CED

Clinical dermatology • Original article

Clinical and Experimental Dermatology

Histopathological study of perilesional skin in patients diagnosed with nonmelanoma skin cancer Z. Apalla,1 P. Calzavara-Pinton,2 A. Lallas,3 G. Argenziano,3 A. Kyrgidis,3 S. Crotti,2 F. Facchetti,4 P. Monari2 and G. Gualdi2 1

First Department of Dermatology, Aristotle University of Thessaloniki, Thessaloniki, Greece; Departments of 2Dermatology and 4Pathology 1-2, Spedali Civili Brescia, Brescia, Italy; and 3Skin Cancer Unit, Arcispedale Santa Maria Nuova, IRCCS, Reggio Emilia, Italy doi:10.1111/ced.12713

Summary

Background. Epidemiological and clinical data suggest that actinic damage to the skin is an important predictor of skin carcinogenesis. Aim. To investigate the association of squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) with sun-damage alterations seen by histopathology. Method. In the current prospective study, perilesional skin of SCC or BCC lesions was evaluated for presence of alterations associated with chronic photodamage. Presence of scarring, perineural/perivascular invasion, haemorrhage/haemorrhagic crust, ulceration/erosion and margin involvement were also assessed. Result. Of 6038 included lesions, 4523 (74.9%) were BCCs and 1515 (25.1%) were SCCs. Presence of actinic damage was five times more frequent in SCC than in BCC (OR = 5.29, 95% CI 4.44–6.00, P < 0.001), and diagnosis of SCC was twice as common in photo-exposed than nonphoto-exposed body sites (OR = 2.34, 95% CI 2.03–2.70, P < 0.001). There were twofold higher odds for actinic damage in SCC compared with Bowen disease (OR = 2.015, 95% CI 1.55–2.61, P < 0.001). Assessing the different BCC histological subtypes, we found that nodular BCC had at least twofold higher odds (OR = 2.63, 95% CI 2.09–3.32), infiltrative BCC had 48% higher odds (OR = 1.487, 95% CI 1.18–1.87) and basosquamous BCC had fourfold higher odds (OR = 4.10, 95% CI 3.01–5.57) of having actinic damage compared with superficial BCC. Conclusions. Histological verification of ultraviolet-associated alterations in the perilesional skin in patients with NMSC in our study confirms the aetiopathogenic link between sun exposure and epithelial carcinogenesis on a histopathological basis. This correlation was stronger for SCCs than for BCCs.

Introduction Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) represent the two commonest subtypes of nonmelanoma skin cancer (NMSC). Although SCC is Correspondence: Dr Zoe Apalla, First Department of Dermatology, Aristotle University of Thessaloniki, 124 Delfon Street, 54643, Thessaloniki, Greece E-mail: [email protected] Conflict of interest: the authors declare that they have no conflicts of interest. Accepted for publication 10 February 2015

ª 2015 British Association of Dermatologists

generally associated with a favourable prognosis, recent evidence suggests that the mortality rates of this cancer might have been underestimated to date.1 Chronic sun exposure has traditionally been considered to play a predominant role in the aetiopathiogenesis of both BCC and SCC, and several epidemiological studies have associated cumulative sun damage with a higher risk for both tumours. This association has been shown to be stronger for SCC.2 In this study, we carried out a histopathological investigation of the association of SCC and BCC with sun-damage alterations.

Clinical and Experimental Dermatology (2016) 41, pp21–25

21

Photodamage and nonmelanoma skin cancer  Z. Apalla et al.

Methods The institutional ethics committee approved the study of the Spedali Civili, Brescia, Italy, and all patients provided written informed consent before enrolment.

Patients This was a prospective study, in which we enrolled patients [3518 (58.3%) men, 2520 (41.7%) women; mean  SD age 70.21  13.48 years, range 18–99] who were scheduled to undergo surgical excision for BCC or SCC. Demographic data, including age, sex and skin phototype, and tumour characteristics, including anatomical site and size, were recorded. All surgical excisions were performed with lateral margins of at least 3 mm of clinically healthy skin. All tumours that were histopathologically diagnosed as BCC or SCC were included in the final analyses. Presence of clear or involved deep and lateral margins, as well as the histopathological subtype of BCC (e.g. superficial, nodular, infiltrative/morphoeaform and basosquamous), were recorded. Histopathology

