Invasive squamous cell carcinoma: comparison ... - Wiley Online Library

CED

Original article

Clinical and Experimental Dermatology

Invasive squamous cell carcinoma: comparison of differentiation grade and tumour depth by anatomical site in 1666 tumours J. H. Pyne,1 E. Barr,1 E. Myint,1 S. P. Clark,1 M. David1 and R. Na2 1

School of Medicine, University of Queensland, Brisbane, Australia; and 2Prince of Wales Clinical School, University of New South Wales, Sydney, Australia

doi:10.1111/ced.13222

Summary

Background. Invasive squamous cell carcinomas (SCCs) presents with different grades of differentiation and depths of invasion. Aim. To compare the grade of differentiation, tumour diameter and tumour depth by anatomical site in invasive SCC. Methods. Retrospective clinical and histopathological data on consecutive cases of SCC came from a clinic in Sydney, Australia were assessed. A multinomial logistic regression model was applied to compare grades of differentiation by age, sex, anatomical sites, and histological tumour maximum diameter and depth. Results. In total, 1666 SCCs were identified, including 82.1% (n = 1367) welldifferentiated, 13.3% (n = 222), moderately differentiated and 4.6% (n = 77) poorly differentiated SCCs. Patients with poorly differentiated tumours were more likely to be older and male (both P < 0.001). The most common site for poor differentiation was the scalp in men (n = 12; 15.6%) and the cheek or chin in women (n = 7; 9.1%). In the multivariate model, compared with well-differentiated SCC, older age was significantly associated with poorly and moderately differentiated SCC (P < 0.01 and P = 0.02, respectively). Larger tumour diameters were related to poor differentiation (P = 0.03). Ear, forehead and chest sites had increased tumour depth and poor differentiation. Conclusions. This study found increased rates of poorly differentiated SCC on the forehead and cheek for both sexes, while men displayed increased rates of poorly differentiated SCC on the bald scalp and the ears. Tumour diameter and depth increased as tumours varied from well-differentiated to moderately differentiated and from moderately differentiated to poorly differentiated. An increase in depth and increased prevalence of poorly differentiated tumours were found on the ears for men and on various facial sites for both sexes.

Introduction Over recent decades, the global incidence of cutaneous SCC has been increasing.1 Chronic ultraviolet light exposure is thought to be the predominant driver of this increased incidence of SCC.2 The majority of invasive SCCs usually begin as SCC in situ3 and arise from Correspondence: Dr John Hamilton Pyne, School of Medicine, University of Queensland, Brisbane, Australia E-mail: [email protected] Conflict of interest: the authors declare that they have no conflicts of interest. Accepted for publication 31 December 2016

ª 2017 British Association of Dermatologists

a precursor actinic keratosis.4 A variety of other aetiological risk factors can be associated with invasive SCC.3 Sequential SCCs on the same patient occur at a higher incidence than an isolated SCC in a comparable general population.5 Organ transplantation recipients often have an escalating incidence of cutaneous SCC with increasing years post-transplantation.6,7 recipients with lighter Fitzpatrick skin type also have an increased risk for SCC following solid organ transplantation.8 Several factors are recognized as adverse prognostic indicators for SCC, including increased tumour diameter9 and depth,10,11 poor differentiation,12 perineural invasion,13 immunosuppression,14 invasion

Clinical and Experimental Dermatology (2018) 43, pp3–10

3

Invasive SCC: comparison of differentiation grade and tumour depth by anatomical site J. H. Pyne et al.

into cranial bone15 and tumour location at various anatomical sites.9 The different grades of differentiation in SCC were originally described by Broders.16 Tumours may present with one or more grades of differentiation within the total tumour mass. Current convention uses subjective17 histopathological criteria to grade SCC along a spectrum, as well-differentiated, moderately differentiated or poorly differentiated tumours. Poor differentiation indicates a shift towards a higher risk of adverse outcomes. Compared with well-differentiated SCC, poorly differentiated SCC has an increased risk for perineural invasion,18 metastasis9 and death.9,19,20 Additional high-risk histopathological features of SCC include desmoplasia, tumour-induced ulceration, endophytic growth and a high mitotic index.21 A recent study reported that the clinical and dermoscopy features of SCC vary with the grade of differentiation.22 Poorly differentiated SCC were more likely to be flat and pink to red compared with welldifferentiated SCC. Dermoscopy identified an increase in incidence of branching and serpentine vessels in the shift from well-differentiated to poorly differentiated SCC23 and an increase in depth of both moderately and poorly differentiated SCC.23 The purpose of this study was to examine variations in the grade of differentiation within invasive SCC, comparing age, sex, maximum histological tumour diameter and depth of invasion by anatomical site. Knowledge of the anatomical locations with a higher incidence of potentially higher-risk SCC may facilitate the early detection of these SCCs.

