The visual identification of precancerous and early-stage cancerous oral lesions is hindered significantly by the difficulty of clinically differentiating these harmful lesions from similar-looking benign lesions.⁶ Precancerous and early-stage cancerous oral lesions vary greatly in their clinical appearance and often lack classic clinical characteristics associated with advanced oral cancers such as ulceration, induration, nodularity, bleeding and cervical adenopathy.⁷ Furthermore, precancerous and early-stage cancerous oral lesions often are asymptomatic—a fact that more than 25 percent of surveyed dentists have failed to recognize.⁸ Thus, the true nature of innocuous-looking, but potentially harmful, oral lesions detected during an oral cancer examination often is not properly ascertained, and patients with such lesions unfortunately are granted a false sense of security.⁹

OralCDx (OralScan Laboratories Inc., Suffern, N.Y.), a computer-assisted method of analysis of an oral brush biopsy, has proven to be a useful adjunct to the oral examination in identifying precancerous and cancerous oral lesions, especially those that are not suspected of being harmful.¹⁰ The primary purpose of this study was to determine if an oral cancer screening program was augmented by testing oral epithelial lesions with the oral brush biopsy system. A secondary purpose was to further ascertain the usefulness of this tool for application in the dental office.

	METHODS AND MATERIALS

TOP ABSTRACT METHODS AND MATERIALS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
All dentists and dental hygienists attending the ADA Health Foundation Health Screening Program at the ADA’s 1999 and 2000 annual sessions were eligible to enroll in the study. The screening program was conducted over a four-day period at each annual session. Thirty-one volunteer licensed dentists performed all of the oral examinations and staffed the head and neck screening station. The majority of dentists were staff members at a Department of Veterans Affairs hospital, while U.S. Army dentists, private practitioners and dental school faculty, including specialists in oral medicine, oral max-illofacial surgery and periodontics, also participated. When indicated, the dentist performed oral brush biopsies, and dental hygienists assisted in the examination and preparation of brush biopsy specimens. A brief instructional demonstration of the oral brush biopsy technique was provided to the dentists who performed the examinations.

Dentists collected data on each participant’s age, sex, and tobacco and alcohol use histories. They recorded all oral lesions that had an epithelial abnormality of undetermined etiology; they excluded oral lesions with obvious etiologies such as an aphthous ulceration or frictional keratosis on the buccal mucosa from cheek biting. They noted the clinical characteristics of each lesion, including the predominant color, morphology, signs and symptoms, and location.

Participants who had oral epithelial abnormalities were given the opportunity to have their oral lesions tested using the OralCDx system. All oral brush biopsies were performed with a sterile instrument specially designed to obtain a complete transepithelial specimen. Neither topical nor local anesthetic was used.

All brush biopsy specimens were analyzed at OralScan Laboratories Inc. in Suffern, N.Y. The brush biopsy results were classified into one of four categories: "negative," no epithelial abnormality; "atypical," abnormal epithelial changes of uncertain diagnostic significance; "positive," definitive cellular evidence of epithelial dysplasia or carcinoma; and "inadequate," incomplete transepithelial biopsy specimens (did not contain an adequate representation of cells from all three epithelial layers of the oral mucosa). Participants who had "atypical" or "positive" brush biopsy results were advised to have their lesions evaluated and to have an incisional biopsy performed, so the lesions could be completely characterized histologically.

	RESULTS

TOP ABSTRACT METHODS AND MATERIALS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
A total of 2,053 people attended the ADA Health Foundation Health Screening Program at the ADA’s 1999 and 2000 annual sessions. A total of 930 dentists and dental hygienists elected to be screened for oral cancer at the head and neck station. Of this group, 89 participants (9.7 percent) who had a total of 93 oral lesions that displayed an epithelial abnormality were identified. All of these participants agreed to undergo an oral brush biopsy with computer-assisted analysis to determine the significance of their oral lesions. Of the 93 oral lesions present in these participants, 62 (66 percent) developed in men and 31 (34 percent) in women. The participants’ ages ranged from 23 to 69 years (mean 49). Of the 89 participants, 82 did not use tobacco products in any form or had not during the past 10 years; the remaining seven participants smoked less than one pack of cigarettes per day. No participant was considered a heavy user of alcohol (7 ounces of alcohol or more per day). The demographic features of study participants are summarized in Table 1.

The seven brush biopsy specimens tested by OralCDx were classified as abnormal; six were "atypical," and one was "positive" (Table 3, page 360). Of the six "atypical" oral lesions, three underwent scalpel biopsy and histologic examination; two of these proved to be dysplastic, and one was benign. (Figure 1 [page 361] shows cellular abnormalities in one of the "atypical" brush biopsies, and Figure 2 [page 362] demonstrates dysplasia as seen in the subsequent incisional biopsy). Of the remaining three "atypical" oral lesions, two spontaneously resolved, and one was to be watched by the participant. The one "positive" oral lesion was subsequently proven by scalpel biopsy and histologic examination to be dysplastic.

View larger version (69K): [in this window] [in a new window]   Figure 1. An "atypical" OralCDx (OralScan Laboratories Inc., Suffern, N.Y.) result showing epithelial cells with an increase in the nuclear-to-cytoplasmic ratio and mild hyperchromasia.

