Accuracy in melanoma detection: A 10-year multicenter survey Giuseppe Argenziano, MD, Lorenzo Cerroni, MD, Iris Zalaudek, MD, Stefania Staibano, MD, Rainer Hofmann-Wellenhof, MD, Nicola Arpaia, MD, Renato Marchiori Bakos, MD, PhD, Brigitte Balme, MD, Jadran Bandic, MD, Roberto Bandelloni, MD, Alexandra M. G. Brunasso, MD, Horacio Cabo, MD, David A. Calcara, BS, Blanca Carlos-Ortega, MD, Ana Carolina Carvalho, MD, Gabriel Casas, MD, Huiting Dong, MD, DMSc, Gerardo Ferrara, MD, Raffaele Filotico, MD, Guillermo G omez, MD, Allan Halpern, MD, Gennaro Ilardi, MTD, PhD, Akira Ishiko, MD, PhD, Gulsen Kandiloglu, MD, Hiroshi Kawasaki, MD, Ken Kobayashi, MD, Hiroshi Koga, MD, Ivanka Kovalyshyn, MD, David Langford, MB, ChB, Xin Liu, MD, Ashfaq A. Marghoob, MD, Massimo Mascolo, MD, Cesare Massone, MD, Laura Mazzoni, MD, Scott Menzies, MBBS, PhD, Akane Minagawa, MD, Loredana Nugnes, MD, Fezal Ozdemir, MD, Giovanni Pellacani, MD, Stefania Seidenari, MD, Katherine Siamas, MD, Ignazio Stanganelli, MD, William V. Stoecker, MD, Masaru Tanaka, MD, Luc Thomas, MD, Philipp Tschandl, MD, and Harald Kittler, MD Reggio Emilia, Naples, Bari, Genoa, Benevento, Meldola, Modena, Italy; Graz and Vienna, Austria; Porto Alegre, Brazil; Lyon, France; Belgrade, Serbia; Buenos Aires, Argentina; Rolla, Missouri; Mexico City, Mexico; Camperdown, Australia; Zhengzhou, Yongcheng, China; New York, New York; Tokyo and Matsumoto, Japan; Bornova Izmir, Turkey; and Merivale Christchurch, New Zealand Background: Early excision is the only strategy to reduce melanoma mortality, but unnecessary excision of benign lesions increases morbidity and healthcare costs. Objective: To assess accuracy in melanoma detection based on number-needed-to-excise (NNE) values over a 10-year period. Methods: Information was retrieved on all histopathologically confirmed cutaneous melanomas or melanocytic nevi that were excised between 1998 and 2007 at participating clinics. NNE values were calculated by dividing the total number of excised lesions by the number of melanomas. Analyses included changes in NNE over time, differences in NNE between specialized clinical settings (SCS) versus non-specialized clinical settings (NSCS), and patient factors influencing NNE. Results: The participating clinics contributed a total of 300,215 cases, including 17,172 melanomas and 283,043 melanocytic nevi. The overall NNE values achieved in SCS and NSCS in the 10-year period were 8.7 and 29.4, respectively. The NNE improved over time in SCS (from 12.8 to 6.8), but appeared unchanged in NSCS. Most of the effect on NNE in SCS was due to a greater number of excised melanomas. Higher NNE values were observed in patients younger than 40 years and for lesions located on the trunk. Limitations: No data concerning the use of dermatoscopy and digital monitoring procedures were collected from the participating centers. Conclusion: Over the 10-year study period, accuracy in melanoma detection improved only in specialized clinics maybe because of a larger use of new diagnostic techniques such as dermatoscopy. ( J Am Acad Dermatol 2012;67:54-9.) Key words: clinical diagnosis; dermatoscopy; melanoma; number needed to excise; skin cancer. Institutional affiliations for all authors may be found in the online version of this article at http://www.eblue.org. Funding sources: None. Conflicts of interest: None declared. Accepted for publication July 18, 2011. Reprint requests: Giuseppe Argenziano, MD, Dermatology Unit, Medical Department, Arcispedale Santa Maria Nuova, Viale 54 Risorgimento 80 - 42100 Reggio Emilia, Italy. E-mail: g.argenziano@ gmail.com. Published online October 10, 2011. 