- Home>
- Specialty Care>
- Dermatology>
- News in Context
The Melanoma Epidemic: A Recurrent, Inextricable Tangle of Phenomena and Epiphenomena
Are we seeing a true rise in the incidence of melanoma, or does it just look that way?
A recent article in the New York Times broaches the old controversy regarding the increases over recent decades in U.S. melanoma rates: Is the increase in disease or in detection? The author based her discussion in part on a paper by investigators at Dartmouth Medical School who compared rates of skin biopsy in Medicare records with rates of melanoma using the Surveillance, Epidemiology, and End Results (SEER) Program database of the National Cancer Institute.
They obtained annual, population-based rates of skin biopsy performed in patients aged 65 or older for each of the geographic areas covered by the SEER program from 1986 through 2001. Over this period, the average rate of biopsy in the nine studied areas increased 2.5-fold, from 2847/100,000 to 7222/100,000. In the same period, average melanoma incidence in these areas increased 2.4-fold, from 45/100,000 to 108/100,000. The researchers applied logistic regression analysis to two possible scenarios: The first assumed no change in melanoma occurrence and found that every additional 1000 biopsies identified 12.6 new melanomas. The second assumed a true increase in melanoma occurrence, with every additional 1000 biopsies identifying 6.9 new cases of melanoma.
The researchers reiterate that mortality from melanoma has not changed significantly over this period and argue that the pattern "suggests over-diagnosis, largely as a result of increased diagnostic scrutiny and not an increase in the true occurrence of the disease," which led to the commentary by the New York Times writer.
Comment: This is an age-old controversy, and several points need scrutiny. First, an increase in melanoma has been observed worldwide, and to reach the same conclusion, the authors would have to assume that over-biopsy is a global event. Second, this study would have benefited from adjustments in biopsy rates by race. As melanoma is largely restricted to the white population, it is critical to know whether the increase in biopsies is balanced across the races. Third, it is important to understand the source of the SEER data: The NCI registry collects cancer information from designated local registries across the country. The only thing we know for sure is that the number of registered melanomas has increased over the past several decades; all subsequent indicators --melanoma incidence, numbers of cases, and so forth -- are estimates derived from census-based population figures. Multiple factors could enhance the apparent rate of melanoma over the past several decades: 1) an actual increase in the number of tumors; 2) an increase in the efficiency of reporting and documentation at the SEER registries; 3) an increased awareness of pigmented lesions by the public (e.g., the shift from "bleeding mole" to the ABCD rule), resulting in earlier and more frequent evaluation of worrying lesions; 4) a lowered threshold for biopsy prompted by patient anxiety or physician fear of litigation; 5) a lowered threshold for melanoma diagnosis by dermatopathologists likewise fearful of litigation; and 6) a shift from a "wait and watch" to a "remove and diagnose" mindset encouraged by increased surgical training or financial incentive. Undoubtedly, we could posit many more hypotheses to explain the increasing number of registered melanomas in 14 cancer registries around the country.
A recent report from the Manitoba Cancer Registry (see J Am Acad Dermatol 2005; 53:320) showed that BCCs and SCCs have also increased dramatically in the Medicare-eligible population between 1986 and 2000; thus, increases in the rates of all skin cancers parallel increases in the rates of skin biopsies. Without knowing the diagnostic breakdown, we cannot assume that melanomas account for a proportional fraction of the procedures. This controversy is unlikely to be resolved in the near future. The probable answer is that melanoma was underreported 30 years ago when centralized efforts were not as consolidated as they are today. Until more studies differentiate among the separate effects contributing to increased cancer registration, the question remains an inextricable tangle of phenomena and epiphenomena.
— Hensin Tsao, MD, PhD
Published in Journal Watch Dermatology August 23, 2005
Citation(s):
Kolata G. Melanoma Is Epidemic. Or Is It? New York Times 2005 Aug 9; (http://www.nytimes.com/2005/08/09/health/09skin.html?ex=1124337600&en=ba2a43159fd7034a&ei=5070&emc=eta1)
Welch HG et al. Skin biopsy rates and incidence of melanoma: Population based ecological study. BMJ 2005 Aug 4; (http://bmj.bmjjournals.com/cgi/rapidpdf/bmj.38516.649537.E0v1)