Two independent histopathologists evaluated perilesional skin for presence of the following pre-defined criteria: (i) actinic keratosis (AK) and (ii) actinic damage [confirmed by the presence of elastosis (nodular/diffuse) and at least two of the following features: hyperkeratosis, acanthosis, lentigines, epidermal atrophy, epidermal hyperpigmentation, fibroblasts, hyperplasia, ephelides, lentigines, telangiectasias, venous lakes, purpura or comedones]. Taking into account the lack of standardized qualitative and quantitative methods for determining actinic damage,3–5 we decided to use the aforementioned criteria. A possible limitation of this study is that differentiation between an AK and the in situ component of the SCC is very difficult, especially on a background of solar-damaged skin. In the tumours characterized by lateral margin involvement, the specimen for assessment was taken from a portion of the margins that was free of tumour. The histopathologists also assessed the presence of (iii) scarring, (iv) perineural invasion (PNI), (v) perivascular invasion, (vi) haemorrhage/haemorrhagic crust and (vii) ulceration/erosion.

Statistical analysis

All variables were tested for normality. Colinearity was assessed via a correlation matrix, using Spearman

22

Clinical and Experimental Dermatology (2016) 41, pp21–25

rho (q) correlation coefficient. Pearson v² and Mann–Whitney U-test were used for nonparametric comparisons. Dichotomous outcome variables were set to diagnosis of each tumour (SCC or BCC) separately for risk analysis. Relative risks were calculated for all dichotomous variables. Crude and adjusted OR and the corresponding 95% CIs were calculated by univariate and conditional multivariate logistic regression, respectively. a level was set at 0.05, while an a level of 0.10 was used as cut-off for variable removal in the automated model selection for multivariate logistic regression. The Type I error probability associated with all tests in this study was set to P = 0.05. All statistical calculations were made with SPSS software (v17.0; SPSS Inc. Chicago, IL, USA).

Results Patient demographics

In our patient group, the majority were male, and the female patients (mean age 69.30  15.0 years) were younger than the male patients (mean age 70.79  12.3 years, Mann–Whitney U-test, P = 0.13). Of the 6038 patients with NMSC, 4523 (74.9%) had BCC and 1515 (25.1%) had SCC. Tumour location

The head and neck (H&N) was the most frequently affected anatomical site (3936 tumours; 65.1%), followed by the trunk (1262 tumours; 20.9%), arms (494 tumours; 8.2%) and legs (340 tumours; 5.6%), while the information about tumour localization was missing for 10 cases (0.2%). A more detailed description of the tumour locations is shown in Table 1. Histopathological findings

The frequencies of the histopathological features found in the perilesional skin of the surgical specimens are shown in Table 2. Of 6038 speciments examined, 1632 (27.0%) exhibited at least one of the two criteria for photodamage (AK and actinic damage): 791 specimens (13.1%) showed the presence of AK, 752 (12.5%) exhibited actinic damage and 87 (1.4%) had both. Presence of actinic damage was five times more frequent in SCC than in BCC (OR = 5.29, 95% CI 4.44–6.00, P < 0.001). The diagnosis of SCC was twice as common in photoexposed than nonphoto-exposed sites (OR = 2.34, 95% CI 2.03–2.70, P < 0.001). Invasive SCC was more

ª 2015 British Association of Dermatologists

Photodamage and nonmelanoma skin cancer  Z. Apalla et al.

Table 1 Location of tumours. Frequency

Frontal/templar Cheek/nasolabial fold Chin Nose Upper lip/prolabio Lower lip/prolabio Upper eyelid Lower eyelid Ear Scalp Neck Back Chest/abdomen Arms Hand Legs Feet Genitals Total

n

%

954 777 237 696 73 41 112 120 275 445 206 702 560 418 76 318 22 6 6038

15.8 12.9 3.9 11.5 1.2 0.7 1.9 2.0 4.6 7.4 3.4 11.6 9.3 6.9 1.3 5.3 0.4 0.1 100.0

Table 2 Frequencies of the histopathological criteria found in the perilesional skin of the surgical specimens in 6038 patients with NMSC.

