermittent sun exposure—high levels of recreational sun exposure and sunburns—even though each indicator is inversely associated with melanoma mortality [Table 1 of (1)]. The additional point estimates above are consistent with the full model presented in Table 2 of our article (1) and continue to point to independent associations between solar elastosis, other sun exposure variables, and death from melanoma.
REFERENCES
(1) Berwick M, Armstrong BK, Ben-Porat L, Fine J, Kricker A, et al. Sun exposure and mortality from melanoma. J Natl Cancer Inst 2005;97:195–9.
(2) Lemish WM, Heenan PJ, Holman CD, Armstrong BK. Survival from preinvasive and invasive malignant melanoma in Western Australia. Cancer 1983;52:580–5.[CrossRef][Web of Science][Medline]
(3) Whiteman DC, Parsons PG, Green AC. p53 expression and risk factors for cutaneous melanoma: a case-control study. Int J Cancer 1998;77:843–8.[CrossRef][Web of Science][Medline]
(4) Purdue MP, From L, Kahn LJ, Armstrong BK, Kricker A, Gallagher RP, et al. Etiologic factors associated with p53 immunostaining in cutaneous malignant melanoma. Int J Cancer 2005 May 17 [Epub ahead of print].
(5) Hornbuckle J, Culjak G, Jarvis E, Gebski V, Coates A, Mann G, et al. Patterns of metastases in familial and non-familial melanoma. Melanoma Res 2003;13:105–9.[Medline]
Related Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2005 97: 1158.
J Natl Cancer Inst 2005 97: 1158-1159.
J Natl Cancer Inst 2005 97: 1159.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||