© 2007 Acta Dermato-Venereologica. ISSN 0001-5555
Acta Derm Venereol 87
Acta Derm Venereol 2007; 87: 330–334
The aim of this retrospective survey was to determine
recurrence rates after treatment of basal cell carcinomas
in a single academic dermatology department. A total of
1016 patients with 1593 histologically verified basal cell
carcinomas (n=1212 primary and n=381 relapsing) were
included. Tumour localization, T-stage and the method
of treatment were significant predictors of the risk of re-
currence (forward Cox regression, p <0.001). The relapse
rate for primary basal cell carcinomas on the scalp was
highest (odds ratio (OR)=2.8, 95% confidence interval
(CI) 1.5–5.3). T2 and T3 tumours showed a 2- and 3-fold
increased relapse rate, respectively, compared with T1
basal cell carcinomas. Radiotherapy and surgical ex-
cision had the lowest relapse rates, whereas curettage
and photodynamic therapy resulted in 5-year relapse
rates of up to 70%. Patients with chronic skin diseases
had a 50% lower risk of relapse than healthy patients
(OR=0.5, CI=0.3–0.8). Recurrent basal cell carcinomas
had a higher relapse rate than primary lesions (OR=1.8,
CI= 1.4–2.2). Patients treated in a specialized skin cancer
unit had a 6.4-fold (CI=2.4–17.4) higher cure rate com-
pared with those treated by less experienced physicians.
Thus, in an uncontrolled, real-life situation curettage or
photodynamic therapy are associated with significantly
higher relapse risk than excision and radiotherapy and
therefore should not be used for high risk primary tu-
mours or recurrent tumours. Treatment in the setting of
a specialized skin cancer unit yields a much lower relapse
rate. Key words: basal cell carcinoma; recurrence rate;
risk factors; immunosuppression; surgical excision; curet-
tage; radiotherapy; photodynamic therapy.
(Accepted December 7, 2006.)
Acta Derm Venereol 2007; 87: 330–334.
Robert Gniadecki, MD, DMSc, Department of Derma-
tology D, Bispebjerg Hospital, Bispebjerg bakke 23,
DK-2400 Copenhagen NV, Denmark. E-mail: rg01@bbh.
Basal cell carcinoma (BCC) is the most common skin
neoplasm, with a 70% increase in incidence over the last
three decades (1). Mortality rates due to BCC are low, but
its high incidence and morbidity translates into a high eco-
nomic burden (1). BCC can cause extensive destruction
of tissue, particularly on the face (2) where early, radical
treatment reduces scarring and disfigurement (3).
Earlier studies identified several prognostic factors
that should be taken into account when choosing the
appropriate treatment for BCC (4). Most (86%) of BCC
are found in the head region, where the tumours are
also most difficult to treat (3). The particularly high-
risk areas are the central region of the face involving
the inner canthus, nostrils, and peri-auricular area (3).
Tumours on the scalp and forehead are considered to
pose a middle risk of recurrence, whereas the low-risk
areas comprise the neck, trunk and extremities (5). In
addition to the localization, the tumour size (T stage)
has prognostic significance (3). Tumour recurrence
increases by approximately 7% for each millimetre of
tumour diameter (6). Other factors are: male sex (7),
sun-sensitive skin (8), age over 60 years at first pre-
sentation (8), histological type (4) and immunological
The presence or absence of risk factors is taken into
consideration when choosing the therapeutic option. In
Europe, especially in the UK, Scandinavia and France,
curettage with or without cautery or simple excision are
often the first choice of treatment, particularly for low
risk tumours less than 1.5 cm in diameter. It has been
argued that these simple and cheap techniques are ade-
quate and probably preferable to the more costly Mohs
micrographic surgery (MMS) (10–12). Several studies
have demonstrated the very low recurrence rate of curet-
tage, comparable to that of MMS and not exceeding
5% (13, 14). In other studies, however, the cure rate of
curettage does not exceed 75% for the difficult to treat
Other treatment options comprise radiotherapy, photo-
dynamic therapy (PDT), cryotherapy with liquid nitrogen
or imiquimod. The selection among these of the therapeu-
tic modalities depends on the predicted tumour aggres-
siveness and local preferences in different centres.
This study reappraised the importance of risk factors
and treatment modality on the recurrence of BCC treated
in an academic setting where MMS is not performed. The
results suggest that the therapeutic outcome is often worse
in real life than it could be deduced from the available
clinical studies. In particular, treatment of higher risk fa-
cial tumours by curettage or PDT resulted in a surprisingly
high 5-year recurrence rate when the treatment was perfor-
med by inexperienced physicians. The results also suggest
how the treatment results by traditional modalities such
as curettage or PDT may be improved.
