Cutaneous squamous cell carcinoma (cSCC) accounts for 20%‒30% of non-melanoma skin malignancies.1 cSCC encompasses a broad spectrum of clinical entities that vary in terms of cellular atypia and dysplasia and range from precursor lesions to in situ cSCC and invasive cSCC.2 There are three main subtypes of cSCC: Bowen’s disease (BD), keratoacanthoma (KA), and invasive cSCC.3

Dermoscopic findings of BD (i.e., in situ SCC) include a pink or pigmented background accompanied by clustered glomerular vessels and dotted vessels, yellow-white squams, and erosion.4 KA shares common clinical and histopathological features with well-differentiated cSCC.5 KA very rarely transforms to invasive cSCC and thus its accurate diagnosis prevents overtreatment.6

Invasive cSCC has specific dermoscopic signs related to its differentiation and depth of invasion.7 The dermoscopic structures seen in cSCC can be summarized as follows: 1) Keratin/central keratin, 2) Vascular structures, 3) Ulceration, and 4) White structureless areas.8 As invasive cSCC differentiates further, the keratin structures disappear, a polymorphic pattern of linear vessels covers the lesion, and ulceration develops.9 Histopathological differentiation is an important risk factor in cSCC tumor aggressiveness and recurrence risk.3,10 Early differential diagnosis of the variants of cSCC is especially important considering the metastatic potential of invasive cSCC.11

This present study aimed to examine the distribution of dermoscopic findings in cases of invasive cSCC with different histopathological differentiation, as well as compare dermoscopic findings according to different morphological characteristics in cSCC patients and evaluate the dermoscopic findings of patients presenting with nodular invasive cSCC compared to those of patients with KA. The authors also aimed to further contribute to the literature by reviewing the dermoscopic findings of BD.

The sociodemographic and clinical characteristics of these patients were consistent with previous studies indicating that most cSCC patients are older and male, and lesions are most commonly located in the head and neck region.16,17 The incidence of actinic keratosis increases with age, and this increase is higher in the 70-year age range.18 Multiple actinic keratosis is one of the risk factors for the development of invasive cSCC.19 In a study, the number of multiple actinic keratoses accompanying invasive cSCC was reported as 11.5%.17 This rate was found to be higher in the present study. This may be related to the average age in this study being 69.05 and patients being regularly followed up for actinic keratosis and cSCC.

In different studies, dermoscopic pigmentation in cSCCs is reported to vary between 0.01% and 25%.20 Its prevalence was found to be higher (5.5%) in pigmented BD.21 In the present study, this rate was found to be 6.7% and it was most commonly observed in BD. When evaluating dermoscopic findings, it should be kept in mind that pigmentation may rarely be seen in KA and invasive cSCC.

Among the dermoscopic findings of cSCC, the most common is linear irregular vessels.3,16 If linear vessels have more than one turn, it is referred to as a serpentine vessel.14,15 The linear irregular vessel structure is also known as a serpentine vessel.22,23 In a study by Lallas et al., the dermoscopic findings of invasive cSCC included radial linear irregular and dotted vessels.16 Although the studies use different nomenclatures of dermoscopic terms, the dermoscopic features of serpentine and dotted vessels should be considered in the diagnosis of cSCC, consistent with the literature.

Rosendahl et al. determined that the vessel arrangement in KA and invasive cSCC was most frequently random, i.e., nonspecific.8 In the present study, a similar distribution was observed in subtypes other than BD. Several studies have demonstrated glomerular vessels and a clustered vascular arrangement to be important findings in the dermoscopic diagnosis of BD.3,9 As a result, glomerular vessels and clustered vascular arrangement help to differentiate from KA and invasive cSCC when diagnosing BD.

Well-differentiated invasive cSCCs show an exophytic growth pattern and are present with white structures accompanied by squam on dermoscopy.16,23 Sgouros et al. detected the dermoscopic finding of white perifollicular openings more frequently in well-differentiated invasive cSCC.3 Manfredini et al. reported that the dermoscopic finding of white structureless areas was more common in well-differentiated invasive cSCC.24 In a study conducted in 2024, it was determined that dermoscopic findings such as centrally distributed keratin, white circles, and whitish perivascular halo were more commonly observed in well-differentiated invasive cSCC compared to non-well differentiated invasive cSCC.25

Lallas et al. emphasized that a central distribution of scales or keratin on dermoscopic images was a negative predictor of the diagnosis of poorly differentiated invasive cSCC.16 In this study, dermoscopic findings of the central keratin plug, keratin crust/scale, and white structureless areas were seen more frequently in well-differentiated invasive cSCCs.