0190-9622/$36.00 Ó 2011 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2011.07.019 Argenziano et al 55 J AM ACAD DERMATOL VOLUME 67, NUMBER 1 BACKGROUND management (pigmented lesion clinic). A nonspeEarly excision is the only strategy to reduce cialized clinical setting (NSCS) was defined as a clinic mortality associated with melanoma, but unnecesnot dedicated specifically to skin cancer, in which a sary excision of benign lesions increases morbidity primary care physician, a dermatologist or any other and raises healthcare costs associated with melaspecialist might treat patients with various dermatonoma screening.1 One of the most useful metrics logic conditions, including skin tumors. for measuring accuracy in melanoma detection is From the databases of the participating centers we the number needed to excise extracted information on all (NNE), calculated as the numskin tumors that were diagCAPSULE SUMMARY ber of melanocytic lesions nosed histopathologically as excised for every confirmed cutaneous melanoma or melOne of the most useful metrics for melanoma. NNE values vary anocytic nevus and that were measuring accuracy in melanoma according to clinician experexcised between 1998 and detection is the number needed to tise, with reported values 2007. The data collected inexcise (NNE), calculated as the number of ranging from 20 to 40 for cluded the age and sex of the melanocytic lesions excised for every general practitioners at nonpatient, location of the exconfirmed melanoma. specialized clinics, from 19 to cised lesion, and Breslow The NNE values achieved in specialized 28 for general practitioners at thickness in the case of clinical settings (SCS) and nonskin cancer clinics, and from melanoma. specialized clinical settings (NSCS) in the 4 to 18 for dermatologists at 10-year period were 8.7 and 29.4, specialized clinics.2,3 Statistical analysis respectively. The NNE improved over The introduction of derThe number needed to extime in SCS (from 12.8 to 6.8), but matoscopy into clinical cise (NNE) was calculated by appeared unchanged in NSCS. practice was aimed specifidividing the total number of cally at helping clinicians excised lesions by the numMost of the effect on NNE in SCS was to improve performance in ber of melanomas. Changes due to a greater number of excised melanoma detection. Two in NNE values over time were melanomas. Higher NNE values were meta-analyses performed in evaluated separately for speobserved in patients younger than 40 both experimental and clincialized and non-specialized years and for lesions located on the ical settings have shown centers using the Cochrantrunk. that, when used by experts, Armitage trend test. When dermatoscopy is associated the test for trend was signifiwith a significant improvement of sensitivity for cant (P\.05), the slope was estimated by using linear melanoma.4,5 In two additional studies, one ranregression. The estimation of the slope provided a domized and one retrospective, experts using derquantitative interpretation of the magnitude of the matoscopy were able to improve the NNE value by trend, that is, the mean yearly increase (positive decreasing the number of unnecessary excisions of slope) or decrease (negative slope) of the dependent benign lesions.3,6 However, data are lacking that variable. A logistic regression model was used for might reveal whether dermatoscopy could similarly multivariate analysis. Statistical analyses were perimprove accuracy of melanoma detection in nonformed by using SPSS 18.0 (SPSS, Chicago, IL) and specialized clinical settings. StaXact (Cytel, Cambridge, MA) statistical software We conducted a multicenter survey to investigate packages. All P values reported are two tailed and a (1) changes in NNE values over a 10-year period P value less than .05 indicates statistical significance. (from 1998 to 2007), (2) differences in NNE values at The large number of melanomas with missing specialized versus nonspecialized clinics, and (3) information concerning thickness prevented our patient factors influencing NNE values. ability to analyze trends about melanoma thickness. d d d METHODS RESULTS Clinics were recruited to participate in the survey through e-mail solicitation to board members of the International Dermoscopy Society (http://www. dermoscopy-ids.org/). Recruitment was targeted to include cases from both specialized and nonspecialized clinics. A specialized clinical setting (SCS) was defined as a clinic dedicated to skin cancer Twenty-three of 40 centers that were solicited by e-mail agreed to participate in the survey. Participating clinics consisted of 21 clinical centers and 2 dermatopathology units from 13 countries (Argentina, Australia, Austria, Brazil, China, France, Japan, Italy, Mexico, New Zealand, Serbia, Turkey, and the United States). Clinical centers were 56 Argenziano et al J AM ACAD DERMATOL JULY 2012 Abbreviations used: NNE: NSCS: SCS: number needed to excise nonspecialized clinical setting specialized clinical setting hospitals (4 centers), academic (13 centers), and private units (4 centers). The two dermatopathology units (established in Graz, Austria, and Naples, Italy) were both academic referral centers serving as regional collectors of specimens from academic and private clinicians. Whereas the clinical centers participating in this study were all categorized as SCS, the two dermatopathology units (in Graz and Naples) received specimens from both SCS and NSCS clinics. The participating clinics contributed a total of 300,215 histopathologically confirmed cases, including 17,172 melanomas and 283,043 melanocytic nevi. The overall NNE values achieved in SCS and NSCS in the 10-year period were 8.7 and 29.4, respectively. Notably, diagnostic assessment as measured by NNE clearly improved over time in SCS, but appeared unchanged in NSCS (Table I and Fig 1). From 1998 to 2007 the mean NNE values in SCS decreased from 12.8 to 6.8, with a reduction per year of 0.6 (95% confidence interval [CI]: 0.4-0.7; Armitage test for trend: P \.001). In NSCS the NNE showed no significant trend over time, decreasing slightly from 31.9 to 28.5 (Armitage test for trend: P = .45). Changes over time in the numbers of excised melanomas and nevi differed between SCS and NSCS. As shown in Fig 2, the total number and the proportion of excised melanomas significantly increased by 1.4% per year in SCS (trend test, calculated dividing the number of melanomas by all excised lesions: P \ .001), whereas the number of excised melanomas decreased by 0.03% per year in NSCS (trend test: P = .12). Parallel to that, the proportion of excised nevi (Fig 3) decreased by 0.7% per year in SCS (trend test: P \ .001), but remained basically stable in NSCS. As shown in Fig 4, the total number of melanomas increased in direct proportion with patient age, being 3 times higher in patients older than 60 years (6487 melanomas) compared with patients between 51 and 60 years of age (2383 melanomas). The numbers of melanomas excised in the remaining age groups were as follows: ages 41-50 years, 1855; ages 31-40, 1416; ages 21-30, 656; and ages 0-20 years, 155 melanomas. For the remaining 4220 melanomas, patient age was not available. In both SCS and NSCS, the highest overall number of excised nevi Table I. Number of excised melanomas (including Breslow thickness) and melanocytic nevi and NNE achieved in SCS and NSCS between 1998 and 2007 SCS 10 years 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 NSCS 10 years 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Melanoma Nevi NNE MM in situ MM \1 mm MM [1 mm NA 9910 596 738 820 832 867 925 954 1261 1377 1540 7262 650 804 846 765 754 728 735 617 613 750 76,183 7023 7288 7059 7227 7112 6922 7711 8277 8559 9005 206,860 20,067 22,217 22,202 20,227 20,297 21,341 21,369 19,114 19,367 20,659 8.7 12.8 10.8 9.6 9.7 9.2 8.5 9.1 7.6 7.2 6.8 29.4 31.9 28.6 27.2 27.4 27.9 30.3 30.1 32.0 32.6 28.5 1428 58 89 83 109 117 105 107 247 253 260 2669 191 236 299 290 313 239 276 278 234 313 2734 153 155 228 237 204 265 257 368 416 451 870 25 47 45 41 37 36 55 190 175 219 1165 61 73 106 88 100 98 109 159 172 199 504 37 40 43 35 24 29 38 62 98 98 4583 324 421 403 398 446 457 481 487 536 630 3219 397 481 459 399 380 424 366 87 106 120 MM, Melanoma; NA, thickness not available; NNE, number needed to excise (total/melanoma); NSCS, non-specialized clinical setting; SCS, specialized clinical setting. J AM ACAD DERMATOL Argenziano et al 57 VOLUME 67, NUMBER 1 Fig 1. Trends over time of NNE in SCS and NSCS. Fig 3. Proportion of nevi excised over time in SCS and NSCS. Fig 4. Numbers of nevi and melanomas excised in SCS and NSCS, by age group. Fig 2. Trends over time of excised melanomas in SCS and NSCS. was from patients between 31 and 40 years of age (12,297 nevi in SCS and 51.928 nevi in NSCS), followed by the 21 to 30 years of age