Oral Warty Dyskeratoma—A Systematic Review of the Literature

Objective: To systematically review the clinicopathological features of oral warty keratoma based on published literature. Materials and Methods: PubMed and Scopus databases were searched for reports of oral warty dyskeratoma. Of the 52 identified articles, only 25 articles (43 cases) satisfied the selection criteria (case report/series in the English language reporting clinicopathologically diagnosed oral warty dyskeratoma/oral focal acantholytic keratosis/oral isolated dyskeratosis follicularis in humans). Risk of bias was assessed using the Joanna Briggs institute critical appraisal checklist for case reports and case series. Results: Most cases had well-circumscribed, white, nodular verruco-papillary lesions with a central depressed/crater-like area. Alveolar ridges were the most frequent sites of occurrence and tobacco was the most commonly associated risk factor. Histopathologically, the most pathognomonic feature was the supra-basal clefting. The cleft had dyskeratotic acantholytic cells (corps ronds, and grains). Below the cleft were projections of the connective tissue villi lined by basal cells. The basal cells in a few cases exhibited hyperplasia in the form of budding into the stroma, but epithelial dysplasia was not reported. The surface epithelium had crypts filled with keratin debris. Conclusion: Oral warty dyskeratoma is a rare solitary self-limiting benign entity, which due to its clinical and histopathological resemblance and associated habit history could be misdiagnosed as leukoplakia or carcinoma. None of the assessed articles provided molecular data, which in turn could be the reason for the lack of insight into the etiopathogenesis of this enigmatic lesion.


Introduction
Warty dyskeratoma represents a benign solitary tumor of the skin, vulva, and oral cavity [1]. The cause of oral warty dyskeratoma (OWK) is yet to be determined, although most cases have shown an association with tobacco smoking/chewing, local chronic trauma/irritation from a sharp tooth, or an ill-fitting denture [2]. Clinically, the lesion is largely asymptomatic, thus most cases are diagnosed during a routine oral examination. Based on the clinical appearance OWK could be classified as a verruco-papillary lesion [3]. The shape of the lesion is often nodular and well-circumscribed. Few cases have also shown a relatively smooth surface in the form of a papule, or a patch [1,2,4,5]. The color of the lesion is white, which is attributed to the hyper-keratinized state of the surface epithelium. The white color and the associated tobacco habits often lead to a provisional diagnosis of

Materials and Methods
The International prospective register of systematic reviews (PROSPERO) was searched for systematic reviews on oral warty dyskeratoma. There were none. Thus, the present review was registered in PROSPERO (registration number: CRD42020213865). PRISMA guidelines [6,7] were strictly adhered to in the qualitative analysis ( Figure 1). The systematic review was conducted in three steps: (1) PubMed and Scopus databases were searched using the keywords 'oral warty keratoma; oral focal acantholytic keratosis; oral isolated dyskeratosis follicularis'. The search was done in May 2021.
(2) The identified articles were screened for potential duplicates and relevance to the topic using their titles and abstracts.
(3) The full text of the articles selected from screening was assessed using the following selection criteria: Inclusion criteria: Case report/series in the English language reporting OWK/oral focal acantholytic keratosis/oral isolated dyskeratosis follicularis in humans. The diagnosis must have been determined using both clinical and histopathological examinations.
Exclusion criteria: Cases wherein the diagnosis lacked clinical and/or histopathological examination.
The publications satisfying the above criteria were included in the qualitative analysis. Data including the number of cases, clinical and histopathological features, the treatment rendered, and the follow-up data were extracted. Step 2 and 3 of the review were conducted by two reviewers (KHA and ATR). The inter-observer reliability in steps 2 and 3 was determined by assessing the kappa coefficient (κ).
Risk of bias assessment: The Joanna Briggs institute critical appraisal checklist for case reports and case series [8] was applied. The risk of bias was categorized as high when the study attained up to 49% score yes, moderate when the study attained 50 to 69% score, yes, and low when the study attained more than 70% score yes.

Results
A total of 52 published studies were identified. Title and abstract screening led to the exclusion of 21 articles. The remaining 31 articles were manually cross-referenced to identify further potential articles. The cross-referencing led to the identification of four articles whose titles and abstracts were relevant to the topic of interest. Based on the fulltext assessment of the (31 + 4) 35 articles, only 25 articles [1,2,4,5, satisfied the selection criteria and were included in the systematic review. Figure 1 summarizes the search strategy used in the present review. The κ values for steps 2 and 3 of the review were 0.96 and 0.98, indicating good inter-observer reliability. Table 1 summarizes the data extracted from the included studies. A total of 43 cases were retrieved from the 25 articles [1,2,4,5, included in the qualitative analysis.
