Treatment Pattern and Outcomes in Verrucous Carcinoma of Oral Cavity: A Single Institutional Retrospective Analysis from a Tertiary Cancer Center and Review of Literature

Abstract

Verrucous carcinoma (VC) is a locally invasive uncommon histopathological variant of oral squamous cell cancer. There is paucity of literature regarding control rates in these cases. We intend to report the outcomes in terms of administered treatment and control rates. 28 patients of oral cavity verrucous carcinomas treated at our institute from March 2014 to December 2018 were reviewed retrospectively. Demographic profile, histopathological features and clinical outcomes were analyzed. Statistical analysis was performed with SPSS for Mac (version 23.0). Median age was 54 years (range 31–75) with M:F ratio of 25:3. Buccal mucosa was the most common site. All patients underwent surgical resection except one. Of these, 24 had neck dissection; 12 had supra-omohyoid neck dissection, eleven had modified neck dissection and one patient underwent radical neck dissection. Three patients had their histology upgraded to squamous cell carcinomas in the post-operative histopathology. The post-operative staging was as follows: 21% stage I and 35% stage II. One patient opted for non-surgical approach and received radical concurrent chemoradiotherapy. Median follow up was 12 months (range 6–36). Two patients had local failures and one had a regional failure. No distant metastasis was found. There was one death. 14-Months survival rate was 92%. Estimated 18 month loco-regional control rate was 92%. Curative surgical resection remains the cornerstone for VC of oral cavity. Any change of histopathology post-operatively to squamous cell carcinoma is a poor prognostic sign and needs appropriate adjuvant treatment.

Introduction

Squamous cell carcinomas of the head and neck are one of the most commonly encountered cancers [1]. Verrucous carcinoma (VC) is one of the more uncommon and rare variants. It was first described by Ackerman [2]. It can be described as a slow growing clinico-pathological entity which is locally invasive, and very unlikely to metastasize. It generally appears as a painless, thick white exophytic plaque. It is well demarcated and lacks ulceration. It bluntly invades underlying soft tissues and may become fixed to the periosteum with subsequent invasion of bone. The prevalence of VC varies from 2 to 12% in various published literature [3].

VC can occur in various sites, including head and neck, skin, genitalia (including penis, vagina and vulva), esophagus and extremities [4]. Of the various sites in head and neck region, oral cavity is one of the most common sites for VC followed by larynx. In oral cavity, buccal mucosa is the most common subsite to be affected [5].

The primary management of VC is surgical resection [6]. Surgery includes a wide local excision. The role of neck dissections in VC is not very clear. Several authors favor the use of neck dissection, while some tend to avoid it [7]. After surgery, the role of adjuvant therapy is not well defined. The role of radiotherapy (RT) and chemotherapy is controversial. Some authors consider that RT is an over-treatment and can be detrimental [8], while many clinicians advocate the use of RT in high risk features like pathological T3/T4 cases, close or positive margin, lympho-vascular invasion, perineural invasion, positive nodes and presence of extra-nodal extension [8–10]. A foci of invasive squamous cell carcinoma might be seen in these types of tumors and it may lead to recurrence if adequate local treatment is not undertaken. So, RT may play a role in decreasing the incidence of local recurrence in these cases.

Reports on outcome of oral cavity VC is limited. No definite consensus based or evidence-based management approach exists for VCs, given the rarity of occurrence and inadequate reporting of existing cases. Therefore, we aimed to evaluate the cases of oral cavity VCs treated at our center and report the treatment patterns and outcomes in this cohort of patients.

Materials and Methods

Data of forty-four patients with histo-pathology of VC was taken out from our central pathology review. Of these, nine patients were excluded as they were not treated at our center. Of the remaining thirty-five, seven patients were excluded because of sites other than head and neck (two penis, two vagina, one vulva, one vagina and one skin). The final analysis was conducted on the remaining twenty-eight patients.

It was a retrospective study of patients treated at a tertiary cancer center from March 2014 to December 2018. Institutional review board approval was obtained for the study. The patient demographic profile, addiction patterns, clinico-pathological factors, histopathological features, staging and outcomes were analyzed. The staging was done as per AJCC 7th edition [11]. The surgical practices followed were wide local excision with/without neck dissection. The detailed surgical practices followed in the cases and the types of neck dissection were documented. All the histo-pathological slides were reviewed and assessed at the central pathology of the institute. The decision to administer adjuvant therapy was based upon the post-operative histopathological findings. The patterns of presentation and the failure patterns for the cohort were analyzed.

