- Email: [email protected]
Clinical and histologic characteristics of clinically unsuspected melanomas
Clinics in Dermatology (2014) 32, 324–330
COMMENT AND CONTROVERSY Edited by Stephen P. Stone, MD
Clinical and histologic characteristics of clinically unsuspected melanomas Heidi M. Hermes, MD a , Joya Sahu, MD b , Laurel R. Schwartz, MD b , Jason B. Lee, MD b,⁎ a
Department of Dermatology, Johns Hopkins University School of Medicine, Baltimore, MD Department of Dermatology and Cutaneous Biology, Jefferson Medical College of Thomas Jefferson University, 833 Chestnut Street, Philadelphia, PA 19107 b
Abstract Thin melanomas are recognized and captured by clinicians at an alarming rate, whereas thick melanomas remain underrecognized. Improved recognition of thick melanomas will require further understanding of their clinical and histologic characteristics at various stages of development because emerging data suggest that the thin melanomas being captured today may not represent the forerunners of the thick melanomas. In this retrospective analysis, pathology requisition forms from melanomas diagnosed by histopathology were examined for submitted clinical diagnosis, patient characteristics, melanoma thickness, and biopsy method. Three hundred eighty-five melanomas were identified from 2003 to 2011. Most lesions (71.7%) were clinically suspected to be melanocytic. The mean depth in this group was 0.62mm. Of the unsuspected cases (28.3%), the most common submitted diagnoses were basal cell carcinomas and seborrheic keratoses, consistent with previous reports. The mean depth in the unsuspected group was 1.64mm, and more frequently extended to the deep margin (51.8% vs 25.4% of the time). Shave biopsy was the overwhelming preferred method of biopsy (79.5% overall). Compared with thin melanomas, thick melanomas are underrecognized by physicians due to their lack of characteristic morphologic features; consequently, they are more frequently associated with suboptimal biopsies. © 2014 Elsevier Inc. All rights reserved.
Introduction Despite the dramatic increase in the incidence of thin melanomas around the world in the last several decades, a corresponding anticipated decline in the incidence of thick melanomas and their associated high mortality have not been observed. Emerging data suggest the existence of three heterogeneous forms of melanomas:(1) fast-growing aggressive melanomas minimally associated with sun exposure and ⁎ Corresponding author. Thomas Jefferson University Hospital, 833 Chestnut Street, Suite 740, Philadelphia, PA 19107. E-mail address: [email protected] (J.B. Lee). 0738-081X/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clindermatol.2013.10.003
melanocytic nevi, (2) slow-growing melanomas associated with intermittent sun exposure and melanocytic nevi, and (3) slow-growing indolent melanomas associated with chronic sun exposure occurring on the head and neck.1–3 Because much of the increase in the incidence of melanoma can be accounted for by stage I disease that has a 5-year survival rate approaching 97%,4,5 the current public and physician efforts appear to have resulted in capturing the thin, slow-growing melanomas (types II and III), but not the thick, fast-growing ones (type I).6,7 Due to their lack of impact on melanoma mortality, the ever-increasing in situ and thin melanomas being diagnosed today have been described as histologically malignant but biologically benign, inert, or indolent.8–10 For
Characteristics of clinically unsuspected melanomas some, length bias and overdiagnosis, contributed by increases in scrutiny, biopsy rates, and cancer registration,9–13provide the explanations for the current incidence and mortality trends.9–13 Diagnostic drift, that is, labeling lesions as malignant that previously would have been labeled as benign, has been argued as a significant reason for the overdiagnosis.14–17 The stable incidence of thick melanomas in the face of capturing thin melanomas on a large scale suggests that the thin and thick melanomas captured today may represent different forms of melanomas rather than the former representing the forerunners of the latter.1,7,18 Currently, thin melanomas are highly recognized and captured at an ever-increasing rate, whereas thick melanomas with poor prognosis are underrecognized. For thick melanomas, in part due to their rarity, epidemiologic data and clinical characteristics have not been well established and extensively described. Improved recognition of thick melanomas will require further understanding of their clinical and histologic characteristics at various stages of their development; moreover, whether thick melanomas are indeed underrecognized needs further confirmation. In this retrospective analysis of histologically confirmed melanomas, clinically unsuspected and suspected cases are compared with respect to their clinical characteristics, biopsy techniques, and depths.
Materials and methods After receiving the approval of the investigational protocol by the Thomas Jefferson University Institutional Review Board (Philadelphia, PA), we performed a search of the Jefferson Dermatopathology Center database to identify all cases diagnosed histologically as melanoma during the period of 2003 through 2011. Melanoma insitu was specifically excluded. The requisition form and report for each melanoma were reviewed to obtain and tabulate age and sex of the patient, sites, clinical diagnosis, biopsy method, margin information, and Breslow’s depth. Cases were divided into two groups for analysis based on whether a melanocytic lesion was suspected in the submitted clinical differential diagnosis. If a melanocytic lesion was mentioned on the requisition form, such as atypical, dysplastic, or Clark nevus, lentigo; lentigo maligna; or melanoma, the case was categorized into the suspected group, whereas those cases with no mention of a melanocytic lesion were categorized into the unsuspected group. Statistical analysis was performed using GraphPad online software (La Jolla, CA). ttest was used to analyze continuous variables among groups; Fisher exact test was used to analyze categorical variables.
Results The retrospective review identified 391 cases of melanomas, 6 cases of which were excluded, because no diagnosis was provided on the requisition form, leaving 385 cases for
325 Table 1
Patient characteristics All Cases
Age (range), y Sex, n Men Women Not reported, n
67 (21-95) 242 148 1
Suspected Group 65 (21-87) 166 109 1
Unsuspected Group 71 (43-97) 74 35 0
analysis. Dermatologists referred all cases included in the study. Of the 385 cases that were included in the study, 240 were men and 144 were women, whereas the patient’s sex was not reported in 1 case. Overall age at diagnosis ranged from 21 to 95 years, with the average being 67 years. The suspected group was composed of 276 cases (71.7%), whereas 109 cases (28.3%) comprised the unsuspected group. The mean age of the suspected and unsuspected groups was 65 and 71 years, respectively. The age difference of the two groups was statistically significant (pb 0.001), with the unsuspected group being older. There was no difference in the sex distribution between the two groups. Table 1 summarizes patient characteristics. Figure 1 demonstrates the clinical diagnoses as submitted by the clinician on the requisition form. In the majority of cases, one diagnosis per specimen was submitted. There were cases, however, where two to four diagnoses were submitted (Table 2). Melanoma appeared on the differential 22% of the cases overall. Basal cell carcinoma was the second most frequent entity on the requisition form (16%). See legend for Figure 1 for list of all diagnoses submitted. In the suspected group, melanoma was the most commonly mentioned diagnosis (30.4%), followed by atypical, dysplastic, or Clark nevus (20.6%), and followed by melanoma in situ or lentigo maligna (13%), nevus not otherwise specified (10%), and lentigo (7%). Spitz nevus accounted for only two cases (0.5%). Other diagnoses commonly included on the differential in this group were seborrheic keratosis (7%) and basal cell carcinoma (6%). In the unsuspected group, basal cell carcinoma was mentioned most frequently (42.6%), followed by seborrheic keratosis (19.9%) and squamouscell carcinoma (15.6%). Interestingly, neoplasm not otherwise specified was more frequent in this group (6.4%). By definition, no melanocytic lesions were included on the differential in this group. The depths of the melanomas ranged from 0.1 to 12 mm, and the mean depth was 0.90 mm for all melanomas. The mean depth of the suspected group was 0.62 mm, whereas the mean depth in the unsuspected group was 1.64 mm. This difference was statistically significant (pb 0.001), indicating that the unsuspected group had a greater depth at initial biopsy. Extension to the base of biopsy was found to be 25.4% in the suspected group compared with 51.8% in the unsuspected group, which is also statistically significant (pb 0.001). Extension to the peripheral margin was more
326
H.M. Hermes et al.