Methods The University of Queensland provided ethics approval, and informed consent was obtained from each participant. This retrospective cohort study was conducted in a single designated skin cancer practice in Sydney, Australia, from 2009 to 2015. Clinical and histopathological data were retrospectively collected, including patient age and sex, tumour size, anatomical site, invasion depth and grade of differentiation. All cases were consecutive and excised with a 1–2 mm minimum dermoscopy-guided margin in a vertical plane down to fat as a single surgical procedure. These were diagnostic rather than therapeutic procedures. Further intervention was performed later as required. Preliminary partial punch biopsies and shave excisions were not performed. All excised tissue was submitted for histopathological assessment using

4


haematoxylin and eosin staining. The maximum diameter of each macroscopic tumour was measured after storage in 10% formalin prior to routine transverse ‘bread-loaf’ sectioning. When the grade of differentiation varied within a tumour mass, the grade closest to poor differentiation was provided as the grade of differentiation for that case. Exclusion criteria

Only cases of histopathologically confirmed invasive SCC were included in the study. All cases of SCC in situ without an invasive component were excluded, as were all cases of keratoacanthoma and all known cases of residual or recurrent SCC from procedures prior to or during the study. All cases were marked with the procedure date on a body map for that patient at the time of excision. This body map was checked before later procedures to ensure that no residual or recurrent tumours entered the study.

Results In total, 1666 SCCs were identified in 842 patients [535 men (63.5%) and 307 women (36.5%)]; of these patients, 152 patients had 2 SCCs each and 132 had ≥ 3 SCCs each. The median age for the first SCC was 71 years [interquartile range (IQR) 62–80 years]. The majority of cases (76%) were diagnosed in patients aged ≥ 65 years. There was a significant (P < 0.001) increase in the median age during the shift from welldifferentiated (72 years) to moderately differentiated (76 years) to poorly differentiated (79 years) tumours (Table 1). A post hoc analysis using Bonferroni correction showed all paired differences to be statistically significant, except that between moderate and poor grades (P = 0.49). Among patients with multiple tumours, men were more likely to develop multiple tumours than women (P < 0.001). The recorded grades of differentiation were: 82.1% well-differentiated (n = 1367), 13.3% moderately differentiated (n = 222) and 4.6% poorly differentiated (n = 77) tumours. The variation in the prevalence of SCC grade by anatomical site is shown in Table 1. More than 60% of the well-differentiated SCCs were located at sites other than the head and neck, while the rates for moderately and poorly differentiated tumours were 45% and 30%, respectively. Poorly differentiated SCCs were most frequent on the scalp (21%; n = 12), cheek/chin (20%; n = 11), forehead (13%; n = 7) and ear (9%; n = 5) in men, and on the cheek combined with chin sites (19%; n = 4)



Table 1 Characteristics of squamous cell carcinomas by grade of differentiation. Grade of differentiation Characteristics Age, years† Sex‡ Male Female Macroscopic MD, mm† SCC depth (mm)† Site‡ Head and neck Scalp Forehead Ear Eyelid Nose Lip Cheek/chin Neck Not head/neck Shoulder Chest/abdomen Back Upper arm/elbow Forearm/wrist Hand Thigh/knee Leg Foot

Total (n 1666) 73 (64–81) 1202 464 8 1.2 691 96 166 105 24 65 28 150 57 975 40 76 48 82 186 188 72 274 9

Well (n 1367)

Moderate (n 222)

Poor (n 77)

P*

72 (64–79)

76 (68–83)

79 (73–84)

< 0.001

(72.1) (27.9) (5–10) (0.6–2.0)

964 403 7 1.1

(70.5) (29.5) (5–10) (0.5–1.8)

182 40 8 2.1

(82.0) (18.0) (6–11) (1.5–3.0)

56 21 10 2.5

(72.7) (27.3) (8–13) (1.8–3.3)

(41.5) (5.8) (10.0) (6.3) (1.4) (3.9) (1.7) (9.0) (3.4) (58.5) (2.4) (4.6) (2.9) (4.9) (11.2) (11.3) (4.3) (16.5) (0.5)

