Acquired Blue Nevi in Older Individuals Retrospective Case Series from a Veterans Affairs Population, 1991 to 2013

Research

Case Report/Case Series Acquired Blue Nevi in Older Individuals Retrospective Case Series From a Veterans Affairs Population, 1991 to 2013

Erik S. Cabral, MD; Frank W. Chen, BA; Barbara M. Egbert, MD; Susan M. Swetter, MD

IMPORTANCE Apart from the atypical mole phenotype, development of new melanocytic nevi in older individuals is uncommon and considered worrisome for melanoma. We performed a retrospective case series in a Veterans Affairs population from 1991 to 2013 to characterize blue nevi (BN) by patient age at biopsy, location, self-reported duration, and relation to prior or subsequent development of cutaneous melanoma. Author Affiliations: Department of OBSERVATIONS A total of 204 BN were identified in 194 predominantly male patients Dermatology, Veterans Affairs Palo (90.7%) who had a mean (SD) age of 62.8 (14.4) years. Clinical duration of 10 years or less Alto Health Care System, Palo Alto, California (Cabral, Swetter); was reported by 90.3% of patients with available data (32.0%). Histopathologic examination Department of Pathology Services, classified 74.0% of BN as common, 1.5% as cellular, and 24.5% as combined type. No Veterans Affairs Palo Alto Health Care malignant BN were identified; however, 18 primary melanomas were diagnosed, most System, Palo Alto, California (Egbert); Department of Dermatology, (72.2%) prior to blue nevus biopsy, including 38.9% in situ and 61.1% with mean (SD) Breslow Pigmented Lesion and Melanoma thickness of 1.02 (0.99) mm. Program, Stanford University Medical Center, Stanford, California (Cabral, CONCLUSIONS AND RELEVANCE The later patient-reported onset of BN suggests a potential Chen, Egbert, Swetter); Bennett Surgical Center, Santa Monica, alternative mechanism of nevogenesis compared with common acquired nevi and differs California (Cabral). from prior reports of BN development in younger adults. The lack of association with Corresponding Author: Susan M. melanoma in older individuals suggests that most benign-appearing BN may be safely Swetter, MD, Department of observed, even in a cohort at higher risk for skin cancer. Dermatology, Pigmented Lesion and Melanoma Program, Stanford University Medical Center, 900 Blake JAMA Dermatol. 2014;150(8):873-876. doi:10.1001/jamadermatol.2013.7366 Wilbur Dr, W0103, Stanford, CA Published online April 30, 2014. 94305-5356 ([email protected]).

lue nevi (BN) are benign pigmented lesions with rare ment alteration, to malignant entities, such as pigmented basal malignant potential. They are considered a subset of cell carcinoma or melanoma. B dermal dendritic melanocytic proliferations, which de- Apart from the setting of the atypical mole or dysplastic rive their dark blue or bluish black color from the scattering nevi syndrome, development of new melanocytic nevi after of light by dermal melanin, known as the Tyndall effect.1,2 The age 50 years is uncommon and raises concern about the diag- melanocytes in BN typically stain positive for Human Mela- nosis of cutaneous melanoma. Based on our clinical observa- noma Black-45 (HMB-45), suggesting that BN stem from lation of the increased frequency and later onset of BN in a pre- tent dendritic melanocytes trapped in the dermal layer dur- dominantly elderly population, we sought to characterize BN ing their migration from the neural crest to the epidermis.3 clinically and histopathologically to elucidate patient- There are 3 main histologic subtypes of BN: common (includ- reported age of onset, the relationship to prior, concomitant, ing sclerotic type), combined, and cellular. Common BN gen- or subsequent primary melanoma, and their malignant po- erally present as solitary and small (1-5 mm) dark blue mac- tential, including the proportion of cases identified as malig- ules or dome-shaped papules. Blue nevi are also commonly nant BN or blue nevus–like melanoma. observed as part of combined nevi, most often in conjunction with a compound or intradermal nevus, and less frequently, with a Spitz or dysplastic nevus. Cellular BN are more Report of Cases likely than common BN to be elevated, larger in size, and have greater malignant potential.1 A retrospective review of BN identified from 1991 to 2013 Patients may report BN as an aesthetic concern because in the Veterans Affairs Palo Alto Health Care System they are often mistaken for cosmetic or traumatic tattoos. For (VAPAHCS) Pathology Service database was performed, the clinician, the differential diagnosis ranges from benign en- with clinical chart review to determine patient outcome, tities, such as common nevi, tattoo, or postinflammatory pig- reported duration of the lesion, and other clinical character-

jamadermatology.com JAMA Dermatology August 2014 Volume 150, Number 8 873

Downloaded From: https://jamanetwork.com/ on 09/28/2021 Research Case Report/Case Series Acquired Blue Nevi in Older Individuals

istics. The VAPAHCS electronic medical record did not A total of 204 cases of biopsy-proven BN were identified become widely available until 1997, thereby limiting clinical in 194 patients who had a mean (SD) age of 62.8 (14.4) years, data annotation for 33 patients. Clinical outcome data were with most biopsies being performed to exclude the diagno- assessed through May 30, 2013. The study was approved by sis of melanoma based on patient report of short or Stanford University and VAPAHCS institutional review unknown duration of the lesion. Patients were predomi- boards. nantly male (90.7%) and white, typical of the VAPAHCS population, although race/ethnicity was not noted in all cases. The mean duration of patient follow-up was 9.0 years Table 1. Clinical and Histopathological Characteristics of 204 Blue Nevi in 194 Patients (range, 0-22 years). In terms of histologic subtypes, 151 of 204 BN (74.0%) Characteristic No. (%) were diagnosed as common, 50 of 204 (24.5%) were diag- Sex nosed as combined (BN with intradermal nevus in 84.0% Male 176 (90.7) and compound nevus in 16.0%, 2 of which were also noted Female 18 (9.3) to be mild to moderately dysplastic), and 3 of 204 (1.5%) as Age at diagnosis, mean (SD), y 62.8 (14.4) cellular. Most BN were located on the extremities (93 of 204 Pathologic diagnosis [45.6%]), followed by the head and neck (64 of 204 [31.3%]), Common 151 (74.0) and the back (35 of 204 [17.2%]). Of the 15 BN on the hands, Cellular 3 (1.5) 8 (53.3%) were on the dorsal surface, while only 2 of 9 BN on Combined 50 (24.5) the feet (22.2%) were located dorsally. For the 64 head and Body site neck BN, 33 (51.6%) were located on the scalp, 16 (25.0%) on Scalp 33 (16.2) the forehead-temple-eyebrow-eyelid region, 6 (9.4%) on the Posterior auricular region 1 (0.5) preauricular-cheek region, 3 (4.7%) on the ear, 2 (3.1%) in Forehead-temple-eyebrow-eyelid region 16 (7.8) the oral mucosa, 2 (3.1%) on the neck, and 1 each (1.6%) on Mucosal, conjunctiva 1 (0.5) the posterior auricular region and conjunctival mucosa. Cel- Preauricular-cheek region 6 (2.9) lular blue nevi were located on the scalp, upper back, and Ear 3 (1.5) dorsal hand. Mucosal, oral 2 (1.0) Clinical and histologic features are summarized in Table 1 Neck 2 (1.0) and representative histopathologic images are shown in the Arm 58 (28.4) Figure. Hand 15 (7.4) The estimated duration of the BN prior to biopsy diagnosis was based on patient-reported onset of the lesion. Clinical Back 35 (17.2) duration was available for 65 BN in 62 patients, of whom 56 Chest 7 (3.4) (who had a mean [SD] age of 57.8 [15.9] years) reported acquir- Trunk 5 (2.5) ing the nevus 10 years or less prior to diagnostic biopsy. Nine Leg 11 (5.4) patients reported the lesion as being of “new” onset. Only 10 Foot 9 (4.4) patients reported a recent change in size or symptoms of the

Figure. Representative Hematoxylin-Eosin–Stained Histopathologic Features

A, Blue nevus, sclerotic type with dendritic melanocytes interspersed in dense collagen. B, Combined nevus with nevocellular intradermal nevus and dendritic melanocytes (original magnification ×10 for both).

874 JAMA Dermatology August 2014 Volume 150, Number 8 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/28/2021 Acquired Blue Nevi in Older Individuals Case Report/Case Series Research

Table 2. Clinical and Histopathological Characteristics of 18 Melanoma in 16 Patients Diagnosed as Having Blue Nevi

Patient No./ Blue Nevi Melanoma Breslow Depth, Timing of Melanoma Sex/Age, y Location Location mm Diagnosis, y 1/M/76 Trunk Nose 3.70 After, 10 2/M/50 Oral mucosa Back 0.79 After, 8 3/M/74 Right mid back Upper back 0.76 Before, 2 4/M/49 Left shoulder Left arm 1.90 Before, 3 5/M/75 Chest Trunk 0.30 Before, 0.5 6/M/76 Left arm Right forearm 0.25 Before, 3 7/M/74 Leg Chest MIS Same time 8/M/65 Temple Chest 0.60 Same time 9/M/79 Left side of back Left lower back 0.95 Before, 1 10/M/82 Back Arm MIS Same time 11/M/63 Mid back 1. Left lower back 1. MIS Before, 0.5 for both 2. Right cheek 2. 0.70 12/M/63 Scalp Back 0.45 Before, 9 13/M/79 Chest Arm MIS Before, 3 14/M/89 Back Neck MIS Before, 2 15/M/76 Left side of back 1. Left upper back 1. MIS Before, 1 for both 2. Right shoulder 2. 0.85 Abbreviation: MMIS, malignant 16/M/66 Left posterior arm Left deltoid MIS Before, 13 melanoma in situ.

lesion prior to biopsy, including 3 from the cohort with re- play a role in acquired BN based on this sex and age ported clinical duration of 10 years or less, 2 from the cohort predilection.6 In our cohort of predominantly older men (mean of duration greater than 10 years, and 5 of unspecified duration. age, >60 years), we identified that BN were commonly re- Eighteen primary melanomas were identified in 16 pa- ported to develop later in life in the subset of patients in whom tients with BN, 7 of 18 (38.9%) of which were melanoma in situ estimated duration of the lesion was possible. and 11 of 18 (61.1%) were thin invasive melanoma, with a mean Similar to prior reports, 45.6% of the BN cases in our co- (SD) Breslow thickness of 1.02 (0.99) mm. Most melanomas (13 hort were located on the extremities (including the dorsal of 18 [72.2%]) were diagnosed prior to BN biopsy (mean, 3.0 hands and feet) and 16.2% on the scalp, although BN were com- years), and 3 of 18 (16.7%) were biopsied at the same time. Mela- monly noted in other regions of the head and neck (31.3%) and nomas were mainly located on the back (6 of 18 [33.3%]) and on the back (17.2%). Notably, there were 2 cases of BN arising arms (6 of 18 [33.3%]), followed by the chest (2 of 18 [11.1%]), in the oral mucosa and 1 on the conjunctival mucosa, which and neck, nose, preauricular-cheek region, and trunk, 1 of 18 have rarely been reported in the literature.4 for each (5.6% per site). These results are summarized in Furthermore, in the subset of patients in whom clinical du- Table 2. ration of the lesion was reported, 90.3% noted onset in later adulthood, suggesting a potentially different mechanism of nevogenesis compared with common acquired nevi, most of 5 Discussion which develop prior to age 40 years. Our findings support the concept that BN are not uncommon in older individuals and, To our knowledge, this study represents the largest BN co- combined with lack of malignant BN in our cohort, suggest that hort described according to histopathologic subtype, patient most banal-appearing BN in older individuals can be safely fol- characteristics, estimated duration of lesion and relation to ma- lowed with clinical observation rather than biopsied to ex- lignant BN as well as to prior, concomitant, or subsequent cu- clude melanoma. However, this recommendation is contin- taneous melanoma. Our cohort was composed of predomi- gent on clinical stability of the lesion with no patient report nantly older white men, typical of the VAPAHCS patient of recent change as well as dermoscopic examination of the demographics. Notably, it highlights the occurrence of be- lesion, which can aid in the diagnosis of BN. nign BN in an older-age demographic markedly different from Blue nevi remain an important entity in the differential the younger, female patient population previously described diagnosis of melanoma, especially in older patient popula- in the literature. tions at high risk for skin cancer, such as patients in the VA Common BN are reported to arise most frequently in Healthcare System. In a study assessing VA cancer incidence women and younger adults (mean age, 38.6 years).4 They oc- using 2007 VA Central Cancer Registry data, melanoma was cur most often on the scalp, as well as the dorsal surfaces of the fifth most common cancer, accounting for 3.4% of hands and feet.3 Cellular BN may be congenital or acquired and cases.7 Melanomas that arise in association with BN, typi- have a higher incidence in females (2.2:1) and in adults younger cally referred to as “malignant BN” or “blue nevus–like than 40 years, with usual location on the buttocks or scalp.5 melanoma,” are most commonly associated with the cellu- Some authors have hypothesized that female hormones may lar type and have only been rarely reported to arise from

