STUDY. Surgical Margins for Lentigo Maligna and Lentigo Maligna Melanoma

Transcription

1 Surgical Margins for Lentigo Maligna and Lentigo Maligna Melanoma The Technique of Mapped Serial Excision STUDY Shyamala C. Huilgol, FACD; Dinesh Selva, FRANZCO; Celia Chen, MBBS; Dudley C. Hill, FACD; Craig L. James, FRCPA; Amanda Gramp, FRCPA; Raman Malhotra, FRCOphth Objectives: To assess the margins required for excision of lentigo maligna (LM) and lentigo maligna melanoma (LMM) by the technique of mapped serial excision (MSE), and to assess the efficacy of MSE. Design: An interventional, prospective, noncontrolled case series. Setting: Tertiary referral, dermatologic surgery unit. Patients: Consecutive patients with head and neck LM or LMM who underwent MSE between March 1, 1993, and October 31, Intervention: The MSE of LM or LMM. Main Outcome Measures: The number of 5-mm levels for excision of LM and LMM and recurrence. Results: One hundred sixty-one LMs or LMMs in 155 patients were treated. Thirty percent (37 of 125) of LMs required more than 5-mm margins. For LMMs less than 1 mm in Breslow thickness, 12% (4/32) required more than 10-mm margins. For primary tumors, 20% of LMs (18 of 91) required more than 5-mm margins, while 10% of LMMs less than 1 mm in Breslow thickness (2 of 21) required more than a 10-mm margin. For recurrent tumors, 56% of LMs (19/34) required more than a 5-mm margin. Mean follow-up of 38 months (range, months) showed 4 recurrences (2%) after MSE. The extrapolated recurrence at 5 years was 5.0%. Conclusions: The current recommendations of 5-mm margins for LM and 10-mm margins for LMM less than 1 mm in Breslow thickness are often insufficient. Our results demonstrate the importance of margin-controlled excision, particularly in recurrent lesions. The use of MSE offers a high cure rate, in conjunction with tissue conservation. Arch Dermatol. 2004;140: From the Department of Dermatology (Drs Huilgol and Hill) and Oculoplastic and Orbital Unit, Department of Ophthalmology (Drs Selva, Chen, and Malhotra), Royal Adelaide Hospital, University of Adelaide; Adelaide Pathology Partners (Dr James), Gramp Skin Pathology (Dr Gramp), and Hill Day Surgery, Wakefield Clinic (Drs Huilgol and Hill), Adelaide, Australia. The authors have no relevant financial interest in this article. LENTIGO MALIGNA (LM) IS A form of melanoma in situ and most commonly presents as a very slowly progressive, irregularly pigmented macule on the face of elderly individuals with sun-damaged skin. In Australia, the annual incidence of LM is the highest in the world and has been estimated at 1.3:100000, with the most significant risk factor being UV radiation exposure. 1,2 Lentigo maligna has a risk of progression to lentigo maligna melanoma (LMM), a subtype of invasive melanoma. Estimates of this risk have ranged as high as 30% to 50%, 3 but the true risk is likely to be lower than this. In the United States, Newell et al 4 reported an estimated annual incidence of LMM of 0.8: in men and 0.6: in women. The treatment options for LM include surgical excision and destructive modalities, while LMM is preferably treated surgically. 5,6 In LM, destructive modalities such as cryotherapy, electrodessication and curettage, laser surgery, radiotherapy, and fluorouracil have high recurrence rates of 20% to 100%. 5,7,8 Surgical excision of LM has the advantage of treating deep periadnexal melanocytes, detecting unsuspected invasive LMM, and permitting histologic assessment of the margins for atypical melanocytes, which commonly extend beyond the clinically apparent border. 9 However, the clinical margins of LM and LMM are often poorly defined and may also be masked by ephelides, pigmented actinic keratoses, lentigines, nevi, and seborrheic keratoses. 10 Confirming the difficulties with both clinical and histologic margin control is a study of 92 patients with LM showing that standard surgical excision with bread-loaf histologic assessment of margins of LM had incomplete excision rates of 9%, with extrapolated 5-year recurrence rates of lesions thought to be completely excised of 31%