Fig. 1 Diagnoses submitted from pathology request forms. Panel A, All submitted diagnoses. Panel B, Suspected group. Panel C, Unsuspected group. Percentage rounded to the nearest whole number.“Others” was defined as entities listed less than five times (b 1% occurrence rate) and in order of decreasing frequencyincludedactinic keratosis, scar, Spitz, poroma, foreign body, dermatofibroma, verruca, prurigo nodule, neurofibroma, insect bite reaction, eczema, and cyst. BCC, basal cell carcinoma; LMM, lentigomaligna melanoma; MIS, melanoma in situ; NOS, not otherwise specified; SCC, squamous cell carcinoma; SK, seborrheic keratosis.
frequent in the suspected group, with 64.2% having positive peripheral margins compared with 36.7% in the unsuspected group, which was not statistically significant (p= 0.02; Table 3). Analysis of depth of lesion by body site or sex of patient revealed no significant trends (data not shown). Table 4 summarizes the frequency of the biopsy techniques of the two groups. The shave biopsy technique
represented the diagnostic procedure of choice for the two groups, 75.7% for the suspected group and 89% for the unsuspected group. Even when melanoma was suspected, shave biopsy was the preferred technique; almost 60% (67/ 112) of the cases were sampled by this technique. Expectedly, excisional biopsy accounted for a higher proportion of the cases in the suspected group.
Characteristics of clinically unsuspected melanomas Table 2
All diagnoses a
One diagnosis Two diagnoses Three diagnoses Four diagnoses Total no.cases (no.of diagnoses)
All
Suspected Group
Unsuspected Group
276 94 14 1 385 (510)
194 71 11 0 276 (369)
82 23 3 1 109 (141)
a Total number of diagnoses submitted as differential diagnosis on the pathology requisition form. The majority had one diagnosis submitted. Total number of diagnoses by group is in parentheses.
Comment Although this study represents the experience of a single institution, which may not have general application, the results corroborate with general trends in melanoma diagnosis and incidence. Similar to the melanoma incidence in the United States and around the world in which thin melanomas (b 1 mm) account for the vast majority of the new cases,1,19–23 more than 70% of the cases in this series were less than 1 mm in depth, with an overall mean depth of 0.90 mm. In this series, all cases were referred by dermatologists, which accounts for an even higher proportion of thin melanomas compared with some other series.23,24 Dermatologists as a group are known to diagnose melanomas at an earlier stage compared with nondermatologists, with a reported mean depth ranging from 0.16 to 0.86 mm25–28; moreover, the proportion of in situ stage was significantly higher than invasive ones when dermatologists detected melanomas.27,28 The proposed explanations of why dermatologists diagnose melanomas at an earlier stage include:(1) having more extensive training and experience in recognizing thin lesions;(2) treating patients harboring more risk factors associated with development of melanoma; and (3) as a group, diagnosing melanomas predominantly of one type with good prognosis, the type that is strongly associated with sun exposure and melanocytic nevi.1,26 It is likely that the combination of these reasons explains the ability of dermatologists to detect melanomas at an earlier stage, but whichever the reason, the trend is clear: led by dermatologists, thinner and thinner melanomas are being diagnosed today more than ever before to the credit of all those who have campaigned for the early detection of the disease. Although dermatologists are proficient in diagnosing thin melanomas, diagnosing thick melanomas still remains a challenge. When histologically confirmed melanomas were clinically unsuspected, they were thicker compared with when they were suspected, 1.64 versus 0.62 mm. The results are consistent with the fact that thin melanomas frequently fulfill the ABCD criteria of melanoma, and thus are readily recognized. Morphologic expression of thick melanomas, on the other hand, is more protean, simulating a wide range of
327 inflammatory and neoplastic diseases. Thick melanomas may present as small, symmetric lesions that are uniform in color defying the ABCD criteria,3,6,18,29 implying that context may be more important than morphologic attributes in recognizing thick melanomas. For this reason, “E” for evolving has been added to the ABCD criteria to emphasize the recognition of changing lesions, because morphologic changes are more frequently observed in melanomas than in melanocytic nevi.30,31 Other screening strategies that rely less on individual morphologic attributes and more on contextual presentation of the patient have been proposed and implemented, which include using the ugly duckling sign, recognizing signature nevi, and lowering the threshold to biopsy in patients with high-risk factors (eg, skin phototype I, personal history of melanoma, and elderly men). Whether these strategies are effective in capturing melanomas with poor prognosis at an earlier stage awaits further validation. As the classification of melanoma transitions from the relative imprecise clinicopathologicbased one to a more precise molecular-based one, additional opportunity to improve the recognition and diagnosis lays ahead,32,33 but for now, thick melanomas continue to elude prompt early diagnosis. Clinical nonmelanocytic simulators of melanoma reported in the literature include seborrheic keratoses, basal cell carcinomas, squamouscell carcinomas, and vascular lesions, among others, with the first two being the most frequently reported (Table 5).34–37 In this series, basal cell carcinoma was the most frequent simulator of melanoma, with seborrheic keratosis coming in second (Figure 1), similar to the results observed in a recent series.37 These observations underscore the importance of recognizing nonmelanocytic simulators of melanoma, because these clinically misdiagnosed lesions are thicker compared with melanocytic simulators. The obvious pitfall of underrecognition is the potential mismanagement as the clinicians, believing that they are either benign or less aggressive malignant lesions, may choose suboptimal biopsy and follow-up care. One strategy to improve the diagnostic specificity of thick melanomas is improving the recognition of nonmelanocytic simulators of melanomas, which should include becoming proficient in dermatoscopy. Although better known for its ability to discriminate melanocytic simulators of melanoma, dermatoscopy is emerging as a powerful tool in improving the diagnostic accuracy of nonmelanocytic simulators of melanoma, especially basal cell carcinoma.38–40 There are limitations of reliance on diagnostic information based on pathology requisition forms, but multiple previous studies have been performed based on similar methods, which provide validity to the method used in this series (Table 5).34,35,37,41–46 Studies emanating from Europe and Australia indicate that excisional biopsy is the preferred diagnostic method when it comes to sampling melanocytic lesions.47–50 In the United States, however, despite the general recommendation of excisional biopsy when sampling a melanocytic lesion,
328
H.M. Hermes et al.
Table 3
Average depth and margin involvement of original biopsy a
All cases Suspected group Unsuspected group p
n
Average Depth (mm)
Base Margin Involved
Peripheral Margin Involved
385 276 109
0.9 0.62 1.64 b 0.001
70 (25.4) 70 (51.8) b 0.001
143 (64.2) 40 (36.7) 0.02
a
The mean of lesions were all less than 1mm thick. Tumors tended to be thicker in the unsuspected group, and they were more likely to extend to the base. Suspected group tended to involve the peripheral margins more often.
shave biopsy represents the preferred biopsy technique performed by clinicians, especially among dermatologists. 37,51 Not surprisingly, shave biopsy represented overwhelmingly the preferred biopsy technique in this series, with more than 80% of the cases having been performed by this method. Although a 1992 survey of US dermatologists indicated that only 10% perform shave biopsy when melanoma is suspected,52 the results of this study and others dispute the past survey results as shave biopsy represents the biopsy technique of choice for dermatologists presumably due to its ease, time, cost, and safety advantages. The potential pitfalls of shave biopsy and other partial biopsy techniques include misdiagnosis and inaccurate histopathologic staging, which may lead to inappropriate and/or suboptimal evaluation, staging, and treatment.50,51 Reported upstaging due to partial biopsies of melanoma ranges from 3% to 25%.37,50,53–56 Compared with a punch biopsy, a shave biopsy may be expected to be less accurate in determining the depth of melanoma due to the shallower nature of the technique, but studies have shown that a shave biopsy is just as accurate, if not more, in determining the final depth of melanoma.50,54,56,57 For thin melanomas, less than 1 mm in depth, initial shave biopsy predicted the final melanoma depth in 97% to 100% of the cases in recent studies.37,55,57 For thick melanomas, larger than 2 mm, partial biopsies, whether shave or punch biopsy, were less predictive of the final melanoma depth and, accordingly, upstaging was observed more frequently in thick than in thin melanomas.50,55–57 Despite the potential mismanagement of melanomas, past meta-analysis and recent studies indicate that the overall prognosis and outcome are not affected by partial biopsies,37,56,58 supporting and validating the practice of shave biopsies by dermatologists for lesions clinically suspected of melanoma; nonetheless, most authors affirm that an excisional biopsy is preferred over any partial biopsies, because it provides the all-important piece of Table 4
prognostic indicator, the Breslow depth, more accurately than any other biopsy method. In this series, when a melanocytic lesion was suspected, the clinicians tended to perform adequate biopsies to ensure that depth of the lesion can be determined, whereas when melanocytic lesion was not suspected, they tended to perform biopsies involving deep margins—25.4% versus 51.8% positive deep margins (Table 3). This result reaffirms that dermatologists are proficient in recognizing thin melanomas and performing biopsies appropriately for their depths. For the unsuspected group, in contrast, clinicians, not cognizant of the therapeutic and staging implications, appeared to have performed biopsies only to determine the diagnosis without regard to the depth of the lesion, resulting in a higher percentage of cases that involved deep margins. Shave biopsy, the preferred technique by dermatologists in this and other series, may suffice for most thin melanomas as previous studies indicate, but an excisional biopsy should be considered when faced with a nodular lesion, especially in the elderly, suspicious for malignancy and/or displaying atypical features of basal cell carcinoma because it may well represent a thick melanoma.
