Management of potentially malignant disorders: evidence and critique

Authors


Giovanni Lodi, Unità di Medicina e Patologia Orale, Dipartimento di Medicina Chirurgia e Odontoiatria, Università degli Studi di Milano, via Beldiletto 1 / 3, Milano 20142, Italia. Tel: +39 02 50319021, Fax: +39 02 50319041, E-mail: giovanni.lodi@unimi.it

Abstract

At a workshop coordinated by the WHO Collaborating Centre of Oral Cancer and Precancer in the United Kingdom, issues related to management of patients affected by oral leukoplakia were discussed by an expert group. The consensus views of the working group are presented here. Although removal of a lesion still seems to be the predominant method of treatment by the majority of relevant health care professionals, no randomized controlled trials have been undertaken to test the hypothesis that excision either by scalpel or laser greatly influences the potential for later malignant transformation within the oral mucosa of an affected individual. Results of observational studies indicate that, although surgery may have a beneficial effect, this is not likely to reduce the risk of later recurrence nor malignant transformation at the same or another site. Medical measures that lessen the size, extent or histopathological features of dysplasia within leukoplakia likewise presently do not seem to be of particular promise, as relapse or later malignant transformation can occur, and there is a risk of adverse effects, particularly with systemic agents (which themselves may be contra-indicated in some individuals). While the risk of malignant transformation, and the development of further potentially malignant disease may theoretically be reduced by cessation of risk activities, such as tobacco usage and alcohol consumption, there remain no good studies that demonstrate that such measures significantly reduce such events.

Introduction

The terminology related to potentially malignant disorders of the oral mucosa (PMD) were discussed by an expert group at a workshop coordinated by the WHO Collaborating Centre for Oral Cancer and Precancer in the United Kingdom (1). The consensus views of the working group are presented in a series of papers. In this report, we review the literature on the reported evidence on the management of oral leukoplakia, the most common PMD encountered in clinical practice.

To critically assess the available literature on management of leukoplakia, or even better, of a patient affected by leukoplakia, it is of primary importance to establish the aims of such management. As leukoplakia is essentially an asymptomatic lesion, outcomes must be considered in terms of recurrence of disease or transformation to oral squamous cell carcinoma (OSCC) at a similar or distant site. Unfortunately, a variable proportion of leukoplakia undergoes malignant transformation (2). A recent study based upon epidemiological data of European patients, concluded that the upper limit of the annual transformation rate of oral leukoplakia is unlikely to exceed 1% (3).

On the basis of such data, it should be deduced that the first aim in the management of leukoplakia is to avoid malignant transformation and, as a consequence, it might be appropriate to consider outcomes of therapy of leukoplakia in terms of later expected OSCC incidence. Elimination or reduction of the size of the lesion (extent) cannot be considered an acceptable indicator of treatment efficacy as recurrence is common event (perhaps as high as 30% in the same or another oral mucosal site) and it cannot be excluded that recurring lesions represent a high risk group. In addition, histopathological change (i.e. lessened or resolution of dysplastic features) cannot be considered a robust and useful outcome. The predictive value, utility, and weaknesses of dysplasia scoring systems are discussed in a separate report in this series (4). Even less useful are the numerous biomarkers proposed as indicator of success in leukoplakia treatment. In fact, although some of them seem to have a good predictive value for OSCC onset (5), none of them has been so far validated as surrogate endpoint for prevention of malignant transformation.