2 Inclusion criteria were histologically confirmed LM or LMM, larger than 1 cm, occurring on the head and neck. All patients gave informed consent before MSE, and all surgical excisions were carried out in a tertiary-referral, private dermatologic day surgery facility by 2 of us (S.C.H. and D.C.H.). Histopathological examination was performed by 2 of us (C.L.J. and A.G.). Lentigo maligna was defined as a form of melanoma in situ with continuous proliferation of atypical melanocytes, singly and in nests, along the basal layer of the epidermis and adnexae; arising in sun-damaged skin with an atrophic epidermis; and with minimal pagetoid invasion into the upper epidermis. Lentigo maligna melanoma was defined as LM with a component of dermal invasion. INTERVENTIONS Figure 1. Lentigo maligna on the right cheek outlined with 5-mm margins. The current Australian guidelines from the National Health and Medical Research Council 12 for primary cutaneous melanoma recommend surgical margins of 5 mm for in situ lesions, 10 mm for invasive tumors less than 1 mm in Breslow depth, and 20 mm for invasive tumors with 1- to 4-mm Breslow depth. These recommendations are based on the National Institutes of Health consensus guidelines 13 and 2 prospective multicenter trials 14,15 and provide an acceptable compromise between achieving local control and tissue conservation. However, these recommended margins are based on trials for melanoma on the trunk and proximal extremities and have not specifically addressed tumors on the head, neck, hands, and feet. Excision margins on the head and neck are often reduced to preserve important anatomic structures and for cosmetic reasons. Therefore, LM and LMM present a unique problem because of the characteristic unpredictability of subclinical extension in head and neck locations where tissue preservation is imperative. The aim of this study was to assess the required surgical margins for excision of LM and LMM, using the technique of mapped serial excision (MSE). We compare the results with recommended margins for these tumors and assess the incidence of recurrence after MSE. METHODS STUDY DESIGN AND PATIENT SELECTION A single-center, prospective study of consecutive patients with head and neck LM or LMM undergoing MSE between March 1, 1993, and October 31, 2002, was conducted. The methods used were previously described by Hill and Gramp. 16 The diagnosis of LM or LMM was histologically confirmed by previous shave, punch, or incisional biopsy. A Wood s light was used to assist in marking the tumor borders and a 5-mm margin was added (Figure 1). In rare cases, such as adjacent to the nostril rim, 5-mm margins could not be achieved around the entire tumor circumference, and the largest possible margin was taken. Excision to the level of the middle to deep subcutaneous fat layer (to include adnexal structures) was performed with the patient under local anesthetic. The excised specimen was mapped, dyed, and sent to the pathology laboratory in formalin. A simple dressing was applied to the wound with an antibiotic ointment, and the patient was examined within 24 hours together with the histology report. The histopathologic examination consisted of orienting a formalin-fixed specimen according to marker dyes, nicks, and sutures and the accompanying color-coded diagram (Figure 2). Vertically cut paraffin-embedded sections cut at 1- to 2-mm intervals (ie, bread-loaf sections) with block identification were stained with hematoxylin-eosin (Figure 3 and Figure 4). The sections were examined by 2 experienced dermatopathologists (C.L.J. and A.G.). The report, including a diagram with the oriented site(s) of any incomplete excision and the distance between the tumor and surgical margins, was faxed to the dermatologist within 24 hours (Figure 5). Where excision was incomplete, as defined earlier, a further targeted 5-mm excision was taken from the involved margin and sent for histopathologic assessment. This was repeated until histologic confirmation of clear margins was achieved. The final defect was repaired either by the dermatologist or by oculoplastic or plastic surgeons. Where excision was complete but margins were noted to be less than 3 mm, one dermatologist (S.C.H.) reexcised the relevant margin to give a total margin of 5 mm. All patients were contacted by telephone and invited to return for clinical review to detect recurrence. The site of previous MSE was examined with magnification, and photographs were compared with previous clinical photographs