Share:
Original Article

Cutaneous Squamous Cell Carcinoma: Clinico-Dermoscopic and Histological Correlation: About 72 Cases

Author Affiliation(s)

Abstract

Introduction: Cutaneous squamous cell carcinoma (cSCC) is the second most common skin cancer. It represents 20% of malignant skin tumors. Dermoscopy is a very useful tool for its diagnosis, which is confirmed by anatomo-pathological study.


Objective: to describe the different dermoscopic structures of cSCC and to investigate their association with clinical form and histopathological grade of differentiation.


Methods: Retro-prospective epidemiological study, collecting patients with squamous cell carcinoma over a period of 5 years. We divided our study population into two groups according to the degree of histological differentiation (well-differentiated and non-well-differentiated) and according to the clinical form (nodulo-ulcerative and nodular) and compared different dermoscopic parameters between these two groups.


Results: Of 72 invasive cSCC, 81.9% were well-differentiated and 18.1% were non-well- differentiated, the clinical form of cSCC was nodulo-ulcerative in 83.3% and nodular in 16.7%. The well-differentiated tumors showed dotted, glomerular and hairpin vessels and the predominance of a white pattern made of centrally distributed keratin as well as white circles, whitish perivascular halo, the distribution of these white structures was radial in nodulo-ulcerarive lesions in contrast to nodular lesions where their distribution as well as that of the keratin was rather diffuse. Non-well-differentiated tumors showed a combined white-red pattern with the predominance of arborizing vessels.


Conclusion:Our results show that dermoscopy can be considered a reliable tool to distinguish between well- and poorly differentiated cSCC, the transition from one to the other is represented by an increase in the amount of mostly arborizing vessels and the decrease in white structures.

Keywords : clinical form, histopathological grade of differentiation, dermoscopy, Cutaneous squamous cell carcinoma

Introduction

Cutaneous squamous cell carcinoma (cSCC) is the second most common skin cancer after basal cell carcinoma [ 1 , 2 ] . It may occur de novo or as a result of precursor lesions [ 1 ] . cSCC accounts for 20% of malignant cutaneous tumors, and its incidence is steadily rising due to population aging and increased emphasis on skin cancer screening [ 2 4 ] . Dermoscopy has become one of the fundamental diagnostic techniques in clinical practice, that allows in vivo evaluation of skin colors and microstructures [ 5 ] . The majority of dermoscopic structures, including white, vascular and pigmented structures have direct histopathologic correlations, making this technique a useful means of communication between pathologists and clinicians [ 6 ] . The degree of histopathological differentiation in cSCC ranges from good to moderate or poor, serving as a significant prognostic factor [ 7 ] . Poor differentiation is an independent risk factor for metastasis, recurrence, and disease-specific mortality, while well-differentiated SCC is associated with a five-year recurrence-free survival rate of 83% [ 8 ] . Thus, early and aggressive management of high-risk cSCC, including poorly differentiated tumors, is recommended to reduce morbidity and mortality [ 8 ] . However, while the dermoscopic criteria of cSCC have been well described, there remains a paucity of research examining the correlation between the histopathological grade of differentiation. Previous studies have reported that the white pattern predominates in well-differentiated cSCC, while the red pattern is more prevalent in poorly differentiated cSCC. Additionally, there is a scarcity of studies investigating the correlation between clinical form and dermoscopic features [ 7 ] .

Objectives

The aim of this study was to describe the different dermoscopic structures of invasive cSCC and investigate their association with clinical form and histopathological grade of differentiation.

Methods

This is a cross-sectional study performed in the dermatology department of HASSAN II University Hospital, Faculty of Medicine and Pharmacy, Sidi Mohamed Ben Abdellah University of Fez, Morocco. The study extended over a period of 5 years, from November 2017 to July 2022.

The inclusion criteria consisted of patients with skin lesions that were histopathologically confirmed as invasive cSCC. Exclusion criteria included patients with poor quality dermoscopic images, lesions histologically proven to be keratoacanthoma or SCC in situ and patients with xeroderma pigmentosum.

Dermoscopic images were captured with a Dermlite DL4 ‘4Gen, at a magnification of 10×, using both polarized and non-polarized light, with and without immersion. Two investigators evaluated the dermoscopic images of all selected lesions according to predefined criteria [ 5 10 ] and which are presented in Table 1 . Additionally, it states that the evaluators were blinded to the histology of the lesion when assessing the dermoscopic features of each lesion.