The data collected was analysed using statistical package for social sciences (SPSS) software for Mac version 23.0. All outcomes were evaluated from the time of registration. Local, regional and loco-regional control was defined as time from registration to failure in tumour bed, regional nodes or both, respectively. Overall survival (OS) was defined as time from registration to death from any cause. Kaplan–Meier method was used for all outcome analysis. p < 0.05 was considered as statistically significant.

Results

Median age was 54 years (range 31–75). Male to female ratio was 25:3. All the patients had tobacco addiction either in chewable form or in the form of smoking. Eighty six percent (24/28) of the patients were tobacco chewers and 53.5% (15/28) were tobacco smokers. Eleven patients (39%) had both the addictions of smoked and smokeless tobacco. Three patients were also addicted to alcohol. The median duration of symptoms after which the patient presented to our clinic was 8 months (range 3–72). The detailed demographic characteristics are described in the Table 1.

Table 1.

Patient demographic characteristics

Clinico-pathological feature analyzed	N (%) (n = 28)
Gender
Male	25 (89%)
Female	3 (11%)
Age
< 40	4 (14%)
40–60	16 (57%)
> 60	8 (29%)
Addictions
Tobacco chewing	24 (86%)
Tobacco smoking	15 (53.5%)
Both chewing and smoking	11 (39%)
Alcoholics	3 (11%)
Site
Buccal mucosa	23 (82%)
Lip	3 (11%)
Tongue	1 (3.5%)
Hard palate	1 (3.5%)
pT stage
pTis	11 (46.5%)
pT1	6 (21%)
pT2	11 (46.5%)
pN stage
pN0	27 (96.5%)
pN1	1 (3.5%)
Overall AJCC staginga
0	11 (39.5%)
I	6 (21.5%)
II	10 (35.5%)
III	1 (3.5%)

pT pathological tumor stage, pN pathological nodal stage

^aAll the staging mentioned are 7th edition AJCC staging

All the patients had a pre-operative histopathology of VC, either primarily performed at our centre or reviewed at our central pathology. All the patients underwent surgical excision with/without neck dissection except one as the patient refused surgery. Of these, four underwent marginal mandibulectomy and 2 had hemi-mandibulectomy. Twenty-four patients (89%) patients had neck dissections. Of these, twelve (50%) were supra-omohyoid neck dissection (SOND), eleven (46.5%) were modified neck dissection (MND) and one (3.5%) patient had radical neck dissection.

Eleven of these patients had post-operative stage of pTis, six had pT1 and eleven had pT2 disease. The median size of the lesion was 3 cm (range 1–6). Only one patient had a positive neck node at level Ib. All others were node-negative. All the patients had clear margins. Risk factors like peri-neural invasion, lympho-vascular invasion and extra-nodal extension were not found in any of these patients.

One patient received neo-adjuvant chemotherapy of 4 cycles of three weekly Paclitaxel (175 mg/m²) and Carboplatin (AUC 6) as the patient presented after 24 months of initial presentation and subsequently had histopathology changed to moderately differentiated squamous cell carcinoma post-operatively. A subset analysis of the twenty-three buccal mucosa patients revealed that 30% (7/23) patients required a form of bone resection, either marginal or segmental. 2/23 (9%) cases had invasive component on final histo-pathology. None of them had bony invasion.

In three of these patients, the post-operative histopathology was changed from VC to oral cavity squamous cell carcinomas (OSCC). One changed to well differentiated carcinoma, one changed to moderately differentiated carcinoma and one changed to poorly differentiated carcinoma. The clinico-pathological details of the cases are enumerated in the Table 2. These patients were subsequently treated as squamous cell carcinomas. The patient with poorly differentiated histopathology received adjuvant post-operative RT alone because of nodal positivity and poor differentiation. The RT dose was 60 Gray in 30 fractions delivered by 3-dimensional conformal radiotherapy. This patient subsequently failed regionally after 14 months.

Table 2.