to detect any signs of recurrence. For those unable to attend, a telephone interview with a set of questions aimed at determining recurrence was conducted. The patients were asked whether there was a new lesion at the site of previous MSE, whether there had been any change in pigmentation or appearance of the scar, and whether any further procedure had been carried out; if so, the reasons for this were noted and the histopathology report was retrieved. OUTCOME MEASURES Each 5-mm surgical excision was defined as a level of excision. The margin of complete excision was defined by the number of 5-mm levels required. The outcome measures were the number 1088

3 A1 A2 A3 B1 B2 C1 C2 D1 D2 Figure 3. Diagram showing the number and location of tissue blocks. Figure 2. Tissue map showing a suture at the 11-o clock position; nicks at the 3-, 6-, 8-, and 11-o clock positions; and blue (dashes), red (solid line), black (double dots), and yellow (broken line) dyes. of 5-mm margin levels required for complete excision of LM and LMM less than 1 mm thick and the incidence of recurrence. STATISTICAL ANALYSIS Continuous data were analyzed by a t test and described as mean and standard deviation. The Wilcoxon rank-sum test was used for nonnormally distributed data. Where appropriate, 95% confidence intervals were stated. Regression with robust standard errors was used to allow for analysis of 2 sites on the same person. A P value of less than.05 was considered statistically significant. Statistical analysis was performed with Stata Statistical Software version 7.0 (Stata Corp, College Station, Tex). RESULTS 1 cm Figure 4. Tissue blocks laid out for embedding. One hundred fifty-five patients (mean age, 68.2±11.8 years; range, years), of whom 73 (47%) were female and 82 (53%) were male, underwent 161 MSE procedures for head and neck LM and LMM. Six patients had 2 LMs or LMMs at separate sites and underwent more than 1 MSE. Of these cases, 141 with shorter follow-up were previously reported in a study comparing periocular and nonperiocular LMs and LMMs, 17 but in this report we include an additional 20 cases and report on all patients with a longer follow-up. NUMBER OF 5-mm LEVELS REQUIRED FOR COMPLETE EXCISION One hundred twenty-five patients had LM; of these lesions, 30% (37/125) required more than 1 level ( 5-mm margins) for complete excision. Thirty-six patients had LMM; 32 (89%) of 36 LMMs were less than 1 mm thick and 12% (4/32) of these lesions required more than 2 levels ( 10-mm margins) for complete excision. The remaining 4 (11%) of 36 LMMs were 1.4 mm, 1.4 mm, 2.9 mm, and 4.0 mm thick. All additional levels for LM and LMM were for marginal in situ disease (Table). There was no significant relationship between preoperative clinical lesion size (both maximum diameter and estimated area were assessed) and the number of 5-mm levels required for complete histologic excision (data not shown). 1089

4 4.5 mm 5.6 mm Recurrent LM was more likely than primary LM to require more than 1 level for complete clear margin excision (P=.01) (Table). PRIMARY LM AND LMM Twenty percent of cases of primary LM required greater than 5-mm margins for complete excision, and 10% of primary LMMs less than 1 mm in Breslow thickness required more than 10-mm margins for complete excision (Table). UNSUSPECTED LMM 5.1 mm 3.9 mm Thirty-six lesions (22%), all of which were referred by other dermatologists, were invasive melanoma (LMM). At least 28 of these (78%) were not considered to be invasive LMM before biopsy and/or MSE. Twenty-four (67%) were referred with a biopsy result of LM but were found to be LMM after MSE. FOLLOW-UP AND RECURRENCE Figure 5. Diagram showing the distance between tumor and surgical margins. Number of 5-mm Levels for Complete Excision of Primary and Recurrent LM and LMM No. (%) No. of 5-mm Levels for Complete Excision Total Primary Recurrent LM (n = 125) (n = 91) (n = 34) 1 88 (70) 73 (80) 15 (44) 2 28 (24) 14 (15) 14 (41) 3 6 (5) 3 (3) 3 (9) 4 3 (2) 1 (1) 2 (6) (30) 18 (20)* 19 (56)* LMM (n = 36) (n = 24) (n = 12) 1 mm Thick (n = 32) (n = 21) (n = 11) 1 19 (59) 12 (57) 7 (64) 2 9 (28) 7 (33) 2 (18) 3 3 (9) 1 (5) 2 (18) 4 1 (3) 1 (5) 0 (0) (12) 2 (10) 2 (18) 1 mm Thick (n = 4) (n = 3) (n = 1) 1 3 (75) 2 (67) 1 (100) 2 0 (0) 0 (0) 0 (0) 3 1 (25) 1 (33) 0 (0) 4 0 (0) 0 (0) 0 (0) Abbreviations: LM, lentigo maligna; LMM, lentigo maligna melanoma. *P =.01. RECURRENT LM AND LMM Forty-six (29%) of the 161 cases had recurred before MSE; of these, 34 (74%) were LM and 12 (26%) were LMM. Previous treatment modalities included surgical excision (32 cases), cryotherapy (12), and argon or carbon dioxide laser therapy (2). Review of previous histology and post-mse histology reports showed that only 1 LM lesion (previously treated by excision without margin control) recurred as LMM. None of these 46 lesions had previously been treated by MSE. One hundred fifty-five cases (96%) were available for follow-up at a mean period of 38±25 months (range, months; median, 18 months). Four patients (2%) could not be contacted for final follow-up, and 2 patients (1%) died during the follow-up period, both as a result of known cardiac disease. Four cases (2%) recurred after MSE at intervals of 12, 31, 39, and 40 months. All 4 were initially primary LM. They all underwent repeat MSE procedures, at which time 2 were histologically confirmed to be LMM (thickness, 1.1 mm and 0.25 mm). To date, no further recurrence has occurred. Kaplan-Meier statistical analysis to extrapolate recurrence rates gave the probability of recurrence at 5 years as 0.05 (95% confidence interval, ). COMMENT The technique of MSE demonstrates that the current recommendations of 5-mm margins for in situ melanoma and 10-mm margins for invasive melanoma less than 1 mm in Breslow thickness 12,13,18,19 are often insufficient for head and neck LM and LMM. Only 80% of primary LM larger than 10 mm and 90% of primary LMM larger than 10 mm and less than 1 mm in Breslow thickness would be managed effectively with the current recommended margins. In recurrent LM, 56% required more than one 5-mm level of excision, compared with 20% of primary LM. These figures show the tendency toward subclinical extension of LM and LMM, particularly in recurrent lesions, and the usefulness of margin-controlled excision in achieving complete tumor eradication. Our finding that complete excision of LM and LMM often requires wider margins than are currently recommended in the National Institutes of Health and National Health and Medical Research Council guidelines supports the results of previous studies. 9-11,20-25 Zitelli et al 10,22 used a modified form of Mohs surgery to excise 231 head and neck tumors in a series of 553 primary cutaneous melanomas. It is not clear how many of these head and neck tumors were of LM-LMM subtype, although it is likely, given 1090

5 the location, that the majority were LM or LMM. A 6-mm margin was adequate in only 75% of these head and neck tumors, while a 1.5-cm margin was required to excise 97% of lesions. Using Mohs surgery, Robinson 23 found that margins of greater than 6 mm were required to excise 13 (81%) of 16 cases of LM. Zalla et al 24 used Mohs surgery with immunostains and found that only 50% of 46 LM cases were clear with margins of 6 mm or less, while only 71% of 14 LMMs less than 1 mm in Breslow thickness were clear with margins of 10 mm or less. Cohen et al 9 performed Mohs surgery in a series of 45 patients with LM and LMM and found that a mean margin of 13 mm was required for complete excision.agarwal-antaletal, 25 usingthe square technique (staged excision with rushed paraffin-embedded en face sections of margins and rushed vertical sections of a central debulking specimen) described by Johnson et al, 26 found that 58% of 92 LM lesions required margins greater than 5 mm. The knowledge that LM and LMM often extend beyondclinicallyapparentmarginshaspromptedvariousmethods of margin control. Mohs micrographic surgery is the best known technique of margin control; however, visualization of atypical melanocytes on frozen sections can be extremely challenging for the Mohs surgeon and/or pathologist. Cohen et al 27 and Barlow et al 28 reported sensitivities of only 73% and 59%, respectively, combined with specificities of only 68% and 81%, respectively, in frozen sections of LM and LMM. In contrast, Zitelli et al 29 reported sensitivity and specificity of 100% and 90%, respectively, in the diagnosis of melanoma on frozen sections. A variety of modified Mohs surgery techniques to overcome the difficulty with frozen-section interpretation have been described. In 1990, Dhawan et al 30 described the first case of LM treated with a modified form of Mohs micrographic surgery, using rushed paraffin-embedded en face sections, and this approach has been subsequently used with success. 