Conclusions What can be done to improve the recognition and capturing of thick melanomas? The current approach to early detection with an emphasis on the ABCD criteria of melanoma and Clark/dysplastic nevi, a presumed precursor to melanoma, has resulted in a marked increase in capturing thin melanomas but not the anticipated decrease in melanoma mortality. Several authors have called for a shift in strategy to improve the detection of thick melanomas at an earlier stage. First, because thick melanomas are more
Biopsy technique breakdown between groups a
All cases Suspected group Melanoma suspected Unsuspected group
n
Shave
Punch
Curette
Excision
385 276 112 109
311 (79.5%) 209 (75.7%) 67 (59.8%) 97 (89.0%)
13 (3.3%) 9 (3.3%) 3 (2.7%) 4 (3.7%)
7 (1.8%) 4 (1.5%) 0 3 (2.8%)
60 (15.3%) 54 (19.6%) 42 (37.5%) 5 (4.6%)
a Shave biopsy was the most common means of biopsy. Subset analysis where melanoma was suspected on the differential revealed that shave biopsy was still the method of choice.
Characteristics of clinically unsuspected melanomas Table 5
329
Nonmelanocytic simulators of melanoma a
Study
Year
Cases Histologically Consistent with Melanoma
No. of Unsuspected Melanomas, n (%)
Basal Cell Carcinoma
Seborrheic Keratosis
Lentigo
Grin et al.40 Witheilerand Cockerell43 Miller and Ackerman45 Zager et al.37 This study
1990 1991 1992 2011 2012
265 1784 354 600 385
51 (19) 583 (32.6) 128 (36) 179 (32) 109 (28.3)
35.0% 12.0% 9.0% 16.0% 16.5%
7.8% 9.0% 13.0% 8.3% 11.2%
5.8% 9.0% 19.0% 2.3% 4.7%
a A summary of previous studies reporting the clinical diagnoses of melanomas diagnosed histologically, based on clinical requisition forms. Common nonmelanocytic simulators include basal cell carcinoma, seborrheic keratosis, and lentigo, reported as percent of total melanocytic cases, which were clinically misidentified.
frequently discovered by patients and their family members than by physicians,25,59–61 implementation of more effective campaigns and measures to educate the public and promote self-skin examination for all age groups have been proposed.62 This form of secondary prevention, which is currently believed to be underused, may yield the highest reward in capturing thick melanomas, because these fastgrowing lesions may not be amenable to screening by health care providers.1,63 Second, targeting groups at a greater risk for development of thick melanomas has been proposed as another strategy to reduce the burden of thick melanomas, which include men older than 65 years and socioeconomically disadvantaged individuals.62,64–66 Both higher risk groups, however, provide challenges for the health care providers because elderly men are less amenable to self-skin examination, and socioeconomically disadvantaged individuals have healthcare access issues. Third, recruitment of more primary care physicians to perform skin examinations has been proposed, for patients not only visit primary care physicians more frequently, but they also have easier access to them.62 Whether such shifts in public campaigns and physician resources will result in a decrease in melanoma-related mortality remains to be proved. The results of this and other studies, however, lend support for a shift in strategy. Under-recognition of thick melanomas revealed in this and other series begs the question of how many of these lesions are actually being left undetected by health care providers because series such as this one can include only those that have been captured and not those left behind. The ongoing molecular, epidemiologic, and clinicopathologic investigations all must converge to further delineate the etiopathogenesis of thick melanomas to improve their detection at an earlier stage.
References 1. Lipsker D, Engel F, Cribier B, et al. Trends in melanoma epidemiology suggest three different types of melanoma. Br J Dermatol. 2007;157: 338-343. 2. Argenziano G, Kittler H, Ferrara G, et al. Slow-growing melanoma: adermoscopy follow-up study. Br J Dermatol. 2010;162:267-273.
3. Martorell-Calatayud A, Nagore E, Botella-Estrada R, et al. Defining fast-growing melanomas: reappraisal of epidemiological, clinical, and histological features. Melanoma Res. 2011;21:131-138. 4. Surveillance, Epidemiology, and End Results Program. SEER stat fact sheet: melanoma of the skin. http://seer.cancer.gov/statfacts/html/ melan.html. Accessed September 12, 2013. 5. Pollack LA, Li J, Berkowitz Z, et al. Melanoma survival in the United States, 1992 to 2005. J Am Acad Dermatol. 2011;65:S78-S86. 6. Demierre MF, Chung C, Miller DR, Geller AC. Early detection of thick melanomas in the United States: beware of the nodular subtype. Arch Dermatol. 2005;141:745-750. 7. Murray CS, Stockton DL, Doherty VR. Thick melanoma: the challenge persists. Br J Dermatol. 2005;152:104-109. 8. Burton RC, Armstrong BK. Recent incidence trends imply a nonmetastasizing form of invasive melanoma. Melanoma Res. 1994;4:107-113. 9. Swerlick RA, Chen S. The melanoma epidemic. Is increased surveillance the solution or the problem? Arch Dermatol. 1996;132: 881-884. 10. Glusac EJ. The melanoma ‘epidemic’, a dermatopathologist’sperspective.J CutanPathol. 2011;38:264–267. 11. Swerlick RA, Chen S. The melanoma epidemic: more apparent than real? Mayo Clin Proc. 1997;72:559-564. 12. Swica Y, Koehler A, Ackerman AB. Lies, damn lies, and statistics: why there is no epidemic of melanoma. Dermatopathol Pract Concept. 2001;7:347-354. 13. Welch HG, Woloshin S, Schwartz LM. Skin biopsy rates and incidence of melanoma: population based ecological study. BMJ. 2005;331:481. 14. Levell NJ, Beattie CC, Shuster S, Greenberg DC. Melanoma epidemic: a midsummer night’s dream? Br J Dermatol. 2009;161:630-634. 15. Shuster S. Malignant melanoma: how error amplification by screening creates spurious disease. Br J Dermatol. 2009;161:977-979. 16. van der Rhee JI, Mooi WJ, Kukutsch NA, et al. Iatrogenic melanoma. Comment on: melanoma epidemic: a midsummer night’s dream? Br J Dermatol. 2010;162:457-458. 17. Frangos JE, Duncan LM, Piris A, et al. Increased diagnosis of thin superficial spreading melanomas: a20-year study. J Am Acad Dermatol. 2012;67:387-394. 18. Chamberlain AJ, Fritschi L, Kelly JW. Nodular melanoma: patients’perceptions of presenting features and implications for earlier detection. J Am Acad Dermatol. 2003;48:694-701. 19. Buettner PG, Leiter U, Eigentler TK, Garbe C. Development of prognostic factors and survival in cutaneous melanoma over 25 years: an analysis of the Central Malignant Melanoma Registry of the German Dermatological Society. Cancer. 2005;103:616-624. 20. Crocetti E, Caldarella A, Chiarugi A, et al. The thickness of melanomas has decreased in central Italy, but only for thin melanomas, while thick melanomas are as thick as in the past. Melanoma Res. 2010;20:422-426. 21. Coory M, Baade P, Aitken J, et al. Trends for in situ and invasive melanoma in Queensland, Australia, 1982-2002. Cancer Causes Control. 2006;17:21-27.
