Incidence and characteristics of thick second primary melanomas: a study of the German Central Malignant Melanoma Registry

Transcription

1 DOI: /jdv JEADV ORIGINAL ARTICLE Incidence and characteristics of thick second primary melanomas: a study of the German Central Malignant Melanoma Registry M. Gassenmaier, 1, * T. Stec, 2 U. Keim, 2 U. Leiter, 2 T.K. Eigentler, 2 G. Metzler, 1 C. Garbe 2, * 1 Department of Dermatology, Eberhard-Karls-University of Tuebingen, Tuebingen, Germany 2 Department of Dermatology, Center for Dermatooncology, Eberhard-Karls-University of Tuebingen, Tuebingen, Germany *Correspondence: M. Gassenmaier and C. Garbe. s: maximilian.gassenmaier@med.uni-tuebingen.de and claus.garbe@med.uni-tuebingen.de Linked article: This article is commented on by K. Peris, pp in this issue. To view this article visit Abstract Background Fast-growing melanomas are thought to be responsible for the stable incidence of thick melanomas. It has been suggested that campaigns for early diagnosis are unlikely to have a major impact on prognosis as rapid vertical growth rather than diagnostic delay is the major determinant for thick melanomas. Objective We investigated the impact of follow-up examinations on the incidence of thick second primary melanomas (SPMs) and analysed their clinic-pathologic characteristics. Methods We analysed a single-centre cohort of 2253 patients of the German Central Malignant Melanoma Registry with prospectively documented follow-up examinations. Results Primary tumour and patient characteristics were well balanced between patients with and without SPMs except for age (median 61 years, interquartile range [IQR] vs. 56 years, IQR 43 67; P = 0.005). Metachronous SPMs occurred in 107 patients (4.7% of total) were thinner than the respective first primary melanoma (FPM) (median Breslow thickness of invasive melanomas 0.40 mm, IQR vs mm, IQR ; P < 0.001) and less often ulcerated (0.9% vs. 15.0%; P < 0.001). Melanomas >2.00 mm occurred in 2.8% of SPMs as compared to 23.4% of FPMs (P < 0.001). Thick SPMs (>1.00 mm; 14.0%) despite close-meshed follow-up examinations were frequently associated with atypical clinical presentation and uncommon histopathologic subtypes. One-third (5/15) of thick SPMs were clinically misdiagnosed as non-melanocytic lesions, most of them as basal cell carcinomas (n = 4). Conclusions Regular total body skin examinations enable a highly efficient detection of early-stage melanomas and reduction of thick melanomas as compared to first primary melanomas. Our data indicate that fast-growing melanomas without opportunity of early detection are rare and cannot explain the stable incidence of thick melanomas. This highlights the importance of close-meshed total body skin examinations in patient groups that are at high risk of first or multiple primary melanomas. Received: 16 February 2018; Accepted: 12 July 2018 Conflicts of interest The authors state no conflict of interest. Funding sources German Central Malignant Melanoma Registry. Introduction Cutaneous melanomas evolve through clinically and histopathologically discernible progression steps. Transformed melanocytes spread first horizontally above the epidermal basement membrane (radial growth phase), then invade the papillary dermis and eventually extend vertically into the deeper dermis (vertical growth phase). 1,2 These steps can evolve with different growth rates and different biological aggressiveness. In the last years, patient interview-based studies have identified fast-growing melanomas with rapid vertical growth and poor prognosis. 3 5 Moreover, fastgrowing melanomas with limited opportunities of early detection are thought to be responsible for the stable incidence of thick