31 Other approaches have included a combination of vertical paraffin sections for central tumor staging and Mohssurgeryfrozen-sectionanalysisofmargins, 10,22,24 Mohs surgery frozen sections of margins in combination with rushed paraffin-section confirmation of tumor-free margins, 9,27 stagedexcisionwithrushedparaffinenfacesections of margins and rushed vertical sections of a central debulkingspecimen, 25,26 verticalcentralparaffinsectionsandrushed paraffin tangential sections of margins in combination with immunohistochemistry, 32 and immunohistochemistry to assess both frozen and paraffin sections. 24,33,34 Recurrence rates for these techniques with follow-up ranging from 2 to 9 years have ranged from 0.5% to 3%. 9,10,22-34 The best results were obtained by Zitelli et al, 10,22 using a modified form of Mohs surgery with frozen-section margin control and vertical paraffin sections of the central tumor, with a recurrence rate of 0.5% at 5 years for 184 in situ melanomas, of which 106 were on the head and neck. Our technique of MSE differs from standard Mohs surgery in certain key aspects. Although the tumor was excised, mapped, and dyed by a Mohs surgeon, numerous vertical bread-loaf paraffin, rather than en face frozen, sections were taken (Figures 1-5). Furthermore, these were assessed by adermatopathologistratherthanthemohssurgeon.themain disadvantage of our technique when compared with Mohs surgery and modified Mohs techniques with paraffin en face sections is that only a percentage of the peripheral margins rather than close to 100% of margins is examined, and this may well account for our small number of recurrences. However, there are advantages of MSE in comparison with Mohs surgery and some of the modified Mohs surgery techniques, including no requirement for a laboratory technician, cryostat, and staining machine for making frozen sections. In addition, other than the greater number of sections required, processing by the pathology service may be carried out in a standard fashion. Furthermore, sections through the tumor permit assessment of marginal melanocytic atypia and hyperplasia in comparison with the central portion of the tumor; the tumor often exists as an island in a sea of sun- and age-related melanocyte abnormality, which must be differentiated, sometimes with difficulty, from the tumor. 35 Sectioning through the tumor out to the surgical margins also allows accurate measurement of the distance between the tumor and the surgical margins. In common with MSE, many of the modified Mohs surgery techniques also describe vertical central sections to permit adequate assessment of adnexal involvement and also the diagnosis and assessment of the depth of invasion in LMM. The use of paraffin sections with dermatopathologist interpretation generates a pathology report with melanoma prognostic indicators; this can be circulated to other medical personnel and sent to central cancer-reporting agencies. Finally, although a few centers of excellence have reported good results with frozen sections, 10,22 we believe that paraffin sections are to be preferred to frozen tissue sections in the interpretation of melanoma. 27,28 Our follow-up data confirm that MSE has a low recurrence rate and compares favorably with Mohs surgery and modified forms of Mohs surgery; with a mean follow-up of 38 months, only 4 cases (2%) recurred. Kaplan-Meier statistical analysis to extrapolate recurrence rates gave the probability of recurrence at 5 years as Like Mohs surgery, MSE permits tissue conservation, of particular importance on the head and neck. The lesions that we excised with MSE are likely to represent a more difficult group of lesions than unselected cases, as all were treated in a tertiary referral practice. A high proportion (29%) were recurrent after previous treatment. Many had at least a component of lightly pigmented to amelanotic tumor, making margin assessment more difficult. In clinical examination of LM and LMM, there are no reliable predictors for the presence of invasive disease other than the presence of an obvious nodule. 