Flash Player 9 (or above) is needed to view presentations.
We have detected that you do not have it on your computer. To install it, go here.

Like this document? Why not share!

Utility of Lesion Diameter in the Clinical Diagnosis of ...






Total Views
Views on SlideShare
Embed Views



0 Embeds 0

No embeds



Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

Utility of Lesion Diameter in the Clinical Diagnosis of ... Utility of Lesion Diameter in the Clinical Diagnosis of ... Document Transcript

  • STUDY Utility of Lesion Diameter in the Clinical Diagnosis of Cutaneous Melanoma Naheed R. Abbasi, MPH, MD; Molly Yancovitz, MD; Dina Gutkowicz-Krusin, PhD; Katherine S. Panageas, DrPH; Martin C. Mihm, MD; Paul Googe, MD; Roy King, MD; Victor Prieto, MD; Iman Osman, MD; Robert J. Friedman, MD; Darrell S. Rigel, MD; Alfred W. Kopf, MD; David Polsky, MD, PhD Objective: To determine the utility of the current di- Results: Of 1657 biopsied lesions, 853 (51.5%) were ameter criterion of larger than 6 mm of the ABCDE ac- 6 mm or smaller in diameter. Invasive melanomas were ronym for the early diagnosis of cutaneous melanoma. diagnosed in 13 of 853 lesions (1.5%) that were 6 mm or smaller in diameter and in 41 of 804 lesions (5.1%) Design: Cohort study. that were larger than 6 mm in diameter. In situ melano- mas were diagnosed in 22 of 853 lesions (2.6%) that were Setting: Dermatology hospital-based clinics and com- 6 mm or smaller in diameter and in 62 of 804 lesions munity practice offices. (7.7%) that were larger than 6 mm in diameter. Patients: A total of 1323 patients undergoing skin biop- Conclusion: The diameter guideline of larger than 6 mm sies of 1657 pigmented lesions suggestive of melanoma. provides a useful parameter for physicians and should continue to be used in combination with the A, B, C, and Main Outcome Measure: The maximum lesion di- E criteria previously established in the selection of atypi- mension (diameter) of each skin lesion was calculated cal lesions for skin biopsy. before biopsy using a novel computerized skin imaging system. Arch Dermatol. 2008;144(4):469-474 I N 1985, SEVERAL OF US DEVISED THE teria remain important to public educa- ABCD criteria for melanoma tion in the early recognition of melanoma. screening1 to facilitate the early di- An important challenge to the ABCDE cri- agnosis of cutaneous melanoma. teria has been the recent recognition of The ABCD criteria are evidence- small melanomas ( 6 mm in diameter). based guidelines established to remind phy- Several authors7-11 have described such tu- Author Affiliations: Ronald O. sicians that Asymmetry, Border irregular- mors in US, Australian, Italian, and Is- Perelman Department of ity, Color variegation, and Diameter larger raeli patient cohorts. Our 2004 publica- Dermatology, New York than 6 mm are features characteristic of tion, which reviewed all available small University School of Medicine, melanoma. In recent publications, our group melanoma data from patient cohorts of New York (Drs Abbasi, more than 30 persons, concluded that Yancovitz, Osman, Friedman, small-diameter melanomas likely com- Rigel, Kopf, and Polsky), See also pages 476 and 538 pose 3% to 14% of all cutaneous melano- Electro-Optical Sciences Inc, Irvington (Dr Gutkowicz- mas but rarely result in recurrence, me- reviewed evidence supporting the concept Krusin), and Department of tastasis, or death.2,7-12 Nevertheless, the of lesion change in the diagnosis of cuta- existence of small-diameter melanomas Epidemiology and Biostatistics, neous melanoma and formally recom- complicates the evaluation of pigmented Memorial Sloan-Kettering Cancer Center, New York, mended expansion of the ABCD criteria to lesions suggestive of melanoma. Some au- New York (Dr Panageas); include an E criterion for Evolution.2,3 Rap- thors7,9 have argued that a D criterion of Department of Dermatology, idly growing melanomas, especially nodu- larger than 6 mm may be misleading be- Harvard Medical School, lar melanomas, frequently lack ABCD fea- cause strict adherence to this guideline will Boston, Massachusetts tures.4,5 Thus, the addition of an E criterion result in the failure to biopsy small mela- (Dr Mihm); Knoxville greatly aids in the diagnosis of banal- nomas. Yet others13 point out that more Dermatopathology Laboratory, appearing melanomas whose sole but criti- than 99% of atypical melanocytic lesions Knoxville, Tennessee (Drs Googe and King); and cal concerning clinical feature is that of biopsied by dermatologists are benign. The University of Texas M. D. growth or change over time.5 Clearly, additional data are needed to help Anderson Cancer Center, As the incidence of melanoma contin- physicians balance their desire to biopsy Houston (Dr Prieto). ues to increase worldwide,6 the ABCDE cri- atypical pigmented lesions suggestive of (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 469 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.