Factors Affecting the Recurrence Rate of Basal Cell Carcinoma
Fie SAAByE BøGElUND, Peter ASHlEDE PHIlIPSEN and Robert GNIADECKI
Department of Dermatology, University of Copenhagen, Bispebjerg Hospital, Copenhagen, Denmark
Basal cell carcinoma recurrence rate
Acta Derm Venereol 87
MATERIALS AND METHODS
Patients and data capture
All patients treated for BCC in our institution in the period
January 1998 to January 2005 were included in the study. In
all cases the diagnosis was histologically proven. Patients were
categorized as having multiple tumours if they presented with
>2 tumours at the first consultation. For all patients their sex,
age, number of primary and recurrent lesions, lesion localization
and size, concurrent diseases, and the treatment modality were
recorded. Whether the treatment took place in a specialized
cancer treatment unit under the supervision of an experienced
dermatologist was also recorded. After data collection we
excluded 86 tumours treated with topical 5-fluorouracil (once
daily, for 3 months or until development of skin ulceration),
CO2-laser, imiquimod or untreated tumours, due to the low
number of individuals. Tumours were categorized into 3 groups:
T1 tumours: <2 cm in diameter, T2 tumours: ≥2 cm to <5cm in
diameter, and T3 tumours ≥5cm in diameter. The patients were
also categorized into two groups: one group treated by a specia-
list in dermatology and the other by a physician in training.
For statistical analysis a forward Cox regression was used inclu-
ding the following variables: age, sex, tumour size, concurrent
disease, presence of multiple tumours, treatment modality, loca-
lization and treatment in cancer treatment unit. Odds ratio (OR),
95% confidence intervals (CI) and p-values were calculated
with SPSS software (version 13.0, SPSS Inc., Chicago, Illinois,
USA). A p-value <0.05 was considered significant.
Factors determining the recurrence rate of primary
A total of 1016 patients were identified (55.4% women
and 44.6% men) presenting with 1593 tumours. A total
of 1212 tumours were primary BCC and 381 tumours
were the recurrent ones referred to our institution from
private practice for further treatment. The mean age at
the time of diagnosis was 73.3 years (range 23–102 ye-
ars). Sixty-eight of the patients had multiple tumours (a
total of 310 tumours) defined as >2 tumours at the first
treatment. The probability of recurrence of primary BCC
was analysed in 804 patients (1139 tumours) including
these parameters, which showed significance in a single
variable Cox regression model: tumour size and locali-
zation, choice of treatment, and concurrent disease.
As expected, tumour localization was strongly and
independently correlated with the recurrence rate.
Truncal carcinomas had the best prognosis, whereas
carcinomas of the scalp had the highest risk of recur-
rence (OR 2.8, CI 1.5–5.3) (Fig. 1A). Stage T1 had the
lowest recurrence rate, whereas stage T2 and T3 had
OR for relapse of 2.0 (CI 1.4–2.8) and 3.1 (CI 1.4–6.8),
respectively (Fig. 1B). Surgical excision and radio-
therapy had excellent cure rates, whereas curettage and
Fig. 1. Five-year risk of relapse after the
treatment of primary basal cell carcinoma
(BCC). The treatment outcome of 1139
histologically verified BCCs were
analysed by forward Cox regression.
(A) Effect of tumour localization (trunk:
black, lower extremities: red, neck:
green, central face and ears: blue, upper
extremities: stippled black, lateral face:
stippled red, scalp: stippled green). (B)
Effect of T-stage (tumour size) (T1:
black, T2: red, T3: green). (C) Influence
of treatment modality (excision: black,
radiotherapy: red, curettage: green,
photodynamic therapy: blue). (D) Effect
of concurrent diseases (organ transplant
recipients: black, chronic inflammatory
skin disease: red, immunosuppressed
patients (patients with HIV infection or
leukaemia; excluding organ transplant
recipients): green, extracutaneous
F. Saabye Bøgelund et al.
Acta Derm Venereol 87
PDT had significantly higher recurrence rate (OR 2.3
(CI 0.9–2.0) and OR 5.2 (CI 3.5–7.5), respectively) (Fig.
1C). Unexpectedly, we found that the patients suffering
from a chronic inflammatory skin disease (psoriasis,
atopic dermatitis, eczema) (n=125) and the immuno-
suppressed individuals (n=64) had an approximately
50% lower relapse rate that the healthy patients (OR
0.5, CI 0.3–0.8) (Fig. 1D).