View larger version (126K): [in this window] [in a new window]   Figure 2. Histologic photomicrograph of the same lesion shown in Figure 1 demonstrating moderate dysplasia—proliferation of atypical basal cells above the parabasal region extending into the middle one-third of the epithelial cross-section.

	DISCUSSION

TOP ABSTRACT METHODS AND MATERIALS RESULTS DISCUSSION CONCLUSIONS REFERENCES

In the oral cancer screenings, if the OralCDx system had not been used, the examining dentists would have had to rely on clinical inspection alone to determine which oral lesions required incisional biopsy. As a result, the three innocuous-appearing precancerous lesions identified by the oral brush biopsy tool in this study would not have been revealed unless incisional biopsies had been performed on all 93 lesions. By using the oral brush biopsy on all 93 lesions, 86 lesions were determined to have no cellular abnormalities, and only seven required incisional biopsies to detect the three precancerous lesions.

All oral lesions with "atypical" and "positive" results should undergo incisional biopsies to completely characterize them. This is indicated in the product directions for OralCDx and also is a requirement of the ADA Seal Program. Not all of these lesions will be verified as precancerous or cancerous.

The probability that an oral lesion with an abnormal oral brush biopsy result will prove to be precancerous or cancerous is known as the positive predictive value. In an OralCDx multicenter study involving nearly 1,000 patients, the positive predictive value of an "atypical" result was approximately 31 percent (approximately one-third of OralCDx "atypical" cases proved to be pre-cancer or cancer by histologic examination, while approximately two-thirds were "negative").¹⁰

In the oral cancer screening study reported in this article, though the sample size of participants who underwent brush biopsies was low and the specificity, sensitivity and the false-negative rate were not determined, two of the six "atypical" results and the one "positive" result proved to be precancerous. This finding is consistent with the data from the multicenter OralCDx study. In contrast, studies of mammography in screened women older than 35 years of age have reported positive predictive values much lower than 10 percent.^11,12 That is, fewer than 10 in 100 women with abnormal mammograms would expect to have their biopsies confirmed positive. Despite this seemingly low positive predictive value, mammography is universally regarded as the standard of care for breast cancer detection.

As demonstrated in previous oral cancer screening studies, at least one in 20 patients and as many as one in six patients are noted to have a questionable oral lesion.^13–16 In this current study, about one in 10 dentists and dental hygienists screened displayed an oral lesion with an epithelial abnormality. Thus, lesions with intact epithelium—such as mucoceles, fibromas and hemangiomas—were excluded. The high incidence of oral lesions observed in this study was surprising, as it could be assumed that risk factors for oral cancer were far fewer in this group of health care professionals than in the general population, and that their overall oral health status would be significantly better. This study confirms the data from other screening studies^13–16 that have demonstrated that oral lesions are not rare but are, in fact, extremely common, even in people who are at low risk. It suggests that general dentists routinely see patients who have oral epithelial abnormalities and, therefore, have the opportunity to definitively identify precancers and early-stage oral cancers.

Many authors have questioned the value of oral screening programs such as those conducted at state fairs, arguing that people at high risk of developing oral cancer are less likely to attend than are health-conscious people.¹⁷ Although this may be true, in the current study, less than 50 percent of all health-conscious dentists and dental hygienists attending a health screening took advantage of the opportunity to be screened for oral cancer. Perhaps this group of professionals believed their risk of developing oral cancer was low and they could forgo an oral health screening.

As it turns out, 25 percent of oral cancers develop in patients who are at low risk and who do not consume tobacco or heavy amounts of alcohol.¹⁸ The lack of coverage in the popular press regarding both the necessity for oral cancer screening and the hazards of oral cancer has been established by Canto and colleagues.¹⁹ There is an urgency to educate the dental profession and to inform the public about the benefits of oral cancer screening.²⁰

The value of oral cancer screening programs also has been questioned because of the poor compliance on the part of people who have had suspicious oral lesions identified and who have been referred for follow-up biopsies. For example, in a study by Frenandez Garrote and colleagues,²¹ only 28 percent of 30,244 patients complied with referrals to oral surgeons to determine the significance of their lesions. Other investigators have reported similarly poor compliance rates.²² If patients are not informed about the potential danger of their oral lesions, their compliance with follow-up scalpel biopsies will be low. This has been well-documented in patients who participate in screening programs, as well as those who visit private dental offices.^23,24

Oral brush biopsy can be an educational and motivational tool for prevention. Just as radiographs depict carious lesions, the computer images of a brush biopsy specimen determined to be precancerous or cancerous display definitive visual evidence of cellular abnormalities. Therefore, patients who view their images displaying abnormal brush biopsy results are, in theory, more likely to be compliant with their dentists’ recommendations for scalpel biopsy and histologic evaluation than are those whose lesions have been identified by clinical examination alone.²⁵ It is relevant to note that high compliance rates for follow-up have been noted for patients with other abnormal cancer detection tests, including mammography and cervical Pap smears.^26,27

	CONCLUSIONS

TOP ABSTRACT METHODS AND MATERIALS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The results of this study suggest that the oral brush biopsy tool might enhance the oral cancer screening process by determining the significance and potentially harmful nature of identified oral lesions. Because the brush biopsy technique is not difficult to learn, as evidenced by the absence of "inadequate" specimens in this study, it can be integrated easily into the routine practice of dentistry. Additionally, the prospect of early oral cancer and precancer detection that results in prolonged survival can offer dentists and patients hope.