5,6 In a series of 85 patients diagnosed clinically as having LM, more than 50% had invasive LMM. 6 Our study confirms the difficulty in clinically distinguishing LM from LMM; none of the 36 patients with LMM had a tumor nodule and more than 75% were not suspected to be invasive LMM by dermatologists before biopsy and/or MSE. Furthermore, the sampling problems associated with biopsies are highlighted by our finding that 24 (67%) of 36 patients with LMM were referred with a biopsy result of LM that was subsequently found to be LMM after MSE. A lack of prospective longitudinal data makes it difficult to estimate the risk of LM progression to LMM. Wayte and Helwig 6 proposed that 30% to 50% of LM may progress to LMM if untreated. However, Weinstock and 1091

6 Sober 3 estimated, on the basis of statistical analysis, that a patient with LM at 45 years of age has a 4.7% lifetime risk of developing LMM, while a 65-year-old patient has a 2.2% lifetime risk. This decreasing lifetime risk of LM transforming to LMM with increasing age at onset most likely reflects patient mortality from unrelated causes before the LM lesion becomes invasive. Thus, we suggest that factors such as age and general health be taken into consideration before undertaking surgical management. In our study, the mean age of patients was 68 years, and all patients were considered fit for surgery. The unknown natural course, lack of reliable clinical predictors of LM and LMM, and the unpredictable subclinical atypical melanocyte extension make a strong argument for complete obliteration of the lesion as the aim of treatment. A wide variety of destructive and surgical therapies have been used. We believe that surgical excision has several advantages. First, it allows assessment of tumor thickness, which is the single most significant prognostic indicator in the management of melanoma. Second, it allows treatment of deep periadnexal melanocytes and detection of both invasive melanoma and atypical melanocytes beyond the clinically apparent margins. Third, surgical excision does not result in the pigmentary changes sometimes seen in destructive modalities such as radiotherapy and cryotherapy; these pigmentary changes may also make subsequent recurrence difficult to detect. 8,36 In conclusion, the current recommendations of 5-mm margins for in situ melanoma (LM) and 10-mm margins for melanoma less than 1 mm in Breslow thickness are sufficient for only 80% of primary LMs larger than 10 mm and 90% of primary LMMs larger than 10 mm and less than 1 mm in Breslow thickness. These guidelines may, however, be useful initial margins for the technique of MSE. As the natural course of LM is essentially unknown and there are difficulties in clinical differentiation of LM and LMM, we suggest that complete excision is the treatment of choice. The findings of this series of head and neck LM and LMM show that MSE appears to be an effective means of attaining a high cure rate in these tumors with characteristic subclinical extension, while also conserving facial and neck tissue. Accepted for publication April 19, This study was supported by an unrestricted financial grant from Allergan, Sydney, Australia. Correspondence: Shyamala C. Huilgol, FACD, Wakefield Clinic, 270 Wakefield St, Adelaide, South Australia 5000, Australia. REFERENCES 1. Holman CDJ, Mulroney CD, Armstrong BK. Epidemiology of pre-invasive and invasive malignant melanoma in Western Australia. IntJCancer. 1980;25: Little JH, Holt J, Davis N. Changing epidemiology of malignant melanoma in Queensland. Med J Aust. 1980;1: Weinstock MA, Sober AJ. The risk of progression of lentigo maligna to lentigo maligna melanoma. Br J Dermatol. 1987;116: Newell GR, Sider JG, Bergfelt L, Kripke ML. Incidence of cutaneous melanoma in the United States by histology with special reference to the face. Cancer Res. 1988;48: Cohen LM. Lentigo maligna and lentigo maligna melanoma [published correction appears in J Am Acad Dermatol. 1997;36(6, pt 1):913]. J Am Acad Dermatol. 1995;33: Wayte DM, Helwig EB. Melanotic freckle of Hutchinson. Cancer. 1968;21: Mahendran RM, Newton-Bishop JA. Survey of U.K. current practice in the treatment of lentigo maligna. Br J Dermatol. 2001;144: Kuflik EG, Gage AA. Cryosurgery for lentigo maligna. J Am Acad Dermatol. 