Criteria

SCC (n = 1515)

BCC (n = 4523)

Total (n = 6038)

Actinic keratosis Actinic damage/elastosis Ulceration/erosion Perineural invasion Vascular invasion Scar/burn Squamous crust

568 307 257 34 16 171 118

271 571 1184 184 1 393 178

839 878 1441 218 17 564 296

(37.5) (20.3) (17.0) (2.2) (1.1) (11.3) (7.8)

(6.0) (12.6) (26.2) (4.1) (0.02) (8.7) (3.9)

(13.9) (14.5) (23.9) (3.6) (0.3) (9.3) (4.9)

NMSC, nonmelanoma skin cancer.

common in the H&N region (81.3%) than the torso (41.9%), arm (47.6%) or leg (4.1%). For BCC, the superficial type was less common in photo-exposed areas and more common in the torso (fourfold; 423% higher odds; OR = 4.23, 95% CI:3.71– 4.84, P < 0.001), arm (67% higher odds, OR = 1.67, 95% CI 1.37–2.04, P < 0.001) leg (56%, OR = 1.57, 95% CI 1.57–1.23, P < 0.001) compared with the H&N (reference category). Probability of sun damage

In a multivariate adjusted model, we included all variables shown in Table 2. We found that there were two-fold higher odds for actinic damage in SCC compared with Bowen disease

ª 2015 British Association of Dermatologists

(BD) (OR = 2.02, 05% CI 1.55–2.61, P < 0.001). For AK as a perilesional finding, the odds were 20% higher in SCC compared with BD (OR = 1.20, 95% CI 0.97–1.47, P = 0.09). Regarding the perilesional alterations in the different BCC histological subtypes, nodular BCC had at least twofold higher odds (OR = 2.63, 95% CI 2.09– 3.32), infiltrative had 1.5-fold higher odds (OR = 1.49, 95% CI 1.18–1.87) and basosquamous carcinoma fourfold higher odds (OR = 4.10, 95% CI 3.01–5.57) of actinic damage compared with superficial BCC. Both nodular and infiltrative BCC had at least twofold higher odds (OR = 2.61, 95% CI 1.86– 3.68; OR = 2.59, 95% CI 1.89–3.54, respectively) and basosquamous carcinoma almost twofold higher odds (OR = 1.81, 95% CI 1.05–3.14) for AK compared with the superficial subtype. Excision margins were clear in 5138 tumours (85.0%). Of the remaining 15% of tumours, 639 (10.6%), 173 (2.9%) and 92 (1.5%) had lateral, deep, and both lateral and deep margins involved, respectively. Compared with BCC, SCC had an almost twofold higher risk for lateral (OR = 1.86, 95% CI 1.56–2.21, P < 0.001), deep (OR = 2.26, 95% CI 1.66–3.08, P < 0.001), or both types of involved margin (OR = 1.77, 95% CI 1.15–2.74, P = 0.01). Of the 6038 NMSC tumours, 218 (3.6%) had PNI. PNI was more common in BCC (4.1% of 4523 cases) than in SCC (2.2% of 1515 cases; Pearson’s v² P = 0.001). However, when only invasive SCC was taken into consideration, PNI was present in 33 of 779 tumours (4.2%). Analysing the BCC subtypes, PNI was most frequent in basosquamous carcinoma (43/288; 14.9%). With regard to perivascular invasion, only 17 NMSCs (0.3%) were positive. Vascular invasion was more common in SCC (1.1% of 1515 cases) than BCC (only one case; Pearson’s v² P < 0.001). When only invasive SCC was taken into consideration, vascular invasion was present in 16 of 779 cases (4.2%).

Discussion According to the currently available data, actinic damage to the skin represents an important predictor of skin carcinogenesis. The high incidence of both BCC and SCC among adults with multiple AK confirms the latter observation. The majority of the published studies designed to assess this clinical/ epidemiological observation did not provide histological confirmation of the photodamage. The procurement of tissue samples for verification of actinic

Clinical and Experimental Dermatology (2016) 41, pp21–25

23

Photodamage and nonmelanoma skin cancer  Z. Apalla et al.