Presence of potential risk factors: The most common associated risk factor was the habits including consumption of tobacco and alcohol. A total of 13 cases used tobacco [10,11,15,19,[21][22][23]28], one case consumed both tobacco and alcohol [12]. Few cases had a combination of several risk factors, such as tobacco, alcohol consumption with cheek biting in one case [11]; ill-fitting denture, and tobacco usage in one case [21]. Less frequent risk factors included maxillary and mandibular denture (one case) [20], sharp tooth (one case) [2], and snuff dipping (one case) [23].
Size of the lesion: The lesion size ranged from 2 mm to 3 cm. The lesion was less than 1 cm in most (n = 22) of the cases.
Age and gender distribution of the patients: The age ranged from 33 to 81 years, with an average age of 55.7 years. Most (n = 32) of the patients were above 50 years old. Males (n = 26) were affected more often than females (n = 15). In two cases the gender of the patient was not mentioned. A papular elevated lesion with a depressed central area The epithelium was hyperplastic and hyperkeratinized, with a central area composed of the parakeratotic plugged cup-shaped depression. Acantholytic, dyskeratotic cells were noted between the keratotic plug and the depression Villi with cleft and corps ronds were found in the base and lateral wall of the lesion. The epithelial portion surrounding the depressed area was hypergranulotic. The sub-epithelial mucosa of the depressed area had moderate-dense inflammatory infiltrate of lymphocytes, plasma cells, histiocytes, and few eosinophils. A diagnosis of focal acantholytic dyskeratosis/warty dyskeratoma was rendered.
The lesion was completely excised during the biopsy. Follow up details were not provided.
Additional information: The patient was a 49-year-old female. The patients had repeated warty lesions on the skin of the left cheek, which was removed by electrocautery. There was no pain or tenderness in the oral lesion. The duration of the lesion was 12 months. Patients wore complete maxillary and mandibular dentures. Additional information: The patient was a 74-year-old female. The patients did not have any associated habits (smoking, alcohol, etc). Granular mass was noted on the inter-gingival papilla from canine to the second molar. Biopsy of the granular mass revealed keratinizing OSCC. Biopsy of the pre-canine papilla also revealed a portion of the thick hyperkeratotic epithelium with granular vacuolated layer, large irregular kerato-hyaline granules, and cells with indistinct borders. This portion was diagnosed as epidermolytic hyperkeratosis. Due to the appearance of epithelial dysplasia, and invasion into the lamina propria, a histopathological diagnosis of OSCC was rendered. The examination of further sections revealed acanthotic stratified squamous epithelium with numerous thin fibrovascular projections. These projections were covered by ortho-keratinized epithelium with a papillomatous surface. The stratum granulosum was thickened. The crevices between the epithelial projections were plugged with keratin and anucleated squames. Connective tissue villi in some areas were covered by just a single layer of basaloid cells which were hyperplastic and has projections in the lamina propria. One crevice had a cyst-like opening with hyper-keratotic stratified squamous epithelium and was partly filled with keratin. Dyskeratosis resembling single-cell keratinization was also found with numerous abnormal mitotic figures. A mild chronic inflammatory infiltrate was noted in the lamina propria. The diagnosis was changed to oral warty dyskeratoma.
The lesion was completely excised during the biopsy. A 6 years follow-up did not reveal any recurrence Additional information on the 1st patient: The patient was a 58-year-old male, who smoked 20 cigarettes per day. The patient has a root stump in the maxillary 1st molar region accompanied by a discharging sinus. The root stump was removed, after which the sinus opening was replaced by the keratinized lesion. The clinical diagnosis was papilloma, which turned out to be warty dyskeratoma. Additional information on the 2nd patient: The patient was a 51-year-old male, who smoked 20 cigarettes per day for 35 years. Smokers' keratosis was present in the left and right retromolar region. 1 The floor of the mouth Reddish raised lesion of about 6 mm was seen on the floor of the mouth in the skin flaps lateral margin Epithelium had hyper and parakeratosis. Basilar clefting with acantholytic cells and dyskeratosis was noted. Basophilic nuclei were seen in the dyskeratotic cells. A diagnosis of focal acantholytic dyskeratosis/warty dyskeratoma was rendered.
Only an incisional biopsy was made. As the diagnosis was benign, the lesion was untreated. Follow up detail was not provided.