  • opsy of each lesion was not recorded; however, lesions sug- Table 1. Distribution of Pigmented Skin Lesions gestive of melanoma were those that generally possessed 1 or in the Study Cohort more of the ABCDE features or were of particular concern to the patient or physician. Lesion Type No. (%) of 1657 Lesions Physicians who participated in the study were composed of Dysplastic nevi 1016 (61.3) a group of dermatologists in both academic and community prac- Other nevi 226 (13.6) tices (see “Clinical Principal Investigators”) across the United Invasive melanoma 54 (3.3) States. They were chosen for their particular interest in the di- Melanoma in situ 84 (5.1) agnosis of pigmented skin lesions and/or a substantial volume Seborrheic keratosis 120 (7.2) of patients potentially requiring skin biopsies. Of the 1657 le- Other 157 (9.5) sions, 44.7% were examined in academic practices and 55.3% were examined in community practices. The proportion of mela- nomas examined in academic practices was 7.6%, vs 8.9% ex- amined in community practices. The observed differences were small and not statistically significant (P=.32). Institutional re- view board approvals were obtained for all participating sites, and all patients signed written informed consent at enroll- ment. MelaFind is an experimental computerized skin imaging sys- tem devised by EOS to diagnose melanoma.14,15 The probe, just slightly larger than a cordless telephone, acquires digital im- ages of pigmented skin lesions illuminated by different spec- tral bands of light from visible to near infrared. It automati- cally extracts pertinent data for image analysis by the EOS central computer. In addition, EOS is developing an image analysis al- gorithm to serve as an aid for physicians in the diagnosis of mela- noma. The images and histopathologic diagnoses collected in this study were part of that development effort; hence, no di- agnostic information was provided to the physicians from the MelaFind system at any time in the study. For this article, the function of the system was to accurately measure the in vivo diameters of pigmented lesions deemed by dermatologists to be potentially malignant. Diameter measurement is not incor- 1 mm porated into the diagnostic computer algorithm being devel- oped to identify lesions as benign or malignant. The maxi- mum diameter is defined as the longest distance between 2 points Figure 1. Representative MelaFind image. The periphery of the lesion, as identified by the image analysis software, is shown in green. Diameter located on the lesion border. Diameter measurements are re- measurements were based on the lesion periphery. A ruler has been added ported to hundredths of millimeters. Image segmentation is based to the image to demonstrate the scale of the image. on lesion images obtained at wavelengths of 430 and 500 nm. A proprietary mathematical operation automatically defines which pixels are in the lesion and which are not. Additional melanoma with the knowledge that most pigmented skin methodologic details can be found in 2 previous publica- lesions are benign. tions.14,15 All lesions in the cohort were biopsied, and histo- To our knowledge, no large studies have described the pathologic diagnoses were independently rendered by 2 der- relationship between in vivo lesion diameter and fre- matopathologists (one diagnosis was rendered by 1 of us quency of melanoma among lesions biopsied because of [M.C.M.] for all lesions and the second diagnosis was ren- concern about melanoma. The main purpose of our study dered by 1 of 3 of us [P.G., R.K., or V.P.]) without previous discussion of diagnostic features. To resolve occasional differ- was to answer the question of whether the D criterion of ences in diagnoses between the 2 dermatopathologists, con- the current ABCDE criteria should be adjusted down- sultation by a third dermatopathologist (1 of 3 of us [P.G., R.K., ward (eg, changing the diameter guideline from 6 to 5 or V.P.]) was obtained. Such differences occurred in only 6% mm) in light of the known existence of small-diameter of cases, a proportion that compares favorably to the consid- melanomas. Our study elucidates the relationship be- erable disagreement reported across the literature for diagno- tween lesion diameter and the diagnosis of melanoma sis of melanocytic lesions by dermatopathologists.16-18 among 1657 consecutive atypical pigmented skin le- At initiation of this analysis, 1657 lesions had been ac- sions biopsied by dermatologists. crued and appraised in the study. We sorted these lesions into categories based on diameter and diagnosis. In our analysis, we were particularly interested in quantifying the number and pro- METHODS portion of melanomas among this cohort of clinically suspi- cious pigmented lesions and among