PVL

Proliferative Verrucous Leukoplakia: Report of Two Cases and a Discussion of Clinicopathology

Proliferative verrucous leukoplakia (PVL) is a recently delineated but poorly recognized form of multifocal leukoplakia that is premalignant and of unproven origin. PVL generally presents as a simple benign form of hyperkeratosis that tends to spread and become diffuse. Although slow-growing, the disease is persistent and irreversible. Clinically, PVL often presents as an exophytic wart-like form of leukoplakia that appears to be resistant to nearly all forms of therapy. PVL of the oral cavity is best-defined as a continuum of oral epithelial disease with hyperkeratosis at one end of a clinical and microscopic spectrum and verrucous carcinoma or squamous cell carcinoma at the other. The microscopic findings associated with PVL are dependent on the stage of the disease and the adequacy of the biopsy. Microscopic findings can be markedly variable. PVL is a clinicopathologic disorder that includes the microscopic entity known as verrucous hyperplasia as a component of its histopathologic progression. This article reports on two cases of PVL, describes the clinicopathology of the disease process, and presents therapeutic and etiologic considerations.

First described in 1985 by Hansen and colleagues,¹ proliferative verrucous leukoplakia (PVL) is recognized as a precancerous form of multifocal progressive leukoplakia with a strong potential for malignant transformation. The World Health Organization Collaborating Center for Oral Precancerous Lesions maintained at that time that leukoplakia should continue to be defined as a white patch of the oral mucosa that cannot be characterized clinically or pathologically as any other disease entity.² Hansen and colleagues, however, were able to demonstrate that the disease is indeed an idiopathic form of precancerous leukoplakia that fails to mimic the WHO pattern of disease and that it can present as a solitary homogeneous plaque-like oral lesion, an aggressive form of dysplasia, verrucous carcinoma, or squamous cell carcinoma. The vast majority of the lesions described by Hansen and colleagues were of unknown origin and exhibited a strong tendency to develop areas of carcinoma over time.¹ The disease is slow-growing, persistent, and irreversible; and over time it becomes exophytic, wart-like, and resistant to most forms of clinical therapy. Recurrence is common at all stages.

PVL is most commonly identified in elderly women and seems to occur in a unique group of patients who do not give a history of tobacco use or alcohol abuse. The 30 patients that Hansen and colleagues identified were followed for from one to 20 years. Thirteen died of or with their disease, 14 were alive with PVL at the time of the study, and three were alive without PVL at last follow-up, up to 21 years later.¹

Figure 1. Proliferative verruccous leukoplakia continuum. From Greer RO, Hoernig G, McDowell JD, Proliferative verrucous leukoplakia: a clinical and histopathologic disease. Pathology Case Reviews, 4:3-7, 1999.

Hansen and colleagues¹ documented that 63 percent of their PVL patients harbored Candida albicans in the superficial layers of their biopsies. In studies in the laboratory at the University of Colorado, the authors of this article identified Candida albicans in 18 of 21 patients with PVL followed over a 10-year period. Marx³ has recently reported that he was able to identify Candida albicans in 75 percent of biopsy specimens from patients with PVL in a series at the University of Miami Medical School.

All reports in the literature suggest that PVL is a single disease entity that demonstrates a unique spectrum of clinical and histopathologic expression that may terminate in squamous cell carcinoma (Figure 1).

During the past 10 years, the authors have had the opportunity to evaluate 21 cases of PVL at the University of Colorado School of Dentistry. The purpose of this study is to discuss the nature of two cases that have had benefit of long-term follow-up, discuss the clinicopathologic aspects of the disease, delineate differential diagnostic guidelines, and propose etiologic considerations and management modalities.

Case Reports

Case 1

Figure 2. Case 1 — PVL of the maillary gingiva. Note cobbled character of lesions and their linear distribution.

A 56-year-old white male was seen in the Sands House Oral Cancer Clinic at the University of Colorado School of Dentistry for evaluation of multifocal white lesions affecting the maxillary gingival mucosa. The lesions were plaque-like, cobbled, and corrugated (Figure 2). The lesions were not painful, and the patient gave no history of ulceration, hemorrhage, or numbness. The patient was not a tobacco user and admitted to using alcohol only rarely. He was on no medications and indicated that until his dentist brought the lesions to his attention, he was unaware of their presence.