Additional information: The patient was a 63-year-old. Had undergone partial mandibulectomy for OSCC of the anterior floor of the mouth 2 years ago. The excised area was reconstructed using a myocutaneous flap and reconstruction plate. The lesion was completely excised during the biopsy.
2-year follow-up did not reveal any recurrence Additional information on the 1st patient: The patient was a 51-year-old female. The lesion was asymptomatic. The patient had tobacco, alcohol, and cheek biting habit. She was also using an old denture. Additional information on the 2nd patient: The patient was a 33-year-old female. She had a history of cigarette smoking. The lesion was completely excised during the biopsy. Follow up detail was not provided.
Additional information: The patient was a 74-year-old female. The lesion was asymptomatic. The patient also had a tender mucocele (with chronic sialadenitis) in the lower lip Additional information on the 1st patient: The patient was an 81-year-old male. The lesion was asymptomatic. Clinical differential diagnosis was OSCC, chronic ulcer, and granulomatous ulcer. Additional information on the 2nd patient: The patient was a 68-year-old male. The lesion was asymptomatic. The lesion has been for more than 1 year. The clinical diagnosis was hyperkeratosis  Orthokeratinized stratified squamous cell epithelium with a prominent granular layer and a thickened spinous layer. An ortho and para keratin core disrupted the epithelial continuity. Suprabasal cleft with acantholytic cells. Connective tissue villi lines by columnar basal cells were noted at the base of the cleft. The basal cells had centrally or eccentrically placed nuclei. Mild chronic inflammation was seen in the stroma. A diagnosis of focal acantholytic dyskeratosis/warty dyskeratoma was rendered.
The lesion was completely excised during the biopsy.   The surface epithelium had an orthokeratin filled cup-shaped invagination. Suprabasilar clefting was noted in the epithelium adjacent to the invagination. The cleft showed grains (dyskeratotic para keratinized cells). The basal cells below the cleft showed hyperplasia. The underlying stroma had fibrosis and focal chronic inflammatory infiltrate. A diagnosis of warty dyskeratoma was made. A diagnosis of focal acantholytic dyskeratosis/warty dyskeratoma was rendered.
The lesion was completely excised during the biopsy. Follow up period for each case is not mentioned, although it is mentioned that no recurrence was noted.
Additional information: The patient was a 60-year-old male. The duration of the lesion was 3 months. The lesion was asymptomatic. The patient had a 44-year history of smoking 20 cigarettes per day. The lesion felt fibrotic on palpation.
Symptoms: A majority of cases did not report any symptoms from the oral lesion. Only five cases had symptoms including pain (one case) [17], mild tenderness (one case) [1], and soreness (two cases) [22,23].
Duration of lesion: Only 11 studies provided the duration of the lesion, of which seven studies reported less than 1 year. The shortest duration was 2 weeks [27], and the longest duration was 20 years [17].
Clinical diagnosis: The proximity to a sharp tooth led to the provisional diagnosis of a chronic ulcer [2]. The verrucous/pebbly appearance and firm consistency in palpation led to the provisional diagnosis of papilloma [1,22,28]. The excess keratin deposition gave the lesion a white appearance, which in turn was often diagnosed clinically as hyperkeratosis [13,18,22]. The white colour of a patch-like lesion along with an associated habit history led to the provisional diagnosis of leukoplakia [22]. A verrucous, or ulcerated lesion with an associated habit history led to the provisional diagnosis of OSCC [15,28]. One case each was diagnosed as stomatitis nicotine [22], burnt palate associated with slow-healing ulcer [22], fibroma [14], basal cell carcinoma [15], and actinic keratosis [15].
Associated lesion: One case had a history of a warty lesion on the skin of the left cheek [20]. One case had a history of OSCC, which led to mandibulectomy, followed by reconstruction of the resected mandible with a mucocutaneous flap and reconstruction plate [4]. The OWK occurred on the lateral margins of the myocutaneous flap. In one case, there was a co-existing OSCC and epidermolytic hyperkeratosis [29]. In one case, a root stump with discharging sinus was removed following which the sinus was replaced by the warty dyskeratoma [28]. One case had a co-existing smoker's keratosis [28]. One case had co-existing mucocele with sialadenitis in the lower lip [5]. One case had a co-existing verrucous xanthoma, which was diagnosed during the histopathological examination [26].
