BJD

British Journal of Dermatology

D ER M A T O L O G I C A L S U R GE RY AN D L A S E R S

Sentinel lymph node biopsy and risk factors for predicting metastasis in cutaneous squamous cell carcinoma J.T. Krediet,1 M. Beyer,1 K. Lenz,2 C. Ulrich,1 B. Lange-Asschenfeldt,1 E. Stockfleth1,3 and D. Terhorst1 1

Skin Cancer Center Charit
e, Department of Dermatology and Allergy and 2Institute of Biometry and Epidemiology, Charit
e-Universit€atsmedizin Berlin, 10117 Berlin, Germany 3 Clinic for Dermatology, Venerology and Allergology, St. Josef-Hospital Bochum – University Hospital of the Ruhr-Universit€at Bochum, 44791 Bochum, Germany

Summary Correspondence Dorothea Terhorst. E-mail: [email protected]

Accepted for publication 24 October 2014

Funding sources None.

Conflicts of interest None declared. DOI 10.1111/bjd.13508

Background Cutaneous squamous cell carcinoma (cSCC) is a common cancer capable of metastasis. Sentinel lymph node biopsy (SLNB) may be a valuable adjunct for patients with cSCC at high risk for metastasis. However, data on risk factors for metastasis and results of SLNB from patients with cSCC are limited. Objectives To evaluate risk factors for metastasis in patients with cSCC in a large cohort study with long-term follow-up and to determine the value of SLNB. Methods We retrospectively analysed all records of patients who underwent excision of cSCC between January 2005 and August 2009 at a tertiary referral centre. In total, 143 patients were included in the cohort, including 17 patients with SLNB and a follow-up time of ≥ 24 months. Results Tumour thickness > 4 mm and recurrent cSCC were strongly associated with metastatic disease. All metastases in this cohort occurred within 24 months of follow-up. SLNB showed a low sensitivity with regard to the development of metastasis. Six of 17 patients developed metastatic disease despite a negative SLNB. Conclusions Patients with risk factors (cSCC with a tumour thickness > 4 mm or recurrent disease) may develop metastases within the first 2 years despite a negative SLNB. Therefore these patients should be closely monitored during the follow-up. Based on our data SLNB does not provide diagnostic value for patients with cSCC.

What’s already known about this topic?

• • •

Cutaneous squamous cell carcinoma (cSCC) is a common cancer capable of metastasis. Different risk factors such as tumour thickness and tumour diameter have been discussed for determining the risk of metastasis. Sentinel lymph node biopsy (SLNB) may be a valuable adjunct for patients with cSCC at high risk for metastases; however, it is currently described only in case reports and small case series with limited follow-up.

What does this study add?

• • •

© 2014 British Association of Dermatologists

The risk for metastases from a cSCC is associated with tumour thickness > 4 mm and tumour recurrence. All metastases occurred within 2 years after excision. SLNB seems to have a low sensitivity for metastases of cSCC. Despite a negative SLNB, some patients developed metastatic disease, underlining the necessity of close follow-up of high-risk patients in the first 2 years after excision, regardless of SLNB status.