The dermatoscopic evaluation was based on pattern and color analysis, including vessel morphology density and distribution, as well as the presence or absence of different white structures such as scales, keratin, white circles, white structureless areas, white perivascular halos, chrysalis and rosettes. The distribution of these structures and the accuracy of the predominant pattern (red, white or combined white-red pattern) and the presence of other structures like erosions, ulceration and Hemorrhage were also assessed.

The recruited cases underwent total excision or biopsy and were submitted to routine histopathological confirmation using hematoxylin and eosin staining.

In this study, cSCC was categorized into two grades of histological differentiation: well and non-well differentiated. Moderately and poorly differentiated cSCC were evaluated together, due to the small number of cases. The clinical forms were classified as nodular and nodulo-ulcerative lesions.

Statistical Analysis

The data were analyzed using version 26.0 of SPSS for Windows. The mean ± standard deviations for numerical variables and the number and % values for categorical variables were given as descriptive statistics. Comparative dermoscopic analysis between the different dermoscopic parameters among the two histological subgroups and the clinical forms were performed by chi-square test or Fisher exact test depending on the sample size of the parameter to compare. For multivariate analysis, binary logistic regression was used. Values of P less than 0.05 were accepted as significant.

Results

Patients Demographics, Clinical Characteristics and Dermatopathological Findings

Of 160 histologically confirmed cSCC cases, 88 were excluded due to uninterpretable dermoscopic images. A total of 72 cSCC were collected of which 49 were male (68.1%) and 23 were females (31.9%), the mean age at diagnosis was 68.67 years with extremes (25–1). Concerning the origin 46 of our patients (63.9%) came from an urban environment, while 26 (36.1%) from a rural environment. Chronic sun exposure was reported in 49 patients (68.1%) and 5 patients had a history of skin cancer (6.9%). Based on the origin of lesions, 54 (75%) lesions were de novo; 7 patients (9.7%) had developed cSCC on a burn scar, 6 (8.3%) on a chronic wound, 4 (5.6%) on a lichen and in one patient (1.4%) on hereditary epidermal bullosa. Associated actinic keratosis lesions diagnosed through clinical examination and dermoscopy were found in 16 patients (22.2%). In our series 57 patients (79.2%) consulted at the tumor stage, while the remaining patients consulted at an earlier stage. Clinically, 49 patients had a phototype IV (68.1%) and 20 had a phototype III (27.8%). The frequency distribution of the lesions across various sites was as follows: cervicocephalic extremity (N = 32; 44.4%), lower limbs (N = 26; 36.1%), oral mucosa (N = 7; 9.7%), genital mucosa (N = 4, 5.6%) and upper limbs (N = 3; 4.2%). The clinical form of cSCC was nodulo-ulcerative in 60 patients (83.3%) and nodular in 12 patients (16.7%). The mean size of the lesions was 6.52 cm. According to the histopathological grade of differentiation, 59% or 81.9% of the cSCC lesions were well differentiated and 13% or 18.1% were non-well differentiated.

Dermatoscopic Analysis

The analysis of different dermoscopic structures in SCC, according to the clinical form and histopathological grade of differentiation in our series, is presented in Tables 2 , 3 , 4 and 5 .

Correlation Dermoscopy-Histopathological Differentiation Grade ( Tables 2 and 3 )

The most observed dermoscopic aspects in well-differentiated cSCC ( figure 1 ) were: dotted vessels (N = 44, 74.6%), glomerular (N = 36, 61%), hairpin (N = 44, 74.6%), whitish scales (N = 44, 74.6%), yellowish scales (N = 51, 86.4%), keratin (N = 56, 94.9%) with central distribution (N = 39, 69.6%), white circles (N = 44, 74.6%), whitish structureless areas (N = 49, 83.1%), perivascular whitish halo (N = 41, 69.5%) with a predominance of the white pattern (N = 38, 64.4%) and < 50% of vascular structures. All these dermoscopic aspects were significantly correlated with well-differentiated cSCC (P < 0.05).