2 64 Gassenmaier et al. melanomas, although the median melanoma thickness has decreased over the last decades. 6 9 To date, it remains unclear whether fast vertical growth ab initio or diagnostic delay is the major determinant for thick melanomas. Clarification of this issue is crucial for the development of effective intervention strategies in terms of secondary prevention of thick melanomas. In this study, we analysed a cohort of 2253 melanoma patients with prospectively documented follow-up examinations and sought to estimate the impact of follow-up examinations on the incidence of thick second primary melanomas (SPMs). We hypothesized that this setting allows particularly for early detection of melanomas with slow or intermediate growth rate but not for thick melanomas with rapid vertical growth from the outset. 5,6,9 Materials and methods Study population Between 1 January 2002 and 31 December 2006, 2500 patients of the Department of Dermatology of the University Hospital Tuebingen were diagnosed with primary malignant melanoma and recorded in the nationwide German Central Malignant Melanoma Registry after obtaining informed written consent. The Department of Dermatology Tuebingen performs melanoma follow-up cooperatively with local dermatologists, and it has been agreed that any patient who is diagnosed with an invasive melanoma in Tuebingen should be referred to the University Hospital Tuebingen. At the initial consultation at the Department of Dermatology, patients were extensively educated regarding the clinical characteristics of melanoma and its metastases, emphasizing the role of self-examination and the recognition of signs of recurrence. Each follow-up appointment consisted of a complete history, total body skin examination, palpation of the scar of primary resection, in-transit route and draining lymph nodes. Intervals of the physical examinations, imaging studies and blood testing were performed according to the guidelines of the German Society of Dermatology which remained rather constant over time (Table S1 S3). When patients were recorded in the German Malignant Melanoma Registry, all melanomas were reviewed by experienced dermatopathologists of the Department of Dermatology. Exceptions only applied for histopathology reports from selected and experienced external dermatopathologists. In general, follow-up examinations were alternated every other examination between the Department of Dermatology Tuebingen and dermatology practices. However, stage IA patients were often seen exclusively by local dermatologists and stage IV patients exclusively in the Department of Dermatology Tuebingen. If follow-up was performed at patients local dermatologists, doctors were provided with follow-up sheets for documentation of their findings (total body skin examination, ultrasound scans of regional lymph nodes, tumour marker S100b, histopathologic report of excisions). Results of follow-up examinations were prospectively documented, and this study was approved by the ethics board of the University Hospital Tuebingen. After definition of further inclusion criteria (follow-up time of at least 3 months, invasive primary cutaneous melanoma with known Breslow thickness), a total of 2254 patients and 150 second primary melanomas were identified. We did not include noninvasive first primary melanomas in our analysis as followup was not routinely performed in these patients. However, noninvasive SPMs (Clark level I) were included into the analysis. Histopathologic reports and computerized medical records of all patients with SPMs (n = 150) were manually reviewed, and four patients were identified with misdiagnosis of SPMs and violation of the inclusion criteria, respectively: The first primary melanoma (FPM) of one patient was diagnosed before 1 January One patient had a local recurrence of the primary melanoma, one patient received two surgeries for complete resection of the primary tumour, and one patient had a local recurrence of a previously incompletely excised cutaneous metastasis. In total, 146 (6.5%) of 2253 patients developed SPMs until 25 April 2013, 365 patients died and 1022 (44.4 %) were lost to followup during their 10-year follow-up period. Second primary melanomas that were diagnosed synchronously with the FPM or within the first 30 days after the FPM were referred to as synchronous SPMs and all other SPMs were referred to as metachronous SPMs. In patients with synchronous second and third primary melanomas, the thicker tumour was considered the SPM. In one patient, whose melanoma was initially biopsied and excised completely 11 months later, the melanoma growth rate was estimated by measuring the increase in Breslow thickness between the two surgical procedures. Statistics Statistical calculations were performed with IBM SPSS Statistics