Small-diameter malignant melanoma: A common diagnosis in New South Wales, Australia Helen M. Shaw, PhD, and William H. McCarthy, FRACS, MEd
Sydney, New South Wales, Australia Background: The "ABCDE" guide developed at the New York University Medical Center and Harvard Medical School has proved useful in the detection of early malignant melanomas. This guide stipulates that melanomas usually are more than 6 mm in maximal diameter and are elevated. Objective: O u r purpose was to determine whether these size criteria applied in Australia. Methods: T h e records of 1150 melanoma patients at the Sydney Melanoma Unit between June 1989 and September 1991 were reviewed. Results: Almost one third of the 1150 melanomas (358) were 6 mm or less in maximal diameter, and the median thickness of these 358 melanomas was 0.8 mrn. Conclusion: The "ABCDE" guide to the signs of early melanoma may have to be modified to include smaller diameter, clinically impalpable melanomas. There appears to be a close correlation between patient age at diagnosis and both lesion diameter and thickness. Because increases in diameter occurred more rapidly than increases in thickness, subtle increases in lesion diameter may provide a better clue to earlier diagnosis. (J AM ACAD DERMATOL 1992;27:679-82.)
Clinical criteria for distinguishing acquired melanocytic nevi from malignant melanomas emphasize differences based on the dimensions of the neoplasm. It is generally believed that acquired melanocytic nevi are less than 7 mm or even up to 10 mm in maximal diameter s' 2; in contrast, melanomas when first clinically identified are usuaUy more than 6 m m in diameter. 3 Schmoeckel and Braun-Falco, 4 from their examination of several thousand melanomas, found only one case with a diameter less than 5 mm. From this, they concluded that such melanocytic lesions cannot usually be recognized as melanomas because specific clinical and histologic features only become apparent as the tumor enlarges. The "ABCD" rule as developed by Rigel et al. 5 is generally considered to be a simple checklist of the signs of early melanoma (A, asymmetry in shape; B, border irregularity; C, variation in color; D, diameter >6 ram). An "E" for "elevation almost always From the Sydney Melanoma Unit, Royal Prince Alfred Hospital, Camperdown,New South Wales, Australia. Supported by the Melanoma Foundation. Accepted for publication April 20, 1992. Reprintrequests:HelenM. Shaw, PhD,SydneyMelanomaUnit, Royal Prince Alfred Hospital,Camperdown, 2050, Australia. 16/1/38751
present" was subsequently added by Fitzpatrick et al. 3 Kamino et al., 6 in their study of 30 small-diameter lesions, first suggested that although this mnemonic is helpful in distinguishing between acquired melanocytic nevi and melanomas, the figure of more than 6 mm for diameter should be modified. Their findings also indicated that small lesions may have a biologic potential similar to larger lesions. The primary aim of the present study was to demonstrate that at the Sydney Melanoma Unit melanomas are frequently removed before they exceed 6 mm in diameter and that these small lesions are capable of recurring locally or metastasizing. In addition, the possibility is discussed that some tumors may be present for many years before they make themselves conspicuous by noticeable changes in appearance. MATERIAL AND METHODS
A retrospective study was made of the primary cutaneous melanomas in new patients seen at the Sydney Melanoma Unit between June 1989 and September 1991. Almost 80% of these patients had their melanoma first diagnosed during this period. The Sydney Melanoma Unit is the major center in New South Wales to which melanoma patients are referred for additional surgical or other treatment. Consequently,biopsiesare performedon only a small proportion of primary melanomas at the 679
Journal of the 680
American Academy of Dermatology
Shaw and McCarthy
Table I. Clinicalandhistologiccharacterisficsoflesions Median
Dia~ter(mm)
No.
2 3 4 5 6 7 8 9 10 11-15 16-20 >20
9 47 64 137 101 127 113 77 133 214 78 49
1''
Age(yr) 32 (5.8) 32 (5.3) 36 (4.6) 41 (5.5) 43 (3.9) 44 (3.6) 45 (3.8) 46 (4.7) 51 (3.5) 52 (2.8) 57 (4.3) 60 (5.4)
(+ CL)
I
I
I
Thickness (nun)
0.61 0.60 0.78 0.82 0.95 0.90 0.90 1.05 1.12 1.20 1.90 2.50
(0.19) (0.14) (0.29) (0.25) (0.24) (0.28) (0.34) (0.51) (0.39) (0.29) (0.75) (2.28)
CL,95%confidencelimits.