330 22. Linos E, Swetter SM, Cockburn MG, et al. Increasing burden of melanoma in the United States. J Invest Dermatol. 2009;129:1666-1674. 23. Criscione VD, Weinstock MA. Melanoma thickness trends in the United States, 1988-2006. J Invest Dermatol. 2010;130:793-797. 24. Berwick M, Erdei E, Hay J. Melanoma epidemiology and public health. Dermatol Clin. 2009;27:205-214. 25. Carli P, De Giorgi V, Palli D, et al. Dermatologist detection and skin self-examination are associated with thinner melanomas: results from a survey of the Italian Multidisciplinary Group on Melanoma. Arch Dermatol. 2003;139:607-612. 26. Pennie ML, Soon SL, Risser JB, et al. Melanoma outcomes for Medicare patients: association of stage and survival with detection by a dermatologist vs a nondermatologist. Arch Dermatol. 2007;143: 488-494. 27. Kantor J, Kantor DE. Routine dermatologist-performed full-body skin examination and early melanoma detection. Arch Dermatol. 2009;145: 873-876. 28. Kovalyshyn I, Dusza SW, Siamas K, et al. The impact of physician screening on melanoma detection. Arch Dermatol. 2011; 147:1269-1275. 29. Liu W, Dowling JP, Murray WK, et al. Rate of growth in melanomas: characteristics and associations of rapidly growing melanomas. Arch Dermatol. 2006;142:1551-1558. 30. Kittler H, Seltenheim M, Dawid M, et al. Morphologic changes of pigmented skin lesions: a useful extension of the ABCD rule for dermatoscopy. J Am Acad Dermatol. 1999;40:558-562. 31. Abbasi NR, Shaw HM, Rigel DS, et al. Early diagnosis of cutaneous melanoma: revisiting the ABCD criteria. JAMA. 2004;292:2771-2776. 32. Curtin JA, Fridlyand J, Kageshita T, et al. Distinct sets of genetic alterations in melanoma. N Engl J Med. 2005;353:2135-2147. 33. Romano E, Schwartz GK, Chapman PB, et al. Treatment implications of the emerging molecular classification system for melanoma. Lancet Oncol. 2011;12:913-922. 34. Grant-Kels JM, Bason ET, Grin CM. The misdiagnosis of malignant melanoma. J Am Acad Dermatol. 1999;40:539-548. 35. Izikson L, Sober AJ, Mihm Jr MC, Zembowicz A. Prevalence of melanoma clinically resembling seborrheic keratosis: analysis of 9204 cases. Arch Dermatol. 2002;138:1562-1566. 36. Matz H, Orion E, Ruocco V, Wolf R. Clinical simulators of melanoma. Clin Dermatol. 2002;20:212-221. 37. Zager JS, Hochwald SN, Marzban SS, et al. Shave biopsy is a safe and accurate method for the initial evaluation of melanoma. J Am Coll Surg. 2011;212:454-460. 38. Marghoob AA, Braun R. Proposal for a revised 2-step algorithm for the classification of lesions of the skin using dermoscopy. Arch Dermatol. 2010;146:426-428. 39. Hirokawa D, Lee JB. Dermatoscopy: an overview of subsurface morphology. Clin Dermatol. 2011;29:557-565. 40. Rosendahl C, Tschandl P, Cameron A, Kittler H. Diagnostic accuracy of dermatoscopy for melanocytic and nonmelanocytic pigmented lesions. J Am Acad Dermatol. 2011;64:1068-1073. 41. Rampen FH, Rumke P. Referral pattern and accuracy of clinical diagnosis of cutaneous melanoma. Acta Derm Venereol. 1988;68: 61-64. 42. Grin CM, Kopf AW, Welkovich B, et al. Accuracy in the clinical diagnosis of malignant melanoma. Arch Dermatol. 1990;126:763-766. 43. Witheiler DD, Cockerell CJ. Histologic features and sensitivity of diagnosis of clinically unsuspected cutaneous melanoma. Am J Dermatopathol. 1991;13:551-556. 44. Witheiler DD, Cockerell CJ. Sensitivity of diagnosis of malignant melanoma: a clinicopathologic study with a critical assessment of biopsy techniques. Exp Dermatol. 1992;1:170-175.
H.M. Hermes et al. 45. Miller M, Ackerman AB. How accurate are dermatologists in the diagnosis of melanoma? Degree of accuracy and implications. Arch Dermatol. 1992;128:559-560. 46. Morton CA, Mackie RM. Clinical accuracy of the diagnosis of cutaneous malignant melanoma. Br J Dermatol. 1998;138:283-287. 47. Lees VC, Briggs JC. Effect of initial biopsy procedure on prognosis in Stage 1 invasive cutaneous malignant melanoma: review of 1086 patients. Br J Surg. 1991;78:1108-1110. 48. Bong JL, Herd RM, Hunter JA. Incisional biopsy and melanoma prognosis. J Am Acad Dermatol. 2002;46:690-694. 49. Molenkamp BG, Sluijter BJ, Oosterhof B, et al. Non-radical diagnostic biopsies do not negatively influence melanoma patient survival. Ann Surg Oncol. 2007;14:1424-1430. 50. Ng JC, Swain S, Dowling JP, et al. The impact of partial biopsy on histopathologic diagnosis of cutaneous melanoma: experience of an Australian tertiary referral service. Arch Dermatol. 2010;146:234-239. 51. Lowe M, Hill N, Page A, et al. The impact of shave biopsy on the management of patients with thin melanomas. Am Surg. 2011;77:1050-1053. 52. Salopek TG, Slade J, Marghoob AA, et al. Management of cutaneous malignant melanoma by dermatologists of the American Academy of Dermatology. I. Survey of biopsy practices of pigmented lesions suspected as melanoma. J Am Acad Dermatol. 1995;33:441-450. 53. Karimipour DJ, Schwartz JL, Wang TS, et al. Microstaging accuracy after subtotal incisional biopsy of cutaneous melanoma. J Am Acad Dermatol. 2005;52:798-802. 54. Stell VH, Norton HJ, Smith KS, et al. Method of biopsy and incidence of positive margins in primary melanoma. Ann Surg Oncol. 2007;14: 893-898. 55. Moore P, Hundley J, Hundley J, et al. Does shave biopsy accurately predict the final breslow depth of primary cutaneous melanoma? Am Surg. 2009;75:369-373. 56. Egnatios GL, Dueck AC, Macdonald JB, et al. The impact of biopsy technique on upstaging, residual disease, and outcome in cutaneous melanoma. Am J Surg. 2011;202:771-777. 57. Ng PC, Barzilai DA, Ismail SA, et al. Evaluating invasive cutaneous melanoma: is the initial biopsy representative of the final depth? J Am Acad Dermatol. 2003;48:420-424. 58. Pflugfelder A, Weide B, Eigentler TK, et al. Incisional biopsy and melanoma prognosis: facts and controversies. Clin Dermatol. 2010;28:316-318. 59. Brady MS, Oliveria SA, Christos PJ, et al. Patterns of detection in patients with cutaneous melanoma. Cancer. 2000;89:342-347. 60. Carli P, De Giorgi V, Palli D, et al. Patterns of detection of superficial spreading and nodular-type melanoma: a multicenter Italian study. Dermatol Surg. 2004;30:1371-1375. 61. McPherson M, Elwood M, English DR, et al. Presentation and detection of invasive melanoma in a high-risk population. J Am Acad Dermatol. 2006;54:783-792. 62. Geller AC, Swetter SM, Oliveria S, et al. Reducing mortality in individuals at high risk for advanced melanoma through education and screening. J Am Acad Dermatol. 2011;65:S87-S94. 63. Pozo L, Diaz-Cano SJ. Tumor screening and biology in malignant melanomas. Arch Dermatol. 2000;136:934-935. 64. Geller AC, Miller DR, Swetter SM, et al. A call for the development and implementation of a targeted national melanoma screening program. Arch Dermatol. 2006;142:504-507. 65. Janda M, Neale RE, Youl P, et al. Impact of a video-based intervention to improve the prevalence of skin self-examination in men 50 years or older: the Randomized Skin Awareness Trial. Arch Dermatol. 2011;147:799-806. 66. McGuire ST, Secrest AM, Andrulonis R, Ferris LK. Surveillance of patients for early detection of melanoma: patterns in dermatologist vs patient discovery. Arch Dermatol. 2011;147:673-678.