version 23.0 (IBM SPSS, Chicago, IL, USA). Skewed numerical variables were described as median value and interquartile range (IQR), and distribution of tumour and patient characteristics were assessed with Pearson s chi-square test, Fisher s exact test, McNemar test, McNemar Bowker test, Sign test, related-samples Wilcoxon signed rank test and Mann Whitney U-test depending on scaling of the variables and paired or unpaired groups. Patients with unknown variables were excluded from statistical analyses. Throughout the analysis, P-values <0.05 were considered as statistically significant. Results During follow-up (median 73 months, IQR 41 95) of the study population (n = 2253), a total of 146 patients (6.5%) developed SPMs. Almost half of all SPMs (47.9%, n = 70) were detected in the first year of follow-up and 26.7% synchronously in the first 30 days (Fig. 1). Of 146 patients with SPMs, 21 (14.4%)

3 Secondary prevention of thick melanomas 65 Number of patients Metachronous melanomas Synchronous melanomas Years after primary melanoma diagnosis Figure 1 Occurrence of second primary melanomas during follow-up. Distribution of synchronous and metachronous second primary melanomas during follow-up. developed three primary melanomas, five patients (3.4%) developed four primary melanomas and two patients (1.4%) developed five primary melanomas. Patients with SPMs were older at excision of their first primary melanoma (FPM) as compared to patients without SPMs (61 years, IQR vs. 56 years, IQR 43 67; P = 0.005) but primary tumour characteristics did not differ between both groups (Table 1). Moreover, men (60 years, IQR 48 68) were older at FPM diagnosis than women (52 years, IQR 40 66; P < 0.001). Patients with synchronous SPMs (diagnosis within 30 days after the FPM) were excluded from all further analyses in this study as these melanomas were most likely already present when the FPM was diagnosed. Comparison of first and second primary melanomas Table 2 compares first and second primary melanomas of patients with metachronous second primary melanomas. FPMs were more often ulcerated (15.0% vs. 0.9%; P < 0.001) and thicker than SPMs (23.4% >2.00 mm vs. 2.8% >2.00 mm; P < 0.001). Invasive FPMs were twice as thick as invasive FPMs (0.80 mm, IQR vs mm, IQR ; P < 0.001), and nearly one-third (29.9%) of SPMs were in situ melanomas. Superficial spreading melanoma (SSM) was the most common subtype with a higher proportion in SPMs (79.4% vs. 64.5%) and none of the SPMs was of nodular subtype as compared to 16.8% of FPMs (P < 0.05). Overall, first and second primary melanomas were similarly distributed on the body (P = 0.67) and the median time till occurrence of metachronous SPMs was 31 months (IQR 10 60) after the FPM. Characterization and comparison of thin and thick SPMs Table 3 shows the associations between thick (>1.00 mm) and thin SPMs ( 1.0 mm) with clinic-pathologic characteristics. Thick SPMs (median 1.50 mm; min 1.02, max 4.00 mm) comprised 14.0% (n = 15) of all metachronous SPMs and were frequently associated with rare histopathologic subtypes (others than SSM and lentigo maligna melanoma (LMM) 33.3% vs. Table 1 Patient and primary melanoma characteristics of patients with single or second primary melanoma. N = 2253 Patients with single primary melanoma (n = 2107) Patients with second primary melanoma (n = 146) Gender 0.143* Female 1070 (50.8%) 65 (44.5%) Male 1037 (49.2%) 81 (55.5%) Localization 0.130* Head and neck 352 (16.7%) 17 (11.6%) Trunk 833 (39.5%) 71 (48.6%) Upper extremities 323 (15.3%) 22 (15.1%) Lower extremities 599 (28.4%) 36 (24.7%) Tumour thickness Median 0.9 (IQR ) 0.9 (IQR ) 0.75** 1.00 mm 1160 (55.1%) 79 (54.1%) mm 471 (22.4%) 35 (24.0%) mm 318 (15.1%) 21 (14.4%) >4.00 mm 158 (7.5%) 11 (6.5%) Clark level 0.47* I II 467 (22.2%) 34 (23.3%) III 550 (26.1%) 36 (24.7%) IV 840 (39.9%) 61 (41.8%) V 58 (2.8%) 1 (0.7%) Unknown 189 (8.9%) 15 (9.6%) Histopathologic subtype 0.072* SSM 1370 (65.0%) 100 (68.5%) NM 225 (10.7%) 23 (15.8%) LMM 208 (9.9%) 13 (8.9%) ALM 96 (4.6%) 5 (3.4%) Others 122 (5.8%) 2 (1.4%) Unknown 86 (4.1%) 3 (2.1%) Ulceration 0.77* Yes 327 (15.5%) 24 (16.4%) No 1780 (84.5%) 122 (83.6%) Age at primary melanoma diagnosis Median (years) 56 (IQR 43 67) 61 (IQR 51 67) 0.005** >60 years 874 (41.5%) 75 (51.4%) 60 years 1233 (58.5%) 71 (48.6%) P-values (P) refer to Pearson s chi-square test* and Mann Whitney U-test**, respectively. ALM, acral lentiginous melanoma; IQR, interquartile range; LMM, lentigo maligna melanoma; NM, nodular melanoma; SSM, superficial spreading melanoma. 