Grade of dysplastic nevus and association with biopsy margin positivity, frequency of excision after positive biopsy margin, frequency of residual nevus in excision specimen, and clinically significant change in diagnosis upon excision.
A and B, Initial biopsy shows moderate to severely dysplastic nevus with predominantly nested melanocytic hyperplasia, atypical, concentric, and lamellar fibroplasia, and mild host inflammatory infiltrate (hematoxylin-eosin, original magnifications ×10 [A] and ×20 [B]). C and D, Excision specimen showing scar on the right with adjacent nested and lentiginous severely atypical melanocytic hyperplasia on the left (C, hematoxylin-eosin, original magnification ×4) consistent with melanoma in situ (D, hematoxylin-eosin, original magnification ×40).
A and B, A component of the lesion appears as a mildly to moderately dysplastic nevus (hematoxylin-eosin, original magnifications ×10 [A] and ×20 [B]). C and D, The dysplastic nevus is associated with melanoma in situ showing moderate to severe lentiginous melanocytic hyperplasia, dyshesion, focal pagetoid spread, mild dermal fibrosis, and melanophages (hematoxylin-eosin, original magnifications ×10 [C] and ×20 [D]).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Reddy KK, Farber MJ, Bhawan J, Geronemus RG, Rogers GS. Atypical (Dysplastic) Nevi: Outcomes of Surgical Excision and Association With Melanoma. JAMA Dermatol. 2013;149(8):928–934. doi:10.1001/jamadermatol.2013.4440
To evaluate the effect of surgical excision, performed after biopsy diagnosis of dysplastic nevus, on final diagnosis, melanoma prevention, and melanoma detection.
Design, Setting, and Participants
Outcome study using retrospective review conducted in an academic dermatopathology practice (Boston Medical Center Skin Pathology Laboratory) that receives specimens from community and academic practices across the United States. Consecutive patient pathology samples of dysplastic nevi and cutaneous melanomas evaluated between September 1, 1999 and March 1, 2011, and identified using systematized nomenclature of medicine codes were included.
Main Outcomes and Measures
In dysplastic nevi cases, the rate of clinically significant change in diagnosis and the rate of melanoma detection as a result of excision. In melanoma cases, the rate and characteristics of association with dysplastic nevus.
Of dysplastic nevi, 196 of 580 (34%) showed a positive biopsy margin, increasing with grade of atypia (P < .001); 127 of 196 with positive biopsy margin received excision (65%), performed more often as grade of atypia increased (P < .001). Two excisions (2 of 127, 1.6%) resulted in a clinically significant change in diagnosis, from biopsy-diagnosed moderately-to-severely dysplastic nevi before excision to melanoma in situ after excision. In melanomas (n = 216), in situ and superficial spreading subtypes were more often associated with dysplastic nevi (20% and 18%, respectively) (P = .002), most often of moderate-to-severe or severe grade.
Conclusions and Relevance
Excision of biopsy-diagnosed mildly or moderately dysplastic nevi is unlikely to result in a clinically significant change in diagnosis, and risk of transformation to melanoma appears very low. Moderately-to-severely and severely dysplastic nevi are more often associated with melanoma, and excision may be beneficial for melanoma detection or prevention.
Clark et al1 described the atypical, or dysplastic, nevus as having unusual clinical and pathologic features found in individuals or families predisposed to melanoma. A National Institutes of Health Consensus Conference2 further defined atypical nevi as having diameter greater than 5 mm, color or border irregularity, and certain histologic features. Evidence supports observations that the presence of atypical nevi is associated with increased overall melanoma risk.3
Atypical nevi are common, with an incidence of 2% to 8% in white patients.4,5 Many are biopsied to evaluate for melanoma.4 Management of a biopsy-proven dysplastic (atypical) nevus with a positive histologic margin remains ill defined.6 Particularly after a biopsy finding of mildly or moderately dysplastic nevus with positive histologic margin, there is significant variation and lack of consensus in subsequent management.6
Surgical excision of the dysplastic nevus biopsy site with a 2- to 3-mm margin of normal skin is frequently performed. The goal of excision is to confirm the biopsy diagnosis, to ensure complete removal due to concerns of future malignant transformation, or both. These goals include melanoma detection and prevention. The lifetime transformation risk of an “average” dysplastic nevus into melanoma is estimated at 1 in 10 000, though risk likely varies with grade of atypia.7 In addition, excision is sometimes performed to eliminate risk of a recurrent nevus, a benign lesion that rarely may be difficult to distinguish from melanoma.8 Disadvantages of excision include risks of scarring and surgical complications and utilization of health resources.