British Journal of Dermatology (2015) 172, pp1029–1036

1029

1030 SLNB and risk factors for cutaneous SCC, J.T. Krediet et al.

Cutaneous squamous cell carcinomas (cSCCs) are among the most common human tumours that are capable of metastasizing. The incidence of cSCC is increasing due to an ageing population and increasing ultraviolet exposure. Furthermore, immunosuppression is a risk factor, as a high incidence of cSCC has been demonstrated in organ transplant recipients.1 cSCCs metastasize in 3–6% of patients, primarily via the lymph vessels.2,3 This percentage may rise to 40% in high-risk cSCC.4–6 However, factors indicating a higher risk of metastasis in cSCC have been controversially discussed over the last few decades. In several studies different histological and clinical features have shown association with metastatic and/or recurrent disease. Tumour thickness ≥ 4 mm, recurrent cSCC, poorly differentiated tumours and immunodeficiency have been discussed as risk factors for metastatic disease. Several other factors such as perineural invasion, desmoplastic growth, dedifferentiation, horizontal tumour size and tumour location on the ear, lip or non-sun-exposed areas have also been associated with metastatic or recurrent disease.7–9 Until now, most studies have declared patients as being at ‘high risk’ if either one, two or more risk factors are present.10,11 The current TNM (tumour, nodes, metastasis) classification for cSCC of the Union for International Cancer Control (2009) and American Joint Committee on Cancer (2010) includes horizontal tumour diameter and infiltration of extradermal structures: T1, ≤ 2 cm maximal diameter; T2, > 2 cm; T3, deep infiltration; and T4, infiltration of the base of the skull or the spine.12,13 The recently published update of the guideline of the German Society of Dermatology additionally recommends the evaluation of tumour thickness to provide a better estimate of the risk of metastasis, according to the results of a prospective analysis of risk factors determining the prognosis of cSCC.2,14 Excision of the sentinel lymph node (SLN), defined as the first tumour-draining lymph node, designated as sentinel lymph node biopsy (SLNB), has been established as a safe and effective tool for guiding treatment and determining prognosis in malignant melanoma.15 There is some evidence that the performance of SLNB followed by lymph node dissection in patients with melanoma might change the course of the disease; that is, the SLNB may have a therapeutic effect.16 Furthermore, the SLNB has become a useful procedure for other nonmelanoma skin cancers, especially Merkel cell carcinoma.17 It allows an exact nodal staging and may lead to further therapy (e.g. radiation, radical lymphadenectomy, extended lymphadenectomy). The recently published guideline for the management of cSCC in high-risk patients from the French dermatology association tends to encourage SLNB within the context of clinical trials and evaluation studies.18 The German guideline states that an SLNB can be considered in cSCC with a higher risk of metastasis.2 Despite these recommendations there are no satisfying datasets analysing the role of SLNB in cSCC. SLNB in patients with high-risk cSCC has so far been reported only in case reports or case series.4,11,19–24 The lack of clearly defined indications for British Journal of Dermatology (2015) 172, pp1029–1036

the use of SLNB procedures has led to studies of heterogeneous groups of patients with a varying number of risk factors.4,19–24 Schmitt et al. reanalysed the case series already published in the literature using an alternative staging system and found that most cSCCs associated with positive SLNB findings occurred in T2 lesions > 2 cm in diameter.25 All of the published studies used only a short follow-up time.4,19–24 The aim of this study was to evaluate factors in patients with cSCC that increase the risk of metastasis in a large cohort study with long-term follow-up. And we specifically defined the value of SLNB regarding its sensitivity, negative predictive value and impact on prognosis and outcome.

Patients and methods Total cohort: analysis of risk factors for metastasis and local recurrence Patient records from the Skin Cancer Center at the Department of Dermatology and Allergy, Charit
e-Universit€atsmedizin Berlin (Berlin, Germany) were screened for patients with cSCC using the clinical tumour documentation software GTDS (Giessen Tumour Documentation System; University Giessen, Germany). All 446 records of patients who underwent surgical removal of a skin tumour suspicious of being a primary cSCC in the period January 2005 to August 2009 were examined further. Only patients whose diagnosis had been established histologically and immunohistochemically were further included in the study. Patients who had tumour dissemination at the time of excision or missing clinical data concerning past medical history were excluded from the study. Furthermore, only patients who had at least one follow-up visit were included. The final cohort selected in this way consisted of 143 patients. From this cohort the following data were collected: age at the time of diagnosis, sex and immunosuppression (Table 1). The following data were collected regarding the tumour: TNM stage, and tumour differentiation, site, horizontal diameter and vertical thickness. The occurrence of metastatic disease and/or local regional recurrence and various risk factors until August 2013 associated with these events were examined in this cohort. Local regional recurrence was defined as the development of tumour at the same site where the earlier tumour was located. Metastasis occurred in the draining lymph nodes and/or distant organs or surrounding tissue, such as bone. Risk factors and definition of high risk The following parameters were classified as risk factors: tumour diameter ≥ 20 mm in greatest dimension; risk tumour site of ear, lip or non-sun-exposed areas (trunk, lower limb); recurrent tumour; immunosuppression; moderate or poor differentiation; and tumour thickness ≥ 4 mm or Clark Level ≥ 4. Patients with two or more risk factors were classified as high-risk patients for metastasis. From patients who © 2014 British Association of Dermatologists

SLNB and risk factors for cutaneous SCC, J.T. Krediet et al. 1031 Table 1 Characteristics of patients with squamous cell carcinoma