4.3%; P = 0.001). There was a trend that thick SPMs were more often ulcerated than thin SPMs, occurred more often in men and the head and neck area, were diagnosed somewhat later after FPM diagnosis as compared to thin SPMs, had a longer latency to the last follow-up examination and a higher rate of irregular follow-up. However, these differences did not reach statistical significance. Moreover, Breslow thickness of the respective FPMs was similar in both groups (median 0.80 mm). P

4 66 Gassenmaier et al. Table 2 Comparison of first and second primary melanoma characteristics of patients with second primary melanomas. N = 107 First primary melanoma (n = 107) Second primary melanoma (n = 107) Gender Female 49 (45.8%) Male 58 (54.2%) Localization 0.67* Head and neck 14 (13.1%) 13 (12.1%) Trunk 52 (48.6%) 51 (47.7%) Upper extremities 20 (18.7%) 14 (13.1%) Lower extremities 21 (19.6%) 29 (27.1%) Tumour thickness Median (mm; invasive melanomas) 0.80 (IQR ) 0.40 (IQR ) Clinic-pathologic characteristics of thick SPMs Follow-up records of all 15 patients with thick SPMs were reviewed to analyse their clinic-pathologic characteristics (Table 4). Overall, six SPMs were clinically misdiagnosed as non-melanocytic or benign lesions due to their atypical clinical appearance, four of them as basal cell carcinomas. Three SPMs had uncommon histopathologic subtypes (desmoplastic, nevoid and dermal melanoma), and another three SPMs were amelanotic and minimally pigmented, respectively. Fast-growing melanomas as defined by melanomas with vertical growth rates of at least 0.5 mm/month could be definitely excluded in two P <0.001** In situ 32 (29.9%) <0.001*** 1.00 mm 58 (54.2%) 60 (56.1%) mm 24 (22.4%) 12 (11.2%) mm 17 (15.9%) 3 (2.8%) >4.00 mm 8 (7.5%) Clark level <0.001*** I 32 (29.9%) II 27 (25.2%) 35 (32.7%) III 26 (24.3%) 17 (15.9%) IV 42 (39.3%) 20 (18.7%) V 1 (0.9%) Unknown 11 (10.3%) 3 (2.8%) Histopathologic subtype <0.05* SSM 69 (64.5%) 85 (79.4%) NM 18 (16.8%) LMM 11 (10.3%) 8 (7.5%) Others 6 (5.6%) 9 (8.4%) Unknown 3 (2.8%) 5 (4.7%) Ulceration <0.001**** Yes 15 (15.0%) 1 (0.9%) No 91 (85.0%) 106 (99.1%) P-values (P) refer to McNemar Bowker test*, related-samples Wilcoxon signed rank test**, Sign test*** and McNemar test****, respectively. IQR, interquartile range; LMM, lentigo maligna melanoma; NM, nodular melanoma; SSM, superficial spreading melanoma. patients: tumour thickness of patient 4 increased 1.14 mm in 11 months and patient 7 was diagnosed with a 1.50 mm amelanotic secondary nodular SSM that has been clinically misdiagnosed as basal cell carcinoma 3.5 months prior to excision. The thickest SPM was diagnosed in a patient with locoregional advanced disease 38 months after the last documented followup examination. Discussion It has been suggested that the intrinsic growth rate rather than diagnostic delay is the major determinant for thick melanomas and that campaigns for early diagnosis are unlikely to have a major impact on prognosis. 10 This hypothesis is challenged by the present study which demonstrates that aggressive vertical growth is not a hallmark of early-stage melanomas and that regular total body skin examinations, as recommended by the German Society of Dermatology, can effectively prevent the occurrence of thick SPMs. Up to 8% of all melanoma patients develop multiple primary melanomas. 11 Primary tumour and patient characteristics in our study were well balanced between patients with and without SPMs except for age. This suggests that patients with SPMs do not develop less aggressive melanomas than patients with single primary melanomas. These findings are in line with earlier reports 11,12 and the work of Hwa et al. 13 who could further demonstrate that the biology of primary tumours, as measured by the presence or absence of mitoses, did not differ between patients with and without SPMs. This observation is important as patients with SPMs are known to have a higher density of common 14 and dysplastic nevi 11 and some studies 15,16 have highlighted that thick and aggressive melanomas develop more often in patients with low nevus count. However, this correlation could not be confirmed by other studies 17 including two studies investigating risk factors for fast growing melanomas. 