For non-well differentiated cSCC ( Figures 2 and 3 ), arborizing vessels were the most common dermoscopic feature (N = 9; 69.3%) and was associated with a higher risk of developing non well differentiated squamous cell carcinoma OR = 14.8 [2.43; 90.37]. The other dermoscopic features that exhibited a statistically significant difference in non-well differentiated SCC included a vessel quantity higher than 50% (N = 9; 69.3%) with a predominance of the combined white-red pattern (N = 8, 61.5%).

Linear irregular vessels, vascular polymorphism, rosettes, chrysalis, presence of ulceration and hemorrhagic spots were present in both well- and non-well-differentiated groups with comparable percentages ( Table 2 ) without statistically significant difference (P > 0.05)

Correlation Dermoscopy-Clinical Form ( Tables 4 and 5 )

The most observed dermoscopic aspects in nodulo-ulcerative variant ( Figure 4 ) were: opaque yellowish scales (N = 52; 86.7%), hemorrhagic spots (N = 54; 90.0%), the presence of ulceration (N = 60; 100% OR = 0.015 [0.001; 0.178]), the central distribution of keratin N = 39; 72.2% OR = 11.120 [1.184; 104.483]) and the radial distribution of different white structures with a predominance of the white pattern (N = 36; 60%). All these dermoscopic aspects showed a statistically significant correlation with the nodulo-ulcerative form, with a P-value < 0.05. Other aspects were also predominant in this form, but the correlation was statistically non-significant, such as hairpin vessels (N = 42; 70%), whitish structureless areas (N = 47; 78.3%) and peri-vascular whitish halo (N = 41; 68.3%).

Concerning the nodular form ( Figure 3 ), the distribution of keratin and different white structures was diffuse, with percentages of (N = 7; 70.0%) and (N = 9; 75.0%) respectively, with the predominance of the white-red pattern, being statistically significant (P < 0.05) for each feature. However, the presence of comma vessels, arborizing vessels and regular linear vessels was more present in this form, with percentages of 33.3%, 50.0% and 25.0% respectively. However, the P value for these features was > 0.05, indicating a non-significant correlation.

Conclusions

The incidence of cutaneous squamous cell carcinoma (cSCC) has been increasing worldwide for several decades and is estimated to be 15 to 35/100,000 people per year [ 2 ] . It is more common in males than in women and affects the older population (80% of cases are in adults over 60), due to the correlation with cumulative sun exposure [ 3 ] . The epidemiological characteristics of the patients in our study are consistent with previous data.

Invasive cSCC can develop de novo, or from pre-cancerous lesion or certain chronic inflammatory skin areas, such as chronic wounds, burns, scars, ulcers etc [ 3 , 4 ] , as was the case in some of our patients.

Dermoscopy is an essential in-vivo, non-invasive diagnostic method that provides the visualization of morphological characteristics not detectable with the naked eye [ 11 ] . Only a limited number of studies have been conducted primarily focusing on the correlation between degree of differentiation and dermoscopy.

In this study, we identified useful clues that may suggest the degree of differentiation of cSCC. The presence of previously described white structures, including scales, centrally distributed keratin, white circles, perivascular whitish halo and whitish structureless areas were strongly associated with well-differentiated cSCC, while the diffuse distribution of keratin and other white structures were more likely to be present in non-well-differentiated cSCC. This finding aligns with the results reported by Saleh HM et al, Lallas et al, and Zalaudek et al reported that these whitish structures were associated with well- or moderately differentiated variants and that a central distribution of scales or keratin was associated with a 36-fold reduced possibility for poor differentiation [ 7 , 8 , 12 ] . In the latter, signs of keratinization are not found [ 8 , 12 , 13 ] , while the moderately differentiated variant would show less expression of dermoscopic white indices compared to the well differentiated cSCC [ 7 ] .

The vascular findings observed in our analysis of well-differentiated tumors are characterized by dotted, glomerular, and hairpin vessels, with the presence of a whitish perivascular halo. In contrast, non-well differentiated tumors exhibited predominantly branched or arborizing vessels. These findings are consistent with the results of Pyne et al [ 14 ] , who also demonstrated through their study that non-well differentiated tumors tended to be histologically deeper than well differentiated lesions. This deeper invasion may account for the higher prevalence of linear and branching vessels, which are more commonly observed as tumors increase in thickness and contribute to tumor development.