The management of dysplastic nevi significantly impacts individual and public health. The impact of surgical excision of dysplastic nevi on goals of melanoma detection and prevention has not been well-studied. Our objectives were to evaluate the impact of surgical excision of biopsy-diagnosed dysplastic nevi on final diagnosis, on melanoma detection, and on melanoma prevention through analysis of concordance of biopsy and excision diagnoses and examination of the risk of dysplastic nevus transformation to melanoma.
Study approval was granted by the Boston Medical Center institutional review board. The Boston Medical Center Skin Pathology Laboratory database was searched using systematized nomenclature of medicine (SNOMED) codes for dysplastic nevi and for primary cutaneous melanoma. All cases were diagnosed by a board-certified dermatopathologist. Dysplastic nevus grade of atypia was diagnosed using criteria summarized by Arumi-Uria et al.9 Biopsy reports dated between September 1, 1999, and March 1, 2011, reporting a pathologic diagnosis of dysplastic nevus were reviewed retrospectively in reverse chronologic order until 580 cases were reviewed. Dysplastic nevi characteristics of gross pathologic size, grade of atypia, anatomic location, and biopsy margin positivity were recorded. In cases reporting a dysplastic nevus with a positive biopsy margin, records were reviewed for surgical excision. Surgical excision of the lesion with a 2- to 3-mm margin of normal skin followed by closure of the skin is the standard method for removal of a biopsy-diagnosed dysplastic nevus. If an excision was performed, date of excision, presence of residual lesion on pathologic examination, and final pathologic diagnosis were recorded. Concordance of biopsy diagnoses with excision diagnoses was recorded. Clinically significant changes in diagnosis upon excision were recorded, defined as severe atypia ormelanoma upon excision of biopsy-diagnosed mildly-to-moderately or moderately dysplastic nevi, and defined as melanoma upon excision of biopsy-diagnosed moderately-to-severely or severely dysplastic nevi.
Pathology reports dated between September 1, 1999, and March 1, 2011, reporting a diagnosis of primary cutaneous melanoma were also reviewed, until a limit of 216 cases. Melanoma type, depth, presence of associated scar, and presence of associated dysplastic nevus were recorded. If an associated dysplastic nevus was present, grade of atypia was recorded. In cases of melanoma with associated scar, records were reviewed, and any history of biopsy-diagnosed dysplastic nevus at the melanoma site was recorded.
Observed characteristics of dysplastic nevi and primary cutaneous melanomas were summarized using counts and proportions. For dysplastic nevi, biopsy margin positivity, frequency of excision after positive biopsy margin, frequency of residual nevus upon excision, and clinically significant change in diagnosis upon excision were summarized within categories of dysplastic nevus grade. For cases of dysplastic nevus with positive biopsy margin, anatomic location and size were also summarized within categories of dysplastic nevus grade. In primary cutaneous melanoma cases, presence and grade of associated dysplastic nevus, presence of associated scar, and history of biopsy-proven dysplastic nevus at the melanoma site were summarized within categories of melanoma type and melanoma depth. The χ2 Fisher exact and Cochran-Armitage trend tests were used to compare proportions and evaluate trends. All statistical analyses were performed using the SAS statistical package, version 9.2 (SAS Institute). All P values calculated are 2-sided.
A total of 580 cases reporting a biopsy diagnosis of dysplastic nevus were reviewed (Table 1). Nearly all were biopsied by shave biopsy technique. Standard shave biopsy of atypical nevi includes partial-thickness dermis and a 1- to 2-mm margin of normal skin. Overall, 196 of 580 (34%) reported a positive biopsy margin. A positive biopsy margin was more often reported as the degree of atypia worsened (mild, 11%; mild to moderate, 19%; moderate, 40%; moderate to severe, 62%; and severe, 85%) (P < .001 for test of trend) (Figure 1). As a result of this observation, further analysis for factors affecting margin positivity was conducted. In dysplastic nevi with positive biopsy margins, gross pathologic size and anatomic location at the head or neck were also positively correlated with grade of atypia and with biopsy margin positivity, with the exception of mildly-to-moderately dysplastic nevi, which showed slightly larger size and greater frequency on the head or neck than would be predicted according to the trend (Table 2).