4,5 Liu et al. 4 investigated characteristics of rapidly growing melanomas and found that higher tumour thickness, older age, fewer freckles and symmetry of the lesion but not mitotic rate, nevus count, gender, amelanosis and histopathologic subtype remained as independent associations in an overall multivariate linear regression model. The low incidence of thick and ulcerated SPMs 11,18 24 and the high rate of SPMs that were diagnosed synchronously or within the first year of follow-up in our study is consistent with the previous reports. 11,25,26 For instance, 21% of FPMs but only 3% of SPMs were thicker than 2.00 mm in a study at the Memorial Sloan-Kettering Cancer Center. 11 These results suggest that many SPMs are probably already (sub)clinically present at excision of the FPM and are early diagnosed during follow-up. Strikingly, none of the SPMs in our study was of nodular subtype. This finding is remarkable as the overall incidence of thick primary melanomas has been stable for decades despite national screening and educational efforts and a fast-growing melanoma subtype, often referred to as nodular melanoma (NM), is

5 Secondary prevention of thick melanomas 67 Table 3 Comparison of thin and thick second primary melanomas. N = 107 Thin SPM ( 1.00 mm; n = 92) Thick SPM (>1.00 mm; n = 15) % of total n % of total n Median time between FPM and SPM (months) 29 (IQR 10 57) (IQR 17 64) * Median time between last follow-up and SPM (months) Regular follow-up 5 (IQR 3 7) 79 7 (IQR 5 8) * Irregular follow-up 38 (IQR 24 53) (IQR 17 38) * Age at second primary melanoma diagnosis Median (years) 64 (IQR 54 70) (IQR 59 71) * >60 years 63.0% % years 37.0% % 5 Gender 0.30** Female 47.8% % 5 Male 52.2% % 10 Localization 0.34*** Head and neck 9.8% % 4 Trunk 48.9% % 6 Upper extremities 13.0% % 2 Lower extremities 28.3% % 3 Histopathologic subtype 0.001*** SSM 82.6% % 9 NM LMM 7.6% 7 6.7% 1 Others 4.3% % 5 Unknown 5.4% 5 Ulceration 0.14*** Yes 6.7% 1 No 100% % 14 Tumour thickness FPM Median (mm) 0.80 (IQR ) 0.80 (IQR ) 0.50* 1.00 mm 54.3% % mm 23.9% % mm 14.1% % 4 >4.00 mm 7.6% 7 6.7% 1 P-values (P) refer to Mann Whitney U-test*, Pearson s chi-square test** and Fisher s exact test***, respectively. FPM, first primary melanoma; IQR, interquartile range; LMM, lentigo maligna melanoma; NM, nodular melanoma; SPM, second primary melanoma; SSM, superficial spreading melanoma. P thought to be responsible for an unchanged melanoma mortality. 4 9 According to Clark, NMs grow vertically from the outset in the absence of a radial growth phase and without evidence of intraepidermal growth beyond the width of three rete ridges on either side. 27,28 However, the current classification of cutaneous melanoma into the subtypes SSM, LMM, NM and ALM has been criticized for several reasons 2,29 : multivariate regression models have failed to reveal differences in prognosis of the four major subtypes if melanoma thickness was taken into account. 30,31 Already in 1980, Ackerman proposed a unifying concept of malignant melanoma as all melanomas would evolve in a similar way: melanomas spread first horizontally and may eventually extend vertically into the dermis. NMs are the same except that they enter the vertical growth phase upon a briefer horizontal growth phase. 2 Considering that NMs are virtually never found to be Clark level II, 27,32 Heenan 33,34 concluded that NM is not a distinct disease entity but rather a common end stage of other histopathologic subtypes. This concept is further corroborated by studies of Bastian who showed distinct patterns of somatic genetic alterations in melanomas on skin with and without chronic sun-induced damage, mucosal and acral melanomas and demonstrated that NMs clustered with these groups but did not have features of an own entity. 35 Moreover, genetic studies have highlighted the necessity of multiple independent mutations in melanoma evolution and have shown that activating mutations of the Mitogen-Activated Protein Kinase pathway