Sex, male/female Age (years), mean (range) Metastatic disease, yes/no Immunosuppression, yes/no Tumour diameter (cm2) Mean Median (range) Tumour thickness (mm) Mean Median (range) Tumour staging T1 T2 T3 T4 NA Nodes staging N0 N1 Metastasis staging M0 M1 Tumour site Upper limb Lower limb Trunk Ear Lip Nose Rest of head Tumour differentiation Good Moderate Poor NA Number of high-risk features 0 1 2 3 4 High-risk status (yes/no) Follow-up time (months), mean (range)

Total cohort (n = 143)

Metastatic (n = 17)

Nonmetastatic (n = 126)

Recurrence (n = 8)

No recurrence (n = 135)

111/32 73 (24–102) 17/126 35/108

15/2 70 (24–84)

96/30 73 (29–102)

6/11

30/96

7/1 73 (66–83) 6/2 2/6

104/31 73 (24–102) 11/124 34/101

4
6 2
0 (0
1–100)

4
9 4 (1–25)

4
5 1
5 (0
1–100)

2
6 3
0 (0
5–4
0)

4
7 2
0 (0
1–100)

3
7 3
0 (0
5–8
0)

4
7 5
1 (1
2–7
6)

3
3 3
0 (0
5–8
0)

4
1 4
5 (1
2–7
0)

3
6 3
0 (0
5–8
0)

94 16 4 3 26

11 1 0 3 2

83 15 4 0 24

5 0 0 1 2

89 16 4 2 24

132 11

6 11

126 –

5 3

127 8

136 7

10 7

126 –

5 3

131 4

10 18 9 14 14 6 72

1 2 1 4 1 0 8

9 16 8 10 13 6 64

0 0 0 1 1 0 6

10 18 9 13 13 6 66

26 23 6 88

6 7 1 3

20 16 5 85

2 5 0 1

24 18 6 87

32 53 37 14 7 58/85 25 (1–98)

0 1 5 8 3 16/1 26 (2–62)

32 52 32 6 4 43/83 25 (1–98)

0 2 1 4 1 6/2 34 (13–62)

32 51 36 10 6 52/83 24 (1–98)

NA, not available.

underwent SLNB and/or developed metastases, further data were collected: SLN region, performance of complete lymph node dissection, SLN status, metastasis and recurrence of the primary tumour.

underwent SLNB and who developed metastases regardless of their follow-up time. We thus conducted a case series comparing 17 patients (Table 2). Diagnostic and surgical procedures

Subcohort: analysis of risk factors and sentinel lymph node in patients with at least 24 months’ follow-up We analysed the risk profile in patients who underwent SLNB with a follow-up time of at least 24 months, or patients who © 2014 British Association of Dermatologists

SLNBs have been suggested in patients at high risk, thus presenting at least two risk factors. However, in some cases SLNB was suggested to patients who had only one risk factor but a very pronounced primary tumour, for example with strong British Journal of Dermatology (2015) 172, pp1029–1036

British Journal of Dermatology (2015) 172, pp1029–1036

M/67 F/82

M/68

11 12

13

Lower limb Ear Head Head

T2N1M1 T2N0M1 X T2N1M0 X

T2N1M0 X

T2N0M1 X

T2N1M0 X T2N1M0 X

T2N0M0 X T2N0M0 X T2N0M0 X T4N1M0

T2N0M0 X

T3N0M0 X

X X

X

X X X

X

X

X

X

a

X X

X

X

X

X

X X

X

X

X X X

X

X

X X X

Negative Positive

Negative Negative Negative Positive

Negative

Negative

Negative Negative

Negative Negative

Occipital Cervical Cervical

Negative Negative Negative

Pre-/ Negative retroauricular Inguinal Negative

Nuchal Axillar

Cervical Axillary Axillary Inguinal

Inguinal

Inguinal

Cervical Inguinal

NA Inguinal

No No No 1st 2nd 1st 2nd

– – – No – Yes, positive –

1st 1st 1st

No

–

– – –

No

–

1st

No No

– –

–

No No

– –

Yes Yes No

No

Yes

Yes No

No No No No

No

No

Yes No

No No

SLN biopsy Total lymph result node dissection Metastasisa Recurrence

M, male; F, female; TNM, tumour, nodes, metastasis; NA, not available. 1st, within first year after operation; 2nd, within second year after operation.