The patient's referring dentist reported that the lesions had been present for less than a year. The lesions were biopsied and described microscopically as verrucous hyperplasia with focal areas of moderate dysplasia. Tissue sections were stained with periodic acid Schiff (PAS) stains and shown to harbor Candida albicans. All lesions were excised completely, and the patient was lost to follow-up.

Figure 3. Case 1 — PVL transition to verrucous carcinoma of the vental tongue. The lesion is exophytic, nodular and ulcerated.

Two years later, the patient returned with a new exophytic lesion affecting the ventral tongue (Figure 3). The new lesion was approximately 2 cm in diameter, nodular, and focally hemorrhagic. The lesion was biopsied, diagnosed as verrucous carcinoma, and treated by a wide surgical excision. Once again, PAS stains were positive for Candida albicans. The patient has now been followed an additional eight years. Since that time, he has presented with an additional white reticular lesion affecting the right buccal mucosa. That lesion was biopsied six months prior to this report and found to demonstrate mild dysplasia. That lesion was totally excised. PAS stains for Candida albicans were negative. The patient continues to be monitored closely and has not developed any additional lesions.

Case 2

A 39-year-old white female who gave a history of never using tobacco products or alcohol presented to the Sands House Oral Cancer Clinic at the University of Colorado School of Dentistry as a referral from a southern California dentist who requested that she be evaluated for an oral lesion by the School of Dentistry's Oral and Maxillofacial Diagnostic Clinic upon moving to Colorado.

Figure 4. Case 2 — PVL, ventral surface of tongue. Note the white streaklike, somewhat lichenoid character of the lesion.

The patient indicated that a white lesion affecting the buccal mucosa had been excised approximately four years earlier and diagnosed as benign hyperkeratosis. At the time of her initial presentation to the clinic, she had a white plaque-like lesion affecting the left ventral surface of the tongue (Figure 4). The lesion was not painful, and the patient gave no history of trauma or injury to the area. The lesion was incisionally biopsied, and a diagnosis of severe epithelial dysplasia was rendered. The patient was rescheduled for surgery to have the tongue lesion removed.

Figure 5. Focus of squamous cell carcinoma arising in the earlier PVL seen in Figure 4. This well-differentiated cancer demonstrates muscular invasion and significant keratin pearl formation.

At the time of definitive surgery, a diagnosis of squamous cell carcinoma was rendered (Figure 5). The first biopsy was retrospectively subjected to PAS staining for Candida albicans. The stains were negative; however, the definitive surgical specimen did show evidence of Candida albicans with PAS staining.

The patient was placed on long-term follow-up and followed for four years and six months with no evidence of residual disease. Four years and nine months into the follow-up, the patient presented with new areas of papillary hyperkeratosis affecting the mandibular gingiva (Figure 6).

Figure 6. Case 2 — PVL, gingiva. Note the papillary character of the lesion.

On excisional biopsy, the lesions demonstrated mild epithelial dysplasia microscopically. PAS stains were negative for Candida albicans. Ten years after the initial biopsy, the patient continues to be monitored closely at three-month intervals, without evidence of new disease.

Discussion

PVL is considered by most authorities to be a persistent progressive preneoplastic process. The disease is thought to have a broad, continuous spectrum of clinical and histologic features with high potential for malignant transformation. The authors believe that the two cases of PVL presented here represent the full spectrum of PVL from benign hyperkeratosis to squamous cell carcinoma.

The most common sites for PVL have been reported to be the buccal mucosa and palate,^1,4 followed by the alveolar mucosa, tongue, floor of the mouth, gingiva, and lip. Hansen1 reports that in early PVL, clinical findings consist of multifocal white plaques of the type presented in the two cases reported here. As PVL progresses, solitary lesions spread and similar lesions appear elsewhere on the oral mucous membrane, often accompanied by diffuse papillary and erythematous changes. The authors' experience suggests that the natural progression of the disease almost always includes some verrucoid or papillary pattern. In late PVL, some areas become clinically indistinguishable from verrucous carcinoma or papillary squamous cell carcinoma.

