ARTICLE
Auteur(s) : Falk Roeder, Robert Krempien, Florian
Sterzing, Angela Funk, Martina Treiber, Jürgen Debus, Marc Bischof
Department of Radiooncology, University of Heidelberg, Im
Neuenheimer Feld 400, D-69120 Heidelberg, Germany
accepté le 7 Février 2007
Merkel cell carcinoma (MCC) is a rare malignant skin tumour which
was first described 1972 [1]. The current estimated incidence is
0.24 per 100,000 person-years [2], with most cases occurring in
elderly patients with an average age of 70 to 75 years [2, 3].
Sunlight appears to be a factor in the cause of MCC, as primary
tumours are most commonly found in sun-exposed areas like the head
and neck or extremities. Immunosuppression appears to be another
risk factor, as shown by the greater than 13-fold increased risk
for the appearance of MCC in patients with HIV [4-6].The cell of
origin is thought to be the dermal neurotactile cell, first
described by Friedrich Sigmund Merkel in 1875 [7]. Merkel cells
(MC) are small cells of neuroendocrine origin, found in the basal
layer of the epidermis, where they form MC-axon-complexes with
primary nerve endings and work as slow-acting mechanoreceptors
[8-10]. The histopathologic diagnosis of MCC on light microscopy is
challenging and, therefore, immunohistochemical markers like
cytokeratin-20, neurofilament protein, or thyroid transcription
factor 1 are used in the differential diagnosis especially to
distinguish between MCC, which expresses both neuroendocrine and
cytokeratin markers, and other small cell neuroendocrine tumours
[11].Typically, MCC shows a tendency to rapid progression (figure 1), early
and frequent metastases to regional lymph nodes, and development of
distant metastases in approximately 50% of patients [8, 12].
Relapse after initial treatment is also frequent, as shown in a
previous study which found 43% recurrences after a median time of 9
months [13]. Staging of MCC in most published series is done
according to a simple three-part staging system with stage I
characterising localised disease, stage II regional disease
(involvement of regional lymph nodes), and stage III distant
metastasis. The prognosis of MCC is strongly associated with stage
at presentation [2].Treatment concepts vary because of the small
number of patients world wide. Surgical excision of the primary
tumour [8, 12, 13], sentinel lymph-node biopsy [14], and
lymphadenectomy in node-positive disease [8, 12, 15] are
recommended. In vitro studies have shown that MCC is a
radiosensitive tumour [16]. Adjuvant radiotherapy to the primary
site and draining lymph node areas appears to improve regional
control and survival [20, 22-24]. However, as a result of the small
patient numbers, evidence from randomised trials does not exist. In
this study, we report the results with radiotherapy in the
treatment of MCC.
Patients and methods
A search of the database of the University of Heidelberg identified
a total of 39 patients with histologically proven MCC who were
treated at the institution from 1981 to 2006. The median age of
patients was 75 years (range 47-93 years). Median follow-up
was 24 months (range 2-106 months). Detailed patient
characteristics are provided in table
1-3.
Treatment concepts were not uniform, because patients were
referred from different centres and regions over a period of
25 years. Fifteen patients were presented with stage I
disease: twelve directly after wide excision of the tumour, and 3
after surgery of a local recurrence (table
2). Twenty-one patients were presented with stage II
disease (table 3): ten patients were
referred to our study in a primary situation. Seven of these
patients received a wide tumour excision and a lymphadenectomy of
the regional lymph nodes. Two patients with unknown primary tumour
localisation received a lymphadenectomy only. One patient showed an
unresectable MCC of the nasopharynx. The remaining eleven patients
in stage II were presented after an initial treatment with surgery
alone. Six patients had at least one surgical excision of a local
recurrence. Five patients showed an inoperable disease at the time
of presentation. Three patients presented with stage III disease
and metastases in distant lymph nodes, brain and orbit.
Radiotherapy was performed with a linear accelerator in 37 of
the 39 patients. Two patients were treated with a cobalt-60 unit.
Field arrangements, the technique of irradiation, and the energies
used depended on the anatomical position of the target volume.
Photon beams with energies from 6 to 42 MV, and electron beams with
energies from 4 to 15 MeV, were used. Median radiation doses of 58
Gy to the primary tumour site and 50 Gy to the adjacent lymph nodes
were applied in fraction doses of 2 Gy. Twenty-four patients
received an adjuvant extended-field radiotherapy which included
tumour bed and regional lymph nodes. Four patients were treated
with an adjuvant involved-field technique restricted to the tumour
bed. Lymph nodes only were irradiated in 2 patients with unknown
primary tumour. Six inoperable patients were treated with
extended-field radiotherapy. Patients in stage III received
palliative radiotherapy.