Surgical treatment

Although surgery is the first choice in the management of oral leukoplakia by most relevant specialists (6, 7), the hypothesis that removing potentially malignant oral lesions by different surgical techniques (scalpel, laser, and cryosurgery) can prevent the onset of oral cancer, remains unproved. To date there is only one reported randomized controlled trial (RCT) evaluating surgical treatment of leukoplakia (8). This study did not provide particularly helpful information on the effectiveness of surgical excision of leukoplakia in preventing OSCC as while comprising two active arms (CO2 and Er:YAG laser), no control arm (no treatment) was included and the duration of follow-up was probably too short (24–96 weeks) to assess the incidence of later OSCC. Aside from this single RCT, all the other data come from follow-up studies, mostly retrospective (i.e. based on clinical files revision), describing rates of malignant transformation in patients who underwent surgical (e.g. scalpel excision) or laser treatment of oral leukoplakia (9–22) (Table 1). Results from such studies are hardly comparable because of differences in diagnostic and inclusion criteria, follow-up time intervals, patient characteristics, and surgical techniques employed; and perhaps as a consequence have highly variable results and sometimes conflicting conclusions (18, 23). It must be added that the interpretation of data is often made even harder by the quality of the report. The retrospective design of most of the surgical interventional studies creates bias. For example, the selection of patients may have been determined by the site, size, nature or histopathology of the lesion, as well as the health and wishes of the patients. In one of the few studies that compared the incidence of cancer in a group undergoing surgical excision with different techniques and a group untreated, the authors concluded that, ‘there was no obvious difference in the malignant transformation rate between patients who received any surgical treatments (5.5%, 5/91) and those who did not (7.8%, 4/51)’ (18). It is noteworthy that similar conclusions were reached about 40 years ago by Einhorn and Wersall who reported that, ‘There is no evidence that the incidence of oral carcinoma can be diminished by surgical removal of the leukoplakia’ (24). The same authors added, ‘this does not mean that (surgical removal) should be abandoned, mainly for histologic diagnosis.’ Again these researchers were far-sighted. In fact an important issue in discussing the role of surgical removal in the management of leukoplakia is the potential relevance of excisional biopsy as a diagnostic tool. Indeed some authors have reported high frequencies (>10%) of OSCC in the specimens of lesions that were excised following an incisional biopsy that seemed to suggest that there was no OSCC in the region of interest (21, 25). Although such findings are not consistent with the majority of studies, they do suggest that even if excisional biopsy of leukoplakia it is not effective as an intervention of primary prevention (i.e. to prevent malignant transformation), it may have a role as intervention of secondary prevention (i.e. to detect very early cancer undetected by an incisional biopsy).

Table 1.   Transformation rate in groups of patients with leukoplakia undergoing surgery
TechniqueNo. of patientsFollow-up: mean (range)Recurrences or new lesionsPatients who developed OSCCReference
  1. E, excision; V, vaporization.

  2. aNumber of patients not available.

CO2 laser, V 39 (69 lesions)42 m (2–102)33 / 69 lesions (47.8%) 5 (20%) (9)
CO2 laser, E 54 2 years 2 patients (3.7%) 0(10)
CO2 laser, E 28 5.4 years (3–9.5) 5 patients (17.8%) 3 (10.7%)(11)
CO2 laser, V 1429 m (12–41) 8 patients (57.1%) 1 (7.1%)(12)
CO2 laser, V or E 44 lesionsa10 m (4–24) 6 / 44 lesions (13.6%) 0(13)
Blade 89 patients 6.8 years (1.5–18.6)12/89 (13.5%)11 / 89 (12.35%)(14)
Various 97 lesionsaUnknown but >6 m31 / 97 lesions (32%) 3 (higher than 4%)(15)
Blade 53Unknown but >12 m 9 patients (17.0%) 0(16)
CO2 laser, E103 lesionsa 5.3 years (0.5–12)10 / 103 lesions (10%) 0(17)
Various 91 4 years (0.5–16) 5 (5.5%)(18)
Cryosurgery 60Unknown (2.5–4.5 years)12 patients (20%) 4 (7%)(19)
CO2 and Nd:YAG laser, V or E 5532 m (6–178)21 patients (38.2%) 5 (9%)(20)
CO2 laser, V or E 5518 m (1–44)12 patients (22%) 4 (7%)(21)
CO2 laser, E20052 m (1–219)28 / 282 lesions (10%) 3 (6%)(22)

Among the different surgical techniques proposed for the treatment of leukoplakia, laser surgery has probably received the greatest attention, there being reports of this over the past 30 years1. Unfortunately, as with other surgical techniques, most studies have major methodological flaws and are very low in the hierarchy of evidence (Table 2).