alterations involving the perilesional skin in our study should help to shed further light into the pathogenesis of SCC and BCC. Based on our results, ultraviolet radiation (UVR)related skin changes of the perilesional skin were five times more frequent in SCC compared with BCC. The latter observation is in line with the findings of other investigators. Foote et al. reported a lower BCC : SCC ratio in their study population compared with the general population in the same geographical region, indicating a higher risk for SCC among individuals with AK.2 Taking into account that AK and SCC are considered parts of the keratinocyte skin cancer continuum, a higher incidence of SCC among individuals with multiple AK is not unexpected. Padillia et al. studied the gene expression profiles of AK and SCC, and found that genes that were upregulated in AK and SCC were downregulated in normal skin, whereas genes that were downregulated in AK and SCC were upregulated in normal skin.6 In a genomewide study in patients with organ transplant, several pathways activated in SCC were found to be already activated in AK, confirming the assumption that AK is a precursor lesion to SCC. These data indicate that the two entities are also closely related genetically.7 Furthermore, the fact that SCC was twice as common in photo-exposed than nonphoto-exposed body sites, together with the age-related rising incidence in our study, highlights the strong pathogenetic role of high lifetime sun exposure and SCC.8 In contrast to SCC,9 approximately 25% of all BCC were found to arise on less sun-exposed body sites, mainly on the trunk. These data, together with the recorded plateau of BCC incidence with age, are in line with the reported stronger association with recreational than with occupational or total sun exposure as a risk factor of BCC.10 A recent systematic review revealed a relatively modest association between markers of chronic UVR exposure and BCC. Analytically, presence of > 10 AK was most strongly associated with BCC, resulting in a five-fold increase in risk (OR = 4.97; 95% CI 3.26– 7.58), while other features of photodamage, including solar elastosis, solar lentigines and telangiectasia, had a weaker association (ORs around 1.5).11 The aforementioned data are further supported by molecular studies demonstrating strong evidence for the association of sunlight and SCC (thymidine dimers and C>T transitions in p53 and other important molecules), whereas it is still unclear whether there is such an association for BCC (lack of UVB signatures at the PTCH/SMO sites).12,13

24

Clinical and Experimental Dermatology (2016) 41, pp21–25

Some interesting findings arose in our study from the correlation of specific histological subtypes of BCC with the presence of AK and photodamage in the perilesional skin. Superficial BCC coexisted with perilesional alterations less frequently than any other BCC subtype. This observation is not unexpected, as superficial BCC is found mainly on the trunk and less often on photoexposed areas. By contrast, basosquamous carcinoma had the highest odds of actinic damage and AK. Traditionally, basosquamous cancinoma is classified under the umbrella of BCC. However, its specific dermoscopic14 and histopathological characteristics, combined with its higher metastatic potential, indicate a tumour type with a more aggressive behavioural pattern, more closely resembling SCC.15 Coexistence of basosquamous carcinoma with a higher incidence of solar damage compared with other BCC subtypes in our study may indicate a closer relationship of this entity to SCC than to BCC. In the studied group, the risk of SCC over BCC increased with age. Using 0–29 years as the reference age group, the 60–69 years group had a 7.5-fold risk for SCC (P < 0.05), while the 69–79, 80–89 and 90– 99 years groups had a risk 11.8-fold (P = 0.02), 17fold (P < 0.01) and 28-fold (P = 0.001) higher, respectively. This is in agreement with epidemiological data reporting a decreased ratio of BCC : SCC with age among individuals with multiple AKs. In the study of Foote et al., older age independently predicted both BCC and SCC development, whereas patient sex was significantly related only to SCC. Older age, male sex, red hair and prolonged residency in a sunny region were other factors associated with a higher risk for SCC.2 Presence of PNI in SCC is related to poor outcome,16 and increased rate of disease-specific death.17,18 In line with the latter, we found a fivefold higher risk for margin involvement when PNI was present in SCC. In addition, our results further support the previous suggestion that PNI represents a poor prognostic factor for local control of BCC.19 Specifically, of the BCC subtypes in our study, PNI was significantly more frequent in basosquamous carcinoma and infiltrative BCC, which are regarded as high-risk subtypes for local recurrence. Furthermore, the presence of PNI in infiltrative BCC posed a 2–4-fold higher risk for margin involvement.

Conclusion Our study provides histological confirmation of UVassociated alterations in the perilesional skin of patients

ª 2015 British Association of Dermatologists

Photodamage and nonmelanoma skin cancer  Z. Apalla et al.

with NMSC, further supporting the aetiopathogenic link of sun exposure and epithelial carcinogenesis. The latter correlation was stronger for SCC than for BCC, indicating a higher risk for SCC among patients with solar damage and clinical signs of field cancerization. In the same context, the current observations support the theory of AK and SCC being parts of the same keratinocyte disease continuum. Close monitoring of elderly patients with extreme photodamage is highly recommended for early detection and treatment of NMSC.