515 63 126 85 17 42 24 108 50 852 35 62 40 71 161 172 57 245 9

(37.7) (4.6) (9.2) (6.2) (1.2) (3.1) (1.8) (7.9) (3.7) (62.3) (2.6) (4.5) (2.9) (5.2) (11.8) (12.6) (4.2) (17.9) (0.7)

123 20 31 15 5 18 2 27 5 99 4 9 6 7 21 14 13 25 0

(55.4) (9.0) (14.0) (6.8) (2.2) (8.1) (0.9) (12.2) (2.3) (44.6) (1.8) (4.1) (2.7) (3.2) (9.5) (6.3) (5.9) (11.3) (0)

53 13 9 5 2 5 2 15 2 24 1 5 2 4 4 2 2 4 0

(68.8) (16.9) (11.7) (6.5) (2.6) (6.5) (2.6) (19.5) (2.6) (31.2) (1.3) (6.5) (2.6) (5.2) (5.2) (2.6) (2.6) (5.2) (0)

0.002 < 0.001 < 0.001 < 0.001

MD, maximum diameter. *difference between well-differentiated, moderately differentiated and poorly differentiated squamous cell carcinomas based on the Kruskal–Wallis non-parametric test, Pearson v² test or Fisher exact test as appropriate; †continuous variables are presented as median (IQR); ‡categorical variables presented as n (%).

and upper arm (14%; n = 3) in women. Only welldifferentiated SCCs were present on the ear, shoulder and back of women and on the feet of both sexes. All poorly differentiated SCCs on the chest (men n = 4 and women n = 1) were noted to be within the midline ‘V’-shaped area of chronic sun exposure created by wearing open-necked clothing. Tumour maximum diameter and tumour depth

In 1662 of the 1666 tumours (the other 4 tumours had diameter data missing), the overall median maximum diameter was 8 mm (IQR 5–10 mm) and the median tumour depth was 1.2 mm (IQR 0.6–2.0 mm) (Table 1). Maximum diameter by anatomical site and grade of differentiation is displayed in Fig. 1. The median maximum diameter was inversely proportional to differentiation grade, being 7 mm (IQR 5–10 mm) for well-differentiated SCC, 8 mm (IQR 6–11 mm) for moderately differentiated SCC and 10 mm (IQR 8–13 mm) for poorly differentiated SCC, showing a significant (P < 0.001) progressive increase in


respective tumour diameters (Table 1). Post hoc testing of paired differences between grades indicated that this progressive trend was significant at each particular shift, with the median maximum diameter being significantly (P < 0.01) larger for moderately differentiated compared with well-differentiated SCCs, and for poorly compared with moderately differentiated SCCs (P = 0.02). Fig. 2 displays the depth data for anatomical site and grade of differentiation. Similarly, there was an inverse correlation between tumour depth and differentiation, with median depth being 1.1 mm (IQR 0.5–1.8 mm) for well-differentiated, 2.1 mm (IQR 1.5–3.0 mm) for moderately differentiated and 2.5 mm (IQR 1.8–3.3 mm) for poorly differentiated SCC, showing a significant (P < 0.001) progressive increase in respective tumour depth (Table 1). The shift towards poorly differentiated SCC was significant for head and neck sites, increased tumour diameter and increased depth in the univariate models (Table 2). Using well-differentiated SCC as the reference in the multinomial regression model, older people were more likely to develop moderately and poorly


5


Anatomic site (n)

(a)

Well

Scalp (63) Ear (85) Eyelid (17) Nose (42) Forehead (126) Lip (24) Cheek/chin (108) Neck (50)

Moderate


Poor


0

10

20

30

Maximum diameter (mm)

(b)

Anatomic site (n)

Well

Shoulder (35) Chest/abdomen (62) Back (40) Upper arm/elbow (71) Forearm (159) Hand (171) Thigh/knee (57) Leg (245) Foot (9)

Moderate


Poor


Figure 1 (a,b) Maximum histological sur-

0

10

20

30

Maximum diameter (mm)

differentiated SCC compared with well-differentiated SCC after adjustment for all significant covariates (Table 2). Compared with women, men had a significantly higher risk of having moderately differentiated compared with well-differentiated SCC (OR = 1.57, 95% CI 1.05–2.34, P = 0.03) and a significantly lower risk of developing poorly differentiated compared with moderately differentiated SCC (OR = 0.50, 95% CI 0.27–0.95, P = 0.03). After adjustment for all confounders, the shift towards poorly differentiated SCC was significant for head and neck sites and for