jamadermatology.com JAMA Dermatology August 2014 Volume 150, Number 8 875

Downloaded From: https://jamanetwork.com/ on 09/28/2021 Research Case Report/Case Series Acquired Blue Nevi in Older Individuals

common BN.8,9 Although malignant BN have been charac- the relatively homogeneous nature of our VAPAHCS study terized as more aggressive in nature than conventional cohort. However, while the older age distribution does not melanoma,10 a recent case series of 23 patients suggests that provide further insight into the incidence of BN in a younger clinical outcomes were similar when matched with common demographic, it does serve to highlight the apparent devel- melanoma subtypes for Breslow thickness and ulceration.11 opment of benign BN in individuals older than 50 years. In a study by Van Raamsdonk et al,12 83% of BN were found Second, age of onset and/or duration of BN were deter- to harbor activating somatic mutations in the GNAQ gene mined by patient recollection of such, with potential recall that encodes the ras-like domain. Blue nevi do not seem to bias as patients may have had difficulty remembering when harbor oncogenic mutations that are commonly found in their BN first developed. Incomplete documentation of other nevi or melanoma, including BRAF,13 NRAS,14 or patient report regarding BN duration in 68.0% of our cases c-kit.15 In our cohort with a mean of 9.0 years clinical follow- also limits firm conclusions on identifying the precise age at up, there were no reports of melanoma arising from com- onset of BN. mon, combined (including with dysplastic features), or cellular BN. However, as with other melanocytic lesions, BN should be biopsied in cases with rapid growth or atypical Conclusions clinical and/or dermoscopic features to exclude malignant transformation. This large retrospective case series of predominantly older men In our BN cohort, 8.2% of patients were also diagnosed as at high risk for skin cancer suggests that BN may occur later having in situ or thin cutaneous melanoma, an incidence rate in life than previously reported and that they do not seem to which is higher than that reported by the VA Central Cancer have increased malignant potential based on age, male sex, and Registry but likely reflects the increased skin cancer risk of pa- prior or subsequent cutaneous melanoma development. Larger, tients followed by dermatology. In addition, most melano- population-based studies are necessary to further character- mas (72.2%) occurred before the BN diagnosis, which limits our ize the incidence and duration of BN in older individuals as well ability to draw conclusions regarding the association of BN with as their association with malignant BN and subsequent devel- subsequent melanoma development, particularly in a popu- opment of cutaneous melanoma. In conjunction with recent lation at higher risk based on age, increased lifetime sun ex- genetic studies demonstrating different mutations in BN com- posure, and fair skin phenotype. However, our findings rein- pared with other nevus subtypes and melanoma, our find- force the concept that BN represent a distinct subtype of ings support the concept that benign BN are not uncom- melanocytic neoplasms with rare malignant potential. monly acquired in older age individuals and that they may be Our study has several limitations. First, it is difficult to safely clinically followed in the absence of concerning clini- generalize our findings to a broader population because of cal and/or dermoscopic features.