One consistent feature documented in PVL is that a large percentage of cases demonstrate infection with the fungal organism Candida albicans.^1,3 Some investigators report that this association is the key to neoplastic transformation, suggesting that the endogenous production of nitrosamines by Candida organisms is the neoplastic stimulus in PVL, ³ while other investigators suggest that there is not enough evidence for a causal claim since Candida may be found in biopsies of normal mucosa.

One of the difficulties in diagnosing PVL has been the lack of clinicopathologic continuity associated with its microscopic diagnosis. The lesion can easily be misinterpreted as a form of simple hyperkeratosis microscopically, especially when its multifocal nature is not passed on to the pathologist.

Recently Murrah and Batsakis⁵ have suggested that verrucous hyperplasia, a term first coined by Shear and Pindborg⁶ and a well-recognized histologic precursor of verrucous carcinoma, is indeed no more than a late-stage and microscopically definable component of PVL.

In a review of 68 cases of verrucous hyperplasia, Pindborg and Shear reported that 40 percent of histologically definable verrucous hyperplasia cases underwent malignant transformation to either verrucous carcinoma or squamous cell carcinoma.⁶ Murrah and Batsakis⁵ champion the fact that verrucous hyperplasia as a unique histologic component of PVL be restricted to lesions of the oral mucosa, sinonasal cavity, and laryngeal mucosa that meet the criteria of an atypical verrucoid growth with keratotic hyperplasia, mucosal clefting, keratotic clefts, parakeratin plugging, and irreversibility. These authors further suggest that verrucous hyperplasia can sometimes be indistinguishable from verrucous carcinoma and note that both disorders require initial aggressive surgical management.

The diagnosis of PVL remains, in large measure, a clinicopathologic one. However, there are unique and reproducible epidemiologic features that distinguish this unusual disorder. In contrast to many forms of oral epithelial dysplasia and leukoplakia, PVL tends to occur more frequently in women than men; and PVL has an affinity for the mucosa of patients with no history of tobacco use.^1,4,7

Silverman and colleagues profiled patients such as these in an assessment of 257 leukoplakic patients followed for a mean period of 8.4 years. They noted that isolated cases initially thought to represent nothing more than simple hyperkeratoses tended to become PVL over time.⁸

Etiology

Considering the difficulty involved in establishing an early diagnosis of PVL, researchers have recently undertaken investigations that attempt to identify a molecular basis for the disease. Flow cytometric analysis has been used by Kahn and colleagues⁴ to study four cases of PVL in an attempt to determine if flow cytometry might be useful in early diagnosis.

These investigators performed flow cytometry on 27 formalin-fixed and paraffin-embedded tissue samples and found DNA aneuploid cell lines in each of the PVL cases studied. A DNA index range of 1.1 to 2.6 was reported. In all four cases studied, the abnormal cell line DNA index appeared to be maintained throughout the sampling period. These results suggest that flow cytometric analysis may be useful in the early recognition of PVL and support the concept that flow cytometric analysis may serve to support aggressive intervention at an early stage in the disease. This is an important intervention since PVL can recur at a faster rate than non-PVL forms of leukoplakia.

Studies in the authors' laboratory over the past decade suggest that the histologic features seen in PVL -- including its often verrucoid clinical appearance and its uniquely identifiable pathologic stages, including verrucous hyperplasia and verrucous carcinoma -- favor a disease-associated infection by human papillomavirus (HPV).⁹ A host of studies have identified HPV infection of oral lesions of all types including leukoplakias^9-11 over the past 10 years. The most compelling work in this arena related to PVL is perhaps that of Palefsky¹¹ and Shroyer and colleagues^12,13 who have shown an association between HPV infection and PVL using molecular biological techniques. Shroyer and colleagues^12,13 evaluated 17 verrucous carcinoma cases seen at the neoplastic end of the PVL spectrum using biotin-labeled probes. These investigators found HPV 6 and 11 DNA in 41 percent of cases they studied by in situ hybridization. They also studied 22 cases of non-PVL-associated oral lesions, including 12 squamous cell carcinomas, and found HPV 16 in five of 12 cases.

To date, more than 80 distinct types of HPV have been identified. A host of studies involving cervical cancer support the premise that HPV 16 and 18 have an affinity for potentiating malignant disease in that site.^14,15 The characterization of HPV infection in PVL, however, has only recently begun. Studies have shown lesser amounts of HPV in oral squamous cell carcinomas than in cervical cancer. This finding may be related to the fact that oral neoplasms harbor low copy numbers of HPV or that oral tissues subjected to HPV DNA typing may have been inadequately sampled.