6


face diameters of squamous cell carcinoma (SCCs) by anatomical sites and grade of differentiation on (a) head and neck, and (b) trunk and limb.

increased tumour depth. Poorly differentiated SCCs displayed significantly larger diameters than differentiated moderately and well-differentiated SCCs (Fig. 1)

Discussion We found that older age, anatomical sites on the head and neck, larger tumour diameter, and increased invasion depth were associated with a higher grade of SCC. Facial sites of both men and women and the bald scalp of men were found to have a higher incidence of



Table 2 Factors related to grade of differentiation in univariate and multivariate multinomial logistic regression models. Model 1*

Model 2†

Moderate vs. well Variable Univariate Age, years Sex Female Male Anatomical site Not head/neck‡ Head and neck§ Macroscopic MD SCC depth Multivariate Age (single years) Sex Female Male Anatomical site Not head and neck‡ Head and neck§ Macroscopic MD SCC depth

Poor vs. well

Poor vs. moderate

OR (95% CI)

P

OR (95% CI)

P

OR (95% CI)

P

1.03 (1.02–1.05)

< 0.001 < 0.001

1.06 (1.03–1.08)

< 0.001 0.68

1.02 (0.99–1.05)

0.12 0.09

Reference 1.90 (1.33–2.73)

Reference 1.11 (0.67–1.87) < 0.001

Reference 2.06 (1.54–2.74) 1.06 (1.03–1.10) 2.24 (1.89–2.65) 1.02 (1.01–1.04)

< 0.001 < 0.001 0.004 0.03

Reference 1.57 (1.05–2.34)

< 0.001 Reference 3.65 (2.23–5.99) 1.13 (1.08–1.18) 2.66 (2.08–3.39) 1.03 (1.00–1.05)

< 0.001 < 0.001 0.02 0.44

Reference 0.79 (0.43–1.44) < 0.001

Reference 2.29 (1.66–3.14) 1.00 (0.96–1.04) 2.18 (1.77–2.68)

Reference 0.59 (0.32–1.08)

0.99 < 0.001

0.04 Reference 1.78 (1.03–3.08) 1.06 (1.01–1.11) 1.19 (1.02–1.38) 1.01 (0.98–1.03)

0.58 0.03

Reference 0.50 (0.27–0.95) < 0.001

Reference 5.11 (2.88–9.04) 1.06 (1.01–1.11) 2.54 (1.97–3.27)

0.02 0.02

0.03 < 0.001

0.01 Reference 2.23 (1.24–4.01) 1.06 (1.00–1.11) 1.16 (1.03–1.32)

0.03 0.02

MD, maximum diameter; SCC, squamous cell carcinoma. *Multinomial logistic regression model used well-differentiated SCC as the reference group to which the moderately and poorly differentiated SCCs were compared (both analyses), adjusted for age, sex, anatomical site (head and neck, and sites that were not head/neck), maximum diameter and SCC depth (multivariate analysis); †multinomial logistic regression model used moderately differentiated SCC as the reference group to which the moderately and poorly differentiated SCCs were compared (both analyses), adjusted for age, sex, anatomical site (head and neck, and sites that were not head/neck) maximum diameter and SCC depth (multivariate analysis); ‡anatomical sites other than the head and neck included the shoulder, chest/abdomen, back, upper arm/elbow, forearm/wrist, hand, thigh/knee, leg and foot; §head and neck included anatomical sites of the scalp, forehead, ear, eyelid, nose, lip, cheek/chin and neck.

moderately and poorly differentiated SCCs. In men, poorly differentiated SCCs on the scalp occurred exclusively on bald areas. The proportions of the three grades of differentiation varied by anatomical site. The variation in the grade proportions recorded does not imply a precursor role or time-frame for a shift from one grade of differentiation to another. Currently, it is unknown if the proportions of the differentiation grades and their spatial distribution within a given invasive SCC vary with tumour progression from early to late disease. In our study with detailed data and a larger sample size, we found that the higher incidence of SCCs on the lower extremity in women was limited to welldifferentiated SCCs, whereas in men, the leg had a lower incidence of well-differentiated and poorly differentiated SCCs (with only four of the latter recorded). Despite the higher prevalence of SCC on the lower extremity, the adjusted results indicated that SCCs on the head and neck shift towards higher grades.