Our results are not in agreement with those of Saleh et al [ 7 ] , who documented that well-differentiated lesions had a higher incidence of linear irregular vessels followed by glomerular vessels and hairpin vessels. According to Lallas et al [ 13 ] , well-differentiated cSCC is characterized by elongated peripheral vessels reminiscent of the arborescent vessels of basal cell carcinoma, while poorly differentiated cSCC shows a polymorphic vascularization. Similar observations were made by Zalaudek [ 12 ] in poorly differentiated squamous cell carcinoma, where a reddish background accompanied small-caliber linear vessels, hairpin vessels, or glomerular vessels. In contrast, the well-differentiated subtype primarily exhibited hairpin and dot vessels.

We found that the quantity of vessels was significantly correlated with the degree of differentiation, which is consistent with the results of Lallas et al [ 8 ] . They reported that tumors with over 50% of the lesion area covered by vessels had a 30- to 120-fold higher possibility of being poorly differentiated. They also added that vessel caliber was also found to be a significant predictor of degree of differentiation, since a small caliber vessel had a threefold higher probability of poor differentiation while those with large caliber vessels had an 83% lower probabilities of being poorly differentiated [ 8 ] . Lallas et al also revealed that the presence of a predominantly red color, attributed to the absence of desquamation and keratin and the presence of bleeding and/or dense vasculature, was in favor of poor differentiation, while a predominantly white or white-yellow color reduced the probability of a poorly differentiated cSCC [ 8 ] . These results are consistent with ours and those of Saleh et al [ 7 ] .

Concerning the correlation of dermoscopic aspects and the clinical form, to the best of our knowledge, only one study has investigated this correlation. This gap in the literature motivated our research, as we aim to contribute significantly to the field. Saleh et al [ 7 ] examined 17 lesions representing three clinical forms of cSCC: nodular, nodulo-ulcerative, and ulcerative. They found that nodular and nodulo-ulcerative lesions showed mainly linear irregular vessels followed by glomerular vessels. In ulcerative lesions, glomerular vessels dominated, with a vessel quantity greater than 50% for nodulo-ulcerative and ulcerative tumors and less than 50% for the nodular form. In our study, linear irregular vessels were the most frequently observed, followed by dotted vessels and then glomerular vessels in both studied forms, Additionally, we found a higher presence of hairpin vessels in the nodulo-ulcerative form and an increased occurrence of arborescent vessels in the nodular form, with a vessel quantity less than 50% in the nodulo-ulcerative tumors, contrary to the above results. For whitish structures, whitish perivascular halo and whitish structureless areas were present in both forms, but more found in the nodulo-ulcerative form. Nodular lesions exhibited a dominant presence of white circles, along with a diffuse distribution of keratin. In contrast, the nodulo-ulcerative form showed a central distribution of keratin. These results were not consistent with those of Saleh et al [ 7 ] , who detected a central distribution of keratin in both nodular and nodulo-ulcerative forms, the presence of white circles in all nodular lesions and in more nodulo-ulcerative than ulcerative lesions with the absence of perivascular whitish halo in the latter form and its presence in the other two forms.

In the same study [ 7 ] , they concluded that nodular and nodulo-ulcerative lesions were significantly associated with well-differentiated cSCC, whereas ulcerative lesions were significantly associated with moderately differentiated cSCC. This association between clinical form and degree of differentiation was not investigated in our study.

The main limitation of our study is the combined analysis of non-well-differentiated tumors, which was due to the small number of poorly differentiated (N = 3) and moderately differentiated (N = 10) cSCC. In addition, the dermoscopic images were interpreted retrospectively, which introduces a potential bias. However, it is important to emphasize that we applied rigorous criteria in the selection process, excluding any images of poor quality. As a result, the number of cases included in the analysis was reduced.

Our study highlights the potential of dermoscopy as a non-invasive technique to provide clinicians with valuable insights into the histopathological differentiation of SCC. In addition, our study emphasizes the importance of collaboration between clinicians and anatomopathologists when an immunohistochemical complement is required, particularly in cases of poorly differentiated tumor proliferation. By engaging in discussions and considering immunohistochemical markers specific to poorly differentiated squamous cell carcinomas, particularly when a dermoscopic red pattern with a predominance of arborizing vessels is visualized. This collaborative approach not only improves patient care but also carries economic benefits by optimizing the selection and utilization of immunohistochemical resources.