COMMENT AND CONTROVERSY Edited by Stephen P. Stone, MD
Clinical and histologic characteristics of clinically unsuspected melanomas Heidi M. Hermes, MD a , Joya Sahu, MD b , Laurel R. Schwartz, MD b , Jason B. Lee, MD b,⁎ a
Department of Dermatology, Johns Hopkins University School of Medicine, Baltimore, MD Department of Dermatology and Cutaneous Biology, Jefferson Medical College of Thomas Jefferson University, 833 Chestnut Street, Philadelphia, PA 19107 b
Abstract Thin melanomas are recognized and captured by clinicians at an alarming rate, whereas thick melanomas remain underrecognized. Improved recognition of thick melanomas will require further understanding of their clinical and histologic characteristics at various stages of development because emerging data suggest that the thin melanomas being captured today may not represent the forerunners of the thick melanomas. In this retrospective analysis, pathology requisition forms from melanomas diagnosed by histopathology were examined for submitted clinical diagnosis, patient characteristics, melanoma thickness, and biopsy method. Three hundred eighty-five melanomas were identified from 2003 to 2011. Most lesions (71.7%) were clinically suspected to be melanocytic. The mean depth in this group was 0.62mm. Of the unsuspected cases (28.3%), the most common submitted diagnoses were basal cell carcinomas and seborrheic keratoses, consistent with previous reports. The mean depth in the unsuspected group was 1.64mm, and more frequently extended to the deep margin (51.8% vs 25.4% of the time). Shave biopsy was the overwhelming preferred method of biopsy (79.5% overall). Compared with thin melanomas, thick melanomas are underrecognized by physicians due to their lack of characteristic morphologic features; consequently, they are more frequently associated with suboptimal biopsies. © 2014 Elsevier Inc. All rights reserved.
Introduction Despite the dramatic increase in the incidence of thin melanomas around the world in the last several decades, a corresponding anticipated decline in the incidence of thick melanomas and their associated high mortality have not been observed. Emerging data suggest the existence of three heterogeneous forms of melanomas:(1) fast-growing aggressive melanomas minimally associated with sun exposure and ⁎ Corresponding author. Thomas Jefferson University Hospital, 833 Chestnut Street, Suite 740, Philadelphia, PA 19107. E-mail address: [email protected] (J.B. Lee). 0738-081X/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clindermatol.2013.10.003
melanocytic nevi, (2) slow-growing melanomas associated with intermittent sun exposure and melanocytic nevi, and (3) slow-growing indolent melanomas associated with chronic sun exposure occurring on the head and neck.1–3 Because much of the increase in the incidence of melanoma can be accounted for by stage I disease that has a 5-year survival rate approaching 97%,4,5 the current public and physician efforts appear to have resulted in capturing the thin, slow-growing melanomas (types II and III), but not the thick, fast-growing ones (type I).6,7 Due to their lack of impact on melanoma mortality, the ever-increasing in situ and thin melanomas being diagnosed today have been described as histologically malignant but biologically benign, inert, or indolent.8–10 For
Characteristics of clinically unsuspected melanomas some, length bias and overdiagnosis, contributed by increases in scrutiny, biopsy rates, and cancer registration,9–13provide the explanations for the current incidence and mortality trends.9–13 Diagnostic drift, that is, labeling lesions as malignant that previously would have been labeled as benign, has been argued as a significant reason for the overdiagnosis.14–17 The stable incidence of thick melanomas in the face of capturing thin melanomas on a large scale suggests that the thin and thick melanomas captured today may represent different forms of melanomas rather than the former representing the forerunners of the latter.1,7,18 Currently, thin melanomas are highly recognized and captured at an ever-increasing rate, whereas thick melanomas with poor prognosis are underrecognized. For thick melanomas, in part due to their rarity, epidemiologic data and clinical characteristics have not been well established and extensively described. Improved recognition of thick melanomas will require further understanding of their clinical and histologic characteristics at various stages of their development; moreover, whether thick melanomas are indeed underrecognized needs further confirmation. In this retrospective analysis of histologically confirmed melanomas, clinically unsuspected and suspected cases are compared with respect to their clinical characteristics, biopsy techniques, and depths.
Materials and methods After receiving the approval of the investigational protocol by the Thomas Jefferson University Institutional Review Board (Philadelphia, PA), we performed a search of the Jefferson Dermatopathology Center database to identify all cases diagnosed histologically as melanoma during the period of 2003 through 2011. Melanoma insitu was specifically excluded. The requisition form and report for each melanoma were reviewed to obtain and tabulate age and sex of the patient, sites, clinical diagnosis, biopsy method, margin information, and Breslow’s depth. Cases were divided into two groups for analysis based on whether a melanocytic lesion was suspected in the submitted clinical differential diagnosis. If a melanocytic lesion was mentioned on the requisition form, such as atypical, dysplastic, or Clark nevus, lentigo; lentigo maligna; or melanoma, the case was categorized into the suspected group, whereas those cases with no mention of a melanocytic lesion were categorized into the unsuspected group. Statistical analysis was performed using GraphPad online software (La Jolla, CA). ttest was used to analyze continuous variables among groups; Fisher exact test was used to analyze categorical variables.
Results The retrospective review identified 391 cases of melanomas, 6 cases of which were excluded, because no diagnosis was provided on the requisition form, leaving 385 cases for
325 Table 1
Patient characteristics All Cases
Age (range), y Sex, n Men Women Not reported, n
67 (21-95) 242 148 1
Suspected Group 65 (21-87) 166 109 1
Unsuspected Group 71 (43-97) 74 35 0
analysis. Dermatologists referred all cases included in the study. Of the 385 cases that were included in the study, 240 were men and 144 were women, whereas the patient’s sex was not reported in 1 case. Overall age at diagnosis ranged from 21 to 95 years, with the average being 67 years. The suspected group was composed of 276 cases (71.7%), whereas 109 cases (28.3%) comprised the unsuspected group. The mean age of the suspected and unsuspected groups was 65 and 71 years, respectively. The age difference of the two groups was statistically significant (pb 0.001), with the unsuspected group being older. There was no difference in the sex distribution between the two groups. Table 1 summarizes patient characteristics. Figure 1 demonstrates the clinical diagnoses as submitted by the clinician on the requisition form. In the majority of cases, one diagnosis per specimen was submitted. There were cases, however, where two to four diagnoses were submitted (Table 2). Melanoma appeared on the differential 22% of the cases overall. Basal cell carcinoma was the second most frequent entity on the requisition form (16%). See legend for Figure 1 for list of all diagnoses submitted. In the suspected group, melanoma was the most commonly mentioned diagnosis (30.4%), followed by atypical, dysplastic, or Clark nevus (20.6%), and followed by melanoma in situ or lentigo maligna (13%), nevus not otherwise specified (10%), and lentigo (7%). Spitz nevus accounted for only two cases (0.5%). Other diagnoses commonly included on the differential in this group were seborrheic keratosis (7%) and basal cell carcinoma (6%). In the unsuspected group, basal cell carcinoma was mentioned most frequently (42.6%), followed by seborrheic keratosis (19.9%) and squamouscell carcinoma (15.6%). Interestingly, neoplasm not otherwise specified was more frequent in this group (6.4%). By definition, no melanocytic lesions were included on the differential in this group. The depths of the melanomas ranged from 0.1 to 12 mm, and the mean depth was 0.90 mm for all melanomas. The mean depth of the suspected group was 0.62 mm, whereas the mean depth in the unsuspected group was 1.64 mm. This difference was statistically significant (pb 0.001), indicating that the unsuspected group had a greater depth at initial biopsy. Extension to the base of biopsy was found to be 25.4% in the suspected group compared with 51.8% in the unsuspected group, which is also statistically significant (pb 0.001). Extension to the peripheral margin was more
326
H.M. Hermes et al.
Fig. 1 Diagnoses submitted from pathology request forms. Panel A, All submitted diagnoses. Panel B, Suspected group. Panel C, Unsuspected group. Percentage rounded to the nearest whole number.“Others” was defined as entities listed less than five times (b 1% occurrence rate) and in order of decreasing frequencyincludedactinic keratosis, scar, Spitz, poroma, foreign body, dermatofibroma, verruca, prurigo nodule, neurofibroma, insect bite reaction, eczema, and cyst. BCC, basal cell carcinoma; LMM, lentigomaligna melanoma; MIS, melanoma in situ; NOS, not otherwise specified; SCC, squamous cell carcinoma; SK, seborrheic keratosis.
frequent in the suspected group, with 64.2% having positive peripheral margins compared with 36.7% in the unsuspected group, which was not statistically significant (p= 0.02; Table 3). Analysis of depth of lesion by body site or sex of patient revealed no significant trends (data not shown). Table 4 summarizes the frequency of the biopsy techniques of the two groups. The shave biopsy technique
represented the diagnostic procedure of choice for the two groups, 75.7% for the suspected group and 89% for the unsuspected group. Even when melanoma was suspected, shave biopsy was the preferred technique; almost 60% (67/ 112) of the cases were sampled by this technique. Expectedly, excisional biopsy accounted for a higher proportion of the cases in the suspected group.
Characteristics of clinically unsuspected melanomas Table 2
All diagnoses a
One diagnosis Two diagnoses Three diagnoses Four diagnoses Total no.cases (no.of diagnoses)
All
Suspected Group
Unsuspected Group
276 94 14 1 385 (510)
194 71 11 0 276 (369)
82 23 3 1 109 (141)
a Total number of diagnoses submitted as differential diagnosis on the pathology requisition form. The majority had one diagnosis submitted. Total number of diagnoses by group is in parentheses.
Comment Although this study represents the experience of a single institution, which may not have general application, the results corroborate with general trends in melanoma diagnosis and incidence. Similar to the melanoma incidence in the United States and around the world in which thin melanomas (b 1 mm) account for the vast majority of the new cases,1,19–23 more than 70% of the cases in this series were less than 1 mm in depth, with an overall mean depth of 0.90 mm. In this series, all cases were referred by dermatologists, which accounts for an even higher proportion of thin melanomas compared with some other series.23,24 Dermatologists as a group are known to diagnose melanomas at an earlier stage compared with nondermatologists, with a reported mean depth ranging from 0.16 to 0.86 mm25–28; moreover, the proportion of in situ stage was significantly higher than invasive ones when dermatologists detected melanomas.27,28 The proposed explanations of why dermatologists diagnose melanomas at an earlier stage include:(1) having more extensive training and experience in recognizing thin lesions;(2) treating patients harboring more risk factors associated with development of melanoma; and (3) as a group, diagnosing melanomas predominantly of one type with good prognosis, the type that is strongly associated with sun exposure and melanocytic nevi.1,26 It is likely that the combination of these reasons explains the ability of dermatologists to detect melanomas at an earlier stage, but whichever the reason, the trend is clear: led by dermatologists, thinner and thinner melanomas are being diagnosed today more than ever before to the credit of all those who have campaigned for the early detection of the disease. Although dermatologists are proficient in diagnosing thin melanomas, diagnosing thick melanomas still remains a challenge. When histologically confirmed melanomas were clinically unsuspected, they were thicker compared with when they were suspected, 1.64 versus 0.62 mm. The results are consistent with the fact that thin melanomas frequently fulfill the ABCD criteria of melanoma, and thus are readily recognized. Morphologic expression of thick melanomas, on the other hand, is more protean, simulating a wide range of
327 inflammatory and neoplastic diseases. Thick melanomas may present as small, symmetric lesions that are uniform in color defying the ABCD criteria,3,6,18,29 implying that context may be more important than morphologic attributes in recognizing thick melanomas. For this reason, “E” for evolving has been added to the ABCD criteria to emphasize the recognition of changing lesions, because morphologic changes are more frequently observed in melanomas than in melanocytic nevi.30,31 Other screening strategies that rely less on individual morphologic attributes and more on contextual presentation of the patient have been proposed and implemented, which include using the ugly duckling sign, recognizing signature nevi, and lowering the threshold to biopsy in patients with high-risk factors (eg, skin phototype I, personal history of melanoma, and elderly men). Whether these strategies are effective in capturing melanomas with poor prognosis at an earlier stage awaits further validation. As the classification of melanoma transitions from the relative imprecise clinicopathologicbased one to a more precise molecular-based one, additional opportunity to improve the recognition and diagnosis lays ahead,32,33 but for now, thick melanomas continue to elude prompt early diagnosis. Clinical nonmelanocytic simulators of melanoma reported in the literature include seborrheic keratoses, basal cell carcinomas, squamouscell carcinomas, and vascular lesions, among others, with the first two being the most frequently reported (Table 5).34–37 In this series, basal cell carcinoma was the most frequent simulator of melanoma, with seborrheic keratosis coming in second (Figure 1), similar to the results observed in a recent series.37 These observations underscore the importance of recognizing nonmelanocytic simulators of melanoma, because these clinically misdiagnosed lesions are thicker compared with melanocytic simulators. The obvious pitfall of underrecognition is the potential mismanagement as the clinicians, believing that they are either benign or less aggressive malignant lesions, may choose suboptimal biopsy and follow-up care. One strategy to improve the diagnostic specificity of thick melanomas is improving the recognition of nonmelanocytic simulators of melanomas, which should include becoming proficient in dermatoscopy. Although better known for its ability to discriminate melanocytic simulators of melanoma, dermatoscopy is emerging as a powerful tool in improving the diagnostic accuracy of nonmelanocytic simulators of melanoma, especially basal cell carcinoma.38–40 There are limitations of reliance on diagnostic information based on pathology requisition forms, but multiple previous studies have been performed based on similar methods, which provide validity to the method used in this series (Table 5).34,35,37,41–46 Studies emanating from Europe and Australia indicate that excisional biopsy is the preferred diagnostic method when it comes to sampling melanocytic lesions.47–50 In the United States, however, despite the general recommendation of excisional biopsy when sampling a melanocytic lesion,
328
H.M. Hermes et al.
Table 3
Average depth and margin involvement of original biopsy a
All cases Suspected group Unsuspected group p
n
Average Depth (mm)
Base Margin Involved
Peripheral Margin Involved
385 276 109
0.9 0.62 1.64 b 0.001
70 (25.4) 70 (51.8) b 0.001
143 (64.2) 40 (36.7) 0.02
a
The mean of lesions were all less than 1mm thick. Tumors tended to be thicker in the unsuspected group, and they were more likely to extend to the base. Suspected group tended to involve the peripheral margins more often.
shave biopsy represents the preferred biopsy technique performed by clinicians, especially among dermatologists. 37,51 Not surprisingly, shave biopsy represented overwhelmingly the preferred biopsy technique in this series, with more than 80% of the cases having been performed by this method. Although a 1992 survey of US dermatologists indicated that only 10% perform shave biopsy when melanoma is suspected,52 the results of this study and others dispute the past survey results as shave biopsy represents the biopsy technique of choice for dermatologists presumably due to its ease, time, cost, and safety advantages. The potential pitfalls of shave biopsy and other partial biopsy techniques include misdiagnosis and inaccurate histopathologic staging, which may lead to inappropriate and/or suboptimal evaluation, staging, and treatment.50,51 Reported upstaging due to partial biopsies of melanoma ranges from 3% to 25%.37,50,53–56 Compared with a punch biopsy, a shave biopsy may be expected to be less accurate in determining the depth of melanoma due to the shallower nature of the technique, but studies have shown that a shave biopsy is just as accurate, if not more, in determining the final depth of melanoma.50,54,56,57 For thin melanomas, less than 1 mm in depth, initial shave biopsy predicted the final melanoma depth in 97% to 100% of the cases in recent studies.37,55,57 For thick melanomas, larger than 2 mm, partial biopsies, whether shave or punch biopsy, were less predictive of the final melanoma depth and, accordingly, upstaging was observed more frequently in thick than in thin melanomas.50,55–57 Despite the potential mismanagement of melanomas, past meta-analysis and recent studies indicate that the overall prognosis and outcome are not affected by partial biopsies,37,56,58 supporting and validating the practice of shave biopsies by dermatologists for lesions clinically suspected of melanoma; nonetheless, most authors affirm that an excisional biopsy is preferred over any partial biopsies, because it provides the all-important piece of Table 4
prognostic indicator, the Breslow depth, more accurately than any other biopsy method. In this series, when a melanocytic lesion was suspected, the clinicians tended to perform adequate biopsies to ensure that depth of the lesion can be determined, whereas when melanocytic lesion was not suspected, they tended to perform biopsies involving deep margins—25.4% versus 51.8% positive deep margins (Table 3). This result reaffirms that dermatologists are proficient in recognizing thin melanomas and performing biopsies appropriately for their depths. For the unsuspected group, in contrast, clinicians, not cognizant of the therapeutic and staging implications, appeared to have performed biopsies only to determine the diagnosis without regard to the depth of the lesion, resulting in a higher percentage of cases that involved deep margins. Shave biopsy, the preferred technique by dermatologists in this and other series, may suffice for most thin melanomas as previous studies indicate, but an excisional biopsy should be considered when faced with a nodular lesion, especially in the elderly, suspicious for malignancy and/or displaying atypical features of basal cell carcinoma because it may well represent a thick melanoma.