6 68 Gassenmaier et al. Table 4 Identification of possible fast-growing melanomas among thick second primary melanomas. N = 15 Patient Age (years) Histopathologic subtype Breslow thickness (mm) Ulceration Localization Last documented follow-up examination before SPM diagnosis (months) 1 65 SSM 4.00 Yes Scalp 38 SPM with multiple palpable lymph node metastases (max. 2.3 cm in diameter) on both sides of the neck and numerous satellite-metastases covering an area of 15 x 11 cm. No documented regular followup 2 67 Desmoplastic melanoma 3.10 No Upper arm 6 Clinical suspicion of basal cell carcinoma 3 52 Nevoid melanoma 3.00 No Back 5 Clinical suspicion of traumatized nevus 4 71 Amelanotic melanoma 1.67 No Scalp 8 Clinical suspicion of basal cell carcinoma. Biopsy revealed 0.53 mm thick melanoma. Complete excision was performed 11 months later and showed the final tumour thickness of 1.67 mm 5 53 Dermal melanoma 1.60 No Scalp 3 Clinical impression of skin-coloured papule 6 67 SSM 1.50 No Lower leg 7 Melanoma mm in diameter 7 62 Amelanotic secondary nodular SSM 8 59 Secondary nodular nevoid SSM Details 1.50 No Shoulder 8 Clinical suspicion of basal cell carcinoma. Surgery was performed 3.5 months later 1.50 No Shoulder 8 Newly arisen pigmented lesion with small amelanotic part. Clinical suspicion of SPM 9 79 SSM 1.42 No Forearm 17 Melanoma mm in diameter. No documented regular follow-up ALM 1.20 No Toe 7 Mitotic rate: 1 mitosis/mm 2. Located in the fourth interdigital space of the left foot LMM 1.20 No Auricle 38 No documented regular follow-up SSM 1.12 No Thigh 4 Pigmented lesion mm in diameter. Highly suspicious for melanoma in dermoscopy SSM 1.12 No Back 14 Patient refused continuation of follow-up at the Department of Dermatology and missed a scheduled follow-up examination 9 months prior to SPM diagnosis Minimally pigmented SSM 1.10 No Shoulder 14 Flat reddish plaque. Clinical suspicion of basal cell carcinoma SSM 1.02 No Shoulder 1 Dysplastic nevus syndrome. Synchronous diagnosis of second and third (Breslow thickness: 0.26 mm) primary melanoma 36 days after first primary melanoma ALM, acral lentiginous melanoma; LMM, lentigo maligna melanoma; NM, nodular melanoma; SPM, second primary melanoma; SSM, superficial spreading melanoma. and telomerase accumulate during early stages of progression (intermediate lesions and melanoma in situ). Additional mutations affecting the Cyclin Dependent Kinase Inhibitor 2A gene and chromatin remodelling complexes then emerge at the transition to invasive melanomas. 36,37 Earlier published studies indicate that the favourable phenotype of SPMs results predominantly from early detection and not from intrinsic differences of first and second primary melanomas. Murali et al. and Vecchiato et al. compared first and second primary melanomas and found no significant differences regarding the presence of mitoses, 22 mitotic rate, 18,22 tumourinfiltrating lymphocytes, 22 presence of regression, 18,22 satellites 22 and pigmentation. 22 Furthermore, numerous studies have consistently shown that in particular elderly patients, men and patients without physical screening in the last years present with thick and advanced melanomas. 4,5,17,38 So far, the initial growth phase of melanomas has not been elucidated in detail and previous studies investigating the kinetics of melanoma growth were limited by the assumption that melanomas are invasive from the outset and that the rate of

7 Secondary prevention of thick melanomas 69 growth is constant over time. 3 5 Moreover, the calculated growth rate of patient interview-based studies solely depends on patient recall and perception that are highly subjective. 3 A more objective approach was used by Lin et al. 39 who calculated the growth rate of 51 initially misdiagnosed melanomas by measuring the increase in melanoma thickness between two sequential biopsies. Of 34 melanomas that were in the in situ phase when the initial biopsy was obtained, 11 were still in the in situ phase when the second biopsy was performed. Melanomas that progressed from the in situ phase to invasive tumours exhibited a low median vertical growth rate of 0.04 mm/month. Similarly, Guerry et al. 1 found that at least 87% of 697 investigated primary melanomas exhibited a radial growth phase and estimated that this growth phase had a duration measured in years and should be sufficiently long-lived to allow diagnosis and surgical cure. These findings are further corroborated by the Gompertz function, according to which tumour growth follows a sigmoid curve where growth is slowest at the start and end of a growth period. 40 Taken together, these observations could explain why close-meshed follow-up examinations are highly effective in the prevention of thick melanomas. An important limitation of our work is the rather small size and focus on patients with SPMs. Although several studies have highlighted the biological similarity of first and second primary melanomas, we cannot exclude a covert selection bias which limits a generalization of our conclusions. Moreover, tumour biology and mutational spectrum of cutaneous melanomas depend on different UV-exposures in the populations and it has been shown that UVA-irradiation induces melanoma invasion. 41 As a consequence, the results of our German cohort may not necessarily be transferable to other parts of the world. On the other hand, the study is strengthened by the prospective documentation of follow-up examinations that provide far more objective and accurate information than patient interviews. This allowed us to determine the date of the last inconspicuous total body skin examination before the SPM diagnosis and to analyse the clinic-pathologic characteristics of thick SPMs. In conclusion, this study provides evidence that fast-growing melanomas without opportunity of early detection are rare and cannot explain the stable incidence of thick melanomas. The fact that follow-up examinations enable a highly efficient diagnosis of SPMs early in their course (86.0% thinner than 1.00 mm) highlights the importance of close-meshed total body skin examinations in patient groups that are at high risk of first or multiple primary melanomas. Moreover, particular attention should be paid during follow-up to SPMs presenting as atypical lesions. Acknowledgements We thank Professor Peter Martus for statistical advice, the patients participating in the German Central Malignant Melanoma Registry and everyone who has been involved in data collection, documentation and maintenance of the data base. References 1 Guerry D 4th, Synnestvedt M, Elder DE, Schultz D. Lessons from tumor progression: the invasive radial growth phase of melanoma is common, incapable of metastasis, and indolent. J Invest Dermatol 1993; 100: 342S 345S. 2 Ackerman AB. Malignant melanoma. A unifying concept. Am J Dermatopathol 1980; 2: Grob JJ, Richard MA, Gouvernet J et al. The kinetics of the visible growth of a primary melanoma reflects the tumor aggressiveness and is an independent prognostic marker: a prospective study. Int J Cancer 2002; 102: Liu W, Dowling JP, Murray WK et al. Rate of growth in melanomas: characteristics and associations of rapidly growing melanomas. Arch Dermatol 2006; 142: Martorell-Calatayud A, Nagore E, Botella-Estrada R et al. Defining fastgrowing melanomas: reappraisal of epidemiological, clinical, and histological features. Melanoma Res 2011; 21: Lipsker DM, Hedelin G, Heid E, Grosshans EM, Cribier BJ. Striking increase of thin melanomas contrasts with stable incidence of thick melanomas. Arch Dermatol 1999; 135: Garbe C, McLeod GR, Buettner PG. Time trends of cutaneous melanoma in Queensland, Australia and Central Europe. Cancer 2000; 89: Criscione VD, Weinstock MA. Melanoma thickness trends in the United States, J Invest Dermatol 2010; 130: Shaikh WR, Xiong M, Weinstock MA. The contribution of nodular subtype to melanoma mortality in the United States, 1978 to Arch Dermatol 2012; 148: Richard MA, Grob JJ, Avril MF et al. Melanoma and tumor thickness: challenges of early diagnosis. Arch Dermatol 1999; 135: Ferrone CR, Ben Porat L, Panageas KS et al. Clinicopathological features of and risk factors for multiple primary melanomas. JAMA 2005; 294: Doubrovsky A, Menzies SW. Enhanced survival in patients with multiple primary melanoma. Arch Dermatol 2003; 139: Hwa C, Price LS, Belitskaya-Levy I et al. Single versus multiple primary melanomas: old questions and new answers. Cancer 2012; 118: Schuurman MS, de Waal AC, Thijs EJM, van Rossum MM, Kiemeney L, Aben KKH. Risk factors for second primary melanoma among Dutch patients with melanoma. Br J Dermatol 2017; 176: Cadby G, Ward SV, Cole JM, Moses EK, Millward M, Palmer LJ. The association of host and genetic melanoma risk factors with Breslow thickness in the Western Australian Melanoma Health Study. Br J Dermatol 2014; 170: Chamberlain AJ, Fritschi L, Giles GG, Dowling JP, Kelly JW. Nodular type and older age as the most significant associations of thick melanoma in Victoria, Australia. Arch Dermatol 2002; 138: Geller AC, Elwood M, Swetter SM et al. Factors related to the presentation of thin and thick nodular melanoma from a population-based cancer registry in Queensland Australia. Cancer 2009; 115: Vecchiato A, Pasquali S, Menin C et al. Histopathological characteristics of subsequent melanomas in patients with multiple primary melanomas. J Eur Acad Dermatol Venereol 2014; 28: de Giorgi V, Rossari S, Papi F et al. Multiple primary melanoma: the impact of atypical naevi and follow up. Br J Dermatol 2010; 163: Uliasz A, Lebwohl M. Patient education and regular surveillance results in earlier diagnosis of second primary melanoma. Int J Dermatol 2007; 46: Savoia P, Quaglino P, Verrone A, Bernengo MG. Multiple primary melanomas: analysis of 49 cases. Melanoma Res 1998; 8: Murali R, Goumas C, Kricker A et al. Clinicopathologic features of incident and subsequent tumors in patients with multiple primary cutaneous melanomas. Ann Surg Oncol 2012; 19:

8 70 Gassenmaier et al. 23 Moseley HS, Giuliano AE, Storm FK 3rd, Clark WH, Robinson DS, Morton DL. Multiple primary melanoma. Cancer 1979; 43: van der Leest RJ, Liu L, Coebergh JW et al. Risk of second primary in situ and invasive melanoma in a Dutch population-based cohort: Br J Dermatol 2012; 167: Blackwood MA, Holmes R, Synnestvedt M et al. Multiple primary melanoma revisited. Cancer 2002; 94: Cascinelli N, Fontana V, Cataldo I, Balzarini GP. Multiple primary melanoma. Tumori 1975; 61: Clark WH Jr, From L, Bernardino EA, Mihm MC. The histogenesis and biologic behavior of primary human malignant melanomas of the skin. Cancer Res 1969; 29: Clark WH Jr, Elder DE, Guerry D 4th et al. Model predicting survival in stage I melanoma based on tumor progression. J Natl Cancer Inst 1989; 81: Weyers W, Euler M, Diaz-Cascajo C, Schill WB, Bonczkowitz M. Classification of cutaneous malignant melanoma: a reassessment of histopathologic criteria for the distinction of different types. Cancer 1999; 86: Koh HK, Michalik E, Sober AJ et al. Lentigo maligna melanoma has no better prognosis than other types of melanoma. J Clin Oncol 1984; 2: Sondergaard K, Schou G. Survival with primary cutaneous malignant melanoma, evaluated from 2012 cases. A multivariate regression analysis. Virchows Arch A Pathol Anat Histopathol 1985; 406: Warycha MA, Christos PJ, Mazumdar M et al. Changes in the presentation of nodular and superficial spreading melanomas over 35 years. Cancer 2008; 113: Heenan PJ, Holman CD. Nodular malignant melanoma: a distinct entity or a common end stage? Am J Dermatopathol 1982; 4: Heenan PJ. Nodular melanoma is not a distinct entity. Arch Dermatol 2003; 139: Curtin JA, Fridlyand J, Kageshita T et al. Distinct sets of genetic alterations in melanoma. N Engl J Med 2005; 353: Shain AH, Yeh I, Kovalyshyn I et al. The genetic evolution of melanoma from precursor lesions. N Engl J Med 2015; 373: Shain AH, Bastian BC. From melanocytes to melanomas. Nat Rev Cancer 2016; 16: Aitken JF, Elwood M, Baade PD, Youl P, English D. Clinical whole-body skin examination reduces the incidence of thick melanomas. Int J Cancer 2010; 126: Lin MJ, Mar V, McLean C, Kelly JW. An objective measure of growth rate using partial biopsy specimens of melanomas that were initially misdiagnosed. J Am Acad Dermatol 2014; 71: Winsor CP. The Gompertz curve as a growth curve. Proc Natl Acad Sci USA 1932; 18: Kamenisch Y, Baban TS, Schuller W et al. UVA-irradiation induces melanoma invasion via the enhanced warburg effect. J Invest Dermatol 2016; 136: Supporting information Additional supporting information may be found online in the Supporting Information section at the end of the article: Table S1. Follow-up scheme between 1994 and Table S2. Follow-up scheme between 2003 and Table S3. Follow-up scheme from 2013 onwards.