its smaller subset of me- Our group, in collaboration with investigators at Electro- lanocytic neoplasms (n = 1380). Sensitivity (true positive/ Optical Sciences Inc (EOS), collected and analyzed diameter [true positive false negative]) and specificity (true negative/ measurements and histopathologic diagnoses of 1657 consecu- [false positive true negative]) were calculated for various cut tive pigmented skin lesions in 1323 patients accrued as part of points for the diagnosis of melanoma (invasive and in situ). The the second phase of the multicenter MelaFind Study.14 Pa- following definitions were used: true positive was defined as mela- tients of any age, race, ethnicity, or sex with pigmented le- nomas with diameters greater than the cutoff value, false posi- sions suggestive of melanoma and requiring skin biopsy were tive was defined as nonmelanomas with diameters greater than eligible for enrollment in the study. The specific reason for bi- the cutoff value, true negative was defined as nonmelanomas (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 470 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.
  • Table 2. Proportion of Melanomas Among All Lesions Stratified by Lesion Diameter Diameter, mm Total Lesions a Invasive Melanomas b In Situ Melanomas b Total Melanomas b 1.01-2.00 1 (0.1) 0 0 0 2.01-3.00 94 (5.7) 2 (2.1) 2 (2.1) 4 (4.3) 3.01-4.00 220 (13.3) 1 (0.5) 9 (4.1) 10 (4.5) 4.01-5.00 281 (17.0) 5 (1.8) 5 (1.8) 10 (3.6) 5.01-6.00 257 (15.5) 5 (1.9) 6 (2.3) 11 (4.3) 6.01-7.00 192 (11.6) 7 (3.6) 9 (4.7) 16 (8.3) 7.01-8.00 186 (11.2) 4 (2.2) 11 (5.9) 15 (8.1) 8.01-9.00 127 (7.7) 4 (3.1) 12 (9.4) 16 (12.6) 9.01-10.00 89 (5.4) 4 (4.5) 6 (6.7) 10 (11.2) 10.00 210 (12.7) 22 (10.5) 24 (11.4) 46 (21.9) Total No. (%) a 1657 54 (3.3) 84 (5.1) 138 (8.3) a Data are given as number (percentage) of the 1657 total lesions. b Data are given as number (percentage) of total lesions in each diameter category. 25.0 In situ lesions Invasive lesions 21.9% 20.0 15.0 Melanoma, % 12.6% 11.2% 10.0 8.3% 8.1% 4.5% 4.3% 5.0 4.3% 3.6% 0 2.01-3.00 3.01-4.00 4.01-5.00 5.01-6.00 6.01-7.00 7.01-8.00 8.01-9.00 9.01-10.00 >10.00 Diameter, mm Figure 2. Proportion of melanomas among all lesions. Each category on the x-axis represents a range of lesion diameters. The proportion of melanomas among biopsied lesions within these ranges is expressed as percentages indicated above each bar and on the y-axis. A break point in the proportion of melanomas among biopsied lesions is apparent at diameters larger than 6.0 mm. with diameters less than the cutoff value, and false negative was study, 138 (8.3%) were histologically diagnosed as mela- defined as melanomas with diameters less than the cutoff value. noma, including invasive (n=54) and in situ (n=84) le- A receiver operating characteristic (ROC) curve was esti- sions. Other diagnoses included dysplastic nevi (n=1016), mated nonparametrically to visually summarize the accuracy other melanocytic nevi (eg, Spitz and congenital) (n=226), of the predictions.19 The unit of analysis throughout the study was pigmented lesion(s), not individual patients. In patients and other nonmelanocytic lesions (eg, seborrheic kera- with more than 1 clinically suspicious pigmented lesion, each tosis) (n=277) (Table 1). was assessed separately with regard to its diameter and diag- Eight hundred four lesions (48.5%) were greater than nosis. In this study the patients contributed 1.25 lesions per 6 mm in diameter, and 853 (51.5%) were 6 mm in di- patient. When we accounted for multiple lesions per patient, ameter or less. A representative MelaFind image used to the area under the curve for the ROC curve was exactly the same calculate lesion diameter is shown in Figure 1. Within (0.68; 95% confidence interval, 0.63-0.73) as when we used each 1-mm diameter range from 2.01 to 6.00 mm, the the regular nonclustered methods.20 proportion of melanomas did not vary significantly, re- maining stable at 3.6% to 4.5% (Table 2 and Figure 2). RESULTS However, we observed a nearly 100% increase in the pro- portion of melanomas when comparing the 5.01- to A total of 852 women and 805 men participated in the 6.00-mm category (4.3%) to the 6.01- to 7.00-mm cat- study. The mean age of the women was 43.5 years (me- egory (8.3%). Among lesions 6.01 mm in diameter or dian age, 42 years); the mean age of the men was 50 years larger, the proportion of melanomas ranged from 8.1% (median age, 51 years). Among the 1657 lesions under to 21.9%, roughly increasing as lesion diameters in- (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 471 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.
  • Table 3. Proportions of Melanoma, Dysplastic Nevi, and Other Nevi Among 1380 Melanocytic Neoplasms Stratified by Lesion Diameter Diameter, mm Melanoma a Dysplastic Nevi a Other Nevi a Total Lesions b 1.01-2.00 0 0 1 (100) 1 (0.1) 2.01-3.00 4 (5.2) 56 (72.7) 17 (22.1) 77 (5.6) 3.01-4.00 10 (5.3) 144 (76.2) 35 (18.5) 189 (13.7) 4.01-5.00 10 (4.0) 202 (80.5) 39 (15.5) 251 (18.2) 5.01-6.00 11 (4.9) 183 (81.0) 32 (14.2) 226 (16.4) 6.01-7.00 16 (9.9) 112 (69.1) 34 (21.0) 162 (11.7) 7.01-8.00 15 (10.6) 107 (75.9) 19 (13.5) 141 (10.2) 8.01-9.00 16 (16.2) 68 (68.7) 15 (15.2) 99 (7.2) 9.01-10.00 10 (13.7) 54 (74.0) 9 (12.3) 73 (5.3) 10.00 46 (28.6) 90 (55.9) 25 (15.5) 161 (11.7) Total No. (%) 138 (10.0) 1016 (73.6) 226 (16.4) 1380 a Data are given as number (percentage) of total lesions for the indicated diameter category. b Data are given as number (percentage) of the 1380 total lesions. There was a generally constant proportion of melano- Table 4. Sensitivity and Specificity of Selected Lesion mas (4.0%-5.3%) among lesions 6 mm in diameter or Diameters Derived From Receiver Operating smaller but a significantly larger and increasing propor- Characteristic Curve Analysis tion of melanomas (9.9%-28.6%) among lesions with di- Sensitivity Specificity ameters greater than 6 mm. Cutoff, mm (n=138) a (n=1519) a Although it was not the intent of this study to ana- lyze the diameter criterion as a stand-alone screening test 4.0 90 (84-94) [124] 20 (18-22) [301] 5.0 83 (75-88) [114] 38 (35-40) [572] for melanoma, we performed an ROC analysis to assess 6.0 75 (67-81) [103] 54 (51-57) [818] the clinical utility of various diameter measurements in 7.0 63 (55-71) [87] 65 (63-68) [994] the diagnosis of melanoma. Specifically, we calculated the 8.0 52 (44-61) [72] 77 (74-79) [1165] sensitivity and specificity of diameters of 4, 5, 6, 7, 8, 9, 9.0 41 (33-49) [56] 84 (82-86) [1276] and 10 mm for the diagnosis of melanoma (Table 4). 10.0 33 (26-42) [46] 89 (87-91) [1335] As diameter increased from 4 to 10 mm, sensitivity de- a Data are given as percentage (95% confidence interval) [number of creased for the diagnosis of melanoma and specificity in- lesions]. creased. The 6-mm diameter criterion yielded a sensi- tivity of 75% and a specificity of 54%. In comparison, a 5-mm cutoff had a sensitivity of 83% and a specificity of creased. We analyzed the proportions of invasive and in 38% and a 7-mm cutoff had a sensitivity of 63% and a situ melanomas separately and found that invasive mela- specificity of 65%. nomas were diagnosed in 13 of 853 lesions (1.5%) that were 6 mm in diameter or smaller and in 41 of 804 le- COMMENT sions (5.1%) that were larger than 6 mm in diameter, whereas in situ melanomas were diagnosed in 22 of 853 lesions (2.6%) that were 6 mm in diameter or smaller and Our study design is unique in several ways. First, we used in 62 of 804 lesions (7.7%) that were larger than 6 mm computer technology to measure the diameters of pig- in diameter. Recognizing the importance of Breslow thick- mented lesions with great accuracy and precision, thus ness to prognosis in melanoma, we also analyzed mean considerably reducing the possibility of interobserver vari- tumor thickness for invasive melanomas in each inter- ability introduced by human measurements of skin le- val listed in Table 2 but could not identify trends or a sions. Second, we measured pigmented lesions in vivo reasonable cutoff based on tumor thickness (data not and reported biopsy results of all lesions studied. Third, shown). we studied a large cohort of suspicious but mostly be- We were also interested in determining whether the nign lesions to better understand the frequency of mela- striking increase in melanoma proportion among le- nomas among clinically suspicious atypical pigmented sions greater than 6 mm would also be observed when lesions. With the exception of 1 study, 9 other au- the analysis was restricted only to melanocytic neo- thors7-11,21 have focused their attention on populations plasms (nevi and melanoma). Table 3 reports the fre- of ex vivo biopsy-confirmed small melanomas. Ex vivo quency of melanomas vs nevi of various diameters. Mela- specimen measurement may result in underestimation nomas comprise in situ and invasive lesions; the remaining of lesion diameter by as much as 20%.22,23 The rigor of melanocytic neoplasms comprise dysplastic nevi and other our study design allows much greater insight into the re- nevi, which include Spitz nevi, congenital nevi, and blue lationship between lesion diameter and diagnosis than nevi. We again observed a 100% increase in the propor- any previous study by using accurate lesion measure- tion of melanomas among the 6.01- to 7.00-mm cat- ments rather than the rough estimates not uncommon egory (9.9%) vs the 5.01- to 6.00-mm category (4.9%). in dermatologic practice. (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 472 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.
  • The goal of this study was to determine if a diameter sis indicates that a diameter criterion of larger than 6 mm measurement of greater than 6 mm was still a useful guide- is not sensitive or specific enough to serve as a stand- line in the diagnosis of melanoma. Ideally, we would want alone screening test for melanoma, consistent with the to determine the frequency of melanomas among all pig- observations of Thomas et al. Unfortunately, in the pres- mented skin lesions; however, this would be exceed- ent study we were unable to examine the utility of the ingly difficult. Melanoma is still a relatively rare tumor, A, B, C, or E criteria or combinations thereof; however, and the average adult patient has between 11 and 50 nevi, Thomas et al found that horizontal enlargement had the with 5% of patients having at least 1 atypical nevus.24 Thus, greatest combination of sensitivity and specificity for mela- to conduct a clinical trial to calculate the frequency of noma, followed by the D criterion, so it may be espe- melanomas among all pigmented skin lesions would re- cially helpful to consider lesional evolution (E), such as quire the removal and histopathologic diagnosis of all pig- horizontal enlargement, in lesions 6 mm or smaller. mented skin lesions on each patient, a nearly impos- The significant increase in the proportion of melano- sible and extraordinarily expensive task. The present study mas among lesions greater than 6 mm in diameter sug- took advantage of a cohort of suspicious clinically atypi- gests that a lesion size of 6 mm may have biological sig- cal lesions that were biopsied primarily because of a con- nificance. The current theory regarding cellular senescence cern for melanoma. This study design would likely en- in melanocytic nevi26,27 predicts that nevus cells un- rich the cohort with melanomas compared with a study dergo a certain number of cell divisions, after which cel- that biopsied all pigmented lesions, yet even within this lular mechanisms of long-term growth arrest (eg, tumor more selective cohort, the rate of melanoma was low. Al- suppressor genes) halt proliferation. Loss of these growth though we do not know how many melanomas were arrest mechanisms is common in melanoma28; however, missed by this group of physicians, the physicians who much additional work is needed to correlate the fre- enrolled patients into the MelaFind study were experi- quency of these biological alterations with lesion size. enced dermatologists who are, arguably, less likely to miss In conclusion, the present study reaffirms that a diam- many malignant lesions or to select benign lesions for eter criterion of larger than 6 mm is a useful guideline for further appraisal than their less specialized colleagues. the early detection of cutaneous melanoma. We do not rec- Nevertheless, the fact that most lesions biopsied in the ommend downward revision of the D criterion at this time. MelaFind Study (853 of 1657) were less than 6 mm in In the United States, rates of melanoma and nonmela- diameter suggests a substantial concern among physi- noma skin cancers have markedly increased, and skin bi- cians not to miss small-diameter melanomas. Despite the opsy rates have more than doubled in 20 years.29 In an era inherent selection bias of our study design, our data pro- that demands greater data to support clinical decision mak- vide strong support for the idea that small melanomas ing, the ABCDE criteria are valuable