Conclusions What can be done to improve the recognition and capturing of thick melanomas? The current approach to early detection with an emphasis on the ABCD criteria of melanoma and Clark/dysplastic nevi, a presumed precursor to melanoma, has resulted in a marked increase in capturing thin melanomas but not the anticipated decrease in melanoma mortality. Several authors have called for a shift in strategy to improve the detection of thick melanomas at an earlier stage. First, because thick melanomas are more
Biopsy technique breakdown between groups a
All cases Suspected group Melanoma suspected Unsuspected group
n
Shave
Punch
Curette
Excision
385 276 112 109
311 (79.5%) 209 (75.7%) 67 (59.8%) 97 (89.0%)
13 (3.3%) 9 (3.3%) 3 (2.7%) 4 (3.7%)
7 (1.8%) 4 (1.5%) 0 3 (2.8%)
60 (15.3%) 54 (19.6%) 42 (37.5%) 5 (4.6%)
a Shave biopsy was the most common means of biopsy. Subset analysis where melanoma was suspected on the differential revealed that shave biopsy was still the method of choice.
Characteristics of clinically unsuspected melanomas Table 5
329
Nonmelanocytic simulators of melanoma a
Study
Year
Cases Histologically Consistent with Melanoma
No. of Unsuspected Melanomas, n (%)
Basal Cell Carcinoma
Seborrheic Keratosis
Lentigo
Grin et al.40 Witheilerand Cockerell43 Miller and Ackerman45 Zager et al.37 This study
1990 1991 1992 2011 2012
265 1784 354 600 385
51 (19) 583 (32.6) 128 (36) 179 (32) 109 (28.3)
35.0% 12.0% 9.0% 16.0% 16.5%
7.8% 9.0% 13.0% 8.3% 11.2%
5.8% 9.0% 19.0% 2.3% 4.7%
a A summary of previous studies reporting the clinical diagnoses of melanomas diagnosed histologically, based on clinical requisition forms. Common nonmelanocytic simulators include basal cell carcinoma, seborrheic keratosis, and lentigo, reported as percent of total melanocytic cases, which were clinically misidentified.
frequently discovered by patients and their family members than by physicians,25,59–61 implementation of more effective campaigns and measures to educate the public and promote self-skin examination for all age groups have been proposed.62 This form of secondary prevention, which is currently believed to be underused, may yield the highest reward in capturing thick melanomas, because these fastgrowing lesions may not be amenable to screening by health care providers.1,63 Second, targeting groups at a greater risk for development of thick melanomas has been proposed as another strategy to reduce the burden of thick melanomas, which include men older than 65 years and socioeconomically disadvantaged individuals.62,64–66 Both higher risk groups, however, provide challenges for the health care providers because elderly men are less amenable to self-skin examination, and socioeconomically disadvantaged individuals have healthcare access issues. Third, recruitment of more primary care physicians to perform skin examinations has been proposed, for patients not only visit primary care physicians more frequently, but they also have easier access to them.62 Whether such shifts in public campaigns and physician resources will result in a decrease in melanoma-related mortality remains to be proved. The results of this and other studies, however, lend support for a shift in strategy. Under-recognition of thick melanomas revealed in this and other series begs the question of how many of these lesions are actually being left undetected by health care providers because series such as this one can include only those that have been captured and not those left behind. The ongoing molecular, epidemiologic, and clinicopathologic investigations all must converge to further delineate the etiopathogenesis of thick melanomas to improve their detection at an earlier stage.
References 1. Lipsker D, Engel F, Cribier B, et al. Trends in melanoma epidemiology suggest three different types of melanoma. Br J Dermatol. 2007;157: 338-343. 2. Argenziano G, Kittler H, Ferrara G, et al. Slow-growing melanoma: adermoscopy follow-up study. Br J Dermatol. 2010;162:267-273.
3. Martorell-Calatayud A, Nagore E, Botella-Estrada R, et al. Defining fast-growing melanomas: reappraisal of epidemiological, clinical, and histological features. Melanoma Res. 2011;21:131-138. 4. Surveillance, Epidemiology, and End Results Program. SEER stat fact sheet: melanoma of the skin. http://seer.cancer.gov/statfacts/html/ melan.html. Accessed September 12, 2013. 5. Pollack LA, Li J, Berkowitz Z, et al. Melanoma survival in the United States, 1992 to 2005. J Am Acad Dermatol. 2011;65:S78-S86. 6. Demierre MF, Chung C, Miller DR, Geller AC. Early detection of thick melanomas in the United States: beware of the nodular subtype. Arch Dermatol. 2005;141:745-750. 7. Murray CS, Stockton DL, Doherty VR. Thick melanoma: the challenge persists. Br J Dermatol. 2005;152:104-109. 8. Burton RC, Armstrong BK. Recent incidence trends imply a nonmetastasizing form of invasive melanoma. Melanoma Res. 1994;4:107-113. 9. Swerlick RA, Chen S. The melanoma epidemic. Is increased surveillance the solution or the problem? Arch Dermatol. 1996;132: 881-884. 10. Glusac EJ. The melanoma ‘epidemic’, a dermatopathologist’sperspective.J CutanPathol. 2011;38:264–267. 11. Swerlick RA, Chen S. The melanoma epidemic: more apparent than real? Mayo Clin Proc. 1997;72:559-564. 12. Swica Y, Koehler A, Ackerman AB. Lies, damn lies, and statistics: why there is no epidemic of melanoma. Dermatopathol Pract Concept. 2001;7:347-354. 13. Welch HG, Woloshin S, Schwartz LM. Skin biopsy rates and incidence of melanoma: population based ecological study. BMJ. 2005;331:481. 14. Levell NJ, Beattie CC, Shuster S, Greenberg DC. Melanoma epidemic: a midsummer night’s dream? Br J Dermatol. 2009;161:630-634. 15. Shuster S. Malignant melanoma: how error amplification by screening creates spurious disease. Br J Dermatol. 2009;161:977-979. 16. van der Rhee JI, Mooi WJ, Kukutsch NA, et al. Iatrogenic melanoma. Comment on: melanoma epidemic: a midsummer night’s dream? Br J Dermatol. 2010;162:457-458. 17. Frangos JE, Duncan LM, Piris A, et al. Increased diagnosis of thin superficial spreading melanomas: a20-year study. J Am Acad Dermatol. 2012;67:387-394. 18. Chamberlain AJ, Fritschi L, Kelly JW. Nodular melanoma: patients’perceptions of presenting features and implications for earlier detection. J Am Acad Dermatol. 2003;48:694-701. 19. Buettner PG, Leiter U, Eigentler TK, Garbe C. Development of prognostic factors and survival in cutaneous melanoma over 25 years: an analysis of the Central Malignant Melanoma Registry of the German Dermatological Society. Cancer. 2005;103:616-624. 20. Crocetti E, Caldarella A, Chiarugi A, et al. The thickness of melanomas has decreased in central Italy, but only for thin melanomas, while thick melanomas are as thick as in the past. Melanoma Res. 2010;20:422-426. 21. Coory M, Baade P, Aitken J, et al. Trends for in situ and invasive melanoma in Queensland, Australia, 1982-2002. Cancer Causes Control. 2006;17:21-27.