ARTICLE INFORMATION 2. Phadke PA, Zembowicz A. Blue nevi and related patient with subacute cutaneous lupus Accepted for Publication: August 6, 2013. tumors. Clin Lab Med. 2011;31(2):345-358. erythematosus. Dermatologica. 1989;178(3):171-175. Published Online: April 30, 2014. 3. Zembowicz A, Phadke PA. Blue nevi and 10. Goldenhersh MA, Savin RC, Barnhill RL, Stenn doi:10.1001/jamadermatol.2013.7366. variants: an update. Arch Pathol Lab Med. 2011;135 KS. Malignant blue nevus. Case report and literature (3):327-336. review. J Am Acad Dermatol. 1988;19(4):712-722. Author Contributions: Dr Cabral had full access to all of the data in the study and takes responsibility 4. Murali R, McCarthy SW, Scolyer RA. Blue nevi 11. Martin RC, Murali R, Scolyer RA, Fitzgerald P, for the integrity of the data and the accuracy of the and related lesions: a review highlighting atypical Colman MH, Thompson JF. So-called “malignant data analysis. and newly described variants, distinguishing blue nevus”: a clinicopathologic study of 23 Study concept and design: Cabral, Chen, Swetter. features and diagnostic pitfalls. Adv Anat Pathol. patients. Cancer. 2009;115(13):2949-2955. Acquisition, analysis, or interpretation of data: All 2009;16(6):365-382. 12. Van Raamsdonk CD, Bezrookove V, Green G, authors. 5. Rodriguez HA, Ackerman LV. Cellular blue nevus: et al. Frequent somatic mutations of GNAQ in uveal Drafting of the manuscript: Cabral, Chen, Swetter. clinicopathologic study of forty-five cases. Cancer. melanoma and blue naevi. Nature. 2009;457 Critical revision of the manuscript for important 1968;21(3):393-405. (7229):599-602. intellectual content: All authors. 6. Rubin AI, Laborde SV, Stiller MJ. Acquired 13. Yazdi AS, Palmedo G, Flaig MJ, et al. Mutations Administrative, technical, or material support: dermal melanocytosis: appearance during of the BRAF gene in benign and malignant Cabral. pregnancy. J Am Acad Dermatol. 2001;45(4): melanocytic lesions. J Invest Dermatol. 2003;121(5): Study supervision: Swetter. 609-613. 1160-1162. Conflict of Interest Disclosures: None reported. 7. Zullig LL, Jackson GL, Dorn RA, et al. Cancer 14. Saldanha G, Purnell D, Fletcher A, Potter L, Previous Presentation: This study was presented incidence among patients of the US Veterans Affairs Gillies A, Pringle JH. High BRAF mutation frequency in part as a poster at the American Society for Health Care System. Mil Med. 2012;177(6):693-701. does not characterize all melanocytic tumor types. Dermatologic Surgery Annual Meeting; October 4, 8. Granter SR, McKee PH, Calonje E, Mihm MC Jr, Int J Cancer. 2004;111(5):705-710. 2012; Atlanta, Georgia. Busam K. Melanoma associated with blue nevus 15. Ohashi A, Funasaka Y, Ueda M, Ichihashi M. and melanoma mimicking cellular blue nevus: c-KIT receptor expression in cutaneous malignant REFERENCES a clinicopathologic study of 10 cases on the melanoma and benign melanotic naevi. Melanoma 1. Bogart MM, Bivens MM, Patterson JW, Russell spectrum of so-called “malignant blue nevus.” Am J Res. 1996;6(1):25-30. MA. Blue nevi: a case report and review of the Surg Pathol. 2001;25(3):316-323. literature. Cutis. 2007;80(1):42-44. 9. Modly C, Wood C, Horn T. Metastatic malignant melanoma arising from a common blue nevus in a

876 JAMA Dermatology August 2014 Volume 150, Number 8 jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/28/2021