Authors' reply to: High naevus counts confer a favourable prognosis in patients with melanoma

Abstract

Dear Editor, We welcome the comments by Autier et al. These correspondents have drawn attention to the reported different outcomes for cancers detected by screening.In our study, 1 we reported evidence for better survival in patients with melanoma with many naevi, suggesting that this might reflect a different biology in this subset of melanomas. Autier et al. have suggested that an alternative explanation might be that the better survival observed in patients with many naevi may in fact result from a selection bias due to these tumours being discovered by screening.The intent of screening is indeed to detect cancers earlier and therefore by definition with a better prognosis. The issue here is whether the patients with higher naevus counts recruited to our studies were more likely to be diagnosed early as a result of screening. There are no formal screening programmes for melanoma in either Spain or the United Kingdom. The UK series were unselected patients with melanoma from a defined geographical area. Very few were identified in follow-up, although this is not quantified. All the Spanish patients included in our study underwent sentinel lymph node biopsy. Sentinel lymph node biopsy is only performed for tumours of Stage IB or higher. There is published data which suggest that melanomas diagnosed in follow-up screening were unlikely to have been offered sentinel node biopsy and are therefore unlikely to have been included in our study. 1 In the Barcelona Melanoma Unit, melanomas diagnosed in patients undergoing screening were reported to be all Stage 0 or IA. 2 When using total body photography and digital dermoscopy, none of the 98 melanomas diagnosed during screening required sentinel lymph node biopsy in a mean follow-up of 98 months. 3,4 We suggest that it is therefore unlikely that the observations reported in our study reflect a screening artefact and more likely to represent a true biological effect.Sentinel node biopsy is a staging tool for patients with thicker primaries, and therefore, the Spanish studies describe participants selected for thicker tumours. The mean thickness of cases included in the model in Table 3 was 2.45 mm. The cases included in the model in Table 4 had a mean Breslow thickness of 3.42 mm. In particular, mean Breslow thickness in the melanoma cases included in the model (Table 3) with >50 naevi was 2.07 mm. The effect of increased numbers of naevi on improved survival remained in the positive sentinel group (Fig. 2). We would argue that this provides evidence to support our hypothesis that the protective effect on survival was not related to selection bias.The multivariate models have been performed with Breslow thickness as both linear and categorical variables. In both models, the protective effects of naevus count on melanoma mortality were maintained. The multivariate models had been performed with Breslow thickness either as a linear or categorical variable. In both models, the protective effects of nevus count on melanoma mortality were maintained. We performed a log like...

Adjuvant Treatment of Melanoma: Recent Developments and Future Perspectives

“…Clinically age and phenotype (like mole count) has been demonstrated to play a role in survival in positive SLN patients 70 . Pathologic features of primary like Breslow thickness, ulceration, mitoses and regression have been associated to prognosis as well as the number of lymph node excised and the number of positive ones 5,6,71 .…”