evidence-based guide- are rare, even among a cohort of clinically atypical pig- lines to aid physicians in decisions regarding the biopsy mented lesions that experienced physicians may excise. of pigmented lesions of the skin. We recommend that a diameter criterion of larger than 6 mm remain a part of the ABCDE criteria. Lesion diam- Accepted for Publication: June 20, 2007. eter greater than 6 mm seems to be a significant indica- Correspondence: David Polsky, MD, PhD, Ronald O. Per- tor of increased suspicion for melanoma because our data elman Department of Dermatology, New York Univer- reveal a significant increase in the proportion of mela- sity School of Medicine, 550 First Ave, Room H-100, New nomas among lesions larger than 6 mm when compared York, NY 10016 (David.Polsky@nyumc.org). with lesions 6 mm or smaller (Figure 2 and Table 2). In Author Contributions: Dr Polsky had full access to all the addition, data from the ROC analysis suggest that down- data in the study and takes responsibility for the integrity ward revision of the diameter criterion to 5 mm would of the data and the accuracy of the data analysis. Study con- result in a significant loss of specificity. Because more than cept and design: Abbasi, Prieto, Osman, Friedman, Rigel, 50% of all suspicious lesions biopsied in this study were Kopf, and Polsky. Acquisition of data: Abbasi, Gutkowicz- 6 mm or smaller yet only 1.5% of those were invasive Krusin, Mihm, Googe, King, and Prieto. Analysis and melanoma and 2.6% were in situ melanoma, it seems that interpretation of data: Abbasi, Yancovitz, Panageas, Prieto, many biopsies are being performed on small lesions based Osman, and Polsky. Drafting of the manuscript: Abbasi, on their A, B, C, and E characteristics, with a low speci- Prieto, and Polsky. Critical revision of the manuscript for ficity for detecting melanoma. It is likely that lowering important intellectual content: Abbasi, Yancovitz, Gutkowicz- the D criterion to 5 mm would only further decrease the Krusin, Panageas, Mihm, Googe, King, Prieto, Osman, specificity of such biopsies. Friedman, Rigel, Kopf, and Polsky. Statistical analysis: For pigmented lesions with diameters that measure 6 Panageas. Administrative and technical support: Gutkowicz- mm or less, we believe that the use of the ABCDE2 cri- Krusin. Study supervision: Osman and Polsky. teria in combination can be an effective management strat- Financial Disclosure: Dr Gutkowicz-Krusin is an em- egy. Specifically, identifying lesions that possess more ployee of EOS; Drs Mihm, Googe, King, and Prieto are atypical features (eg, A, B, C, or E) is likely to be more dermatopathologists for the MelaFind Study, sponsored specific for melanoma. Thomas et al25 demonstrated that by EOS; Dr Friedman is a consultant for and share- specificity for the diagnosis of melanoma increases mark- holder in EOS; Dr Rigel is a consultant for EOS; Drs Kopf edly as the number of ABCDE criteria present in a pig- and Polsky are investigators for the MelaFind Study spon- mented lesion increase from 1 or more (36%) to 4 or more sored by EOS; and Dr Polsky is a consultant for EOS. (94%); however, there is a concomitant decrease in sen- Role of the Sponsor: The data were extracted from a clini- sitivity from 97% to 54%, respectively. Our ROC analy- cal study sponsored by EOS; EOS provided data to the (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 473 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.