What’s already known about this topic? ● Several epidemiological studies have associated

cumulative sun damage with a higher risk for both BCC and SCC ● The majority of the published studies designed to assess this clinical/epidemiological observation did not provide histological confirmation of the photodamage.

What does this study add? ● Procurement of tissue samples for examination

and histopathological verification of actinic alterations involving the perilesional skin in our study confirms this association on a histopathological basis, shedding additional light on the pathogenesis of SCC and BCC.

References 1 Karia PS, Han J, Schmults CD. Cutaneous squamous cell carcinoma: estimated incidence of disease, nodal metastasis, and deaths from disease in the United States, 2012. J Am Acad Dermatol 2013; 68: 957. 2 Foote JA, Harris RB, Giuliano AR et al. Predictors for cutaneous basal- and squamous-cell carcinoma among actinically damaged adults. Int J Cancer 2001; 95: 7–11. 3 Griffiths CE, Wang TS, Hamilton TA et al. A photonumeric scale for the assessment of cutaneous photodamage. Arch Dermatol 1992; 128: 347–51. 4 Griffiths CE. The clinical identification and quantification of photodamage. Br J Dermatol 1992; 127: 37–42. 5 Marks R, Edwards C. The measurement of photodamage. Br J Dermatol 1992; 127: 7–13.

ª 2015 British Association of Dermatologists

6 Padilla RS, Sebastian S, Jiang Z et al. Gene expression patterns of normal human skin, actinic keratosis, and squamous cell carcinoma: a spectrum of disease progression. Arch Dermatol 2010; 146: 288–93. 7 Hameetman L, Commandeur S, Bavinck JN et al. Molecular profiling of cutaneous squamous cell carcinomas and actinic keratoses from organ transplant recipients. BMC Cancer 2013; 13: 58. 8 Kennedy C, Bajdik CD, Willemze R et al. The influence of painful sunburns and lifetime sun exposure on the risk of actinic keratoses, seborrheic warts, melanocytic nevi, atypical nevi, and skin cancer. J Invest Dermatol 2003; 120: 1087–93. 9 NCCI Non-Melanoma Skin Cancer Working Group. The 2002 National Non-Melanoma Skin Cancer Survey. Melbourne: Carlton, National Cancer Control Initiative, 2003. 10 Kricker A, Armstrong BK, English DR, Heenan PJ. Does intermittent sun exposure cause basal cell carcinoma? A case-control study in Western Australia. Int J Cancer 1995; 60: 489–94. 11 Khalesi M, Whiteman DC, Doi SA et al. Cutaneous markers of photo-damage and risk of basal cell carcinoma of the skin: a meta-analysis. Cancer Epidemiol Biomarkers Prev 2013; 22: 1483–9. 12 Kubo Y, Matsudate Y, Fukui N et al. Molecular tumorigenesis of the skin. J Med Invest 2014; 61: 7–14. 13 Chen AC, Halliday GM, Damian DL. Non-melanoma skin cancer: carcinogenesis and chemoprevention. Pathology 2013; 45: 331–41. 14 Giacomel J, Lallas A, Argenziano G et al. Dermoscopy of basosquamous carcinoma. Br J Dermatol 2013; 169: 358–64. 15 Betti R, Crosti C, Ghiozzi S, Cerri A et al. Basosquamous cell carcinoma: a survey of 76 patients and a comparative analysis of basal cell carcinomas and squamous cell carcinomas. Eur J Dermatol 2013; 23: 83–6. 16 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. 17 Leibovitch I, Huilgol SC, Selva D et al. Cutaneous squamous cell carcinoma treated with Mohs micrographic surgery in Australia, II: perineural invasion. J Am Acad Dermatol 2005; 53: 261–6. 18 Clayman GL, Lee JJ, Holsinger FC et al. Mortality risk from squamous cell skin cancer. J Clin Oncol 2005; 23: 759–65. 19 Balamucki CJ, Mancuso AA, Amdur RJ et al. Skin carcinoma of the head and neck with perineural invasion. Am J Otolaryngol 2012; 33: 447–54.

Clinical and Experimental Dermatology (2016) 41, pp21–25

25

Histopathological study of perilesional skin in patients diagnosed with nonmelanoma skin cancer.

Epidemiological and clinical data suggest that actinic damage to the skin is an important predictor of skin carcinogenesis...
69KB Sizes 2 Downloads 7 Views