Lesion Selection by Melanoma High-Risk Consumers During Skin Self-examination Using Mobile Teledermoscopy

Mobile teledermatoscopy (MTD) for the early detection of skin cancer uses smartphones with dermatoscope attachments to magnify, capture, and transfer images remotely.¹ Using the asymmetry–color variation (AC) rule, consumers achieve dermoscopy sensitivity of 92.9% to 94.0% and specificity of 62.0% to 64.2% for melanoma.²

This pilot randomized trial assessed lesions of concern selected by consumers at high risk of melanoma using MTD plus the AC rule (intervention, n = 10) or the AC rule alone (control, n = 12) during skin self-examination (SSE). Also measured were lesion location patterns, lesions overlooked by participants, provisional clinical diagnoses, likelihood of malignant tumor, and participant pressure to excise lesions.

Ethics approval, informed consent, and intervention group (n = 10) characteristics were described previously.¹ All participants were provided with an AC rule fact sheet and standardized SSE instructions. Participants underwent clinical skin examinations (CSEs) 3 to 6 months after SSEs to assess lesions of concern found during SSE and additional lesions potentially overlooked.

Participants’ characteristics were similar in the intervention and control groups: overall, 60% male; working full time, 68%; personal history of melanoma, 73%; and body areas with moles, 59%. During SSE, 107 lesions were identified (66 in the intervention group and 41 in the control group; Figure, A), with patterns of body lesion locations similar for both groups. Figure, B and C, compares lesions identified during SSE and provisional clinical diagnosis during CSE. Likelihood of malignant tumor and pressure by participants to excise lesions during CSE are listed in the Table. Forty-two additional lesions not pointed out by participants were noted during CSE (20 in the intervention group and 22 in the control group), including 1 clinically presenting as melanoma (dysplastic nevus), 2 basal cell carcinomas (1 confirmed in the intervention group and 1 resolved before surgery in the control group), and 1 squamous cell carcinoma (confirmed in the intervention group) (Table). On average, participants’ SSE in both groups missed 2 lesions (intervention median [SD], 2 [1.43]; control median [SD], 2.09 [0.93]).

Consumer-selected lesions were unlikely to be malignant, although more than one-third were dysplastic nevi. During CSE, the dermatologist detected other higher-priority skin lesions. These lesions were in hard-to-see body areas and might have been missed during SSE.

Participants in both groups selected lesion locations that reflect the SSE primary body areas (arms, face, and front of legs) reported by Mujumdar et al.³ Both groups also selected lesions on the back, shoulders, and legs, reflecting findings by Carli et al.⁴ Our participants did not select lesions in sexually sensitive⁵ or harder-to-see areas.

Previously, Boone et al⁵ found a lower proportion of missed lesions in partner-assisted compared with unassisted SSEs. We instructed participants to select 3 to 5 lesions during SSE, which may have contributed to participants missing lesions and explain some discrepancies between participant and dermatologist assessment.

Future studies need to instruct participants to also submit location photographs of lesions to aid re-identification during CSE. Consumers with many moles, such as participants in this study, may find it difficult to discriminate lesions of concern. Such individuals benefit from regularly scheduled CSEs; however, between visits they could use MTD to follow specific lesions designated by the dermatologist. More research is needed on the interface of MTD with cognitive processes to select lesions of concern. Factual knowledge is gained from the AC rule, but knowledge alone may not allow discrimination between benign and malignant skin lesions.⁶

Future studies of MTD could benefit from targeting partner-assisted SSEs, increasing the number of SSEs to generate more lesions submitted for telediagnosis, assessing the effect of dermatologists’ feedback between SSE rounds, and submitting lesion location photographs. The process of lesion selection decision making using MTD or other lesion selection aids merits further investigation.

Corresponding Author: Monika Janda, PhD, Queensland University of Technology, Victoria Park Rd, Kelvin Grove, QLD 4061, Brisbane, Australia (m.janda@qut.edu.au).

Accepted for Publication: August 22, 2013.

Published Online: February 12, 2014. doi:10.1001/jamadermatol.2013.7743.

Author Contributions: Drs Janda and Soyer had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Janda, Loescher, Soyer.

Acquisition of data: Janda, Banan, Soyer, Horsham.

Analysis and interpretation of data: Janda, Loescher, Horsham, Soyer.

Drafting of the manuscript: Janda, Loescher, Horsham.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Janda.

Obtained funding: Janda, Soyer.

Administrative, technical, or material support: Janda, Banan, Horsham.

Study supervision: Janda, Soyer.

Conflict of Interest Disclosures: Dr Soyer is a cofounder and shareholder of e-derm-consult GmbH and is a shareholder and reports for MoleMap by Dermatologists Pty Ltd. No other disclosures were reported.

Funding/Support: This study was supported in part by Queensland University of Technology Institute of Health and Biomedical Innovation and National Health and Medical Research Council of Australia Career Development Fellowship 1045247 (Dr Janda). Dr Soyer holds a National Health and Medical Research Council of Australia Practitioner Fellowship 1020145.

Role of the Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.