330 22. Linos E, Swetter SM, Cockburn MG, et al. Increasing burden of melanoma in the United States. J Invest Dermatol. 2009;129:1666-1674. 23. Criscione VD, Weinstock MA. Melanoma thickness trends in the United States, 1988-2006. J Invest Dermatol. 2010;130:793-797. 24. Berwick M, Erdei E, Hay J. Melanoma epidemiology and public health. Dermatol Clin. 2009;27:205-214. 25. Carli P, De Giorgi V, Palli D, et al. Dermatologist detection and skin self-examination are associated with thinner melanomas: results from a survey of the Italian Multidisciplinary Group on Melanoma. Arch Dermatol. 2003;139:607-612. 26. Pennie ML, Soon SL, Risser JB, et al. Melanoma outcomes for Medicare patients: association of stage and survival with detection by a dermatologist vs a nondermatologist. Arch Dermatol. 2007;143: 488-494. 27. Kantor J, Kantor DE. Routine dermatologist-performed full-body skin examination and early melanoma detection. Arch Dermatol. 2009;145: 873-876. 28. Kovalyshyn I, Dusza SW, Siamas K, et al. The impact of physician screening on melanoma detection. Arch Dermatol. 2011; 147:1269-1275. 29. Liu W, Dowling JP, Murray WK, et al. Rate of growth in melanomas: characteristics and associations of rapidly growing melanomas. Arch Dermatol. 2006;142:1551-1558. 30. Kittler H, Seltenheim M, Dawid M, et al. Morphologic changes of pigmented skin lesions: a useful extension of the ABCD rule for dermatoscopy. J Am Acad Dermatol. 1999;40:558-562. 31. Abbasi NR, Shaw HM, Rigel DS, et al. Early diagnosis of cutaneous melanoma: revisiting the ABCD criteria. JAMA. 2004;292:2771-2776. 32. Curtin JA, Fridlyand J, Kageshita T, et al. Distinct sets of genetic alterations in melanoma. N Engl J Med. 2005;353:2135-2147. 33. Romano E, Schwartz GK, Chapman PB, et al. Treatment implications of the emerging molecular classification system for melanoma. Lancet Oncol. 2011;12:913-922. 34. Grant-Kels JM, Bason ET, Grin CM. The misdiagnosis of malignant melanoma. J Am Acad Dermatol. 1999;40:539-548. 35. Izikson L, Sober AJ, Mihm Jr MC, Zembowicz A. Prevalence of melanoma clinically resembling seborrheic keratosis: analysis of 9204 cases. Arch Dermatol. 2002;138:1562-1566. 36. Matz H, Orion E, Ruocco V, Wolf R. Clinical simulators of melanoma. Clin Dermatol. 2002;20:212-221. 37. Zager JS, Hochwald SN, Marzban SS, et al. Shave biopsy is a safe and accurate method for the initial evaluation of melanoma. J Am Coll Surg. 2011;212:454-460. 38. Marghoob AA, Braun R. Proposal for a revised 2-step algorithm for the classification of lesions of the skin using dermoscopy. Arch Dermatol. 2010;146:426-428. 39. Hirokawa D, Lee JB. Dermatoscopy: an overview of subsurface morphology. Clin Dermatol. 2011;29:557-565. 40. Rosendahl C, Tschandl P, Cameron A, Kittler H. Diagnostic accuracy of dermatoscopy for melanocytic and nonmelanocytic pigmented lesions. J Am Acad Dermatol. 2011;64:1068-1073. 41. Rampen FH, Rumke P. Referral pattern and accuracy of clinical diagnosis of cutaneous melanoma. Acta Derm Venereol. 1988;68: 61-64. 42. Grin CM, Kopf AW, Welkovich B, et al. Accuracy in the clinical diagnosis of malignant melanoma. Arch Dermatol. 1990;126:763-766. 43. Witheiler DD, Cockerell CJ. Histologic features and sensitivity of diagnosis of clinically unsuspected cutaneous melanoma. Am J Dermatopathol. 1991;13:551-556. 44. Witheiler DD, Cockerell CJ. Sensitivity of diagnosis of malignant melanoma: a clinicopathologic study with a critical assessment of biopsy techniques. Exp Dermatol. 1992;1:170-175.
H.M. Hermes et al. 45. Miller M, Ackerman AB. How accurate are dermatologists in the diagnosis of melanoma? Degree of accuracy and implications. Arch Dermatol. 1992;128:559-560. 46. Morton CA, Mackie RM. Clinical accuracy of the diagnosis of cutaneous malignant melanoma. Br J Dermatol. 1998;138:283-287. 47. Lees VC, Briggs JC. Effect of initial biopsy procedure on prognosis in Stage 1 invasive cutaneous malignant melanoma: review of 1086 patients. Br J Surg. 1991;78:1108-1110. 48. Bong JL, Herd RM, Hunter JA. Incisional biopsy and melanoma prognosis. J Am Acad Dermatol. 2002;46:690-694. 49. Molenkamp BG, Sluijter BJ, Oosterhof B, et al. Non-radical diagnostic biopsies do not negatively influence melanoma patient survival. Ann Surg Oncol. 2007;14:1424-1430. 50. Ng JC, Swain S, Dowling JP, et al. The impact of partial biopsy on histopathologic diagnosis of cutaneous melanoma: experience of an Australian tertiary referral service. Arch Dermatol. 2010;146:234-239. 51. Lowe M, Hill N, Page A, et al. The impact of shave biopsy on the management of patients with thin melanomas. Am Surg. 2011;77:1050-1053. 52. Salopek TG, Slade J, Marghoob AA, et al. Management of cutaneous malignant melanoma by dermatologists of the American Academy of Dermatology. I. Survey of biopsy practices of pigmented lesions suspected as melanoma. J Am Acad Dermatol. 1995;33:441-450. 53. Karimipour DJ, Schwartz JL, Wang TS, et al. Microstaging accuracy after subtotal incisional biopsy of cutaneous melanoma. J Am Acad Dermatol. 2005;52:798-802. 54. Stell VH, Norton HJ, Smith KS, et al. Method of biopsy and incidence of positive margins in primary melanoma. Ann Surg Oncol. 2007;14: 893-898. 55. Moore P, Hundley J, Hundley J, et al. Does shave biopsy accurately predict the final breslow depth of primary cutaneous melanoma? Am Surg. 2009;75:369-373. 56. Egnatios GL, Dueck AC, Macdonald JB, et al. The impact of biopsy technique on upstaging, residual disease, and outcome in cutaneous melanoma. Am J Surg. 2011;202:771-777. 57. Ng PC, Barzilai DA, Ismail SA, et al. Evaluating invasive cutaneous melanoma: is the initial biopsy representative of the final depth? J Am Acad Dermatol. 2003;48:420-424. 58. Pflugfelder A, Weide B, Eigentler TK, et al. Incisional biopsy and melanoma prognosis: facts and controversies. Clin Dermatol. 2010;28:316-318. 59. Brady MS, Oliveria SA, Christos PJ, et al. Patterns of detection in patients with cutaneous melanoma. Cancer. 2000;89:342-347. 60. Carli P, De Giorgi V, Palli D, et al. Patterns of detection of superficial spreading and nodular-type melanoma: a multicenter Italian study. Dermatol Surg. 2004;30:1371-1375. 61. McPherson M, Elwood M, English DR, et al. Presentation and detection of invasive melanoma in a high-risk population. J Am Acad Dermatol. 2006;54:783-792. 62. Geller AC, Swetter SM, Oliveria S, et al. Reducing mortality in individuals at high risk for advanced melanoma through education and screening. J Am Acad Dermatol. 2011;65:S87-S94. 63. Pozo L, Diaz-Cano SJ. Tumor screening and biology in malignant melanomas. Arch Dermatol. 2000;136:934-935. 64. Geller AC, Miller DR, Swetter SM, et al. A call for the development and implementation of a targeted national melanoma screening program. Arch Dermatol. 2006;142:504-507. 65. Janda M, Neale RE, Youl P, et al. Impact of a video-based intervention to improve the prevalence of skin self-examination in men 50 years or older: the Randomized Skin Awareness Trial. Arch Dermatol. 2011;147:799-806. 66. McGuire ST, Secrest AM, Andrulonis R, Ferris LK. Surveillance of patients for early detection of melanoma: patterns in dermatologist vs patient discovery. Arch Dermatol. 2011;147:673-678.