Clinical presentation: The majority of the OWK had a central depressed area which was described as ulceration, sinus opening, depressed umbilicated centre, crater, cavitation. In a few cases, the central area was described as fissures [18], and fistula [23] was used. The lesioned surface in most cases was verrucous/warty/pebbly (Figure 2). The lesion was described as nodular to papular. The prominence of keratinization in the lesions could be evidenced by the clinical description of the lesion as a white raised lesion with a central depressed area where keratin debris could be picked off [28]. The lesion was well-circumscribed in most cases. Two cases had a white stria radiating from the central portion of the lesion [10,11]. The crusted appearance was also described in a few cases involving the lip [16,17].
Histopathological presentation: The microscopic descriptions were largely uniform. The most common and characteristic feature was the intra epithelial clefting (Figure 3green arrow). The cleft was in most cases noted in the supra-basal region. The bottom of the cleft had connective tissue villi projections lined by a single row of basal cells which in some cases exhibited hyperplasia without atypia (Figure 3-yellow arrow). The upper portion of the cleft had the spinous layer and stratum granulosum. Acanthosis was evident in some cases. The granulosum layer was prominent with large keratohyalin granules in some cases. The cleft itself contained acantholytic cells which had a varying degree of dyskeratosis. The characteristic corps ronds and grains were also described in a few cases [2,5,[11][12][13]15,17,20,21,24,27,28]. The surface epithelium showed a cup-shaped depression corresponding to the areas clinically described as ulceration, sinus opening, depressed umbilicated centre, crater, and cavitation. The depressed epithelium contained ortho/parakeratin debris (Figure 4-green arrow). The pseudo-epithelial proliferation was evident in a few cases (Figure 4-yellow arrow), which had often led to the misdiagnosis of OSCC. Despite a few atypical features, the cases did not report any epithelial dysplasia.
Diagnosis: All the included cases were diagnosed as warty dyskeratoma/focal acantholytic dyskeratosis. Incisional biopsy of two cases was misdiagnosed as OSCC due to the pseudo-epithelial proliferation and dyskeratosis [18]. In those cases, the excisional biopsy revealed the characteristic features of FAD including the suprabasal clefting, keratinfilled debris in the crater of the surface epithelium, corp ronds, grains, and the lack of epithelial dysplasia. Treatment: Excisional biopsy was the most common treatment strategy employed. In one case, an incisional biopsy was performed. Since the biopsy confirmed a diagnosis of warty dyskeratoma, the rest of the lesion was left untreated [4].
Follow up: Only nine articles provided the follow-up period, which ranged from 5 months to 14 years. No recurrence was noted in any of the reported cases.
Risk of bias assessment: Among 20 included case reports, only four case reports [2,10,13,29] provided the follow-up details. Additionally, one case report did not provide the clinical details [29]. Among the 5-case series, none provided any clear data on the consecutive or the complete inclusion of participants. Two case series [18,23] did not provide the followup details. Despite the above-mentioned limitations, the Joanna Briggs institute critical appraisal checklist showed that all the included case reports and series had only a low risk of bias. Tables 2 and 3 summarize the Joanna Briggs institute critical appraisal checklist for case reports and case series, respectively.  Risk of bias categorized as high when the study attained up to 49% score yes, moderate when the study attained 50 to 69% score yes, and low when the study atttained more than 70% score yes.

Discussion
Warty Dyskeratoma was first reported in 1957 on the skin [30]. The first report of an OWK was issued 10 years later by Gorlin et al. [9]. OWK is a relatively less explored entity, which in turn could be largely attributed to its rare occurrence and benign self-limiting nature. Although it is a small non-progressive lesion, its clinical and histopathological resemblance to oral potentially malignant disorders (OPMDs) and OSCC often mandates additional investigation. In addition to its close appearance to a malignant entity, OWK has also been associated with major risk factors for malignancy including tobacco and alcohol [10][11][12]15,19,22,23,28]. Thus, it is necessary to understand the natural history of this enigmatic solitary lesion of the oral cavity. The present systematic review is largely based on individual case reports/case series with only one extensive published case series reporting 15 cases [22]. Given the rarity of the lesion, at present, no original studies exist analysing the varying aspects of OWK. Very few additional investigations (ultramicroscopic and immunofluorescence) are reported in the case reports/series [10,21,24]. Thus, the present systematic review is focussed largely on analysing the various clinical and histopathological features of OWK.
All the cases reported in the review were solitary. The importance of a solitary presentation lies in the close resemblance of OWK with Darriers disease, even entailing the designation of 'pseudo-darriers disease'. Thus, in cases where the OWK-like lesions are multiple as in the present cases, Darriers disease must be considered as a potential diagnosis. OWK can be distinguished from Darriers disease by the absence of germline mutations in ATP2A2 (the gene responsible for Darier disease) and the absence of SERCA2 protein [1,2].