Table 2.   Hierarchy of strength of evidence for treatment
Study design
Systematic reviews of randomized controlled trials
Randomized controlled trial
Non-randomized controlled trial
Observational studies
Case series
Case report

Carbon dioxide, NdYAG, and KTP laser have been employed with various vaporization or excision techniques for the treatment of oral leukoplakia. The main advantages of laser therapy are the potential haemostatic effects and the potential for limited tissue contraction and scarring post-therapy, both of which may permit the treatment of lesions of large dimensions. Other favorable features of laser therapy may include reduced post-operative pain, swelling, and infection. Nevertheless, laser surgery is not the panacea of leukoplakia therapy as treatment must always be preceded by histopathological confirmation of the nature of the lesion, wounds may take longer to re-epithelialize, small granulomas can complicate healing (26) and histopathological confirmation of the nature of the excised lesion is not possible if ablation techniques have been employed. A study comparing different laser techniques, CO2 laser, NdYAG laser, and KTP, demonstrated differences in recurrence rates (34.2%, 28.9%, and 17.0%, respectively); however, only six patients were included in the KTP group whereas the CO2 and NdYAG groups comprised 38 and 36 patients, respectively.(15).

Cryosurgery does not seem to be of particular benefit, recurrence rates of 20–71.4% (15, 19) being reported, along with malignant transformation rates of 7–25% (18, 19). As mentioned previously, the only RCT on surgical treatment of leukoplakia is a two-arm study comparing CO2 and NdYAG lasers. Unfortunately, the small number of patients (10 subjects) and the short follow-up (range 24–96 weeks), do not allow any sound conclusions although the authors stated that, ‘both treatment approaches seem to have limitations to achieve predictable eradication of oral leukoplakia’.

Medical treatment

In the update of a Cochrane review (27), nine RCTs testing medical therapy for management of leukoplakia were found following an extensive literature search. The chemopreventive agents employed included local and systemic vitamin A and retinoids, (28–32), systemic beta carotene (31), lycopene (a carotenoid) (33), ketorolac (as mouthwash) (34), local bleomycin (35), and a mixture of tea used both topically and systemically (36).

Only two studies reported useful data on malignant transformation (31, 35) and unfortunately none of the three treatments tested (topical bleomycin, systemic vitamin A, and systemic beta carotene) were of benefit when compared with placebo (Fig. 1).

Figure 1.

 Medical treatment for the prevention of leukoplakia transformation [from Lodi et al. (27)].

Data on complete resolution of the oral lesions were available from all the nine studies included in the review. Two studies showed a small but significant benefit for the systemic treatment with beta carotene (31) or lycopene (33) when compared with the controls. Vitamin A or retinoids (29–32) were also of some benefit. Unfortunately, the recurrence rates among those who responded to treatment were high (20–64%) when reported (20–64%), as well as adverse effects (up to 100%) (27). The current conclusion of the systematic review is that none of the treatments investigated are effective in preventing malignant transformation of oral leukoplakia.

Cessation of risk activities

Although tobacco is often indicated as the principal risk factor for leukoplakia (37), no studies on groups of smokers are available to assess the effect of cessation of such habits upon the chance of malignant transformation. However, educational programs encouraging smoke cessation can lead to a decrease in the incidence of leukoplakia (38) and smoking cessation may cause resolution of a fair number of leukoplakias (39).

A single study on effect of cessation of smokeless tobacco was found. In that study, a group of military trainees stopped chewing tobacco. After 6 weeks, most leukoplakias (97.5%) had resolved (40). As it is likely that smokeless tobacco leukoplakia is a lesion with specific characteristics, such results cannot be extended to lesions associated with smoked tobacco.