Finally, unidentified HPV subtypes unique to the oral cavity may remain undetected. Although HPV 16 and 18 appear to be the most commonly identified forms of HPV found in certain mucous membrane malignancies, HPV 6, 11, 16, and 18 remain the most consistently identified types, in precancerous lesions of the oral cavity, including PVL.^11-13

In fact, Palefsky et al¹¹ report that HPV 16 plays a fundamental role in the pathogenesis of PVL-associated oral epithelial dysplasia and possibly cancer. These findings are all the more compelling when one considers that these investigators found HPV in only a small fraction of the more common non-PVL-associated lesions they also studied.

The mechanism by which HPV infection contributes to the progression of PVL has not been fully appreciated, but the process is likely related to a series of chromosomal mutations. Credence is given to this mutation theory when it is recognized that the E6 and E7 proteins of HPV 16 bind and inactivate two significant cell cycle regulators, the p53 and retinoblastoma proteins.^16-18 HPV E6 protein and p53 may in fact enhance the chromosomal instability that potentiates the PVL cascade.


Differential Diagnostic and Histopathologic Considerations

Verrucous Hyperplasia

Figure 7. "Sharp variety" of PVL demonstrating narrow keratinized verrucous processes and makred epithelial acanthosis.

Shear and Pindborg6 initially described two classic histologic patterns for a disorder termed verrucous hyperplasia, which is now recognized as a late-stage component of PVL. The first pattern, or "sharp variety," of verrucous hyperplasia features long, narrow, and heavily keratinized verrucous processes (Figure 7). These hyperplastic lesions may or may not present as multifocal homogenous areas of leukoplakia when identified clinically.

A second type, or "blunt variety," of verrucous hyperplasia consists of verrucous processes that are broader, flatter, and not as heavily keratinized as in the sharp variant (Figure 8).

Figure 8. "Blunt" variety of PVL demonstrating broad, flat verrucous processes, and minimal keratin plugging between papillary fronds.

Both histologic subtypes of verrucous hyperplasia can progress to verrucous carcinoma over time.

In their study of 68 verrucous hyperplasia cases, Shear and Pindborg⁶ found that 37 cases were of the sharp variety and 24 were of the blunt type. In nearly all cases, a dense inflammatory infiltrate was seen in the supporting collagenous lamina propria. Dysplastic atypia was identified in 66 percent of cases. These authors emphasize that verrucous hyperplasia is best distinguished from verrucous carcinoma in biopsies taken at the margins of lesions. Verrucous hyperplasia demonstrates verrucous processes where the greater part of the hyperplastic epithelium remains superficial to adjacent normal epithelium. Verrucous carcinoma features verrucous processes that are also superficial, but the broad rete processes of that disease extend deeper than the abutting normal epithelium, often enveloping a margin of normal epithelium into the underlying connective tissue.

Non-PVL-Associated Leukoplakias

Leukoplakia is a term that continues to cause considerable consternation in the literature. The term leukoplakia simplex is most often used to define a localized homogenous form of white hyperkeratoses affecting the oral mucous membrane. The lesion is generally dense and can have a corrugated or smooth surface. In those instances when a solitary oral leukoplakic lesion is recognized in a patient, it may be difficult to determine if the lesion is early PVL without a characteristic history of clinical persistence, irreversibility, focal erythematous components, and an exophytic or wart-like pattern. For this reason, biopsy, long-term follow-up, and close scrutiny of such lesions is mandatory for proper patient management.

Nodular Leukoplakia

Nodular leukoplakia is a term that has been variously referred to in the literature as erosive leukoplakia or speckled leukoplakia. It is a form of leukoplakia that characteristically contains zones of redness or erythema. The lesions of nodular leukoplakia are rarely, if ever, long-standing; and they typically are not multifocal. Nodular leukoplakia generally presents histologically as some form of epithelial dysplasia, or maturational atypia, and from 5 percent to 25 percent of nodular leukoplakias are reported to be dysplastic.¹⁹

Viadent Hyperkeratosis

Viadent hyperkeratosis is a form of leukoplakia that develops in the oral cavity in association with the use of an oral antigingivitis rinse that contains an extract from the blood root plant, Sanguinaria canadensis L. This form of leukoplakic hyperkeratosis is typically identified in the buccal vestibule. In the authors' experience, women with Viadent hyperkeratosis far outnumber men. The authors have had the opportunity to evaluate 65 cases of this unusual form of leukoplakia, recording it most often in the maxillary buccal vestibule.

Damm and colleagues have documented their experience with 88 cases and report that 33 Viadent hyperkeratoses demonstrated basal layer hyperplasia and maturational atypia histologically consistent with dysplasia. The authors' experience, however, has been that none of the Viadent hyperkeratoses that they have identified clinically and examined histologically have shown any form of preneoplastic or dysplastic differentiation. Some lesions are reversible following biopsy, while others persist or recur.

Additional Papillary Lesions

Papillary oral lesions such as squamous papilloma, condyloma accuminatum, molluscum contagiosum, and verruca vulgaris may be difficult to distinguish from some of the phases of PVL. Generally, the diagnostic problems associated with such distinctions are related to the papillary quality of the lesions. Diagnostic problems may arise when the papillary lesions are evaluated by pathologists in the face of minimal or inadequate tissue for examination or insufficient clinical information, or when the tissue has been processed or embedded improperly. Thus, it is exceedingly important that the clinician provide the pathologist with as much information about the lesion as possible so that specific papillary lesions can be distinguished from certain clinical phases of PVL.

Management

Because of the neoplastic risk associated with it, any case suspected of being PVL must be extensively sampled during the evaluation. Medina and colleagues²¹ reviewed 104 cases of verrucous carcinoma and found foci of conventional squamous cell carcinoma in 20 patients. These investigators suggest that such diagnostic errors relate to the problem of insufficient tissue sampling. The same can be said for PVL as it relates to clinical sampling. Lesions suspected of representing PVL should be sampled from several areas if they are multifocal, and several biopsies may have to be taken over time to determine if one is dealing with PVL. The pathologist has a responsibility to contact the clinician for clarification when problems related to clinical sampling arise.

In light of the fact that PVL is a multifocal residual and recurrent disease, many investigators have attempted to develop a standard protocol for management of PVL lesions. Marx has suggested³ that all PVL cases be screened for Candida albicans infection since most investigators have indicated that Candida is present in close to 75 percent of cases.

He further suggests that multiple tissue samples be taken to prevent sampling errors, recommends a course of long-term follow-up, and favors total excision of PVL. Following excision, the patient is placed on a regimen of beta carotene, 30 mg, q.i.d.; Griseofulvin Microsize, 250 milligrams, t.i.d.; Nystatin Oral Suspension, 100,000 units per cc in a full teaspoon that is swished and expectorated, t.i.d.; or, Diflucan, 100 milligrams, p.o., q.d. Marx reports that this protocol has proven beneficial to his patients with PVL, but that long-term studies need to be completed to determine the protocol's ultimate success. Radiation therapy is contraindicated in the management of PVL, as is laser ablation therapy according to Marx, since radiation may pose a risk for a more rapid neoplastic transformation of the disease, and laser ablation carries with it the risk of masking clinical signs of the disease process as well as limiting proper pathologic scrutiny of the tissue.

Summary

PVL is a dynamic process representing a linear progression of oral hyperkeratosis from a clinically benign manifestation to, in some cases, invasive squamous cell carcinoma. The end stage of the progression of PVL is often manifested by clinical lesions that are verrucous or papillary.

This verrucoid or papillary character may ultimately be classified as verrucous hyperplasia, verrucous carcinoma, or papillary squamous cell carcinoma; but, regardless of terminology, papillary lesions tend to favor the neoplastic end of the PVL spectrum. Clinicians and pathologists must remain vigilant in their examination of the multiple molecular pathways by which this disease can develop, and pathologists must recognize its dynamic histopathology. Management of PVL is often frustrating for clinician, patient, and pathologist; and only as all three parties become more cognizant of the disorder will there be a fuller understanding of the disease.

Authors

Robert O. Greer, DDS, ScD, is a professor of pathology in the University of Colorado School of Medicine and chair of the Division of Oral and Maxillofacial Pathology at the University of Colorado School of Dentistry.