M/79 M/69 M/83

F/66 M/51 M/55 M/42

7 8 9 10

15 16 17

M/72

6

M/24

F/76

5

14

M/81 F/60

3 4

X X

Tumour Tumour site: Moderately Tumour diameter ear, lip, non-sun- Recurrent Immunoor poorly thickness ≥ 4 mm ≥ 2 cm exposed area tumour suppression differentiated or Clark level ≥ 4 SLN

T2N0M0 X T2N0M0 X

T2N0M0 T1N0M0

M/66 M/65

1 2

Lip Lower limb Ear Lower limb Lower limb Lower limb Head Trunk Trunk Lower limb Head Upper limb Head

TNM

Sex/age No. (years) Site

Table 2 Sentinel lymph node (SLN)-negative patients with follow-up time ≥ 24 months

1032 SLNB and risk factors for cutaneous SCC, J.T. Krediet et al.

© 2014 British Association of Dermatologists

SLNB and risk factors for cutaneous SCC, J.T. Krediet et al. 1033

induration or ulceration. Another such reason could be that the primary tumour evolved over pre-existing skin damage such as a burn injury. At 16–18 h before excision of the cSCC, 1 mL of technetium-99m-labelled nanocolloids was injected subcutaneously at the tumour margins. At 2–4 h after injection, lymphoscintigraphy was performed to obtain dynamic flow and static images. The skin projection of the SLN was marked with ink. Directly before excision, vital blue dye (patent blue V) was injected intracutaneously around tumours located on the trunk. SLNB was then performed using a handheld gamma camera to detect the SLN. In cases where the primary tumour was located close to the SLN, the tumour was removed first in order to prevent ‘shine-through’. The safety margin for the excision of the primary tumour was situation dependent in the head and neck area and was 1 cm on the rest of the body. All excision margins were controlled micrographically (three-dimensional histology). Harvested SLNs were formalin fixed, serially sectioned (5–7 lm) and stained with haematoxylin and eosin. Immunohistochemical staining with antibodies to pan-cytokeratin was performed when no tumour cells were seen under haematoxylin and eosin staining. Statistical analysis All data were collected via Microsoft Excel, and statistical analysis was performed using SPSS v.21 software (IBM, Armonk, NY, U.S.A.). Clinical and histological characteristics of the patients were summarized with the use of descriptive statistics. The sensitivity of the SLNB was calculated from the number of true positives (patients who had metastatic disease and were SLN positive) divided by the number of patients who had metastatic disease and had SLNB. The negative predictive value of the SLNB was computed by comparing the number of patients with true negative SLNBs with the overall number of SLNB-negative patients. Lymph node metastasis in the same lymph node region that gave a negative SLNB was defined as a false-negative SLN. The Cox proportional hazards regression model was used to analyse the prognostic value of the high-risk features present at the time of diagnosis. Univariate hazard ratios were calculated with 95% confidence intervals. The primary end point was time to metastasis, defined as the time from the date of diagnosis of the primary tumour to the date of diagnosis of metastasis.

Results Occurrence of metastases and association with risk factors In total 143 patients were enrolled in the study, and their characteristics are summarized in Table 1. Seventeen patients developed metastatic disease, and eight patients had local recurrence (5
6%). Metastases were found in 14 cases in the first and in three cases in the second year of follow-up. No metastases were detected thereafter. Overall 94% (n = 16) of © 2014 British Association of Dermatologists

the metastatic patients were high-risk patients. Univariate Cox regression analysis showed a statistically significant correlation for metastasis-free survival with tumour thickness, tumour size, recurrent tumours and patients with high-risk status (Table 3). Table 4 further stratifies the patients at high risk and low risk, and shows the most predominant risk factors. Sensitivity and negative predictive value of sentinel lymph node biopsy All metastases occurred within 24 months of excision. To evaluate the value of the SLNB, we decided to include in the further analysis only patients with a follow-up time of at least 24 months, as well as patients who developed metastatic disease regardless of their follow-up time. Out of 35 patients who underwent SLNB, 17 patients fulfilled this criterion (Table 2). SLNB was positive in two patients, one of whom underwent radical lymphadenectomy, and the histopathological evaluation demonstrated lymph node metastasis in other lymph nodes. Six of 15 patients with a negative SLN developed metastatic disease during follow-up. All of these six patients had at least two risk factors. One of the six developed lymph node metastasis in the lymph node region where the SLN was localized; the SLNB was thus a false negative according to the definition.24,26 Two patients had lymph node metastasis in a lymph node region different from the SLNB compartment. The remaining three patients developed metastases outside of the lymph nodes: one patient with a negative Table 3 Univariate Cox regression models for metastasis-free survival

