Cutaneous melanoma is malignancy of the pigment producing cell, the melanocyte, which is normally located in the basal layer of the epidermis. This tumour is becoming more frequent, especially in light-skinned caucasians. Fifty years ago, melanoma was invariably diagnosed in its late stages, and patients died rapidly from metastatic disease. Since the early signs of the disease have been publicized the 5-year survival rates have more than doubled. Early recognition followed by prompt surgical therapy offers the best chance for cure.

In the United States of America approximately 32000 cases of malignant melanoma are diagnosed annually; there are over 6800 deaths. The frequency in children is less than one case per million per year, but after the age of 15 it is about 100 times more frequent. The incidence in black and east Asian subjects, as well as in dark-skinned caucasians (natives of India, Puerto Rico), is 5 to 10 per cent that seen in light-skinned caucasians.

The incidence of melanoma has increased over the past two decades more rapidly than that of any other tumour. Although the cause of this increase is unknown, the short period over which this change has occurred suggests that some environmental or behavioural change is responsible. The increases are especially noteworthy in adolescents and young adults. In men aged 15 to 35, melanoma is the second most common cause of death from cancer in the United States: in New Zealand it accounts for 23 per cent of all cancers diagnosed in all individuals of this age group. The 5-year survival rate in 1940 was approximately 40 per cent; by the mid 1980s it had increased to about 85 per cent. However, there are still few long-term survivors of disseminated melanoma. The most fre quent site of cutaneous melanoma is the upper back in men and the lower leg from knee to ankle, as well as the upper back, in women. In black, east Asian, and darkly complexioned caucasian subjects, the most common sites are palms, soles, nailbeds, and mucous membranes. Several lines of evidence indirectly link solar exposure to the increasing incidence of melanoma (Table 1) 443. The incidence of melanoma varies inversely with latitude within countries, and the disease is more common in individuals of British extraction in Australia than in the United Kingdom.

Table 2 444 ranks the risk profiles associated with the development of melanoma: the highest risk is associated with a persistently changing mole. Increase in size and change in colour are the two most common characteristics of early cutaneous melanoma. The increased risk associated with increasing numbers of naevi and with sporadic and familial dysplastic naevi is reviewed elsewhere (see Section 32.3 226). Factors associated with only a small increase in relative risk may be important if they affect a substantial proportion of the population. MacKie has proposed an algorithm that can be used in screening to distinguish between high- and low-risk people. Those who freckle, have three or more dysplastic naevi, three or more severe sunburns, or 20 or more naevi have an increased risk of about 600-fold (men) or 200-fold (women) over people with none of these factors.

Most melanomas can be distinguished from non-malignant cutaneous lesions (Fig. 1) 1514. Table 3 445 lists those factors in the history and physical examination which are of diagnostic help, while Table 4 446 lists features distinguishing four types of cutaneous melanoma. Approximately 5 per cent of melanomas may be clinically amelanotic and in these cases the diagnosis can only be made by the pathologist. Even with the best clinical skills, not all melanomas can be diagnosed clinically. A biopsy specimen should be obtained when melanoma is a diagnostic possibility.

Total excisional biopsy with narrow margins is the procedure of choice for the diagnosis of cutaneous melanoma: this provides the entire specimen for histopathological examination and also provides definitive therapy for benign lesions. Removal of a partial biopsy specimen appears to be safe and is a reasonable course of action when total excisional biopsy is not easily accomplished. Either punch (trephine) or incisional biopsy can be used. Final determination of melanoma type and thickness (important for prognosis) must await complete excision, however. When partial biopsies are performed, biopsy of the most raised portion will usually be sufficient for diagnosis. In flat or plaque-type lesions biopsy of the darkest portion most frequently produces a representative specimen.

Several classifications exist for melanoma staging; some are three-stage and others four-stage. The joint AJCC/UICC classificationis likely to be most widely used internationally (Table 5) 447. Although unwieldy, this system allows accurate comparisons to be made between series from different centres.