Analysis showing clinic-pathologic features and outcomes in oral verrucous carcinoma cases with post-operative squamous cell carcinoma histopathology

Sl no	Age/gender	Site	pT/pN/TNM stagea	Post-op HPE	Surgery (with neck dissection details)	Local/regional failure	Survival
1	65/male	Buccal mucosa	pT2/pN1/stage III	Moderately differentiated SCC	WLE + left hemimandibulectomy + Lt MND	Local failure	NAD
2	31/male	Tongue	pT2/pN0/stage II	Poorly differentiated SCC	WLE + right MND	Regional failure (Level IB)	Death at 14 months
3	70/male	Buccal mucosa	pT1/pN0/stage I	Well differentiated SCC	WLE + marginal mandibulectomy + leftt SOND	NAD	NAD

SCC squamous cell carcinoma, WLE wide local excision, NAD no abnormality detected, MND modified neck dissection, SOND supra-omohyoid neck dissection

^aAll the staging mentioned are 7th edition AJCC Staging

There was one regional failure and subsequently the patient expired. There was one death in our study. That patient had VC of tongue, which placed it in a higher risk category for both local and distant recurrence. The post-operative histopathology in that patient was upstaged to grade 3 OSCC. The patient received both surgery and radiotherapy to bilateral face and neck. The patient had regional recurrence in level II cervical nodes and expired after that.

Median follow up was 12 months (range 6–36). Two patients had local failures and one had a regional failure. No distant metastasis were found. There was one death. The 14 months survival rate was 92%. The 12-month local and regional control rates were 100%. The estimated 18 months loco-regional control rate was 92% (Fig. 1).

Fig. 1.

Kaplan–Meier curves depicting overall survival. Time on Y-axis is depicted in months

After exclusion of the three patients having squamous cell carcinoma, a post hoc analysis was performed to report the outcomes in pure oral verrucous carcinomas. There was one local failure in twenty-four analysed cases, translating into crude local control rate of 96% and crude regional control rate of 100% (Fig. 2).

Fig. 2.

Kaplan–Meier curves depicting Loco-regional failure free survival. Time on Y-axis is depicted in months

Discussion

VC is considered to be an indolent variant of OSCC. While some consider it to be part of histologic continuum of leukoplakia to OSCC [12], others consider it to be a different clinic-pathological entity [13]. Clinically, it appears as a wart-like exophytic lesion and histo-pathologically, it is a low-grade, well differentiated squamous cell carcinoma [9]. They lack the usual cytologic criteria of malignancy. There is no significant cellular atypia or minimal individual cell dysplasia. Mitoses are rare; when observed they are located in supra-basal position immediately above the basal cell layer, where they occur normally. The basement membrane remains intact throughout the lesion [9, 14]. The tumour cells have acidophilic cytoplasm with minimal nuclear atypia and features of keratinization, koilocytosis and squamous pearl formation. There is a chance of microinvasion and/or frank invasive squamous cell carcinoma foci being present in the edges or inside the tumors [9, 13, 14]. Slootweg et al. postulated that the existence of verrucous carcinoma should be considered as an expression of ubiquitous pre-malignant change in the whole of oral cavity mucosa [14]. They found around one-third patients (37%) had co-existing squamous lesions whenever verrucous lesion was reported in the oral cavity.

VC has comparable clinical and histological picture proliferative verrucous leukoplakia (PVL), verrucous hyperplasia and hybrid oral verrucous carcinoma. Because of the scarcity of cytologic features of dysplasia, the presence of background inflammation, clinical overlap with PVL, and the risk of progression to OSCC, it is important that adequate tissue be obtained at the time of biopsy of suspected VC [10, 12]. Punch biopsies can be non-informative. Therefore, an incisional biopsy is necessary. It should include a generous portion of the lesion, including the submucosa. Because of the characteristic features of the border between the lesion and normal tissue, the biopsy should include some adjacent normal appearing tissue.

Various published literature has described that VC is more common in males and the common age group it is found is in the fifth and sixth decade [7]. Our study had 89% male patients. Around more than two-thirds patients were below the age of 60 years, which could be because of the fact that we had multiplicity of addictions in many cases. Buccal mucosa is the most common site mentioned in most of the literature as in our study [3, 5, 7].