Wait and see

A possible management approach might be to simply keep a leukoplakia under strict clinical and histological surveillance, with frequent clinic visits and biopsies, but without active intervention. In this case, the clinical aim is to make a diagnosis of malignant transformation as early as possible, to treat the oral cancer at a very initial stage, thus possibly providing the best possible prognosis. An observational retrospective study by Schepman et al. compared the incidence of OSCC in two groups of subjects with leukoplakia, one comprising patients who underwent any active treatment (medical and/or surgical) and the other patients kept under regular clinical follow-up (37). As shown in Fig. 2, there was no significant difference in the risk of malignant transformation between the two groups thus perhaps suggesting that the natural history of leukoplakia might be independent from the treatment, and that there is a subgroup of lesions destined to malignant transformation regardless of the therapy adopted. However, as previously mentioned, observational studies may have serious selection bias. In this case, for example, it is possible that lesions considered at higher risk had a greater likelihood of being actively treated, whereas lesions with a less troubling clinical appearance may have been more likely to be selected for higher surveillance only, thus overestimating the transformation rate in treated lesions and underestimating the same variable in non-treated lesions.

Figure 2.

 Follow-up of 166 patients with oral leukoplakia, of whom 87 had active treatment (intervention) and 79 had not (surveillance) (= 0.18), an ‘event’ is defined as a malignant transformation [from Schepman et al. (23)].

Discussion and critique

Leukoplakia is considered to be the most common potentially malignant disorder of the oral mucosa. Nevertheless, there remains little evidence that there is a reliable and safe method of stopping the recurrence of leukoplakias, and the potential for OSCC development.

As highlighted earlier, there remain considerable flaws in many of the studies of determining the most effective means of treating potentially malignant disease of the oral mucosa. Previous studies did not always detail histopathological responses to therapy, definitions of response to therapy were not always uniform, long-term follow-up was not detailed in any study and there was a frequent risk of bias. Additionally, it is not always possible to determine the exact disease that was being investigated, as leukoplakia is a clinical manifestation of oral epithelial hyperplasia, dysplasia and malignancy, and there is a risk that pre-treatment histopathological assessment may not be representative of the most clinically significant disease (i.e. the most worrisome degrees of dysplasia being potentially underrepresented).

Surgical excision of oral leukoplakia-like lesions does not reduce the risk of subsequent disease (23, 41), indeed the risk of subsequent malignant transformation may not be lowered by surgical removal of a leukoplakia (23, 24). Nevertheless, surgical excision does allow the opportunity for examination to ensure all areas of dysplasia have been identified (usually by histopathological examination) and excised. The value of histopathological examination of frozen sections to ensure adequate clearance of disease at surgical margins [as with OSCC (42)] may not be helpful as it will not detect areas of ploidy (if relevant) – and surgeons may have to be more aggressive in the excision of disease than they usually practice (43). Recurrence of leukoplakia following laser excision is likewise high (7.7–38.1%), and later malignant transformation may be as high as 6% (15, 22), but perhaps may be reduced by use of a micromanipulator and operative microscope as opposed to a conventional handpiece. Post-operative re-epithelialization can be slow following laser excision, there is the need for appropriate health and safety safeguards and the requirement for conventional surgical excision of areas of histopathological interests (26). Cryotherapy is not considered to be a first line therapy of oral leukoplakia and related disorders, by virtue of its lack of widespread clinical availability and risk of post-operative scarring, tissue contraction and importantly the resultant inability to observe signs of clinical recurrence (19, 44, 45). Hence, at the present time, no physical method of local excision of oral leukoplakia guarantees long-term resolution of relevant oral epithelial disease, or the possibility of the development of OSCC.

Photodynamic therapy (PDT) is considered by some specialists to be of future application for the management of oral leukoplakia. In principal, PDT would seem to offer the potential for localized and effective eradication of areas of dysplasia by virtue of the differential uptake of a photosensitizer, but to date there have been few reports of the efficacy of PDT for the treatment of oral epithelial dysplasia (OED) and leukoplakia (46–49). Topical photosensitizers such as 5-ALA lessen the risk of cutaneous photosensitization, but may not cause resolution of the leukoplakia (49). To date, while PDT seems attractive, and may have application for the palliative care of advanced head and neck malignancy (48, 50) and the management of early OSCC (48), there is limited data to suggest that PDT will resolve existing leukoplakias or prevent future similar or neoplastic disease.