Depth recordings showed that the thickest SCCs on the head and neck were poorly differentiated. Regarding anatomical location, the thickest SCCs on the chest and the upper arm were poorly differentiated, whereas those on the leg were moderately and poorly differentiated. The patients in this study were not in the habit of wearing gloves during winter, and they often displayed pronounced chronic solar-damaged skin on the dorsa of both hands. However, there was only one case of moderately and one case of poorly differentiated SCC on the dorsum of the hand for both men and women (four tumours in total). Most lip tumours were excised from the vermilion border, or arose from background actinic chelitis on the mucosal lower lip. Mucosal SCC on the lip could be considered as oral cavity tumours with a worse prognosis than cutaneous lip SCC. We acknowledge that it is unclear whether tumours arising at the vermilion border should be regarded as either mucosal or cutaneous tumours.


7


Anatomic site (n)

(a)

Well

Scalp (63) Forehead (126) Ear (85) Eyelid (17) Nose (42) Lip (24) Cheek/chin (108) Neck (50)

Moderate


Poor


0

2

4

6

8

10

SCC depth (mm)

(b)

Well


Moderate


Poor


Figure 2 (a,b) Depth of squamous cell 0

5

10

15

SCC depth (mm)

Maximum tumour diameter and depth of invasion data and other study findings may not represent the likely larger dimensions for more advanced SCCs treated in tertiary facilities. The vast majority of SCCs in this study were noted to arise within actinic keratoses during histopathological assessment, and the recorded macroscopic tumour diameters do not necessarily represent the true maximum diameters of invasive SCCs. We acknowledge that there is a bias in the higher numbers of precursor actinic keratoses, favouring a higher proportion of well-differentiated tumours in the

8


carcinoma (SCCs) by anatomical sites and grade of differentiation on (a) head and neck, and (b) trunk and limb.

study. Strict regular follow-up may also have resulted in early detection, favouring thinner tumours. Limitations

We concede that some tumours were incompletely excised due to poor clinical, dermoscopic or histopathological margin definition. However, this volume of residual tumour tissue was considered quite small and unlikely to affect the study results. The majority of patients were white with light skin types;



darker skin types may have been under-represented. Most patients were born in Australia and typically had a history of excessive chronic sun exposure. Thus, SCCs with aetiologies other than excessive chronic solar exposure may also be under-represented. Some patients had multiple SCCs at presentation, resulting in a selection bias favouring clinically larger tumours. The scalp and other head sites overlying embryonic clefts may also be over-represented by selection bias. Whether there was a single grade or more than one grade in a tumour was not recorded. Only four patients had undergone organ transplantation. Background iatrogenic or endogenous immunosuppression was not formally recorded, and may have influenced the study findings. The post-excision tumour diameter measurements were smaller than the in vivo diameters owing to tissue shrinkage.

Conclusion This study found that older age, male sex, anatomical sites on the head, increased tumour diameter and increased invasion depth were associated with a higher grade of differentiation within early invasive SCCs. Further, the bald scalp and ears of men and the forehead and chin of both sexes were found to be areas for increased invasion depth and a shift towards poor differentiation.

What’s already known about this topic? • Early primary cutaneous SCC varies in diame-

ter, depth and grade of differentiation.

What does this study add? • Tumour depth, tumour diameter and grade of

differentiation vary by anatomical site in early SCC. • Depth of invasion and grade of differentiation in early SCC vary by patient sex and anatomical site. • Even in early SCC, deeper invasion and increased poor differentiation were favoured on the ear of men. • Increased depth of invasion, location on the ear and a shift towards poor differentiation are risk factors for adverse outcomes in advanced SCC.


• Early effective intervention for SCCs with poten-

tial adverse outcomes based on these characteristics is optimized by early detection of these tumours. • Knowledge of anatomical sites that have a higher incidence of SCC with high risk factors may help identify these tumours early in disease progression.

