Conclusion:
Clinically dysplastic nevi confer substantial risk of melanoma.

STUDY

1. Case-control study conducted in outpatient clinics.

1. Cases: over 700 consecutive patients with newly diagnosed histologically verified melanoma.
2. Controls: over 1,000 individuals without melanoma matched for age, sex, and geographic distribution. Almost all were white.

2. All received complete skin examination, photography of the most atypical nevi, and if the patient was willing, a biopsy of the most atypical nevus.
3. All nevi larger than 2 mm were counted.
4. Criteria for diagnosis of clinically dysplastic nevi:

1. Size 5 mm or larger
2. Entirely flat or having a flat component
3. At least 2 of the following:

1. Variable pigmentation
2. Irregular, asymmetric outline
3. Indistinct borders
   [See illustration p 1440]

RESULTS
1. Adjusted estimated relative risks (RR) of melanoma by nevus type and number:

Number of nevi > 2 mm & < 5 mm Adjusted RR

0 - 24 1.0
> 100 3.4

Number of nondysplastic nevi > 5 mm

0 1.0
> 10 2.3

Dysplastic nevi

None 1.0
1 2
> 10 12

(RRs rose stepwise as numbers of nevi increased. (Note, even one clinically dysplastic nevus conferred a significant risk.)

DISCUSSION

1. According to some authorities, tumor progression consists of a series of qualitatively different proliferative lesions composing a neoplastic system. Few, if any, precursor lesions progress to cancer. Examples of such neoplastic systems include differing types of colonic polyps with progressive genetic variation and colon carcinoma; atypia, dysplasia, and cervical carcinoma; and differing kinds of melanocytic nevi and melanoma. Dysplastic nevi are intermediate between other nevi and melanoma. Dysplastic nevi are the central risk factor for cutaneous melanoma.
2. Although the risk associated with non-dysplastic nevi is significantly lower and present in a smaller proportion of cases, these nevi do appear to confer modest risk of melanoma independent of clinical atypia. This implies a separate pathway in melanoma etiology.
3. On the basis of number and type of nevi, a clinician can assess risk of melanoma.
4. There remains a substantial proportion of melanomas that do not arise in the setting of dysplastic nevi. Among individuals with these melanomas, there appears to be a consistent, modestly elevated risk associated with freckling. Both clinical variables may be surrogates for past sun exposure. Freckling reflects both host susceptibility and sun exposure and increases risk. Risks associated with freckling roughly added to (rather than multiply) risks associated with all types of nevi.
5. Once high risk individuals are identified, screening and prevention programs can be rationally designed for prevention and early detection of this epidemic cancer.

CONCLUSION
Although non-dysplastic nevi confer a small risk, clinically dysplastic nevi confer a substantial risk for melanoma. On the basis of number and type, clinicians can identify a population at high risk of this epidemic cancer for screening and intervention.
JAMA May 14, 1997; 277: 1439-44 Original investigation, first author from National Cancer Institute, Bethesda, MD
An editorial in this issue of JAMA (pp1475-76) comments:
The incidence of cutaneous melanoma has been doubling each decade, likely owing to increased sun exposure. Early detection carries a special urgency, as cure is directly related to the size and depth of the primary lesion.
Dysplastic nevi can be difficult to recognize clinically. They are characterized by their flatness and their asymmetry, border irregularity, color variability, and large diameter (together known as ABCD), the same features that alert clinicians to melanoma. However, even experts have difficulty in determining the difference between benign, dysplastic, and malignant lesions by inspection. Referring a patient to an expert for evaluation and potential biopsy must be foremost in the mind of the primary care physician confronted with a suspicious lesion.