Unit. Nevertheless, the primary lesion biopsy specimens from all patients treated by the Unit are routinely reviewed. All specimens in the present study were obtained by excisional biopsy. In each case, treatment of the primary lesion was wide local excision and closure or wide excision and grafting. The policy of the Sydney Melanoma Unit is to excise lesions 1 mm or less in thickness with a 1 cm margin and those more than i mm thickwith a 2 to 3 cm margin. The thickness of the lesion, not its diameter, determines the width of excision. In general, stage I patients with lesions more than 1.5 mm in thickness had an elective lymph node dissection. The combination of histologic criteria used to diagnose small-diameter melanomas were the same as those used to diagnose larger diameter melanomas. 7 These included epidermal and dermal invasion, mitotic figures in the dermal component, cytologic pleomorphism, lymphatic invasion, and fine or dusty pigment. The following features were available on each lesion: 1. Maximal diameter (to the nearest whole millimeter) as recorded in the macroscopic description 2. Maximal vertical thickness, Clark's level of invasion, and histogenetic type as recorded in the microscopic description Excluded from consideration were lentigo maligna melanomas and Clark level I or in-situ lesions. RESULTS Table I shows the distribution of 1150 lesions according to clinical and histologic features. During this period, 358 or 31.1% of lesions were 6 mm or less in diameter. The correlation between increasing diameter and thickness of lesions and age of patients at diagnosis is shown in Fig. 1. The diameter of lesions in older
patients (60 years or older) were, in general, six times the diameter of lesions in young patients. There was a threefold increase in the thickness of lesions in older patients. Median thickness of the small-diameter lesions was 0.8 m m and nearly three fourths were of the superficial spreading type. In four of the 358 patients with small-diameter lesions (1%), the lesion was either their second or third primary melanoma. None of these 358 patients had the dysplastic nevus syndrome. Median follow-up of the 1150 patients was 20 months (range 2 to 110 months). In this period, 41 of the 358 patients with lesions 6 mm or less in diameter either first presented for treatment with lymph node involvement (stage III of the Union Internationale Contre le Cancer/American Joint Committee on Cancer classification) or were stage I or II patients in whom local recurrences or metastases developed (11.5%). To date, despite the short follow-up on most patients with small-diameter lesions, 11 (3.1%) have died of cancer. DISCUSSION For the most part, the " A B C D E " guide to t h e signs of early melanoma is simple and should be recommended as a means of education for patients and medical personnel. However, from the study of Kamino et al. 6 and the present analysis, it is clear that it may be worthwhile modifying the " D " and "E" of this guide to facilitate early diagnosis. The study of Kamino and Ratech 8 of approximately 3500 patients indicated that nearly 1% o f cases had invasive melanomas less than 6 m m in di-
Volume 27 N u m b e r 5, Part 1 N o v e m b e r 1992
ameter. Almost one third of 1150 melanomas in the Sydney Melanoma Unit had diameters less than that recommended by the guide. This large discrepancy between the figures of Kamino and Ratech 8 and ours may be explained by the fact that in Australia, with an exceptionally high public awareness of melanoma, the most common reason for the early diagnosis of melanoma is a prophylactic mole check by a physician of an educated, concerned patient rather than any particular change in a mole noted by the patient. 9 However, this does not imply that Australian physicians are more adept at diagnosing early melanoma than physicians elsewhere because the Australian quest for detecting lesions at the earliest stage is achieved at a cost of a considerable number of unnecessary excisions) ~ In addition, median thickness of the small-diameter lesions from the Sydney Melanoma Unit was about 0.8 mm and two thirds were of the superficial spreading variety. Therefore they were not always clinically palpable or "obviously or subtly elevated" on examination, as required by Fitzpatrick et al. 3 for their extra "E" in the "ABCD" guidelines because not all had developed an early vertical growth phase.l i The high proportion of superficial spreading melanomas found in the present study fails to support the contention of Schmoecke112 that smalldiameter melanomas were mainly of the nodular type. We propose that a more appropriate "E" of the "ABCDE" guidelines for detecting early melanoma could be "examination" of the patient's entire skin. A high index of suspicion should be attached to a lesion that does not resemble other pigmented lesions in that patient. Schmoeckelt 2 contended that melanomas are still histologically in the stage of melanoma in situ or atypical melanocytic hyperplasia when they are small (horizontal diameter
Sydney, New South Wales, Australia Background: The "ABCDE" guide developed at the New York University Medical Center and Harvard Medical School has proved useful in the detection of early malignant melanomas. This guide stipulates that melanomas usually are more than 6 mm in maximal diameter and are elevated. Objective: O u r purpose was to determine whether these size criteria applied in Australia. Methods: T h e records of 1150 melanoma patients at the Sydney Melanoma Unit between June 1989 and September 1991 were reviewed. Results: Almost one third of the 1150 melanomas (358) were 6 mm or less in maximal diameter, and the median thickness of these 358 melanomas was 0.8 mrn. Conclusion: The "ABCDE" guide to the signs of early melanoma may have to be modified to include smaller diameter, clinically impalpable melanomas. There appears to be a close correlation between patient age at diagnosis and both lesion diameter and thickness. Because increases in diameter occurred more rapidly than increases in thickness, subtle increases in lesion diameter may provide a better clue to earlier diagnosis. (J AM ACAD DERMATOL 1992;27:679-82.)
Clinical criteria for distinguishing acquired melanocytic nevi from malignant melanomas emphasize differences based on the dimensions of the neoplasm. It is generally believed that acquired melanocytic nevi are less than 7 mm or even up to 10 mm in maximal diameter s' 2; in contrast, melanomas when first clinically identified are usuaUy more than 6 m m in diameter. 3 Schmoeckel and Braun-Falco, 4 from their examination of several thousand melanomas, found only one case with a diameter less than 5 mm. From this, they concluded that such melanocytic lesions cannot usually be recognized as melanomas because specific clinical and histologic features only become apparent as the tumor enlarges. The "ABCD" rule as developed by Rigel et al. 5 is generally considered to be a simple checklist of the signs of early melanoma (A, asymmetry in shape; B, border irregularity; C, variation in color; D, diameter >6 ram). An "E" for "elevation almost always From the Sydney Melanoma Unit, Royal Prince Alfred Hospital, Camperdown,New South Wales, Australia. Supported by the Melanoma Foundation. Accepted for publication April 20, 1992. Reprintrequests:HelenM. Shaw, PhD,SydneyMelanomaUnit, Royal Prince Alfred Hospital,Camperdown, 2050, Australia. 16/1/38751
present" was subsequently added by Fitzpatrick et al. 3 Kamino et al., 6 in their study of 30 small-diameter lesions, first suggested that although this mnemonic is helpful in distinguishing between acquired melanocytic nevi and melanomas, the figure of more than 6 mm for diameter should be modified. Their findings also indicated that small lesions may have a biologic potential similar to larger lesions. The primary aim of the present study was to demonstrate that at the Sydney Melanoma Unit melanomas are frequently removed before they exceed 6 mm in diameter and that these small lesions are capable of recurring locally or metastasizing. In addition, the possibility is discussed that some tumors may be present for many years before they make themselves conspicuous by noticeable changes in appearance. MATERIAL AND METHODS
A retrospective study was made of the primary cutaneous melanomas in new patients seen at the Sydney Melanoma Unit between June 1989 and September 1991. Almost 80% of these patients had their melanoma first diagnosed during this period. The Sydney Melanoma Unit is the major center in New South Wales to which melanoma patients are referred for additional surgical or other treatment. Consequently,biopsiesare performedon only a small proportion of primary melanomas at the 679
Journal of the 680
American Academy of Dermatology
Shaw and McCarthy
Table I. Clinicalandhistologiccharacterisficsoflesions Median
Dia~ter(mm)
No.
2 3 4 5 6 7 8 9 10 11-15 16-20 >20
9 47 64 137 101 127 113 77 133 214 78 49
1''
Age(yr) 32 (5.8) 32 (5.3) 36 (4.6) 41 (5.5) 43 (3.9) 44 (3.6) 45 (3.8) 46 (4.7) 51 (3.5) 52 (2.8) 57 (4.3) 60 (5.4)
(+ CL)
I
I
I
Thickness (nun)
0.61 0.60 0.78 0.82 0.95 0.90 0.90 1.05 1.12 1.20 1.90 2.50
(0.19) (0.14) (0.29) (0.25) (0.24) (0.28) (0.34) (0.51) (0.39) (0.29) (0.75) (2.28)
CL,95%confidencelimits.