References 1 Lomas A, Leonardi-Bee J, Bath-Hextall F. A systematic review of worldwide incidence of nonmelanoma skin cancer. Br J Dermatol 2012; 166: 1069–80. 2 Xiang F, Lucas R, Hales S et al. Incidence of nonmelanoma skin cancer in relation to ambient UV radiation in white populations, 1978–2012: empirical relationships. JAMA Dermatol 2014; 150: 1063–71. 3 Crowson AN, Magro CM, Mihm MC. Biopsy Interpretation of the Skin: Primary Non-Lymphoid Cutaneous Neoplasia. Philadelphia: Lippincott Williams & Wilkins, 2010. 4 Fernandez-Figueras MT, Carrato C, Saenz X et al. Actinic keratosis with atypical basal cells (AK I) is the most common lesion associated with invasive squamous cell carcinoma of the skin. J Eur Acad Dermatol Venereol 2015; 29: 991–7. 5 Marcil I, Stern RS. Risk of developing a subsequent nonmelanoma skin cancer in patients with a history of nonmelanoma skin cancer: a critical review of the literature and meta-analysis. Arch Dermatol 2000; 136: 1524–30. 6 Krynitz B, Edgren G, Lindelof B et al. Risk of skin cancer and other malignancies in kidney, liver, heart and lung transplant recipients 1970 to 2008: a Swedish population-based study. Int J Cancer 2013; 132: 1429–38. 7 Ulrich C, Kanitakis J, Stockfleth E et al. Skin cancer in organ transplant recipients: where do we stand today? Am J Transplant 2008; 8: 2192–8. 8 Gogia R, Binstock M, Hirose R et al. Fitzpatrick skin phototype is an independent predictor of squamous cell carcinoma risk after solid organ transplantation. J Am Acad Dermatol 2013; 68: 585–91. 9 Schmults CD, Karia PS, Carter JB et al. Factors predictive of recurrence and death from cutaneous squamous cell carcinoma: a 10-year, single-institution cohort study. JAMA Dermatol 2013; 149: 541–7. 10 Krediet JT, Beyer M, Lenz K et al. Sentinel lymph node biopsy and risk factors for predicting metastasis in cutaneous squamous cell carcinoma. Br J Dermatol 2015; 172: 1029–36. 11 Roozeboom MH, Lohman BG, Westers-Attema A et al. Clinical and histological prognostic factors for local recurrence and metastasis of cutaneous squamous cell


9


12

13

14

15

16 17 18

10

carcinoma: analysis of a defined population. Acta Derm Venereol 2013; 93: 417–21. Brougham ND, Dennett ER, Cameron R et al. The incidence of metastasis from cutaneous squamous cell carcinoma and the impact of its risk factors. J Surg Oncol 2012; 106: 811–15. Carter JB, Johnson MM, Chua TL et al. Outcomes of primary cutaneous squamous cell carcinoma with perineural invasion: an 11-year cohort study. JAMA Dermatol 2013; 149: 35–41. Brin L, Zubair AS, Brewer JD. Optimal management of skin cancer in immunosuppressed patients. Am J Clin Dermatol 2014; 15: 339–56. Farasat S, Yu SS, Neel VA et al. A new American Joint Committee on Cancer staging system for cutaneous squamous cell carcinoma: creation and rationale for inclusion of tumor (T) characteristics. J Am Acad Dermatol 2011; 64: 1051–9. Broders AC. Practical points on the microscopic grading of carcinoma. NY State J Med 1932; 667–71. Weedon D, Strutton G. Skin Pathology, 2nd edn. Edinburgh: Churchill Livingstone, 2002. Campoli M, Brodland DG, Zitelli J. A prospective evaluation of the clinical, histologic, and therapeutic


19

20

21

22

23

variables associated with incidental perineural invasion in cutaneous squamous cell carcinoma. J Am Acad Dermatol 2014; 70: 630–6. Brinkman JN, Hajder E, van der Holt B et al. The effect of differentiation grade of cutaneous squamous cell carcinoma on excision margins, local recurrence, metastasis, and patient survival: a retrospective follow-up study. Ann Plast Surg 2015; 75: 323–6. Kyrgidis A, Tzellos TG, Kechagias N et al. Cutaneous squamous cell carcinoma (SCC) of the head and neck: risk factors of overall and recurrence-free survival. Eur J Cancer 2010; 46: 1563–72. Petter G, Haustein UF. Squamous cell carcinoma of the skin: histopathological features and their significance for the clinical outcome. J Eur Acad Dermatol Venereol 1998; 11: 37–44. Lallas A, Pyne J, Kyrgidis A et al. The clinical and dermoscopic features of invasive cutaneous squamous cell carcinoma depend on the histopathological grade of differentiation. Br J Dermatol 2015; 172: 1308–15. Pyne J, Sapkota D, Wong JC. Squamous cell carcinoma: variation in dermatoscopic vascular features between well and non-well differentiated tumors. Dermatol Pract Concept 2012; 2: 204a05.