Of dysplastic nevi with a positive biopsy margin, 127 of 196 (65%) received surgical excision (Table 1). As grade of atypia increased, frequency of excision after positive biopsy margin increased (mild, 12%; mild to moderate, 53%; moderate, 63%; moderate to severe, 81%; severe, 82%) (P < .001 for test of trend) (Figure 1). Of dysplastic nevi with a positive biopsy margin that received surgical excision, 42 of 127 (33%) showed residual nevus. Rate of biopsy margin positivity sorted by grade of atypia and grade of atypia were not associated with the presence of residual nevus upon excision (P = .68 for test of trend). Quiz Ref IDTwo of 127 surgical excisions (1.6%) led to a clinically significant change in diagnosis. Both were biopsy-diagnosed moderately-to-severely dysplastic nevi that were diagnosed as melanoma in situ upon excision (Figure 2). There were not any cases of biopsy-diagnosed mildly or moderately dysplastic nevi that resulted in a clinically significant change in diagnosis upon excision.
Quiz Ref IDA total of 216 primary cutaneous melanomas were also reviewed (Table 3). Sixteen of 90 superficial spreading melanomas (18%) and 9 of 44 melanomas in situ (20%) were associated with a dysplastic nevus (Table 3 and Figure 3). The association of melanomas in situ and superficial spreading melanomas with dysplastic nevi (25 of 134, 18.7%) was significantly higher than the association between other melanoma types and dysplastic nevi (2 of 82, 2%) (P = .001). None of melanomas showing associated scar without associated dysplastic nevus revealed a history of biopsy-diagnosed dysplastic nevus at the site (Table 3). In melanomas associated with dysplastic nevi, the most frequent grades of atypia were moderate-to-severe and severe (Table 4). Of 27 melanomas associated with dysplastic nevi, 26 of 27 (96%) were 1 mm or less in depth. Melanoma association with dysplastic nevus decreased with increasing melanoma depth (P = .03 for test of trend) (Table 5). Overall, 10.97% of invasive melanomas were associated with a dysplastic nevus.
Atypical nevi are associated with an increased risk of cutaneous melanoma, which may develop in existing nevi or de novo on previously normal skin. Therefore, management of the patient with atypical nevi is recommended to include complete skin examinations at regular intervals and patient education in melanoma prevention and early detection.4 While many physicians also perform surgical excision of dysplastic nevi having a positive histologic margin, outcomes of atypical nevus excision have not been examined in detail, to our knowledge, and the appropriate surgical management of dysplastic nevi has remained poorly defined.
In our examination of the impact of dysplastic nevus excision on final diagnosis and on melanoma detection, varying effects were observed for mildly and moderately dysplastic nevi compared with moderately-to-severely and severely dysplastic nevi. Quiz Ref IDMildly and moderately dysplastic nevi did not show any clinically significant change from biopsy diagnosis nor any evidence of melanoma upon surgical excision. The data suggest that biopsied mildly and moderately dysplastic nevi may not require subsequent surgical excision to confirm the diagnosis and are unlikely to harbor associated melanoma in residual cells after biopsy.Quiz Ref IDIn contrast, moderately-to-severely dysplastic nevi with a positive biopsy margin displayed a 4% rate of melanoma in situ diagnosis upon surgical excision. Moderately-to-severely and severely dysplastic nevi were also most often associated with melanomas. This suggests that in contrast to patients with mildly and moderately dysplastic nevi, patients with moderately-to-severely and severely dysplastic nevi with positive biopsy margins are more likely to benefit from surgical excision for confirmation of diagnosis, melanoma detection, and melanoma prevention.
In our series, 34% of dysplastic nevi showed a positive biopsy margin, consistent with a previous report by Armour et al.10 Dysplastic nevus biopsy margins were more often positive as grade of atypia worsened, and as gross size and frequency of head or neck location increased. Potential theoretical explanations for the increased biopsy margin positivity observed with increasing grade of atypia include increased subclinical extension as atypia worsens, increased gross nevus size as atypia worsens with clinician biopsies not increasing proportionately in size, increased proportion of head or neck lesions having wide atypical melanocytic hyperplasia, and/or medical or legal concerns leading to more thorough margin assessment as grade of atypia increases. Our study confirms that patients having a dysplastic nevus with a positive biopsy margin frequently receive excision, and that frequency of excision appears to increase with grade of atypia. The greatest variability in management was observed for cases of mildly-to-moderately and moderately dysplastic nevi, which received surgical excision after a positive biopsy margin finding in 53% and 63% of cases, respectively.
There have been limited reports describing practice patterns in the management of atypical nevi. In 2002, a survey of dermatologists reported that 67% of respondents preferred to perform surgical excision of a dysplastic nevus that had a positive biopsy margin, with many self-writing in that the decision would depend on the degree of atypia.6 Our findings confirm that excision frequency is positively correlated with degree of atypia, and we have quantified observed excision rates.