Unit. Nevertheless, the primary lesion biopsy specimens from all patients treated by the Unit are routinely reviewed. All specimens in the present study were obtained by excisional biopsy. In each case, treatment of the primary lesion was wide local excision and closure or wide excision and grafting. The policy of the Sydney Melanoma Unit is to excise lesions 1 mm or less in thickness with a 1 cm margin and those more than i mm thickwith a 2 to 3 cm margin. The thickness of the lesion, not its diameter, determines the width of excision. In general, stage I patients with lesions more than 1.5 mm in thickness had an elective lymph node dissection. The combination of histologic criteria used to diagnose small-diameter melanomas were the same as those used to diagnose larger diameter melanomas. 7 These included epidermal and dermal invasion, mitotic figures in the dermal component, cytologic pleomorphism, lymphatic invasion, and fine or dusty pigment. The following features were available on each lesion: 1. Maximal diameter (to the nearest whole millimeter) as recorded in the macroscopic description 2. Maximal vertical thickness, Clark's level of invasion, and histogenetic type as recorded in the microscopic description Excluded from consideration were lentigo maligna melanomas and Clark level I or in-situ lesions. RESULTS Table I shows the distribution of 1150 lesions according to clinical and histologic features. During this period, 358 or 31.1% of lesions were 6 mm or less in diameter. The correlation between increasing diameter and thickness of lesions and age of patients at diagnosis is shown in Fig. 1. The diameter of lesions in older
patients (60 years or older) were, in general, six times the diameter of lesions in young patients. There was a threefold increase in the thickness of lesions in older patients. Median thickness of the small-diameter lesions was 0.8 m m and nearly three fourths were of the superficial spreading type. In four of the 358 patients with small-diameter lesions (1%), the lesion was either their second or third primary melanoma. None of these 358 patients had the dysplastic nevus syndrome. Median follow-up of the 1150 patients was 20 months (range 2 to 110 months). In this period, 41 of the 358 patients with lesions 6 mm or less in diameter either first presented for treatment with lymph node involvement (stage III of the Union Internationale Contre le Cancer/American Joint Committee on Cancer classification) or were stage I or II patients in whom local recurrences or metastases developed (11.5%). To date, despite the short follow-up on most patients with small-diameter lesions, 11 (3.1%) have died of cancer. DISCUSSION For the most part, the " A B C D E " guide to t h e signs of early melanoma is simple and should be recommended as a means of education for patients and medical personnel. However, from the study of Kamino et al. 6 and the present analysis, it is clear that it may be worthwhile modifying the " D " and "E" of this guide to facilitate early diagnosis. The study of Kamino and Ratech 8 of approximately 3500 patients indicated that nearly 1% o f cases had invasive melanomas less than 6 m m in di-
Volume 27 N u m b e r 5, Part 1 N o v e m b e r 1992
ameter. Almost one third of 1150 melanomas in the Sydney Melanoma Unit had diameters less than that recommended by the guide. This large discrepancy between the figures of Kamino and Ratech 8 and ours may be explained by the fact that in Australia, with an exceptionally high public awareness of melanoma, the most common reason for the early diagnosis of melanoma is a prophylactic mole check by a physician of an educated, concerned patient rather than any particular change in a mole noted by the patient. 9 However, this does not imply that Australian physicians are more adept at diagnosing early melanoma than physicians elsewhere because the Australian quest for detecting lesions at the earliest stage is achieved at a cost of a considerable number of unnecessary excisions) ~ In addition, median thickness of the small-diameter lesions from the Sydney Melanoma Unit was about 0.8 mm and two thirds were of the superficial spreading variety. Therefore they were not always clinically palpable or "obviously or subtly elevated" on examination, as required by Fitzpatrick et al. 3 for their extra "E" in the "ABCD" guidelines because not all had developed an early vertical growth phase.l i The high proportion of superficial spreading melanomas found in the present study fails to support the contention of Schmoecke112 that smalldiameter melanomas were mainly of the nodular type. We propose that a more appropriate "E" of the "ABCDE" guidelines for detecting early melanoma could be "examination" of the patient's entire skin. A high index of suspicion should be attached to a lesion that does not resemble other pigmented lesions in that patient. Schmoeckelt 2 contended that melanomas are still histologically in the stage of melanoma in situ or atypical melanocytic hyperplasia when they are small (horizontal diameter