1994; 31: Cohen LM, McCall MW, Hodge SJ, Freedman JD, Callen JP, Zax RH. Successful treatment of lentigo maligna and lentigo maligna melanoma with Mohs micrographic surgery aided by rush permanent sections. Cancer. 1994;73: Zitelli JA, Brown CD, Hanusa BH. Surgical margins for excision of primary cutaneous melanoma. J Am Acad Dermatol. 1997;37: Osborne JE, Hutchinson PE. A follow-up study to investigate the efficacy of initial treatment of lentigo maligna with surgical excision. Br J Plast Surg. 2002; 55: NHMRC and Australian Cancer Network Clinical Practice Guidelines. The Management of Cutaneous Melanoma. Canberra, Australia: NHMRC and Australian Cancer Network, Commonwealth of Australia; NIH Consensus Conference: diagnosis and treatment of early melanoma. JAMA. 1992;268: Veronesi U, Cascinelli N. Narrow excision (1-cm margin): a safe procedure for thin cutaneous melanoma. Arch Surg. 1991;126: Balch CM, Urist MM, Karakousis CP, et al. Efficacy of 2-cm surgical margins for intermediate-thickness melanomas 1 to 4 mm): results of a multiinstitutional randomized surgical trial. Ann Surg. 1993;218: Hill DC, Gramp AA. Surgical treatment of lentigo maligna and lentigo maligna melanoma. Australas J Dermatol. 1999;40: Malhotra R, Chen C, Huilgol SC, Hill DC, Selva D. Mapped serial excision for periocular lentigo maligna and lentigo maligna melanoma. Ophthalmology. 2003; 110: Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg. 1970;172: Cascinelli N, van der Esch EP, Breslow A, Morabito A, Bufalino R, WHO Collaborating Centres for Evaluation of Methods of Diagnosis and Treatment of Melanoma. Stage I melanoma of the skin: the problem of resection margins. Eur J Cancer. 1980;16: Coleman WP III, Davis RS, Reed RJ, Krementz ET. Treatment of lentigo maligna and lentigo maligna melanoma. J Dermatol Surg Oncol. 1980;6: Pitman GH, Kopf AW, Bart RS, Casson PR. Treatment of lentigo maligna and lentigo maligna melanoma. J Dermatol Surg Oncol. 1979;5: Zitelli JA, Brown CD, Hanusa BH. Mohs micrographic surgery for the treatment of primary cutaneous melanoma. J Am Acad Dermatol. 1997;37: Robinson JK. Margin control for lentigo maligna. J Am Acad Dermatol. 1994; 31: Zalla MJ, Lim KK, DiCaudo DJ, Gagnot MM. Mohs micrographic excision of melanoma using immunostains. Dermatol Surg. 2000;26: Agarwal-Antal N, Bowen GM, Gerwels JW. Histologic evaluation of lentigo maligna with permanent sections. J Am Acad Dermatol. 2002;47: Johnson TM, Headington JT, Baker SR, Lowe L. Usefulness of the staged excision for lentigo maligna and lentigo maligna melanoma: the square procedure. J Am Acad Dermatol. 1997;37: Cohen LM, McCall MW, Hodge SJ, Freedman JD, Callen JP, Zax RH. Successful treatment of lentigo maligna and lentigo maligna melanoma with Mohs micrographic surgery aided by rush permanent sections. Cancer. 1994;73: Barlow RJ, White CR, Swanson NA. Mohs micrographic surgery using frozen sections alone may be unsuitable for detecting single atypical melanocytes at the margins of melanoma in situ. Br J Dermatol. 2002;146: Zitelli JA, Moy RL, Abell E. The reliability of frozen sections in the evaluation of surgical margins for melanoma. J Am Acad Dermatol. 1991;24: Dhawan SS, Wolf DJ, Rabinovitz HS, Poulos E. Lentigo maligna: the use of rush permanent sections in therapy. Arch Dermatol. 1990;126: Clayton BD, Leshin B, Hitchcock MG, Marks M, White WL. Utility of rush paraffinembedded tangential sections in the management of cutaneous neoplasms. Dermatol Surg. 2000;26: Stonecipher MR, Leshin B, Patrick J, White WL. Management of lentigo maligna and lentigo maligna melanoma with paraffin-embedded tangential sections: utility of immunoperoxidase staining and supplemental vertical sections. J Am Acad Dermatol. 1993;29: Menaker GM, Chiang JK, Tabila B, Moy RL. Rapid HMB-45 staining in Mohs micrographic surgery for melanoma in situ and invasive melanoma. J Am Acad Dermatol. 2001;44: Gross EA, Andersen WK, Rogers GS. Mohs micrographic excision of lentigo maligna using Mel-5 for margin control. Arch Dermatol. 1999;135: Florell SR, Boucher KM, Leachman SA, et al. Histopathologic recognition of involved margins of lentigo maligna excised by staged excision: an interobserver comparison study. Arch Dermatol. 2003;139: Tsang RW, Liu FF, Wells W, Payne DG. Lentigo maligna of the head and neck: results of treatment by radiotherapy. Arch Dermatol. 1994;130: