SHORT COMMUNICATION

Thin melanoma and late recurrences: it is never too thinand never too late

Antonio Giovanni Richetta • Ugo Bottoni •

Giovanni Paolino • Rita Clerico • Carmen Cantisani •

Marina Ambrifi • Paola Corsetti • Stefano Calvieri

Received: 4 February 2014 / Accepted: 1 March 2014 / Published online: 8 March 2014

� Springer Science+Business Media New York 2014

Abstract In the absence of risk factors, thin melanomas

(TM) present a long-term survival; however, recurrences

may occur. We describe the predictive clinicopathological

features of patients with metastatic TM. Kaplan–Meier

product was performed for the survival analysis, while Cox

proportional hazards regression was used to evaluate the

effect of the clinicopathological features on disease-free

survival (DFS) and overall survival (OS). Median DFS of

the entire cohort was 26 months and three patients devel-

oped late metastases. Nine patients developed extra-nodal

metastases as first recurrence, while cases of positive sen-

tinel lymph node biopsy (SLNB) were not found. DFS and

OS varied according to the clinicopathological features, but

only ulceration remained the main statistical significance

value. According to our results, a hypothetical use of

SLNB in TM without other risk factors is not currently

feasible. No consensus exists as to which patients with TM

are at risk for metastases or late recurrences.

Keywords Thin melanoma � Metastatses � Late

recurrences � Sentinel lymph node biopsy � Survival

Introduction

The incidence of malignant melanoma (MM) has drasti-

cally increased in the past decades. Approximately 70 % of

new cases of MM are thin melanomas (TM), which are

lesions\1 mm in Breslow thickness [1]. In the absence of

other risk factors (such as ulceration and/or mitotic

rate C1/mm2), TM present a 10-year survival rate of 96 %

[2]; however, it is known that a portion of this group could

develop recurrences and die [3, 4]. In this regard, some

authors propose the use of sentinel lymph node biopsy

(SLNB) for primary lesions C0.75 mm (some-

times C0.5 mm), but the universal application of SLNB

would expose a large number of low-risk patients to the

toxicity related to the procedure [5]. Some centers have

tried to select patients with TM at risk to disease pro-

gression according to several parameters, but these factors

varied from series to series [6]. Therefore, we have ana-

lyzed our casuistic of patients with TM to record the per-

centage of disease progression and put in evidence any

significant predictive factor, if present.

Materials and methods

We computer-searched the clinical records of all our

patients registered into a 20-year-old melanoma database,

to identify patients with TM (Breslow thickness B1.0 mm)

who developed metastases. We divided patients into T1a

and T1b, according to the presence of ulceration and/or

mitotic rate C1/mm2 [2]. All patients were treated with a

local re-excision of the scar, whereas the sentinel lymph

node biopsy (SLNB) was performed only for T1b patients.

Disease-free survival (DFS) and overall survival (OS) were

confronted with the same parameters analyzed by Balch

et al. [2] as reported in Table 1. The Kaplan–Meier product

was performed for the survival analysis, while Cox pro-

portional hazards regression was used to evaluate the effect

of the clinicopathological features on DFS and OS.

A. G. Richetta � G. Paolino (&) � R. Clerico � C. Cantisani �M. Ambrifi � P. Corsetti � S. Calvieri

Clinica Dermatologica, La Sapienza University of Rome,

Viale del Policlinico 15, 00186 Rome, Italy

e-mail: gio8519@libero.it

U. Bottoni

University Magna Graecia, V.le Europa, 88100 Catanzaro, Italy

123

Med Oncol (2014) 31:909

DOI 10.1007/s12032-014-0909-4

Results

A total of 22 patients with metastatic TM were found; they

were 1.2 % out of 1.851 patients with MM present in our

database and 2.8 % among melanomas with a Breslow

thickness B1.0 mm. No cases of positive SLNB were

detected in our sample. Thirteen patients (59 %) showed

lymph nodal metastases as first recurrence, while the other

nine patients showed other anatomical sites [such as dermal

(n = 4), lung (n = 3), brain (n = 1) and ovarian (n = 1)

metastases]. Median DFS of the entire cohort was

26 months (ranging between 2 and 135 months) and three

patients developed late metastases (C10 years). Regarding

DFS and initial clinicopathological features, only ulcera-

tion reached statistical significance both at log-rank test

and Cox hazards regression. Also for OS, ulceration

remained the main statistical significant predictor, while

axial site and Breslow thickness reached the significance

only at Cox hazards regression (Table 1).

Discussion

According to our data, shown in Table 1, we can assess

that only ulceration remains the main predictor for tumor

progression in TM, for both DFS and OS (at log-rank test

and Cox hazards regression), while axial site and Breslow

thickness reach the statistical significance value only in the

multivariate analysis, limiting their predictability.

Although clinical and pathological models have been

proposed to identify predictive factors for recurrences in

patients with TM, currently no consensus exists about

which patients with TM are at risk for metastases. In this

regard, most reports that analyzed the risk of progression of

TM and the utility to perform SLNB confirmed the

importance of ulceration and mitotic rate as main predic-

tors [3, 7]. Some authors highlighted an increased risk of

progression for TM C0.75 or C0.5 mm [3, 5, 7], while

others indicated age and anatomic site as further risk fac-

tors [8–10].

In some cases of TM, we observed late metastases.

Metastatic tumor dormancy in MM is defined as MM with

recurrence after more than 10 years from the diagnosis of

primary lesion. In the literature, there have been few series

that have reported on the clinical significance of this small

subgroup of patients with MM [6]. Currently, significant

predictors associated with late recurrences (especially in

TM) have not been identified.

In conclusion, our data confirm the utility to perform

SLNB only in patients with TM stage T1b [2]. Particularly,

the percentage of skip metastases (extra-nodal) found in

our patients is against a hypothetical use of SLNB in all

TM C0.75 or 0.5 mm without other risk factors, especially

ulceration.

Conflict of interest None.

Ethical standards We declare that our study was a performer in

accordance with ethical standards laid down in 1964 Declaration of

Helsinki. We declare also that all persons gave their informed consent

prior to their inclusions in the study.

References

1. Gimotty PA, Guerry D, Ming ME, Elenitsas R, Xu X, Czerniecki

B, et al. Thin primary cutaneous malignant melanoma: a prog-

nostic tree for 10-year metastasis is more accurate than American

Joint Committee on Cancer staging. J Clin Oncol. 2004;22:

3668–76.

2. Balch CM, Soong S, Ross MI, Urist MM, Karakousis CP, Temple

WJ, et al. Final version of 2009 AJCC melanoma staging and

classification. J Clin Oncol. 2009;27:6199–206.

3. Han D, Zager J, Shyr Y, Chen H, Berry LD, Iyengar S, et al.

Clinicopathologic predictions of sentinel lymph node metastasis

in thin melanoma. J Clin Oncol. 2013;31:4387–93.

4. Balch CM, Balch GC, Sharma RR. Identifying early melanomas

at higher risk for metastases. J Clin Oncol. 2012;30:1406–7.

Table 1 Disease-free survival (DFS), overall survival (OS) and

clinicopathological features in patients with metastatic thin melanoma

(TM)

n DFS

(median

months)

pa pb OS

(median

months)

pc pd

Gender

Male 9 39 0.3 0.4 76 0.3 0.7

Female 13 25 124

Age

C61 6 14 0.4 0.3 45 0.08 0.5

B60 16 32 99

Sites

Axial 13 39 0.1 0.4 76 0.2 0.03

Extremities 9 25 78

Ulceration

Presence 4 15 0.01 0.02 28 0.04 0.04

Absence 18 28 85

Mitotic rate

C1/mm2 2 45 0.8 0.9 57 0.3 0.9

\1/mm2 20 28 85

Clark level

II 11 26 0.8 0.7 124 0.2 0.1

III–IV 11 26 78

Breslow

C0.75 mm 13 26 0.4 0.4 76 0.08 0.02

B0.74 mm 9 36 130

pa Kaplan–Meier product and log-rank test between DFS curves; pb

Cox proportional hazards regression between DFS and predictive

factors analyzed; pc OS Kaplan–Meier product and log-rank test; pd

Cox proportional hazards regression between OS and predictors.

n indicates number of patients. In italic, significant values

909 Page 2 of 3 Med Oncol (2014) 31:909

123

5. Quarto G, Sivero L, Benassai G, Bucci L, Desiato V, Perrotta S,

et al. Is 1 mm thickness sec. Breslow the correct cut-off for sentinel

node biopsy in melanoma? Report of six cases of metastasis by thin

melanoma. Ann Ital Chir. 2012;29:83.

6. Faries MB, Steen S, Ye X, Sim M, Morton DL. Late recurrence in

melanoma: clinical implications of lost dormancy. J Am Coll

Surg. 2013;217:27–34.

7. Han D, Yu D, Zhao X, Marzban SS, Messina JL, Gonzalez RJ,

et al. Sentinel node biopsy is indicated for thin melanomas

C0.76 mm. Ann Surg Oncol. 2012;19:3335–42.

8. Venna S, Thummala S, Nosrati M, Leong SP, Miller JR, Sagebiel

RW, et al. Analysis of sentinel lymph node positivity in patients

with thin primary melanoma. J Am Acad Dermatol. 2013;68:

560–7.

9. Green AC, Baade P, Coory M, Aitken JF, Smithers M. Popula-

tion-based 20-year survival among people diagnosed with thin

melanomas in Queensland Australia. J Clin Oncol. 2012;30:

1462–7.

10. Faries M, Wanek L, Elashoff D, Wright BE, Morton DL. Pre-

dictors of occult nodal metastasis in patients with thin melanoma.

Arch Surg. 2010;145:137–42.

Med Oncol (2014) 31:909 Page 3 of 3 909

123