Descriptive statistics, Kaplan-Meier estimation analysis, and
log rank test were applied for statistical analysis. One-, three-
and five-year actuarial rates for overall survival (OS),
disease-specific survival (DSS), loco-regional control (LC), and
distant metastases-free survival were calculated using the
Kaplan-Meier method from the date of radiotherapy applying
STATISTICA version 5.5 (StatSoft Inc®, USA).
Loco-regional control is defined as freedom from recurrence at the
primary tumour site and regional lymph nodes. Disease-specific
survival is defined as the percentage of patients in this series
who have survived the MCC for the study period. Statistical
differences in survival rates were tested with the log rank test.
P-values were two-sided with a p-value of less than 0.05 considered
statistically significant.
Table 1 Patient characteristics
|
n
|
%
|
|
Sex
|
|
|
|
male
|
15
|
38
|
|
female
|
24
|
62
|
|
|
|
|
Localisation
|
|
|
|
head and neck
|
16
|
41
|
|
trunk
|
3
|
8
|
|
upper limb
|
8
|
21
|
|
lower limb
|
8
|
21
|
|
lymph node*
|
3
|
8
|
|
nasopharynx
|
1
|
3
|
|
|
|
|
Stage at first diagnosis
|
|
|
|
stage I
|
27
|
69
|
|
stage II
|
11
|
28
|
|
stage III
|
1
|
3
|
|
|
|
|
Stage at radiotherapy
|
|
|
|
stage I
|
15
|
38
|
|
stage II
|
21
|
54
|
|
stage III
|
3
|
8
|
|
|
|
|
Situation at radiotherapy
|
|
|
|
primary
|
22
|
56
|
|
recurrent
|
17
|
44
|
|
|
|
Table 2 Detailed patient characteristics for stage I
|
Stage I patients (n = 15)
|
n
|
%
|
|
|
|
|
Size of primary tumour
|
|
|
|
< 1 cm
|
2
|
13
|
|
1-2 cm
|
5
|
33
|
|
2-4 cm
|
7
|
47
|
|
> 4 cm
|
1
|
7
|
|
|
|
|
Type of surgery
|
|
|
|
tumour alone
|
11
|
73
|
|
tumour + lymph nodes
|
2
|
13
|
|
tumour + sentinel node biopsy
|
2
|
13
|
|
|
|
|
Resection margin
|
|
|
|
R0
|
13
|
87
|
|
R1
|
2
|
13
|
|
R2
|
0
|
0
|
Table 3 Detailed patient characteristics for stage II
|
Stage II patients (n = 21)
|
n
|
%
|
|
|
|
|
Size of primary tumour
|
|
|
|
No primary
|
3
|
14
|
|
< 1 cm
|
0
|
0
|
|
1-2 cm
|
6
|
29
|
|
2-4 cm
|
7
|
33
|
|
> 4 cm
|
5
|
24
|
|
|
|
|
Size of positive lymph nodes
|
|
|
|
< 2cm
|
6
|
29
|
|
2-4 cm
|
11
|
52
|
|
> 4 cm
|
4
|
19
|
|
|
|
|
Number of positive lymph nodes
|
|
|
|
single node
|
4
|
19
|
|
2-3 nodes
|
6
|
29
|
|
>3 nodes
|
11
|
52
|
|
|
|
|
Type of surgery
|
|
|
|
tumour + lymph nodes
|
9
|
43
|
|
tumour alone
|
4
|
19
|
|
lymph nodes alone
|
2
|
10
|
|
no surgery
|
6
|
29
|
|
|
|
|
Resection margin (primary tumour)
|
|
|
|
R0
|
13
|
100
|
|
R1
|
0
|
0
|
|
R2
|
0
|
0
|
Results
Stage I at presentation for radiotherapy
Fifteen patients presented with stage I disease (figure 2). The actuarial
one-, three- and five-year loco-regional control rates after
adjuvant radiotherapy were found to be 100%, 90%, and 75%,
respectively. Twelve patients (80%) were referred directly after
excision of the primary tumour. Loco-regional control was achieved
in all patients. Three patients (20%) were presented after excision
of a local recurrence. Two of them relapsed loco-regionally 16 and
42 months after adjuvant radiotherapy.
The median overall survival for stage I patients was 46 months,
resulting in actuarial three-, and five-year overall survival rates
of 100% and 50%, respectively. Distant metastases were found in
three patients after 4, 17 and 68 months, resulting in one-,
three-, and five-year distant metastases-free survival rates of
92%, 83%, and 83%, respectively. Metastases were located in distant
lymph nodes, skin, central nervous system, and pancreas. The three-
and five-year disease-specific survival was 100% and 67%,
respectively; the median was not reached.
Stage II at presentation for radiotherapy
Twenty-one patients presented with stage II disease. Ten (48%) of
these patients had initially shown stage I disease and suffered
from a local recurrence with lymph node infiltration (stage II)
after median six months (range 2-12 months), from initial treatment
with surgery alone.
The one-, three-, and five-year actuarial loco-regional control
rate for stage II patients after radiotherapy was 78%. Three of
four loco-regional relapses were found in patients who had at least
one recurrence before radiotherapy (figure 3). The median
overall survival was 32 months, resulting in actuarial one-,
three-, and five-year overall survival rates of 63%, 29% and 14%,
respectively. Distant metastases were found in seven patients after
a median of six months, resulting in one-, three- and five-year
distant metastases-free survival rates of 67%, 58%, and 58%,
respectively. Metastases were located in distant lymph nodes, bone,
liver, skin and bone marrow. The 1-, 3-, and 5-year
disease-specific survival was 84%, 55%, and 55%, respectively.
Stage III at presentation for radiotherapy
Three patients were presented in stage III with distant metastases
(lymph nodes, brain, orbit). Two of them initially had a surgical
excision of the primary tumour (stage I) without adjuvant
radiotherapy, and were referred to our department with distant
metastases and/ or loco-regional progression. Good symptom relief
was achieved with palliative radiotherapy, but all patients died
within 6 months after diagnosis of metastases.
Side effects
Acute mucosal and skin reactions (WHO grade I-II) were observed. No
serious late complications were associated with radiotherapy.
Although the current study is a retrospective, non-randomised
analysis, comparison of the outcome in different stages and
dependence of primary treatment was completed: local control did
not differ significantly between stage I and II patients. But,
loco-regional control was significantly improved in patients who
received surgery and immediate adjuvant radiation as primary
treatment, compared to those initially treated with surgery alone
(p = 0.039). Considering disease-specific survival, patients
presenting with stage I disease at radiotherapy showed a trend
towards improved survival compared to stage II patients (p = 0.16),
although significance was not reached.
Discussion
Merkel cell carcinoma is a rare, aggressive tumour with high rates
of local recurrences and distant metastases [8, 12]. In the current
study, treatment of MCC with surgery followed by adjuvant
radiotherapy resulted in good loco-regional control without major
toxicities. Loco-regional control could be achieved in 95% of
patients who were treated first with surgery and then immediate
radiotherapy, whereas 36% of patients initially treated with
surgery alone developed a loco-regional relapse. The tendency
towards a better loco-regional control with surgery and immediate
adjuvant radiotherapy is supported by findings of previous studies
[17, 19, 20, 23, 27]. Medina-Franco et al. identified (in a review
of 1024 cases) 11 series that compared local relapse rates with and
without adjuvant radiotherapy. They found a highly significant
improvement in local control (p = 0.0001) with adjuvant
radiotherapy [21]. Meeuwissen et al. reported increased control
rates for tumour sites (100% vs. 79%) and regional lymph nodes (82%
vs. 14%) for surgery and radiotherapy compared to surgery alone
[17]. Similar results were published by Morrison et al. [27].
Kokoska et al. reported loco-regional control of 85% after surgery
and radiotherapy compared to 10% after surgery alone [19]. The
study by Allen et al. [13], in which no significant improvement of
local control was found, showed nodal recurrence rates of 26% in
the surgery alone group compared to 13% in the group with adjuvant
radiotherapy. Significance was possibly not achieved due to the
fact that the radiotherapy group was small (17%) compared to the
surgery group (83%) and the series may have not been robust enough
to reveal a distinct difference. Inoperable MCC responds well to
primary radiation, but the probability of recurrence is high [18,
28, 29]. Two of six inoperable patients in our study treated with
radiotherapy alone relapsed after median 7.5 months.
Although there was a significant difference in loco-regional
control between patients in the primary and recurrent situation,
this advantage did not convert into significantly improved
disease-specific and overall survival. This could be explained by
the high percentage of patients with positive lymph nodes at the
time of radiotherapy, because lymph node involvement is a strong
predictor of distant spread and overall survival [8]. In this
context, it is noteworthy that twelve of the seventeen patients
(71%) who were initially treated with surgery alone showed a stage
progression from localized disease (stage I) at first diagnosis to
regional (stage II) or distant spread (stage III) in a recurrent
situation, and thus a worsened prognosis.
Considering overall survival (OS), stage I patients showed an
improved outcome with a five-year OS of 50% compared to 14% for
stage II. These results are consistent with previous studies. In
the series by McAfee et al., five-year OS decreased from 44% to 23%
if nodal involvement was present at diagnosis, although these
differences closely missed significance [26]. Morrisson et al.
reported a significantly improved OS of 40 months in node negative
vs. 13 months in node positive patients [27]. Medina-Franco et al.
[21] found an improved median OS of 97 months in stage I patients
compared to 15 months in stage II patients.
Disease-specific survival (DSS) was used as an additional
survival parameter in this study, because MCC is a disease of
elderly patients (median age of 75 years), and, overall survival
can be strongly affected by other factors in this patient group.
Stage I patients showed an improved five-year DSS of 67% compared
to 55% for stage II patients. The results are similar to a previous
study by McAfee et al., who reported a decrease of five-year DSS
from 58% for stage I to 38% for stage II [26]. Distinct differences
between OS and DSS in the current study, which are pronounced
compared to other series, most likely result from extensive
co-morbidities found in most patients [26]. However, a trend
towards improved outcome was seen for stage I patients, independent
of the survival parameter regarded, which underlines the importance
of the extent of disease for prognosis. For patients with stage III
disease, the prognosis was poor and treatment was only palliative.
Depending on the localisation of distant spread, radiotherapy with
moderate doses can achieve good local control and palliation.
Considering etiological factors, the median age of the patients
in our series was 75 years and corresponded to the literature [2,
3, 24]. In contrast to some previous studies, which reported a male
predominance, 61% of our patients were female [17, 25]. The
distribution of primary tumours to different anatomical sites in
our series was quite similar to the distribution-pattern reported
from the surveillance, epidemiology, and end results program (SEER)
[2]. In 24 of 36 patients (67%), the primary lesion was localized
in sun-exposed areas (head and neck or distal extremities)
supporting the hypothesis of ultraviolet radiation as a factor in
the genesis of MCC. Similar to other studies, a primary skin lesion
could not be found in few patients [24]. Ten patients (26%) showed
additional malignancies in their history, four of them more than
one disease. This association is likely to be explained by the high
median age of the patient group.
Surgery and immediate adjuvant radiotherapy were found to result
in good loco-regional control. Radiation therapy-related toxicities
were low. But, as a result of the small number of patients in this
retrospective study, general treatment recommendations can not be
given. Radiotherapy alone is suggested only in inoperable or
metastatic MCC.
Acknowledgements
Financial support: none.
Conflict of interest: none
References
1 Toker C. Trabecular carcinoma of the skin. Arch Dermatol
1972; 105: 107-10.
2 Agelli M, Clegg LX. Epidemiology of primary Merkel
cell carcinoma in the United States. J Am Acad Dermatol 2003; 49:
832-41.
3 Goessling W, McKee PH, Mayer R. Merkel cell
carcinoma. J Clin Oncol 2002; 20: 588-98.
4 Miller RW, Rabkin CS. Merkel cell carcinoma and
melanoma: etiological similarities and differences. Cancer
Epidemiol Biomarkers Prev 1999; 8: 153-8.
5 Tai PT, Yu E, Winquist E, Hammond A,
Stitt L, Tonita J, Gilchrist J. Chemotherapy in
neuroendocrine/Merkel cell carcinoma of the skin: case series and
review of 204 cases. J Clin Oncol 2000; 18: 2493-9.
6 Engels EA, Frisch M, Goedert JJ,
Biggar RJ, Miller RW. Merkel cell carcinoma and HIV
infection. Lancet 2002; 359: 497-8.
7 Merkel F. Tastzellen und Tastkörperchen bei den
Haustieren und Menschen. Arch Mikr Anat 1875; 11: 636-52.
8 Poulsen M. Merkel-cell carcinoma of the skin. Lancet
Oncol 2004; 5: 593-9.
9 Briggaman RA, Wheeler CEJ. The epidermal-dermal
junction. J Invest Dermatol 1975; 65: 71-84.
10 Ogawa H. The merkel cell as a possible mechanoreceptor
cell. Prog Neurobiol 1996; 49: 317-34.
11 Bobos M, Hytiroglou P, Kostopoulos I,
Karkavelas G, Papadimitriou CS. Immunohistochemical
distinction between merkel cell carcinoma and small cell carcinoma
of the lung. Am J Dermatopathol 2006; 28: 99-104.
12 Koljonen V. Merkel cell carcinoma. World J Surg Oncol
2006; 4: 7.
13 Allen PJ, Bowne WB, Jaques DP,
Brennan MF, Busam K, Coit DG. Merkel cell carcinoma:
prognosis and treatment of patients from a single institution. J
Clin Oncol 2005; 23: 2300-9.
14 Mehrany K, Otley CC, Weening RH,
Phillips PK, Roenigk RK, Nguyen TH. A meta-analysis
of the prognostic significance of sentinel lymph node status in
merkel cell carcinoma. Dermatol Surg 2002; 28: 113-7.
15 Brissett AE, Olsen KD, Kasperbauer JL. lewis
JE, Goellner JR, Spotts BE, Weaver AL, Strome SE. Merkel cell
carcinoma of the head and neck: a retrospective case series. Head
Neck 2002; 24: 982-8.
16 Leonard JH, Ramsay JR, Kearsley JH,
Birrell GW. Radiation sensitivity of merkel cell carcinoma
cell lines. Int J Radiat Oncol Biol Phys 1995; 32: 1401-7.
17 Meeuwissen J, Bourne R, Kearsley J. The
importance of post-operative radiotherapy in the treatment of
merkel cell carcinoma. Int J Radiat Oncol Biol Phys 1995; 31:
325-31.
18 Raaf JH, Urmacher C, Knapper WK, Shiu MH,
Cheng EW. Trabecular (Merkel cell) carcinoma of the skin:
treatment of primary, recurrent and metastatic disease. Cancer
1986; 57: 178-82.
19 Kokoska ER, Kokoska MS, Collins BT,
Stapleton DR, Wade TP. Early aggressive treatment for
Merkel cell carcinoma improves outcome. Am J Surg 1997; 174:
688-93.
20 Herbst A, Haynes H, Nghiem P. The standard of
care for Merkel cell carcinoma should include adjuvant radiation
and lymph node surgery. J Am Acad Derm 2002; 46: 640-2.
21 Medina-Franco H, Urist MM, Fiveash J,
Heslin MJ, Bland KI, Beenken SW. Multimodality
treatment of merkel cell carcinoma: case series and review of 1024
cases. Ann Surg Oncol 2001; 8: 204-8.
22 Shaw JH, Rumball E. Merkel cell tumour: clinical
behaviour and treatment. Br J Surg 1991; 78: 138-42.
23 Veness MJ, Morgan GJ, Gebski V. Adjuvant
locoregional radiotherapy as best practice in patients with merkel
cell carcinoma of the head and neck. Head Neck 2005; 27:
208-16.
24 Veness MJ, Perera L, McCourt J,
Shannon J, Hughes TM, Morgan GJ, Gebski V.
Merkel cell carcinoma: improved outcome with adjuvant radiotherapy.
ANZ J Surg 2005; 75: 275-81.
25 Boyle F, Pendlebury S, Bell D. Further
insights into the natural history and management of primary
cutanous neuroendocrine (Merkel cell) carcinoma. Int J Radiat Oncol
Biol Phys 1995; 31: 315-23.
26 McAfee WJ, Morris CG, Mendenhall CM,
Werning JW, Mendenhall NP, Mendenhall WM. Merkel
cell carcinoma: treatment and outcomes. Cancer 2005; 104:
1761-4.
27 Morrison W, Peters L, Silva E, Wendt C,
Kian Ang K, Goepfert H. The essential role of radiation
therapy in securing locoregional control of merkel cell carcinoma.
Int J Radiat Oncol Biol Phys 1990; 19: 583-91.
28 Pacella J, Ashby M, Ainslie J, Minty C.
The role of radiotherapy in the management of primary cutaneous
neuroendocrine tumours: Experience of the Peter Maccallum Cancer
Institute (Melbourne Australia). Int J Radiat Oncol Biol Phys 1988;
86: 1077-84.
29 Hasle H. Merkel cell carcinoma: The role of primary
treatment with radiotherapy. Clin Oncol 1991; 3: 114-6.
|
Printable version
Version PDF