The thickness of the primary tumour in millimetres (Breslow scale) is the single most important prognostic factor for melanoma patients who have no clinical evidence of disease elsewhere. There is a reasonably direct correlation between the primary tumour thickness (measured vertically from the granular cell layer down to the deepest tumour cell) and survival at 5, 8, or 10 years. There is also an inverse relationship between thickness of the primary tumour and time of death (Table 6) 448. A small number of patients with very thin tumours develop metastatic disease; this may occur 10 years or more after removal of the primary tumour. Since melanoma may have a prolonged natural history, comparison of results at 10 years after diagnosis, is more consistent than comparisons made after only 5 years.

Clark's levels of invasion, which historically preceded thickness measurement, also correlate well with outcome. In the Clark system, level II tumours show penetration of the papillary dermis, in level III the papillary dermis is involved, level IV enters the reticular dermis, and level V enters the subcutaneous fat. Using this system it is difficult to distinguish level III from level IV. In addition, the designation of polypoid tumours of substantial thickness as level III underestimates the risk associated with these lesions.

Several centres have evaluated other factors that add to the ability to predict outcome (Table 7) 449. Almost all studies have shown that women with localized disease survive longer than men: the location of tumours on favourable sites such as the lower extremity may explain this difference, although some studies suggest that location alone is not sufficient to explain this difference in survival. Increased numbers of mitoses, the histological presence of an ulcer, and the presence of microscopic satellites, defined as discrete nests of tumour cells at least 0.05 mm in diameter separated from the main body of the tumour and located in the reticular dermis or subcutaneous fat, are all associated with a worse prognosis. Tumours in certain sites such as scalp, hands and feet, and mucous membranes also appear to have worse prognoses. Recently, Clark et al. have proposed a six-factor model to separate low, intermediate, and high-risk patients that have downward penetration of the tumour (vertical growth phase). The six factors are tumour thickness of 1.70 mm or more, sex, axial plus acral locations versus extremities, regression, tumour infiltration by lymphocytes, and mitotic rate. By objectively evaluating prognostic factors, appropriate surgical programmes can be established and the efficacy of experimental adjuvant therapies can be assessed.

Excision of tumour with 5 cm surgical margins was established in 1907 by Sampson Handley, and was normal practice in many centres for a period of over 70 years. With the institution of the Breslow thickness scale low-risk patients could be identified for whom thinner surgical margins appeared to be safe. In 1977, Breslow and Macht noted that the width of the surgical excision margin did not appear to be an important factor in outcome for melanomas less than 0.75 mm thick. The World Health Organization Melanoma Programme has recently demonstrated that 1 cm margins are effective in the treatment of melanomas 2 mm or less in thickness. Safe and effective margins for melanomas more than 2 mm thick are currently being evaluated (Table 8) 450.

In recent years, closure with split thickness skin graft has become much less common, and most lesions are now closed primarily or using skin flaps. The need to remove underlying fascia as part of the excision of the primary tumour has been controversial: some clinicians regard the fascia as a natural barrier to tumour spread that should be left intact, while others feel that since the fascia may contain tumour cells it should be removed at the time of primary surgery. The best data available at present show no difference in outcome in relation to whether the fascia is removed or not.

Controversy exists over the benefit derived from the removal of non-palpable regional nodes. Patients with melanomas thinner than 1 mm rarely have nodal metastases, making the value of elective nodal dissection in this group low. Similarly, most patients with melanomas thicker than 4 mm fail from systemic metastases, and elective nodal dissection does not appear to improve their prognoses. In a large randomized study in 1977, the World Health Organization Melanoma Programme failed to show a benefit from elective nodal dissection. However, there were several technical problems in the construction of the study, and it is currently not clear whether some patients, such as those with melanoma of 1.5 mm to 4 mm in thickness might benefit from elective nodal dissection. A randomized study is being undertaken in the United States of America in an attempt to clarify this. Until such results are available, elective nodal removal is performed at the discretion of the surgeon, after discussion with the patient.