  • research team but had no role in study design, data analy- relation and DNA ploidy analysis. J Am Acad Dermatol. 1990;22(6, pt 1):1032- 1038. sis, or data interpretation. 11. Bergman R, Katz I, Lichtig C, Ben-Arieh Y, Moscona AR, Friedman-Birnbaum R. Clinical Principal Investigators: Jeffrey Callen, MD, Uni- Malignant melanomas with histologic diameters less than 6 mm. J Am Acad versity of Louisville, Louisville, Kentucky; Armand Dermatol. 1992;26(3, pt 2):462-466. Cognetta, MD, Dermatology Associates of Tallahassee, 12. Schmoeckel C. Small malignant melanomas: clinicopathologic correlation and DNA Tallahassee, Florida; Caron Grin, MD, University of Con- ploidy analysis [reply letter]. J Am Acad Dermatol. 1991;24(6, pt 1):1036-1037. 13. Ackerman AB, Mihara I. Dysplasia, dysplastic melanocytes, dysplastic nevi, the necticut, Hartford; Kenneth Gross, MD, Skin Surgery dysplastic nevus syndrome, and the relation between dysplastic nevi and ma- Medical Group, San Diego, California; Allan Halpern, MD, lignant melanomas. Hum Pathol. 1985;16(1):87-91. Memorial Sloan-Kettering Cancer Center, New York, New 14. Elbaum M, Kopf AW, Rabinovitz HS, et al. Automatic differentiation of mela- York; Peter Lee, MD, University of Minnesota, Minne- noma from melanocytic nevi with multispectral digital dermoscopy: a feasibility apolis; Seth Lerner, MD, Adult and Pediatric Dermatol- study. J Am Acad Dermatol. 2001;44(2):207-218. ogy Specialists, Trumbull, Connecticut; Norman Levine, 15. Gutkowicz-Krusin D, Elbaum M, Jacobs A, et al. Precision of automatic mea- surements of pigmented skin lesion parameters with a MelaFind™ multispectral MD, University of Arizona Health Sciences Center, Tuc- digital dermoscope. Melanoma Res. 2000;10(6):563-570. son; Gary Monheit, MD, Total Skin and Beauty Derma- 16. Corona R, Mele A, Amini M, et al. Interobserver variability on the histopatho- tology, Birmingham, Alabama; Gary Peck, MD, Wash- logic diagnosis of cutaneous melanoma and other pigmented skin lesions. J Clin ington Cancer Institute, Washington, DC; Harold Oncol. 1996;14(4):1218-1223. Rabinovitz, MD, Skin & Cancer Associates, Plantation, 17. Farmer ER, Gonin R, Hanna MP. Discordance in the histopathologic diagnosis of melanoma and melanocytic nevi between expert pathologists. Hum Pathol. Florida; Jennifer Schwartz, MD, University of Michigan, 1996;27(6):528-531. Ann Arbor; Nancy Thomas, MD, PhD, University of North 18. Meyer LJ, Piepkorn M, Goldgar DE, et al. Interobserver concordance in discrimi- Carolina, Chapel Hill; Yardy Tse, MD, Dermatology As- nating clinical atypia of melanocytic nevi, and correlations with histologic atypia. sociates of San Diego, Encinitas, California. J Am Acad Dermatol. 1996;34(4):618-625. 19. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. REFERENCES Biometrics. 1988;44(3):837-845. 20. Obuchowski NA. Nonparametric analysis of clustered ROC curve data. Biometrics. 