Risk features

Patients with metastatic disease, n (%) HR (95% CI)

Tumour thickness is high-risk feature Yes (n = 20) 11 (55) No (n = 22) 9 (41) Tumour size is high-risk feature Yes (n = 48) 9 (19) No (n = 68) 39 (57) Tumour is recurrent Yes (n = 9) 5 (56) No (n = 134) 4 (3) Tumour site is high-risk feature Yes (n = 55) 8 (15) No (n = 85) 47 (55) Immunosuppression Yes (n = 36) 6 (17) No (n = 107) 30 (28) Tumour grading is high-risk feature Yes (n = 29) 8 (28) No (n = 26) 21 (81) Patient is at high risk Yes (n = 58) 16 (28) No (n = 85) 42 (49)

P-value

4
31 (1
20–15
49)

0
025

5
76 (1
53–21
78)

0
025

6
52 (2
22–19
12)

0
001

1
58 (0
61–4
13)

0
35

1
31 (0
49–3
56)

0
59

1
62 (0
56–4
74)

0
38

23
37 (3
09–176
65) 0
002

HR, hazard ratio; CI, confidence interval.

British Journal of Dermatology (2015) 172, pp1029–1036

1034 SLNB and risk factors for cutaneous SCC, J.T. Krediet et al. Table 4 Characteristics of patients with squamous cell carcinoma (SCC): high- and low-risk patients High risk (n = 58) Low risk (n = 85) Sex, male/female 46/12 Age (years), mean (range) 71 (24–92) Immunosuppression is a high-risk feature Yes 25 No 33 Tumour site is a high-risk feature Yes 37 No 21 Tumour size is a high-risk feature Yes 32 No 16 NA 10 Tumour thickness is a high-risk feature Yes 19 No 11 NA 28 Tumour is recurrent SCC as high-risk feature Yes 5 No 53 Grading is high-risk feature Yes 26 No 13 NA 19 Tumour size (cm2) Mean 7
5 Median (range) 4
0 (0
1–100) Tumour thickness Mean 4
2 Median (range) 4
2 (0
5–8
0) Tumour staging T1 32 T2 12 T3 3 T4 3 NA 8 Nodes staging N0 48 N1 10 Metastasis staging M0 52 M1 6 Tumour site Upper limb 2 Lower limb 14 Trunk 8 Ear 9 Lip 6 Nose 1 Rest of the head 18 Tumour differentiation Good 13 Moderate 20 Bad 6 NA 19 Metastatic disease Yes 16 No 42

65/20 74 (29–102) 11 74 18 67

Table 4 (continued) High risk (n = 58) Low risk (n = 85) Number of high-risk features 0 0 1 0 2 37 3 14 4 7 Follow-up time 25 (1–93) (months), mean (range)

32 53 0 0 0 24 (1–98)

NA, not available.

16 52 17

cervical SLN developed lung metastasis, one patient with a primary tumour on the forehead showed metastasis in the orbit, and one patient with a negative cervical SLN developed subcutaneous metastases. The sensitivity and negative predictive value of the SLNB were calculated to be 66% and 93%, respectively. No significant association was found between SLNB status and the anatomical region of the SLN (v2-test, P = 0
81) or the high-risk status of the patients (v2-test, P = 0
49).

1 11 73 4 81 3 13 69

Metastases in patients without sentinel lymph node biopsy

2
5 1
0 (0
2–30) 2
5 2
6 (0
6–7
0) 62 4 1 0 18

Of the patients who did not undergo SLNB, nine developed metastatic disease. Five patients developed metastases of draining lymph nodes. In three patients, metastases to subcutaneous structures were observed, such as osseous and subcutaneous regions. One patient developed metastases of both the lymph nodes and inner organs such as lungs, liver and pancreas. Table 5 depicts all patients with metastasis who did not undergo SLNB.

Discussion

84 1 84 1 8 4 1 5 8 5 54 13 3 0 69 1 84 (continued)

British Journal of Dermatology (2015) 172, pp1029–1036

Our study cohort shows typical characteristics for patients with cSCC regarding sex, age, tumour localization, tumour diameter and histological features such as tumour thickness and differentiation. These are comparable with the features of the study population of Brantsch et al.14 However, we have a higher proportion of immunosuppressed patients in our cohort (24
5%) than the 5% reported by Brantsch et al.14 or the 14
5% in the cohort of Schmults et al.9 This is due to the high number of organ transplant recipients seen at our department, and reflects a referral bias to a specialized centre. The proportion of patients who had local recurrence was also slightly higher in our cohort (5
6%) than in those reported by Brantsch et al.14 (3%) and Schmults et al. (4
6%).9 This unique nature of our cohort might be one of the reasons for a higher percentage of extranodal metastases (35%) compared with the reports by Schmults et al. (10%) and Brantsch et al. (0%). Our study clearly supports the role of recurrent disease, tumour size ≥ 20 mm in the greatest dimension and tumour thickness ≥ 4 mm or Clark level ≥ 4 as high-risk factors for © 2014 British Association of Dermatologists

X X X M/81 9

Ear

T2N1M0

X

X

X X X X

© 2014 British Association of Dermatologists

M, male; F, female; TNM, tumour, nodes, metastasis. a1st, within first year after operation; 2nd, within second year after operation.

X X X X X X X X

X X X

X

T2N0M1 T2N1M0 T2N0M1 T4N1M0 T4N0M1 T2N1M0 T2N1M0 T2N1M1 M/83 M/71 M/68 F/81 M/78 M/84 M/84 M/50 1 2 3 4 5 6 7 8

Head Head Head Lip Head Ear Ear Trunk

TNM No.

Site

1st

Subcutaneous/bone Lymph node submandibular Subcutaneous/orbit Cervical lymph node Subcutaneous Cervical lymph node Cervical lymph node Lung, liver, pancreas, subcutaneous and lymph node Cervical lymph node 1st 1st 1st 2nd 1st 1st 1st 1st X X X X X X X

Recurrent tumour Site: ear, lip, non-sunexposed area Tumour diameter ≥ 2 cm Sex/age (years)

Table 5 All patients with metastasized tumour who did not undergo sentinel lymph node biopsy

Immunosuppression

Moderately or poorly differentiated

Thickness ≥ 4 mm or Clark level ≥ 4

Metastasisa

Site of metastasis

SLNB and risk factors for cutaneous SCC, J.T. Krediet et al. 1035

the development of metastases. The analysis of risk factors was limited by the fact that not all information about the presence or absence of risk factors was available for all patients. In consequence this may have led to an underestimation of the risk status of patients who were classified as low risk, and the metastatic rate in high-risk patients may have been overestimated. However, with our cohort including a high percentage of immunosuppressed patients we can confirm the results of earlier studies with regard to both important risk factors and metastatic rates for high-risk cSCC.4–6,9,14 A false-negative SLNB is defined as a SLN that shows no metastases in the histopathological evaluation and is taken from a lymph node compartment in which a metastasis is detected later on.26 However, other groups have defined a false-negative result as a tumourfree SLNB with metastatic disease at any body site.23 Using the first, stricter, definition we had one of 17 patients with a false-negative SLN (6%). However, five more patients developed metastatic disease at distant body sites despite a negative SLN. In six patients (35%) a negative SLN was thus not associated with a good prognosis, resulting in a low sensitivity for SLNB in cSCCs. In the literature, some studies have shown no false-negative SLNs,11,20 whereas others had a rate of false-negative SLNBs around 10%23,24 or 33%.10 Also, rates for positive SLNs vary from 0%23 to over 4
5%22 up to 44%.11 Our SLN positive rate of 12% (two out of 17 patients) was within this range. All of these reports have shortcomings, such as small numbers of patients, which lead to large sample variation and might account for some of the divergence. Another explanation could be the different inclusion criteria, as no clear-cut guidelines exist on when to carry out an SLNB. Furthermore these studies report only short follow-up times and have no control groups. Also, our study has shortcomings such as limited generalizability due to the high number of immunosuppressed patients in our cohort. Large, prospective studies are needed for final validation of the value of SLNB in SCC. cSCCs are located mostly on the head, where it might be difficult to detect an SLN, although all of the surgeons performing the SLNB in our study were very experienced. Compared with other sites of the body, the head and neck area has some special features such as the complex anatomy of the lymph system, many structures at risk and frequent proximity of the primary lesion to the lymph node basin. The SLN might also be hidden in some deeper structures.27,28 In our study, the false-negative SLN was located in the head area, whereas the two patients with a positive SLN had their primary tumour on the limbs. So these data could suggest that patients with high-risk cSCCs on the extremities might profit more than patients with head and neck cSCC. In 85% of metastatic cSCCs the disease spreads to the lymph nodes, and only then can SLNB be successful in finding metastasis. In 15% of cases the tumours bypass the regional lymph node and spread directly to distant organs by haematogenous metastasis. This seems to have been the case for the patient who developed systemic metastatic disease without any sign of regional lymph node involvement. British Journal of Dermatology (2015) 172, pp1029–1036