Factors against the use of elective nodal dissection include the presence of ambiguous drainage (radionuclide scanning following biopsy can evaluate the direction of drainage), ill health, extreme age, obesity (increased likelihood of lymphoedema) and a negative view by the patient after learning the potential benefits and side-effects of this approach.

In specialized centres perfusion of the affected extremity with phenylalanine mustard at elevated temperature has produced good control of in-transit lesions and satellite lesions. Although such perfusion is reasonably well accepted as a palliative measure, its benefit as an adjuvant in the treatment of primary extremity melanoma has not yet been established in prospective randomized studies. In a recent comparative investigation, melanoma patients matched for current prognostic factors showed a similar outcome after treatment with surgery alone or with perfusion.

Removal of clinically diseased regional nodes appears to be warranted in patients with no significant disseminated disease; in the presence of disseminated disease, a more systemic approach is usually considered. The response to chemotherapy is greatest for disease in the skin, lymph nodes, and lung; tumours in bone, brain, and liver rarely respond. Long-term remission can occasionally be achieved by aggressive surgical removal of an isolated metastasis in the brain or other organ, but multiple metastases often appear within months. Table 9 452 shows possible approaches to the treatment of melanoma in various sites.

Follow-up of melanoma patients is critical to enable detection of early recurrence and for the diagnosis of second or subsequent tumours at an early, curable, stage. Table 10 451 depicts the follow-up schedule currently in use in the Pigmented Lesion Clinic at the Massachusetts General Hospital.

The best hope for cure of cutaneous melanoma lies in early diagnosis and appropriate surgical removal. Public and professional education campaigns have resulted in a progressive fall in median melanoma thickness at diagnosis and an increase in 5-year survival. Awareness of risk factors and of the characteristics of early melanoma should result in continued improvement in survival.

Balch CM, et al., eds. Cutaneous Melanoma, 2nd edn. Philadelphia: J B Lippincott, 1992.

Beahrs OH, Henson DE, Hutter RVP, Myers MH, eds. Manual for Staging of Cancer. American Joint Committee on Cancer, 3rd edn. Philadelphia: J B Lippincott, 1988: 139–144.

Breslow A. Tumor thickness, level of invasion, and node dissection in stage I cutaneous melanoma. Ann Surg 1975; 182: 572–5.

Clark WH Jr., From L. Bernardino EA, Mihm MC, Jr. The histogenesis and biologic behavior of primary human malignant melanoma of the skin. Cancer Res 1969; 29: 705–26.

Clark WH Jr, et al. Model predicting survival in stage I melanoma based upon tumor progression. J Natl Cancer Inst 1989; 81: 1893–904.

Friedman R, Rigel D, Kopf A, Harris M, Baker D. Cancer of the Skin. Philadelphia: W B Saunders, 1991.

Ho VC, Sober AJ. Therapy for cutaneous melanoma: an update. J Am Acad Dermatol 1990; 22: 159–76.

Koh HK, Kligler BE, Lew RA. Sunlight and cutaneous malignant melanoma: evidence for and against causation. Photochem Photobiol 1990; 51: 765–79.

Mackie RM, Freudenberger R, Aitchison TC. Personal risk factor chart for cutaneous melanoma. Lancet 1989; ii: 487–90.

Rhodes AR, et al. Risk factors for cutaneous melanoma: a practical method of recognizing predisposed individuals. JAMA 1987; 258: 3146–54.

Veronesi U., et al. Delayed regional lymph node dissection in stage I melanoma of the skin of the lower extremities. Cancer 1982; 49: 2420–30.

Veronesi U., et al. Thin stage I primary cutaneous malignant melanoma: comparison of excision with margins of 1 or 3 cm. N Engl J Med 1988; 318: 1159–62.