The exact etiology of VC is unknown. It is known to have a definite association with tobacco and alcohol addiction [13, 15–17]. We also observed a similar addiction pattern in our study, wherein all cases had some form of tobacco addiction, smoked or smokeless. In the Indian subcontinent, chewable form of tobacco is more commonly used to compared to smoked forms of tobacco. It is also seen to be associated with premalignant lesions like leukoplakia and erythroplakia, which tend to adversely affect the disease-free survival. Studies have reported the incidence of erythroplakia in the range of 34–48% [5, 18]. We observed leukoplakia in 25% cases. The present study did not find a significant relationship between leukoplakia and disease-free survival, which can be attributed to the low mortality and limited follow-up period. VC is often found to be co-associated with poor oral and dental hygiene, resulting from a multiplicity of factors like poor socio-economic status and various form of addictions [5, 17]. Since it is very commonly seen in people who chew or snuff tobacco, it is also called as snuff dipper’s cancer.

Various treatment modalities have been tried for management including, surgery, radiotherapy, chemotherapy, photodynamic therapy, cryotherapy, laser therapy and treatment with recombinant alpha-interferon [4]. There have been mixed reports about the outcomes of treatment. Surgery is the primary modality of treatment of VC [19]. Since these lesions are clinically ambiguous, surgical excision with appropriate margins is mandated. VC involving tooth bearing mucosa should be treated with excision of underlying alveolus. Also, these lesions can be multifocal or may have a focus of squamous cell carcinoma, which further focus on the role of excision with wide margins [17, 20]. RT, whenever indicated is necessary for management and should be considered. Often multi-modality approach achieves better results when compared to single modality, i.e., surgery alone as was advocated by a study of national survey on head and neck verrucous carcinomas [10]. Few authors have raised suspicion of anaplastic transformation of tumours following RT [12], but these reservations have been refuted and it is now considered safe enough to use RT even upfront, when surgery is not feasible or not indicated [8–10, 20–22]. Several authors have published details about patients of VC treated with RT and having excellent outcomes [5, 7, 23].

The role of neck dissection remains ambiguous since no proper guidelines exist [7]. As per the literature, around 1/6th–1/4th of the VC patients present with palpable neck nodes but, they seldom turn out to be positive [5, 7]. It is generally difficult to perform a neck dissection secondary to a wide local excision as the surgical resection may involve intricate flap reconstruction procedures. Most of the patients in the current study have had neck dissections (89%). Contemporary literature suggests excising one node adjacent to the suspicious node, and in general, SOND is sufficient to address the neck in VC. However, MND can be considered if there is high suspicion of positive nodes being present in the post-operative histopathology. In the present study, no statistically significant difference in survival was found between SOND, MND and radical neck dissections (p-0.23), although it is worth mentioning that only one patient had radical neck dissection. Considering omission of neck dissection in VC continues to be a debatable discussion. There is a general consensus that neck dissection should be done in the primary setting, if there is suspicion of higher grade, large tumours or positive nodes being present. Although a staged neck dissection can also be considered a reasonable approach, but that might delay wound healing and subsequent administration of adjuvant treatment and can result in worse outcome. Therefore, in cases with uncertain clinical node positivity, it is prudent to consider a SOND, rather than leave the neck unaddressed. In our study, all the patients with neck dissections had node-negative disease except one. A thorough literature review suggests that only very few studies have reported the presence of pathologically positive neck nodes [5, 7, 24]. Interestingly enough, that patient was found to be upstaged to moderately differentiated OSCC in the detailed post-operative histopathology.

All the patients in our study had clear margins. There is no consensus with respect to the size of the surgical margins necessary to decrease the risk of recurrence. The general consensus followed for OSCC, that > 5 mm is considered as clear margin is also applied to VCs [5]. Close margin (< 5 mm and > 1 mm) is associated with poor prognosis [25] and the patients with close margins should receive adjuvant RT [7, 24]. Depth of invasion (DOI) in VC is one of the aspects which remains to be addressed. DOI assumes importance in light of the latest AJCC classification, which considers it an important criterion for the pT staging. The median DOI in our study for patients with ≥ pT1 disease was 3 mm (range 1–10). DOI is an important factor in VC as it is in OCSS. One patient was given RT as it had a depth of invasion of 10 mm and was a case of VC tongue.

One of our patients received neo-adjuvant chemotherapy in view of a large lesion present at diagnosis. Post-operative histopathology was found to have one pathological positive node. After surgery, this patient had local recurrence. So, delaying primary surgery also appears to be a risk factor for local recurrence.