group (10.493 nevi in SCS and 44.473 nevi in NSCS) and the 41 to 50 years of age group (8478 nevi in SCS and 36.213 nevi in NSCS). As shown in Fig 5, the mean 10-year NNE was higher for younger patients, particularly patients younger than 40 years of age. The differences in NNE between all age groups were significant (trend test: P \ .001), with an exponential decrease of the NNE for every 10-years age group. In SCS, time trends calculated over the 10-year study period showed a significant decrease of the NNE in age groups 31-40 and 51-60 years (Table II); by contrast, NSCS showed no general time trend. Overall, the most frequent location of melanoma was the trunk (4938 lesions), followed by the head/ Fig 5. NNE values in SCS and NSCS, by age group. neck and the lower limbs (Fig 6; data were unavailable for 3813 melanomas). The trunk was the most frequent location of excised nevi, and excluding the genital region, was the anatomic site with the highest 58 Argenziano et al J AM ACAD DERMATOL JULY 2012 Table II. SCS time trends of NNE in different age groups Age group 0-20 21.30 31-40 41-50 51-60 [60 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 P value 157.7 141.8 24.6 12.0 7.6 2.7 120.4 50.7 30.6 15.2 7.8 2.5 NA 42.3 32.8 11.8 6.2 2.7 191.7 37.1 20.6 13.5 5.0 3.0 263.0 51.8 19.8 13.6 6.1 3.0 126.3 41.2 24.4 11.4 6.6 2.6 107.0 43.4 25.4 14.5 7.2 2.8 76.4 37.4 25.2 10.0 6.1 2.5 175.3 34.3 17.6 10.3 5.3 2.5 164.5 40.4 17.0 11.4 4.8 2.2 .40 .07 .05 .12 .05 .19 NA, Not available; NNE, number needed to excise; SCS, specialized clinical setting. Fig 6. Numbers of nevi and melanomas excised, by anatomic site. Fig 7. NNE values in SCS and NSCS, by anatomic site. NNE in both SCS and NSCS (Fig 7). In SCS, time trends calculated over 10 years showed a significant decrease of the NNE for the trunk, head/neck, and lower limbs (Table III). No general time trends were apparent in NSCS. The introduction of dermatoscopy has enriched the diagnostic armamentarium of clinicians by providing new morphologic clues that are particularly helpful for improving the early detection of melanoma.7 A recent meta-analysis of dermatoscopic studies performed in a clinical setting showed dermatoscopy to be superior to naked-eye examination alone in melanoma detection, with estimated sensitivities of 90% versus 71%, respectively, and estimated specificities of 90% and 81%, respectively.5 The growing trend to use dermatoscopy in SCS may be responsible for the improving NNE obtained in these centers from 1998 to 2007. In an earlier study conducted in an SCS over a 5-year period when dermatoscopy was gradually introduced, the malignant/benign ratio improved from 1:18 to 1:4.3, but only for clinicians who used dermatoscopy.3 No significant improvement was found for clinicians who did not use dermatoscopy. As in our study, improvement in NNE with the use of dermatoscopy over the 5-year study period appeared to be due to an increased proportion of excised melanomas (7.6% of melanomas excised in 1997 compared with 13.4% of melanomas excised in 2001) and a consequent reduction of the proportion of excised nevi. Other than aspects related to the physician’s expertise, various additional factors have a strong DISCUSSION The most striking result of our study is the finding that the NNE decreased significantly over time in SCS, yet remained stable in NSCS. In SCS the NNE decreased from 12.8 to 6.8 in the 10-year study period, whereas it remained essentially unchanged at approximately 29 in NSCS. Most of the effect on NNE in SCS was due to the striking increase in the number of excised melanomas and, as a consequence, to the decreasing proportion of excised nevi. The increased number of melanoma could be related to an increased incidence of this tumor in the general population. However, if this were true, then a similar effect should have also occurred in NSCS where, instead, the number of excised melanomas actually decreased slightly. A more reasonable explanation for the increase in melanoma excisions could be the effect of screening individuals with a higher incidence of melanoma than the general population and the expanding use of dermatoscopy, especially