1036 SLNB and risk factors for cutaneous SCC, J.T. Krediet et al.

In our study, all metastases were detected within the first 2 years after excision, similar to previous studies.14,22 We recommend a close follow-up of high-risk patients especially in the first 2 years at 3-month intervals, both by clinical investigation and by ultrasound. Ultrasonography is harmless and cheap and seems to detect early metastatic nodes more reliably than palpation or more expensive procedures, such as positron emission tomography and computed tomography scans.29 Given the often advanced age and reduced state of health of patients with cSCC, the minimally invasive alternative of an ultrasound-guided fine-needle aspiration of suspicious lymph node lesions should also be discussed. Such follow-up schedules have been evaluated and established for patients with malignant melanoma,30,31 and patients with early detection of lymph node metastases have a significantly better prognosis than those with more advanced nodal disease,32 strongly arguing in favour of a structured follow-up process.

References 1 Terhorst D, Drecoll U, Stockfleth E, Ulrich C. Organ transplant recipients and skin cancer: assessment of risk factors with focus on sun exposure. Br J Dermatol 2009; 161:85–9. 2 Breuninger H, Eigentler T, Bootz F et al. Brief S2k guidelines – cutaneous squamous cell carcinoma. J Dtsch Dermatol Ges 2013; 11:37–45. 3 Czarnecki D, Staples M, Mar A et al. Metastases from squamous cell carcinoma of the skin in southern Australia. Dermatology 1994; 189:52–4. 4 Weisberg NK, Bertagnolli MM, Becker DS. Combined sentinel lymphadenectomy and Mohs micrographic surgery for high-risk cutaneous squamous cell carcinoma. J Am Acad Dermatol 2000; 43:483–8. 5 Clayman GL, Lee JJ, Holsinger FC et al. Mortality risk from squamous cell skin cancer. J Clin Oncol 2005; 23:59–65. 6 Mullen JT, Feng L, Xing Y et al. Invasive squamous cell carcinoma of the skin: defining a high-risk group. Ann Surg Oncol 2006; 13:902–9. 7 Martinez JC, Cook JL. High-risk cutaneous squamous cell carcinoma without palpable lymphadenopathy: is there a therapeutic role for elective neck dissection? Dermatol Surg 2007; 33:410–20. 8 National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology: Basal Cell and Squamous Cell Skin Cancers. Fort Washington, PA: NCCN, 2013. 9 Schmults CD, Karia PS, Carter JB et al. Factors predictive of recurrence and death from cutaneous squamous cell carcinoma: a 10year, single-institution cohort study. JAMA Dermatol 2013; 149:541–7. 10 Kwon S, Dong ZM, Wu PC. Sentinel lymph node biopsy for highrisk cutaneous squamous cell carcinoma: clinical experience and review of literature. World J Surg Oncol 2011; 9:80. 11 Reschly MJ, Messina JL, Zaulyanov LL et al. Utility of sentinel lymphadenectomy in the management of patients with high-risk cutaneous squamous cell carcinoma. Dermatol Surg 2003; 29:135–40. 12 Edge SB, Byrd DR, Compton CC et al. eds. AJCC Cancer Staging Manual. New York: Springer, 2009. 13 Sobin LH, Gospodarowicz MK, Wittekind C. TNM Classification of Malignant Tumours. Chichester: John Wiley & Sons, 2009. 14 Brantsch KD, Meisner C, Sch€ onfisch B et al. Analysis of risk factors determining prognosis of cutaneous squamous-cell carcinoma: a prospective study. Lancet Oncol 2008; 9:713–20.

British Journal of Dermatology (2015) 172, pp1029–1036

15 Wong SL, Balch CM, Hurley P et al. Sentinel lymph node biopsy for melanoma: American Society of Clinical Oncology and Society of Surgical Oncology joint clinical practice guideline. J Clin Oncol 2012; 30:2912–18. 16 Satzger I, Meier A, Hoy L et al. Sentinel node dissection delays recurrence and prolongs melanoma-related survival: an analysis of 673 patients from a single center with long-term follow-up. Ann Surg Oncol 2011; 18:514–20. 17 Maza S, Trefzer U, Hofmann M et al. Impact of sentinel lymph node biopsy in patients with Merkel cell carcinoma: results of a prospective study and review of the literature. Eur J Nucl Med Mol Imaging 2006; 33:433–40. 18 Bonerandi JJ, Beauvillain C, Caquant L et al. Guidelines for the diagnosis and treatment of cutaneous squamous cell carcinoma and precursor lesions. J Eur Acad Dermatol Venereol 2011; 25(Suppl. 5):1–51. 19 Civantos F Jr, Zitsch R, Bared A, Amin A. Sentinel node biopsy for squamous cell carcinoma of the head and neck. J Surg Oncol 2008; 97:683–90. 20 Matthey-Gie ML, Boubaker A, Letovanec I et al. Sentinel lymph node biopsy in nonmelanoma skin cancer patients. J Skin Cancer 2013; 2013:267474. 21 Michl C, Starz H, Bachter D, Balda BR. Sentinel lymphonodectomy in nonmelanoma skin malignancies. Br J Dermatol 2003; 149:763–9. 22 Renzi C, Caggiati A, Mannooranparampil TJ et al. Sentinel lymph node biopsy for high risk cutaneous squamous cell carcinoma: case series and review of the literature. Eur J Surg Oncol 2007; 33:364–9. 23 Sahn RE, Lang PG. Sentinel lymph node biopsy for high-risk nonmelanoma skin cancer. Dermatol Surg 2007; 33:786–92. 24 Wagner JD, Evdokimow DZ, Weisberger E et al. Sentinel node biopsy for high-risk nonmelanoma cutaneous malignancy. Arch Dermatol 2004; 140:75–9. 25 Schmitt AR, Brewer JD, Bordeaux JS, Baum CL. Staging for cutaneous squamous cell carcinoma as a predictor of sentinel lymph node biopsy results: meta-analysis of American Joint Committee on Cancer criteria and a proposed alternative system. JAMA Dermatol 2014; 150:19–24. 26 Nowecki ZI, Rutkowski P, Nasierowska-Guttmejer A, Ruka W. Survival analysis and clinicopathological factors associated with falsenegative sentinel lymph node biopsy findings in patients with cutaneous melanoma. Ann Surg Oncol 2006; 13:1655–63. 27 Gyorki DE, Boyle JO, Ganly I et al. Incidence and location of positive nonsentinel lymph nodes in head and neck melanoma. Eur J Surg Oncol 2014; 40:305–10. 28 Patuzzo R, Maurichi A, Camerini T et al. Accuracy and prognostic value of sentinel lymph node biopsy in head and neck melanomas. J Surg Res 2014; 187:518–24. 29 Hinz T, Voth H, Ahmadzadehfar H et al. Role of high-resolution ultrasound and PET/CT imaging for preoperative characterization of sentinel lymph nodes in cutaneous melanoma. Ultrasound Med Biol 2013; 39:30–6. 30 Livingstone E, Windemuth-Kieselbach C, Eigentler TK et al. A first prospective population-based analysis investigating the actual practice of melanoma diagnosis, treatment and follow-up. Eur J Cancer 2011; 47:1977–89. 31 Garbe C, Paul A, Kohler-Sp€ath H et al. Prospective evaluation of a follow-up schedule in cutaneous melanoma patients: recommendations for an effective follow-up strategy. J Clin Oncol 2003; 21:520–9. 32 Andruchow JL, Veness MJ, Morgan GJ et al. Implications for clinical staging of metastatic cutaneous squamous carcinoma of the head and neck based on a multicenter study of treatment outcomes. Cancer 2006; 106:1078–83.

© 2014 British Association of Dermatologists