As the term warty indicates, the lesion has a verruco-papillary surface in most cases. Thus, the clinical differential diagnosis often includes verrucous hyperplasia (VH), verrucous carcinoma (VC), and papillary OSCC [31]. Unlike VC, VH, and OSCC, the OWK is self-limiting with no potential for malignant transformation. In such cases, the difference in the size could be a major indicator of the nature of the lesion. Most cases of OWK are less than 1 cm, which is contrary to the progressive nature of VC, VH, and OSCC. Additionally, the OWK is well-circumscribed, which is not the case in locally invasive lesions including VC, VH, and OSCC [32]. In addition to the larger verrucous lesions, several virus-associated oral lesions are less than 1 cm in dimension. An example of such a lesion is squamous papilloma. In such cases, the presence of the viral particles in the squamous papilloma lesion would be a distinguishing factor [33]. Electron microscopic studies have confirmed the absence of viral particles in OWK, although the data are based on only two included studies [21,24]. Warty dyskeratoma from extra-oral sites has also been shown to be negative for viral (human papillomavirus) DNA [34]. Thus, although similar in presentation to most virus-induced papillary lesions, OWK may not have a viral etiology.
As OWK is considered to be a result of abnormal keratinization, all the cases presented as white lesions representing excess keratin production. Most cases present as a nodule, although a few cases of patch-like or papular lesions are also reported. A white patch-like lesion is often provisionally diagnosed as leukoplakia [22], especially with an associated tobacco history. In such cases, the absence of epithelial dysplasia, presence of suprabasal clefting, keratin-filled crypts, and the acantholytic dyskeratotic cells (corps ronds, grains) would be the distinguishing features from true tobacco-induced potentially malignant leukoplakic lesions. Few cases have presented data on consistency as 'firm' based on palpation [19]. The firmness can be attributed to the collection of keratin debris in the crypts of the surface epithelium. The most characteristic clinical feature of an OWK was the central areas described as depressed, crater-like, and umbilicated. This area is often filled with varying degrees of keratin debris imparting the white colour and firm consistency.
Microscopic examination reveals a series of depression and elevations in the surface epithelium which are consistent with the verruco-papillary surface of the lesion. The depression is filled with keratin debris, although overall the entire affected surface epithelium has a varying degree of hyper-ortho/parakeratosis. The characteristic histopathological feature of OWK is the cleft noted in the suprabasal region of the epithelium. The cleft is bound on its lower part by a series of projections created by connective tissue villi, which are in turn lined by a single row of basal cells. These basal cells exhibit varying degrees of hyperplasia, although atypia is not reported. The basal cell hyperplasia often leads to budding into the underlying stroma causing the appearance of pseudo-epithelial hyperplasia, which along with dyskeratosis can be misdiagnosed as OSCC [28]. In such cases, the absence of epithelial dysplasia and the presence of characteristic suprabasal clefting and keratin-filled crypts would aid in the diagnosis of focal acantholytic dyskeratosis. Within the suprabasal cleft are the dyskeratotic acantholytic cells (corps ronds and grains) [2,5,12,13,15,17,20,21,24,27,28]. Above the cleft are the thickened layer of spinosum (acanthosis) [11,12,15,16,21,23,27] and granulosum [5,20,23,28].
Although the lesions are relatively small, given their close resemblance to OPMDs and OSCC and the associated tobacco history, most cases were excised completely. None of the cases included in the present review showed any signs of recurrence or malignant transformation.

Conclusions
OWK is a rare benign solitary well-circumscribed, white verruco-papillary lesion involving the keratinized mucosa. The most characteristic clinical feature was the central depression/crater/umbilicated centre. Histopathologically, an OWK can be precisely described as a focal acantholytic dyskeratosis with varying degrees of dyskeratosis. The most characteristic histopathological feature was the supra-basal clefting, keratin-filled crypts, corps ronds, and grains. Their close resemblance to OPMDs and OSCC and the presence of associated tobacco habits mandates a histopathological confirmation of the true nature of the lesion. To date, there is no evidence of recurrence or malignant transformation in an OWK. Thus, despite its likeness to OPMDs and OSCC, an OWK is a benign selflimiting entity that, apart from a histopathological confirmation, requires no therapeutic interventions.

•
OWK is a benign lesion with clinicopathological resemblance to OPMD and OSCC • Common clinical feature includes central depression/crater/umbilicated centre Funding: This research received no external funding.