Our results show that dermoscopy can be considered a reliable tool to distinguish between well- and poorly differentiated cSCC. This distinction is characterized by an increase in predominantly arborizing vessels and a corresponding decrease in white structures as the tumor progresses from a well-differentiated to a poorly differentiated state. Moreover, the nodulo-ulcerative form demonstrates a central distribution of keratin, while the nodular form displays a diffuse distribution. Dermoscopy not only aids in accurate diagnosis but also provides valuable guidance to the anatomo-pathologist. Additionally, the implementation of dermoscopy offers economic benefits by streamlining the diagnostic process and facilitating timely and appropriate patient management, ultimately leading to improved prognosis.

Further studies are desirable to investigate the association between the different dermoscopic criteria and clinical form and even with the histological subtype of cSCC.

References

  1. Carcinome épidermoïde (spinocellulaire) et ses précurseurs Bonerandi JJ, Monestier S. EMC – Dermatologie.2011;6:1-18. CrossRef
  2. Cutaneous Squamous Cell Carcinoma: Review of the Eighth Edition of the American Joint Committee on Cancer Staging Guidelines, Prognostic Factors, and Histopathologic Variants Motaparthi K, Kapil JP, Velazquez EF. Adv Anat Pathol.2017;24(4):171-194. CrossRef PubMed
  3. Cutaneous Squamous Cell Carcinoma: From Pathophysiology to Novel Therapeutic Approaches Fania L, Didona D, Di Pietro FR, et al. Biomedicines.2021;9:171. CrossRef PubMed
  4. Cutaneous Squamous Cell Carcinoma Parekh V, Seykora JT. Clin Lab Med.2017;37(3):503-525. CrossRef PubMed
  5. Non-invasive diagnostic techniques in the diagnosis of squamous cell carcinoma Warszawik-Hendzel O, Olszewska M, Maj M, Rakowska A, Czuwara J, Rudnicka L. J Dermatol Case Rep.2015;9(4):89-97. CrossRef PubMed
  6. Dermoscopy and dermatopathology correlates of cutaneous neoplasms Yélamos O, Braun RP, Liopyris K, et al. J Am Acad Dermatol.2019;80:341-363. CrossRef PubMed
  7. Dermoscopic criteria of squamous cell carcinoma in relation to clinical variant and histopathologic differentiation grade Saleh HM, Mokhtar MM, Mostafa AE. J Egypt Womens Dermatol Soc.2020;17:83-90. CrossRef
  8. The clinical and dermoscopic features of invasive cutaneous squamous cell carcinoma depend on the histopathological grade of differentiation Lallas A, Pyne J, Kyrgidis A, et al. Br J Dermatol.2015;172:1308-1315. CrossRef PubMed
  9. Dermatoscopy of facial actinic keratosis, intraepidermal carcinoma, and invasive squamous cell carcinoma: a progression model Zalaudek I, Giacomel J, Schmid K, et al. J Am Acad Dermatol.2012;66(4):589-597. CrossRef PubMed
  10. Observation of chrysalis structures with polarized dermoscopy Marghoob AA, Cowell L, Kopf AW, Scope A. Arch Dermatol.2009;145(5):618. CrossRef PubMed
  11. Dermoscopic characteristics of nodular squamous cell carcinoma and keratoacanthoma Lin MJ, Pan Y, Jalilian C, Kelly JW. Dermatol Pract Concept.2014;4(2):9-15. CrossRef PubMed
  12. Dermoscopy of actinic keratosis, intraepidermal carcinoma and squamous cell carcinoma Zalaudek I, Argenziano G. Curr Probl Dermatol.2015;46:70-76. CrossRef PubMed
  13. Update on non-melanoma skin cancer and the value of dermoscopy in its diagnosis and treatment monitoring Lallas A, Argenziano G, Zendri E, et al. Expert Rev Anticancer Ther.2013;13(5):541-558. CrossRef PubMed
  14. Squamous cell carcinoma: variation in dermatoscopic vascular features between well and non-well differentiated tumors Pyne J, Sapkota D, Wong JC. Dermatol Pract Concept.2012;2(4):204a05. CrossRef PubMed

Send mail to Author


Send Cancel