Poorly differentiated cSCC presents clinically as a flat lesion and dermoscopically as red in color, often with abnormal vascular structures.16,24 In a study, it is emphasized that the dermoscopic finding of arborizing vessels is important in the diagnosis of poorly differentiated invasive cSCCs. In this study, invasive cSCCs were examined in two groups: well-differentiated and non-well-differentiated.25 Both this study and the study by Sgouros et al.3 examined invasive cSCC across three different histopathological subtypes. No difference was found between these 3 different histopathological subtypes in terms of dermoscopic findings of vascular structures. Among the patients with poorly differentiated cSCC in the present study, ulceration was detected clinically in 75%, and ulceration/bleeding and red background color were more common dermoscopically. The authors believe that further research is needed on this subject.

In the literature, the dermoscopic findings of white circles surrounding follicles and white halos surrounding vessels are not used separately but referred to collectively as white circles.16,22 However, Papageorgiou et al. examined these two dermoscopic findings separately and emphasized that white circles surrounding follicles supported a diagnosis of poorly differentiated invasive cSCC.24 The findings of our study support this. In the same study, white halos surrounding vessels were determined to be an important dermoscopic finding for early invasive cSCC. In the present study, white halos surrounding vessels were most frequently detected in well-differentiated cSCC and were important in differentiating nodular invasive cSCC from KA.

The presence of keratin on dermoscopy has high sensitivity in the differentiation of invasive cSCC and KA from cutaneous basal cell carcinoma, one of the other nonmelanocytic skin cancers.8 However, the positive predictive value of this dermoscopic finding decreases to 50% when distinguishing cSCC and KA from BD. In the present study, among the dermoscopic signs of keratinization, white circles and white structureless areas were less common in patients with BD. As in this study and the study by Rosendahl et al., the dermoscopic findings of white circles and white structureless areas are supportive of cSCC and KA when differentiating from BD.8

In their study, Sgouros et al. emphasized that dermoscopic erosion was mostly seen in patients with BD, while ulceration was mostly seen in patients with invasive cSCC.3 In this study, ulcerations were not seen in any of the patients with BD, whereas erosion was detected more frequently than invasive cSCC and KA. When evaluated together with the literature, these findings stand out as important in the dermoscopic differentiation of BD.

Dermoscopic findings seen in KA include a white background, central keratin plug, and linear irregular or hairpin vessels.3,6 The literature emphasizes the similarity of dermoscopic findings in KA and invasive cSCC.8,26 The most common dermoscopic finding is keratin crust/scale, observed in 100% of KAs and 90% of invasive cSCC.23 The dermoscopic findings of keratin crust/scale and white background were most commonly observed in KA in the present study, consistent with the literature. However, the central keratin plug is seen significantly more in KA than in the other subtypes. As this study supports, the central keratin plug is important in the diagnosis of KA.

The most common morphological presentation of KA is nodular lesions with well-defined margins and central hyperkeratotic plugs.5 In the present study, 69.5% of KA patients presented with nodular lesions. Nodular and ulcerated lesions are morphologies seen in invasive cSCC.3 Dermoscopic differentiation of KA and invasive cSCC is important in patients presenting with similar morphology. Lin et al. argued that the central keratin plug was important in distinguishing KA from nodular cSCC.23 The present findings also support this. In the same study, glomerular vascular structures were reported to be the most common vascular structures in nodular cSCC, while no difference in vascular structures or vascular arrangement was found between cSCC and KA. In the study by Pyne et al.,27 the branching vessel structure was found to be more common in KA than in invasive cSCC. No comments have been made about lesion morphologies in this study. In the present study, branched vascular arrangement was seen more frequently in patients with nodular KA. In the dermoscopic diagnosis of nodular KA, the presence of central keratin plug and branched vascular arrangement structures should be considered.

Limitations of this study are its retrospective design, the relatively small number of examined lesions in the subgroups, and evaluation of the clinical morphology of the lesions based solely on clinical images, without palpation.

In this study, specific dermoscopic findings were obtained for the differential diagnosis of BD and nodular KA. However, specific dermoscopic findings were not detected for moderately differentiated invasive cSCC.

The accuracy of the proposed dermoscopic criteria cannot be generalized as a result of the present study, but it will assist clinicians in this regard due to the limited number of studies in the literature.

Mustafa Ürün: Approval of the final version of the manuscript; study conception, design, and planning; collection, analysis, and data interpretation; writing; critical literature review and critical review of the manuscript.

Yıldız Gürsel Ürün: Approval of the final version of the manuscript; study conception, design, and planning; data interpretation; critical review of the manuscript.

Ömer Faruk Elmas: Approval of the final version of the manuscript; analysis, and data interpretation; critical literature review.

Nuray Can: Approval of the final version of the manuscript; analysis, and data interpretation; writing; critical literature review.

None declared.