To our knowledge, the effect of dysplastic nevus excision on the final diagnosis has been examined previously in only 2 small studies.10,11 Armour and colleagues10 found that 21 of 22 shave biopsies (95%) of atypical nevi were concordant with the excision diagnosis. Study limitations include a small sample size and lack of reporting of grades of atypia examined. Cohen and colleagues11 examined 189 excision specimens obtained after biopsy of atypical melanocytic lesions, many of which were atypical nevi. Of these, 47 of 189 (25%) contained residual nevus cells and 1 of 189 (0.5%) contained melanoma (the initial biopsy diagnosis having been moderately-to-severely atypical nevus). Our data also support a low rate of change in diagnosis upon excision (1.6%) when all grades of atypical nevi are considered together and confirms that moderately-to-severely atypical nevi are more likely to be associated with melanoma.
The association of dysplastic nevi with melanoma has been reported, although without sorting by grade of atypia. Studies have estimated that 60% to 80% of melanomas arise de novo12 and that melanomas arise in association with an atypical nevus in 0.5% to 46% of cases.4 Based on data showing that 11% of invasive melanomas were associated with an atypical nevus, Tsao et al7 estimated the lifetime risk of an “average” atypical nevus transforming to melanoma as approximately 1 in 10 000. Our study confirms an 11% rate of association between dysplastic nevi and invasive melanoma. Several melanoma-associated dysplastic nevi were of unspecified grade, and depending on the true level of atypia, estimated transformation risk may vary accordingly.
Sorted by grade of atypia, our data suggest that the lifetime risk of a moderately-to-severely or severely dysplastic nevus transforming to melanoma is likely clinically significant. In contrast, the lifetime risk of mildly or moderately atypical nevi transforming to melanoma may be similar to that of a normal “typical” nevus.
Our findings are most generalizable to dysplastic nevi that have been biopsied and found to have clinically negative margins but positive histologic margins because it is standard dermatologic practice to remove the entire clinically visible nevus during a shave biopsy procedure. Study limitations include the difficulty of estimating melanoma transformation risk of an atypical nevus retrospectively, though prospective study of this outcome is unlikely to be conducted. Large, prospective clinical trials are encouraged in further studying outcomes of excision of dysplastic nevi.
Overall, our data suggest 2 dysplastic nevi risk categories: (1) mildly, mildly-to-moderately, and moderately dysplastic nevi generally appearing as low-risk lesions, and (2) moderately-to-severely and severely dysplastic nevi appearing to carry a higher risk for coexistent or future melanoma. This suggests that patients with biopsied mildly or moderately dysplastic nevi may be observed clinically, while patients with moderately-to-severely and severely atypical nevi have a higher risk of melanoma and may benefit from excision. Moderately atypical nevi, which have a large variation in management, appear to display similar characteristics and behavior to those of mildly atypical nevi. Together, these data suggest that routine surgical excision of biopsied mildly or moderately atypical nevi with a positive biopsy margin for the purpose of melanoma prevention or detection may not be indicated, Quiz Ref IDwhile surgical excision of biopsied moderately-to-severely or severely atypical nevi with a positive biopsy margin may be beneficial in melanoma prevention and detection. Individual patient factors are important in medical decision making, which remains the responsibility of the treating physician. Given the common biopsy diagnosis of atypical nevus, we encourage further study of outcomes of surgical excision to promote optimum patient care and to maximize melanoma prevention and detection efforts.
Corresponding Author: Kavitha K. Reddy, MD, Laser & Skin Surgery Center of New York, 317 E 34th St, 11th Floor, New York, NY 10016 (firstname.lastname@example.org).
Accepted for Publication: March 28, 2013.
Published Online: June 12, 2013. doi:10.1001/jamadermatol.2013.4440.
Author Contributions: Drs Reddy and Rogers had full access to the all of the data in the study, take responsibility for the integrity of the data and the accuracy of data reporting and analysis.
Study concept and design: Reddy, Rogers.
Acquisition of data: Reddy, Farber.
Analysis and interpretation of data: Reddy, Bhawan, Geronemus, Rogers.
Drafting of the manuscript: Reddy, Farber, Rogers.
Critical revision of the manuscript for important intellectual content: Reddy, Bhawan, Geronemus, Rogers.
Obtained funding: Reddy.
Administrative, technical, and material support: Bhawan.
Study supervision: Geronemus, Rogers.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by a Cutting Edge Research Grant from the American Society for Dermatologic Surgery (Dr Reddy).
Role of the Sponsors: The sponsor did not have any role in the design or conduct of the study; in the collection, analysis, or interpretation of data; or in the preparation, review, or approval of the manuscript.
Additional Contributions: We gratefully acknowledge Robin Ruthazer, MPH, of the Biostatistics Research Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, for her assistance with statistical analysis.
Create a personal account or sign in to: