Supplement: Abstracts from the SCCANZ Skin Cancer Conference, June 3–5, 2011, Gold Coast, Australia

Introduction: Serial dermatoscopic monitoring of pigmented skin lesions may introduce a delay in diagnosis of melanoma. Conversely it is argued that feature-poor melanomas may not otherwise be diagnosed without a large increase in biopsy rates. This study compared average Breslow thickness of melanomas that have been serially monitored with those that have not. Method: A retrospective analysis of a clinical audit of all histopathologically confirmed melanomas detected over a two-year period in a primary care skin cancer clinic was performed. Data collected included patient age and sex, date of first monitoring, date of decision to biopsy, Breslow thickness, and histopathologic diagnosis. Lesions judged to be dermatoscopically borderline for melanoma were serially imaged using a MoleMax I (Derma Instruments, Austria). Monitored lesions were reviewed at three months and then on subsequent patient review throughout the normal course of their care. Incidences of in-situ lesions and averages of Breslow thickness of monitored and unmonitored lesions were calculated and compared. Results: A total of 90 melanomas were detected. 26% were invasive, with an average Breslow thickness of 0.9 millimetres. 33 melanomas had been monitored and 27% of these were invasive with an average Breslow thickness of 0.33 millimetres. 57 melanomas were not monitored and 26% of these were invasive with an average Breslow thickness of 1.2 millimetres. There was no significant difference in the proportion of in-situ lesions detected. Monitored lesions which were invasive were significantly thinner (p <0.04; z=-1.743; Two Independent Sample Wilcoxon Rank Sum Test). Only 7/33 (21%) of monitored melanomas showed change at three months. Conclusion: In a small single centre study, the use of monitoring to detect melanomas resulted in a similar proportion of in-situ melanomas and did not result in an increase in Breslow thickness of invasive melanomas at time of excision, compared with unmonitored lesions. This may be because borderline lesions are detected earlier in their evolution, or may be inherently slower growing. A monitoring interval of 3 months may not be sufficient to detect change in borderline lesions. A further larger study is required to confirm these findings.