in SCS. Argenziano et al 59 J AM ACAD DERMATOL VOLUME 67, NUMBER 1 Table III. SCS time trends of NNE in different body areas Location 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Decrease of NNE per year P value for trend Trunk Head/neck Lower limbs Upper limbs Acral area Genital area 16.7 10.9 10.1 6.2 11.0 14.0 18.5 9.4 8.7 10.4 8.0 NA 16.1 8.1 6.9 8.4 6.7 11.6 14.8 8.9 7.1 9.5 8.0 43.0 16.1 9.0 7.9 6.0 11.5 25.5 14.6 7.5 6.5 5.9 7.2 13.0 13.4 7.8 8.3 7.1 7.4 NA 11.1 7.6 5.6 6.9 6.9 11.7 9.5 6.2 6.4 6.5 6.9 8.1 9.6 5.5 6.7 6.4 5.6 12.8 1.0 0.5 0.3 0.2 0.4 1.2 \.001 \.001 .02 .17 .07 .45 NA, Not available; NNE, number needed to excise; SCS, specialized clinical setting. influence on the NNE, including those related to the lesion and to the patient. In an Australian study involving primary care physicians working in specialized skin cancer clinics, the highest NNE rates occurred with patients younger than 30 years of age (NNE = 123), with patients between 30 and 44 years of age (NNE = 70), and with nevi located on the trunk (NNE = 35).2 Similar findings were seen in another study involving primary care physicians in Australia.8 A greater number of benign lesions were excised per melanoma (NNE = 83) in the youngest patients (aged 10-19 years) compared with those 70 years of age or older (NNE = 11). Similarly, in our study we found higher NNE rates in the youngest age group and in patients with lesions located on the trunk. Various factors may be relevant in interpreting these data. First, the likelihood that melanoma increases with increasing age and is extremely rare in anyone younger than 20 years old. Many patients between the ages of 20 and 50 years have multiple nevi, which are often located on the trunk. Nevi that exhibit atypical clinical features require excision to rule out melanoma; consequently, much of the economic burden of melanoma screening results from excisions and biopsies of benign lesions, especially in patients with multiple nevi.1 With the use of digital monitoring, the number of unnecessary excisions of benign lesions in such patients can be markedly reduced.9,10 Although no data specifically concerning the use of dermatoscopy and digital monitoring procedures were collected from the centers involved in our study, it is likely that these procedures were increasingly adopted in centers dedicated to melanoma screening. This could explain the significant decrease in NNE values observed in SCS over the study period in patients in age groups 31-40 and 51-60 years and in nevi excised from the trunk, head/neck, and lower limbs. In conclusion, the use of dermatoscopy can improve the accuracy in melanoma detection as measured by NNE values. Incorporation of this diagnostic technique in clinical practice should be expanded, not only to improve melanoma detection but also to decrease the excision rate of benign lesions, especially in younger patients. We are indebted to Barbara J. Rutledge, PhD, for editing assistance and to Richard Scolyer and Christine Hill who helped with the data collection. REFERENCES 1. Baade PD, Youl PH, Janda M, Whiteman DC, Del Mar CB, Aitken JF. Factors associated with the number of lesions excised for each skin cancer: a study of primary care physicians in Queensland, Australia. Arch Dermatol 2008;144: 1468-76. 2. Hansen C, Wilkinson D, Hansen M, Argenziano G. How good are skin cancer clinics at melanoma detection? Number needed to treat variability across a national clinic group in Australia. J Am Acad Dermatol 2009;61:599-604. 3. Carli P, De Giorgi V, Crocetti E, Mannone F, Massi D, Chiarugi A, et al. Improvement of malignant/benign ratio in excised melanocytic lesions in the ‘dermoscopy era’: a retrospective study 1997-2001. Br J Dermatol 2004;150:687-92. 4. Kittler H, Pehamberger H, Wolff K, Binder M. Diagnostic accuracy of dermoscopy. Lancet Oncol 2002;3:159-65. 5. Vestergaard ME, Macaskill P, Holt PE, Menzies SW. Dermoscopy compared with naked eye examination for the diagnosis of primary melanoma: a meta-analysis of studies performed in a clinical setting. Br J Dermatol 2008;159:669-76. 