Recurrence rate of primary BCC of the head
Since BCC occurs mostly in the cephalic region and
poses a special therapeutic problem, we undertook the
separate analysis of risk factors in this group of patients.
A total of 534 patients with 601 tumours (385 central
face (chin, nose, lips, eye surroundings and forehead)
and ears, 162 lateral face (cheeks, temples) and 54 at
the scalp) were included. The risk factors resembled
those of the primary BCC elsewhere (Fig. 2A). The
differences were the significantly higher relapse rate
in patients with multiple tumours (OR 1.6, CI 1.1–2.5)
(Fig. 2B). The specific localization in the lateral vs.
central portion of the face was not a significant para-
meter for recurrence. Excision seemed to be a slightly
better treatment modality than radiotherapy (Fig. 2C),
however the statistical significance has not been attain-
ed due to a low number of patients where excision had
been performed (n=26). As in the unselected primary
tumours, curettage yielded a greater than three-fold
increase in recurrence (OR 3.5), whereas PDT was
associated with a five-fold increase in recurrence rate
(OR 5.3, CI 3.1–9.1).
Treatment of relapsing tumours
For the analysis of relapse rates after treatment of
recurrent tumours, the data were pooled from the pa-
tients who recurred after the treatment at our institution
(n=709) and those patients who were referred to us from
the private practice (n=381). The overall relapse rate
of the recurrent tumours was approximately twice as
high as for the primary BCC (OR 1.8, CI 1.4–2.2), and
the facial tumours (OR 2.1, CI 1.4–3.1). The results of
single treatment modalities are shown in Fig. 3.
Relapse rates for the patients treated in a specialized
skin cancer unit
Relapse rates were compared for the patients treated
in a skin cancer unit and those treated in the general
dermatology outpatient clinic in our institution. This
analysis showed the significant therapeutic advantage
Fig. 2. Five-year risk of relapse after the treatment of primary basal cell carcinoma
(BCC) in the face and scalp. A total of 601 histologically verified BCCs were analysed
as in Fig. 1. (A) Effect of T-stage (tumour size) (T1: black, T2: red, T3: green). (B)
Different treatment outcomes in patients presenting with single BCCs (black) vs. multiple
tumours (red). (C) Influence of treatment modality (radiotherapy: red, excision: black,
curettage: green, photodynamic therapy: blue).
Basal cell carcinoma recurrence rate
Acta Derm Venereol 87
of the specialized unit (relapse rate 6.4-fold lower, CI
2.4–17.4), when adjusted for the known significant pa-
rameters comprising the type of treatment, localization
of the tumour, and T-stage (Fig. 4).
The drawback of this study is the possible error re-
sulting from the lack of prospective randomization.
However, we were able to identify the risk factors
influencing treatment outcome in a real life situation.
We confirmed that the risk of recurrence depends on
tumour size, localization, choice of treatment modality
and the clinical experience of the treating physician.
Primary T2 tumours relapse twice as frequently and
T3 tumours three times as frequently as T1 tumours.
BCCs in the head region were more difficult to treat
than the tumours on the trunk and lower extremities.
However, we could not confirm the data from other
studies suggesting that the highest relapse rate occurs
after the treatment of BCC in the central face and ears
vs. the lateral portions of the face. In our analysis the
highest recurrence rate occurred for the tumours on the
scalp and, surprisingly, on the upper extremities. An-
other unexpected finding was that the tumours arising
in the context of other skin diseases (such as eczema
or psoriasis) or general immunosuppression were cha-
racterized by better cure rates than those in otherwise
healthy individuals. The role of immunosuppression
in the pathogenesis of BCC is well established and it
is considered that the tumours in immunosuppressed
patients are more difficult to treat (4, 18). The sur-
prisingly good treatment results in this patient group
may be explained by an easier access to dermatological
care providers and thus earlier diagnosis.
Surgical excision and radiotherapy were the moda-
lities associated with the lowest recurrence rates. In
accordance with other results (19) the failure rate of
radiotherapy was slightly higher (OR 1.3) than that
of surgery, both in unselected BCC tumours and the
tumours on the head. The 5-year recurrence rate has
been reported to be in the range 15–20% (20), which
was also the value obtained in this study. Thus, radio-
therapy can be considered as a simpler alternative for
the treatment of BCC, especially in elderly individuals
where the long-term cosmetic outcome is of a lesser
concern than in younger subjects (21). The results also
show that radiotherapy could be the preferred option for
recurrent tumours, providing comparable, if not better,
results than simple excision, with an approximately
60% 5-year cure rate.
Curettage is the most common treatment modality in
Europe, but the results are highly variable. In skilled
hands curettage can approach the efficacy of surgical
excision (1, 15, 22), with cure rate of 90% or higher.
However, for the inexperienced physicians the recur-
rence rates of over 30% are not uncommon, due to a
poor curettage technique or inappropriate selection of
patients (15, 17, 23). Supervision and more intensive
teaching of dermatology residents results in a much
lower recurrence rates (23) which is supported here by the
data showing a greater than 6-fold decrease in relapses
where the BCC were treated in a setting of a specialized
skin cancer unit in our institution. Curettage does not
seem to be an appropriate modality for the treatment of
recurrent tumours and poses a high risk of relapse.
In our centre the efficacy of cryotherapy and PDT
were comparable, but significantly inferior to excision,
curettage or radiotherapy for unselected tumours and
for head tumours. In particular, attempts to treat re-
Fig. 3. Five-year risk of relapse after the treatment of recurrent basal cell
carcinoma (BCC). Treatment outcome of 1090 recurrent BCCs was analysed
as in Fig. 1. Radiotherapy: red, excision: black, curettage: green, photodynamic
therapy: blue, cryotherapy: stippled black.
Fig. 4. Risk of relapse after treatment of basal cell carcinoma (BCC) in
a specialized skin cancer unit (black line) vs. the treatment in a general
dermatology outpatient ward (red line). Cox regression adjusted significant
variables (T-stage, localization, concurrent disease, treatment modality).
F. Saabye Bøgelund et al.
Acta Derm Venereol 87
lapsing BCC with these modalities resulted in only an
approximately 20% 5-year cure rate. PDT is a relatively
new mode of treatment for non-melanoma skin cancer
and actinic keratoses, showing efficacy similar to that
of cryosurgery of BCC (24). Several studies reported
high cure rates for PDT, ranging from 87% to 53% (1,
25). The cure rate may be increased to 89% by perform-
ing the curettage prior to PDT (26). In this study we
included the tumours that have not been curetted before
PDT, which may partially explain the poor results of this
treatment modality. It is however unknown at present
whether PDT offers any additional benefit over a careful
The results from this study may help further to refine
treatment guidelines for BCC and have already led to
the establishment of the skin cancer unit at our insti-
tution. The high-risk BCC comprising larger than T1
tumours localized on the face or scalp should preferably
be treated with excision or radiotherapy. The primary
goal is the achievement of complete remission, since
treatment of relapsing tumours is associated with a very
poor outcome. PDT should primarily be used for small
tumours in the cosmetically important areas and its use
for relapsing BCC cannot be recommended. The data
do not support the use of curettage for the treatment of
relapsing tumours. The results of this study also show
that the treatment of BCC is still an area of a significant
medical need, requiring novel therapeutic modalities
providing higher cure rates at less expense.
Conflict of interest: None to declare
1. Stockfleth E, Sterry W. New treatment modalities for basal cell
carcinoma. Recent Results Cancer Res 2002; 160: 259–268.
2. Bath-Hextall F, Bong J, Perkins W, Williams H. Interven-
tions for basal cell carcinoma of the skin: systematic review.
BMJ 2004; 329: 705.
3. Netscher DT, Spira M. Basal cell carcinoma: an overview
of tumor biology and treatment. Plast Reconstr Surg 2004;
4. Telfer NR, Colver GB, Bowers PW. Guidelines for the
management of basal cell carcinoma. Br J Dermatol 1999;
5. Silverman MK, Kopf AW, Grin CM, Bart RS, levenstein
MJ. Recurrence rates of treated basal cell carcinomas. Part
2: curettage-electrodesiccation. J Dermatol Surg Oncol
1991; 17: 720–726.
6. Helsing P, Kramer P, Haugstvedt A, Aarebrot S, Todal A,
Moseng D, loeb M. Kirurgisk behandling av basalcelle-
karsinomer. [Surgical treatment of basal cell carcinoma].
Tidsskr Nor Laegeforen 2004; 124: 2740–2742.
7. Silverman MK, Kopf AW, Bart RS, Grin CM, levenstein MS.
Recurrence rates of treated basal cell carcinomas. Part 3: sur-
gical excision. J Dermatol Surg Oncol 1992; 18: 471–476.
8. Lovatt TJ, Lear JT, Bastrilles J, Wong C, Griffiths CE,
Samarasinghe V, et al. Associations between ultraviolet
radiation, basal cell carcinoma site and histology, host
characteristics, and rate of development of further tumors.
J Am Acad Dermatol 2005; 52: 468–473.
9. Weimar VM, Ceilley RI, Goeken JA. Aggressive biologic
behavior of basal- and squamous-cell cancers in patients
with chronic lymphocytic leukemia or chronic lymphocytic
lymphoma. J Dermatol Surg Oncol 1979; 5: 609–614.
10. Essers BA, Dirksen CD, Nieman FH, Smeets NW,
Krekels GA, Prins MH, et al. Cost-effectiveness of Mohs
micrographic surgery vs surgical excision for basal cell
carcinoma of the face. Arch Dermatol 2006; 142: 187–194.
11. Smeets NW, Kuijpers DI, Nelemans P, Ostertag JU,
Verhaegh ME, Krekels GA, et al. Mohs’ micrographic
surgery for treatment of basal cell carcinoma of the face
– results of a retrospective study and review of the literature.
Br J Dermatol 2004; 151: 141–147.
12. Smeets NW, Krekels GA, Ostertag JU, Essers BA, Dirksen
CD, Nieman FH, et al. Surgical excision vs Mohs’ micro-
graphic surgery for basal-cell carcinoma of the face: ran-
domised controlled trial. lancet 2004; 364: 1766–1772.
13. lawrence CM. Mohs’ micrographic surgery for basal cell
carcinoma. Clin Exp Dermatol 1999; 24: 130–133.
14. Thissen MR, Neumann MH, Schouten LJ. A systematic
review of treatment modalities for primary basal cell car-
cinomas. Arch Dermatol 1999; 135: 1177–1183.
15. Kopf AW, Bart RS, Schrager D, lazar M, Popkin Gl. Curet-
tage-electrodessication treatment of basal cell carcinomas.
Arch Dermatol 1977; 113: 439–443.
16. Reymann F. Treatment of basal cell carcinoma of the skin
with curettage. II. A follow-up study. Arch Dermatol 1973;
17. Barlow JO, Zalla MJ, Kyle A, DiCaudo DJ, lim KK,
Yiannias JA. Treatment of basal cell carcinoma with curet-
tage alone. J Am Acad Dermatol 2006; 54: 1039–1045.
18. Madan V, Hoban P, Strange RC, Fryer AA, Lear JT. Genetics
and risk factors for basal cell carcinoma. Br J Dermatol
2006; 154 Suppl 1: 5–7.
19. Avril MF, Auperin A, Margulis A, Gerbaulet A, Duvillard
P, Benhamou E, et al. Basal cell carcinoma of the face:
surgery or radiotherapy? Results of a randomized study.
Br J Cancer 1997; 76: 100–106.
20. Zagrodnik B, Kempf W, Seifert B, Muller B, Burg G,
Urosevic M, Dummer R. Superficial radiotherapy for
patients with basal cell carcinoma: recurrence rates, histo-
logic subtypes, and expression of p53 and Bcl-2. Cancer
2003; 98: 2708–2714.
21. Silverman MK, Kopf AW, Gladstein AH, Bart RS, Grin
CM, levenstein MJ. Recurrence rates of treated basal cell
carcinomas. Part 4: X-ray therapy. J Dermatol Surg Oncol
1992; 18: 549–554.
22. Werlinger KD, Upton G, Moore Ay. Recurrence rates of pri-
mary non-melanoma skin cancers treated by surgical excision
compared to electrodessication-curettage in a private derma-
tological practice. Dermatol Surg 2002; 28: 1138–1142.
23. Alexiades-Armenakas M, Ramsay D, Kopf AW. The appropri-
ateness of curettage and electrodesiccation for the treatment
of basal cell carcinomas. Arch Dermatol 2000; 136: 800.
24. Wang I, Bendsoe N, Klinteberg CA, Enejder AM,
Andersson-Engels S, Svanberg S, et al. Photodynamic
therapy vs. cryosurgery of basal cell carcinomas: results
of a phase III clinical trial. Br J Dermatol 2001; 144:
25. Naidenov N, Dencheva R, Tsankov N. Recurrence rate of
basal cell carcinoma after topical aminolevulinic acid-based
photodynamic therapy. Acta Dermatovenerol Croat 2004;
26. Soler AM, Warloe T, Berner A, Giercksky KE. A follow-up
study of recurrence and cosmesis in completely responding
superficial and nodular basal cell carcinomas treated with meth-
yl 5-aminolaevulinate-based photodynamic therapy alone and
with prior curettage. Br J Dermatol 2001; 145: 467–471.