1. Friedman RJ, Rigel DS, Kopf AW. Early detection of malignant melanoma: the 1997;53(2):567-578. role of physician examination and self-examination of the skin. CA Cancer J Clin. 21. Helsing P, Loeb M. Small diameter melanoma: a follow-up of the Norwegian Mela- 1985;35(3):130-151. noma Project. Br J Dermatol. 2004;151(5):1081-1083. 2. Abbasi NR, Shaw HM, Rigel DS, et al. Early diagnosis of cutaneous melanoma: 22. Golomb FM, Doyle JP, Grin CM, Kopf AW, Silverman MK, Levenstein MJ. De- revisiting the ABCD criteria. JAMA. 2004;292(22):2771-2776. termination of preexcision surgical margins of melanomas from fixed-tissue 3. Rigel DS, Friedman RJ, Kopf AW, Polsky D. ABCDE: an evolving concept in the specimens. Plast Reconstr Surg. 1991;88(5):804-809. early detection of melanoma. Arch Dermatol. 2005;141(8):1032-1034. 23. Silverman MK, Golomb FM, Kopf AW, et al. Verification of a formula for deter- 4. Liu W, Dowling JP, Murray WK, et al. Rate of growth in melanomas: character- mination of preexcision surgical margins from fixed-tissue melanoma specimens. istics and associations of rapidly growing melanomas. Arch Dermatol. 2006; J Am Acad Dermatol. 1992;27(2, pt 1):214-219. 142(12):1551-1558. 24. Schafer T, Merkl J, Klemm E, Wichmann HE, Ring J. The epidemiology of nevi ¨ 5. Lipsker D. Growth rate, early detection, and prevention of melanoma: mela- and signs of skin aging in the adult general population: results of the KORA- noma epidemiology revisited and future challenges. Arch Dermatol. 2006;142 survey 2000. J Invest Dermatol. 2006;126(7):1490-1496. (12):1638-1640. 25. Thomas L, Tranchand P, Berard F, Secchi T, Colin C, Moulin G. Semiological value 6. Thompson JF, Scolyer RA, Kefford RF. Cutaneous melanoma. Lancet. 2005;365 of ABCDE criteria in the diagnosis of cutaneous pigmented tumors. Dermatology. (9460):687-701. 1998;197(1):11-17. 7. Gonzalez A, West AJ, Pitha JV, Taira JW. Small-diameter invasive melanomas: 26. Bennett DC. Human melanocyte senescence and melanoma susceptibility genes. clinical and pathologic characteristics. J Cutan Pathol. 1996;23(2):126-132. Oncogene. 2003;22(20):3063-3069. 8. Shaw HM, McCarthy WH. Small-diameter malignant melanoma: a common di- 27. Michaloglou C, Vredeveld LC, Soengas MS, et al. BRAFE600-associated senescence- agnosis in New South Wales, Australia. J Am Acad Dermatol. 1992;27(5, pt 1): like cell cycle arrest of human naevi. Nature. 2005;436(7051):720-724. 679-682. 28. Sharpless E, Chin L. The INK4a/ARF locus and melanoma. Oncogene. 2003;22(20): 9. Bono A, Bartoli C, Moglia D, et al. Small melanomas: a clinical study of 270 con- 3092-3098. secutive cases of cutaneous melanoma. Melanoma Res. 1999;9(6):583-586. 29. Welch HG, Woloshin S, Schwartz LM. Skin biopsy rates and incidence of mela- 10. Kamino H, Kiryu H, Ratech H. Small malignant melanomas: clinicopathologic cor- noma: population based ecological study. BMJ. 2005;331(7515):481. (REPRINTED) ARCH DERMATOL/ VOL 144 (NO. 4), APR 2008 WWW.ARCHDERMATOL.COM 474 Downloaded from www.archdermatol.com on November 10, 2010 ©2008 American Medical Association. All rights reserved.