John D. McDowell, DDS, MS, is an assistant professor of oral medicine, diagnosis and radiology at the University of Colorado School of Dentistry.

George Hoernig, HT, is a professional research assistant, Division of Oral and Maxillofacial Pathology, University of Colorado School of Dentistry.

References
1. Hansen LS, Olson JA, Silverman S Jr, Proliferative verrucous leukoplakia. A long-term study of 30 patients. Oral Surg Oral Med Oral Pathol 60:285-98, 1985.

2. Kramer IRH, Lucas RB, et al, Definition of leukoplakia and related lesions, an aid to studies on oral precancer. Oral Surg Oral Med Oral Pathol 46:518-39, 1978.

3. Marx R, A discussion of proliferative verrucous leukoplakia: Thirty-first annual refresher course in oral and maxillofacial surgery. University of Colorado, Nov 19, 1998.

4. Kahn MA, Dockter ME, Hermann-Petrin JM, Proliferative verrucous leukoplakia. Four cases with flow cytometric analysis. Oral Surg Oral Med Oral Pathol 78:469-75, 1994.

5. Murrah V, Batsakis JG, Proliferative leukoplakia and verrucous hyperplasia. J Otol Rhinol Laryngol 103:660-3, 1994.

6. Shear M, Pindborg JJ, Verrucous hyperplasia of the oral mucosa. Cancer 46:1855-62, 1980.

7. Pogel MA, Sublingual keratosis and malignant transformation. J Oral Pathol Med 8:176-8, 1979.

8. Silverman S Jr, Gorsky M, Lozada F, Oral leukoplakia and malignant transformation: A follow-up study of 257 patients. Cancer 53:563-8, 1984.

9. Greer RO, Eversole LR, Crosby LK, Detection of human papilloma virus-genomic DNA in oral epithelial dysplasias, oral smokeless tobacco-associated leukoplakias and epithelial malignancies. J Oral Maxillofac Surg 48:1201-5, 1990.

10. deVillers EM, Weidauer H, Oho H, Papillomavirus DNA in human tongue carcinomas. Int J Cancer 36:575-82, 1985.

11. Palefsky JM, Silverman S Jr, et al, Association between proliferative verrucous leukoplakia and infection with human papillomavirus type 16. J Oral Pathol Med 24:193-7, 1995.

12. Shroyer KR, Greer RO, Detection of human papillomavirus DNA by in situ DNA hybridization and polymerase chain reaction in premalignant and malignant oral lesions. Oral Surg Oral Med Oral Pathol 71:708-13, 1991.

13. Shroyer KR, Greer RO, et al, Detection of human papillomavirus DNA in oral verrucous carcinoma by polymerase chain reaction. Modern Pathol 6:669-72, 1993.

14. Crum C, Miato M, et al, Cervical papillomaviruses segregate within morphologically distinct precancerous lesions. J Virol 54:675-81, 1985.

15. Lorincz AT, Reid R, et al, Human papillomavirus infection of the cervix: Relative risk associations of 15 common anogenital types. Obstet Gynecol 79:328-37, 1992.

16. Scheffner M, Werness BA, et al, The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 63:1129-36, 1990.

17. Werness BA, Levine AJ, Howley PM, Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science 248:76-9, 1990.

18. Munger K, Werness BA, et al, Complex formation of human papillomavirus proteins with retinoblastoma tumor suppressor gene product. Embo J 8:4099-105, 1989.

19. Neville BW, Damm DD, et al, Oral and Maxillofacial Pathology. WB Saunders Co, Philadelphia, 1995, p 285.

20. Damm DD, Curran A, et al, Leukoplakia of the maxillary vestibule -- an association with Viadent? Oral Surg Oral Med Oral Path, 87:61-6, 1999.

21. Medina JE, Dichel W, Luna MA, Verrucous-squamous carcinomas of the oral cavity. Arch Otolaryngol 110:437-40, 1984.

To request a printed copy of this article, please contact: Robert O. Greer, DDS, ScD, Division of Oral and Maxillofacial Pathology, University of Colorado Health Sciences Center, 4200 E. Ninth Ave., Campus Box C285, Denver, CO 80262.