6. Carli P, de Giorgi V, Chiarugi A, Nardini P, Weinstock MA, Crocetti E, et al. Addition of dermoscopy to conventional naked-eye examination in melanoma screening: a randomized study. J Am Acad Dermatol 2004;50:683-9. 7. Argenziano G, Soyer HP, Chimenti S, Talamini R, Corona R, Sera F, et al. Dermoscopy of pigmented skin lesions: results of a consensus meeting via the Internet. J Am Acad Dermatol 2003;48:679-93. 8. English DR, Del Mar C, Burton RC. Factors influencing the number needed to excise: excision rates of pigmented lesions by general practitioners. Med J Aust 2004;180:16-9. 9. Kittler H, Binder M. Risks and benefits of sequential imaging of melanocytic skin lesions in patients with multiple atypical nevi. Arch Dermatol 2001;137:1590-5. 10. Menzies SW, Emery J, Staples M, Davies S, McAvoy B, Fletcher J, et al. Impact of dermoscopy and short-term sequential digital dermoscopy imaging for the management of pigmented lesions in primary care: a sequential intervention trial. Br J Dermatol 2009;161:1270-7. 59.e1 Argenziano et al J AM ACAD DERMATOL JULY 2012 Authors’ institutional affiliations: Dermatology Unit, Medical Department, Arcispedale Santa Maria Nuova, Reggio Emilia, Italy (Dr Argenziano); Department of Dermatology, Medical University of Graz, Graz, Austria (Drs Cerroni, Zalaudek, Hofmann-Wellenhof, and Massone); Department of Biomorphological and Functional Sciences, Pathology Section, University of Naples Federico II, Naples, Italy (Drs Staibano, Ilardi, Mascolo, and Nugnes); Second Unit of Dermatology, Department of Internal Medicine, Immunology, and Infectious Diseases, Policlinico, University of Bari, Bari, Italy (Drs Arpaia and Filotico); Department of Dermatology, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil (Dr Bakos); Department of Dermatology, Lyon 1 University, Centre Hospitalier Lyon Sud, Pierre B enite, France (Drs Balme and Thomas); Skin Cancer Unit, Special Hospital for Plastic and Reconstructive Surgery, ORS Hospital Belgrade, Belgrade, Serbia (Dr Bandic); Department of Pathology and Department of Dermatology, Galliera Hospital, Genoa, Italy (Drs Bandelloni and Brunasso); Dermatology Section, Instituto de Investigaciones Medicas A. Lanari, University of Buenos Aires, Buenos Aires, Argentina (Dr Cabo); The Dermatology Center, Rolla, Missouri (Mr Calcara and Dr Stoecker); Department of Dermatology, Hospital de Especialidades Centro Medico Nacional ‘‘La Raza’’, Mexico City, Mexico (Dr Carlos-Ortega); Sydney Medical School, University of Sydney, Sydney Melanoma Diagnostic Centre, Royal Prince Alfred Hospital, Camperdown, Australia (Drs Carvalho and Menzies); Hospital Aleman, Buenos Aires, Argentina (Dr Casas); Department of Dermatology, The First Teaching Hospital, University of Zhengzhou, Zhengzhou, Henan province, P. R. China (Dr Dong); Anatomic Pathology Unit, Gaetano Rummo General Hospital, Benevento, Italy (Dr Ferrara); Department of Pathology, Hospital de Especialidades Centro Medico Nacional ‘‘La Raza’’, Mexico City, Mexico (Dr G omez); Memorial Sloan-Kettering Cancer Centre, New York, New York (Drs Halpern, Kovalyshyn, Marghoob, and Siamas); Keio University, Tokyo, Japan (Drs Ishiko and Kawasaki); Department of Pathology, Faculty of Medicine, University of Ege, Bornova Izmir, Turkey (Dr Kandiloglu); Department of Dermatology, Tokyo Women’s Medical University Medical Center East, Tokyo, Japan (Drs Kobayashi and Tanaka); Department of Dermatology, Shinshu University, Matsumoto, Japan (Drs Koga and Minagawa); Molecheck, Aikmans Rd Clinic, Merivale Christchurch, New Zealand (Dr Langford); Department of Dermatology, General Hospital, Yongcheng Coal Group Corporation, Henan General Coal Group Corporation, Yongcheng, Henan province, P. R. China (Dr Liu); Skin Cancer Unit, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori, Meldola (Forlı-Cesena), Italy (Drs Mazzoni and Stanganelli); Department of Dermatology, Faculty of Medicine, University of Ege, Bornova Izmir, Turkey (Dr Ozdemir); Department of Dermatology, University of Modena and Reggio Emilia, Modena, Italy (Drs Pellacani and Seidenari); Department of Dermatology, Division of General Dermatology, Medical University of Vienna, Vienna, Austria (Drs Kittler and Tschandl).