As noted earlier, a spectrum of different systemic and topical retinoids, and other agents, have been investigated in RCTs for efficacy in the treatment of potentially malignant disease, but there remain limited data as to their exact efficiency, and their long-term local and systemic consequences are largely unknown.

Blocking Epidermal Growth Factor Receptor (EGFR) related pathways via inhibition of cyclo-oxygenase (COX) 2 and EGFR tyrosine kinases is attractive, particularly in the light of in vitro and animal studies of OSCC (51), however, the application of COX-2 inhibitors for the prevention and treatment of leukoplakia, particularly in persons with tobacco-related risks of arteriosclerosis, would seem to be very limited. Similarly, the use of topical non-specific COX inhibitors (e.g. ketorolac) may be limited by poor penetration of the agent into areas of hyperkeratosis (34).

A viral basis for some leukoplakia-type lesions has been suggested, as not all affected persons have the commonly identifiable risk factors of tobacco usage and/or alcohol consumption. Human papillomaviruses (HPVs), in particular high risk types 16 and 18, have been detected in up to 85% of examined leukoplakia-type lesions (52), although the mean prevalence of HPV carriage in such disease is 30%. Similarly, HPV has been variably detected in other potentially malignant oral mucosal disease. However, the exact association of HPV with OED still remains unclear, although perhaps as with head and neck SCC (53, 54), a subset of patients may have virally associated disease amenable to specific therapies. At present, agents such as imiquinod are available for the treatment of cutaneous HPV infection (55), but are not licensed for oral topical application, hence it is not known if such an approach would be of therapeutic benefit for OED (56). Preventative strategies such as vaccination may be applicable, indeed when such vaccines do become widely available, it will be possible to assess the real impact of HPV upon the development of OED, and whether oral leukoplakia is prevented (57, 58).

Antifungal strategies may seem attractive for the treatment of leukoplakia as correlations between the intra-lesional presence of candida and degree of OED (59, 60) and between the frequency of oral yeast and likelihood of OED or SCC (61) have also been reported. Likewise, it is suggested that chronic hyperplastic candidosis (CHC) may have some malignant potential (62, 63). Nevertheless, although there have been suggestions that both topical and systemic therapies may lessen clinical signs of CHC (64, 65), there is no good data indicating that topical or systemic antifungals reliably resolve leukoplakias nor lessen the risk of malignancy in the mouths of affected individuals. Similarly, although links between T. pallidum infection and OSCC risk have been postulated, antibacterial strategies would presently seem to have no logic for the management of oral leukoplakia.

As with other disorders because of alcohol, and particularly tobacco, it may be possible to lessen the risk of leukoplakia (and OSCC) by cessation of tobacco usage and/or alcohol consumption. However, it is estimated that it may be 10–15 years before the risk of OED lessens significantly (66). A major hindrance to even exploring the potential benefits of cessation programs remains the poor uptake and long-term compliance of patients. Similarly, while dietary supplementation (i.e. enhanced intake of fresh fruit and vegetables) may theoretically lessen the malignant potential of the oral epithelium (67), it may be difficult for patients to comply with such change, particularly if economically deprived.

Conclusion

Until appropriate long-term studies of effective therapy of well-defined disease are undertaken, it must be assumed that at the very least leukoplakias should be removed in their entirety, patients regularly monitored for further relevant mucosal change, and directed to avoid the major risk factors of oral epithelial dysplasia, in particular tobacco usage and alcohol consumption. Until the question of how best to identify lesions with OED is answered (as discussed in previous accompanying papers in this series) it would seem best to presume that all isolated white lesions are potentially malignant and thus treat them all similarly. The long-term specialist monitoring of patients with previous leukoplakia (like that of oral lichen planus) is unclear, however to ensure effective utilization of resources such monitoring should be shared by both primary and secondary health care providers, this being optimized by appropriate education of the former by the latter.

Footnotes

  • 1

    Searching Pubmed using the two terms “leukoplakia” and “laser” 115 records were obtained on October 2006

Ancillary