Accurate diagnosis of VC remains a problem. It needs to be differentiated from other similar appearing benign lesions like verrucous hyperplasia and keratosis and malignant lesions like well differentiated OSCC [14]. VCs have a very indolent course of progression and carry excellent prognosis. Although some authors report local recurrence rates as high as 28%, many authors have concluded that it has much better prognosis compared to OSCC [7, 24]. The existing treatment guidelines are primarily derived from the management of OSCCs.

It is noteworthy that there is a risk of local recurrence in VC even after clear resection margins. The risk is also significant for in situ carcinomas. This can be partly explained by the fact that in situ carcinomas involve a wider mucosal area than in depth. The pre-operative biopsy may not always be representative of the actual disease and high degree of suspicion is necessary to properly diagnose and differentiate these lesions from squamous cell carcinoma. There have been reports where the initial biopsy was suggestive of VC, but the post-operative histopathology turned out to be squamous cell carcinoma [7]. The pattern of invasion and the risk of progression to squamous cell carcinoma dictate an aggressive approach with complete surgical excision. The conversion of VC to OSCC is a largely under-reported domain. The current study found three such patients whose histopathology changed from pre-operative status of VC to OSCC post-operatively. Franklyn et al. reported one local failure with 24 months median follow up in thirty patients, although four of these patients were reported to have hybrid oral VC [7]. Our study had two patients with local recurrence. One of the patients had a focus of moderately differentiated OSCC, and the other had the symptoms of an exophytic lesion for 24 months before presenting for surgery, which possibly resulted in field cancerization and higher risk of local recurrence compared to others who presented in shorter intervals.

Various authors have tried to assess the outcomes in oral verrucous carcinomas. The major studies assessing the role of surgery and radiotherapy in the management of oral VCs are retrospective analysis and the details including the patient profiles, treatment and outcomes are outlined in Table 3. The crude local control rate in the post hoc analysis of verrucous carcinoma alone, that is 96%, corroborate with that reported in the literature. Franklyn et al. have reported similar crude failure rate of 4.6% [7]. As has been reported by several other studies, buccal mucosa VC generally carries a better prognosis as compared to sites like lower alveolus and tongue because of higher risk of local and regional recurrence [20].

Table 3.

Major studies addressing oral verrucous carcinomas (studies reporting more than 20 patients have been considered)

Study	No. of pts	Site	Treatment	Median FU	Outcomes
Jyothirmayi et al. [23]	42	Buccal mucosa—50.9% Tongue—20.8%	Radical radiotherapy	56 months	5-Year DFS 66%, 5 year actuarial survival—86%
Vidyasagar et al. [21]	107	Buccal mucosa—65.4%	Radiotherapy	NA	5-Year overall survival—49%
Walvekar et al. [5]	101	Buccal mucosa—61.6% Alveolus + GBS–27.7%	Surgery alone	4.6 years	5-Year DFS—77.6%, Crude local control rate—68%
Rekha et al. [3]	133	Buccal mucosa—52.6% Alveolus—19.5%	Surgery ± radiotherapy	NA	Crude recurrence rate—2.25%
Franklyn et al. [7]	30	Buccal mucosa—56% Tongue–20%	Surgery ± radiotherapy	24 months	Crude local failure rate 4.6%

NA not available, GBS gingivobuccal sulcus, DFS disease free survival

Strengths of the study include the fact that, verrucous carcinomas being an uncommon tumor, with no definite guidelines for treatment, institutional retrospective reports need to be published to accumulate data on behavior and outcomes of oral VCs. Limited follow up duration and retrospective nature of the study are the limitations of this report.

Conclusion

Management of VC continues to be ambiguous because of its indolent behavior. Notwithstanding the potential for slow growth and lack of metastasis, aggressive treatment is needed. Proper coordination between onco-pathologist, surgeon and radiation oncologist is needed to achieve the best results. Any change in histopathology post-operatively to squamous cell carcinoma is a poor prognostic sign and appropriate adjuvant treatment modality need to be considered in such a scenario.

Funding

No Funding was received for the present work.

Compliance with Ethical Standards

Conflict of interest

All authors declare that they have no conflict of interest.

Ethical Approval

Since this is a retrospective study, ethical approval has been waived off by the institutional ethical committee.

Informed Consent

Since this is a retrospective study, an informed consent specifically for this study was not taken individually from the patients. However, all patients included in this study had given consent for the treatment administered in this study (as part of the routine institutional policy).