The Role of the Brush Biopsy in the Early Detection of Oral Pre-cancers and Cancers

By John A. Svirsky, DDS, MEd, James C. Burns, DDS, MS, PHD, Dennis G. Page, DDS, MS and Louis M. Abbey, DMD, MS

The good news for dentists and their patients is that the poor prognosis from oral cancer is going to improve in the near future. This is because dentists are now able to identify unsuspected early and potentially curable oral cancers and more importantly, their precursor lesion, dysplasia, which when treated, prevents its progression to oral cancer.

Consider the findings from a screening program in which, approximately 1,000 patients, consisting of dentists and dental hygienists, underwent an oral cancer examination.1 Roughly 10 percent of them had a small red or white spot in their mouth, and when these spots were tested with the brush biopsy, three of the dentists proved to have dysplasia. None of these pre-cancers looked suspicious nor was any of the dentists aware of the findings in their own mouth. Fortunately for these dentists, their benign looking spots were tested with the brush biopsy, and their pre-cancers were correctly identified at an early, easily treatable stage.

As research investigators in the multicenter brush biopsy clinical trial and having had the experience of lecturing on the technique to thousands of dentists, we would like to emphasize that dentists and dental hygienists see potentially harmful spots on a daily basis. Unexplained white and red spots, like the ones detected in the study referenced above, may seem innocuous. However, guessing that the spot is unimportant may be dangerously incorrect and result in delayed diagnosis and mutilating surgery.


Until recently, the only method of diagnosing an oral cancer or pre-cancer was to biopsy a suspicious lesion by traditional methods. Although biopsying all suspicious lesions is exactly what we continue to stress to our students and fellow dentists, this advice has its limitations. More than half of all patients with oral cancer have evidence of metastases at time of diagnosis and only a 50 percent 5-year survival rate.2 The reason for this, in part, lies in the fact that oral cancers are readily detected by clinical examination in their late stages but not in their early stages.

The diagnosis of an oral cancer is, accordingly, made only when an oral lesion is suspicious enough that it causes the dentist to refer the lesion for biopsy. For example, clinical features and symptoms of a suspicious lesion include large size, ulceration, and pain, and although these are all indications for immediate biopsy, an oral cancer that exhibits any of these features, all too often, is already considered advanced. Fortunately, dentists encounter these types of suspicious lesions perhaps only a few times per year, at most (Figs. 1 & 2).

By contrast, early oral cancers and pre-cancers often do not look suspicious at all.3 In fact, the opposite is true, and they look identical to the harmless-appearing oral lesions that dentists encounter almost daily (Figs. 3-5). They are often small or even tiny in size. They look innocent, appearing often as a flat, white or red area. They do not cause pain, and in most cases, patients are not even aware of their presence. Consequently, these cancers go undetected until they progress to a stage with suspicious features and serious consequences. Common sense would dictate that every “big cancer” starts as a “small cancer.”

Dentists, often unknowingly, have been faced with the impossible task of biopsying every unexplained tiny harmless-looking oral lesion to find the few pre-cancers and cancers amongst this group. And what has made this diagnostic task impossible is the fact that these types of harmless-looking lesions are so common, developing in five percent–10 percent of the population.1,4 If a dentist examines 40 to 50 new or recall patients per week, that would translate into a handful of patients per week with a harmless looking lesion. Although most of these lesions are entirely benign and harmless, clearly, some may not be.

The only way to tell which nonsuspicious lesion needs additional evaluation and treatment is to biopsy every lesion, which is impractical considering the fact that dentists would need to refer several patients each week for an invasive biopsy that has potential morbidity. Consequently, many unsuspected pre-cancers and early cancers are left to progress to a more advanced and deadly stage, ultimately developing signs and symptoms of suspicion, finally triggering the need for a biopsy. This may explain why, in the study by Wildt et al.,5 the delay in the diagnosis of oral cancer by dentists was longest for patients with small, nonsuspicious oral lesions.

The brush biopsy empowers dentists with a tool that can be used to painlessly and accurately evaluate commonly encountered harmless-looking lesions.6,7 Dentists no longer have to speculate about which oral lesions are benign and require no treatment, and which are potentially malignant, despite the benign appearance of the lesion. Early detection of unsuspected oral cancers and their precursor lesions is now possible, and fortunately for patients, it is also painless and effortless.8

The early detection of oral cancer clearly saves lives. Patients fortunate enough to be identified with early-stage oral cancers have significantly improved survival rates and undergo much less extensive surgical procedures than patients with late-stage lesions.


With the advent of the brush biopsy, we have developed a new protocol for dentists to follow as they perform oral examinations and encounter oral abnormalities. These guidelines are based upon recommendations from the American Dental Association (see sidebar). Oral lesions that appear suspicious of a malignancy should be referred for immediate biopsy. If a lesion is caused by an infection such as herpes or Candidiasis, then treatment should be instituted promptly.

Oral lesions with an obvious etiology such as trauma or aphthous ulceration require palliative treatment and observation to ensure that the lesion resolves. And if a lesion looks harmless and its cause unexplained, then a brush biopsy should be performed to determine if it is potentially pre-cancerous or cancerous. This advice ensures that all suspected, dangerous lesions are diagnosed and treated as soon as possible and all unsuspected, dangerous lesions are detected early when they can be most easily treated or even cured.


There has been a sharply increasing trend in the incidence of tongue cancer in young patients since the mid 1970s.9 According to the cancer surveillance database from the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) Program, the incidence of head and neck cancer has remained stable in patients older than 40 years of age comparing the periods from 1973-1984 to 1985-1997, while tongue cancer in adults younger than 40 years increased approximately 60 percent during the same period.9

Although tobacco and alcohol have long been implicated as traditional risk factors for oral cancer in adults of any age, studies have demonstrated that a significant number of young patients, under age 40, with oral cancer have never smoked or consumed alcohol. Investigators have theorized that the duration of exposure to carcinogens such as alcohol and tobacco in young oral cancer patients may be too short for malignant transformation.10,11 In addition, according to the American Cancer Society, the incidence of oral cancer in women has increased significantly, and females now account for about one third of all oral cancer cases. These facts underscore the need for a thorough oral cavity examination in all patients and not just those considered at high risk.


The computer-assisted analysis of the oral brush biopsy, also known as OralCDx, was introduced in 2000. It was the recipient of American Dental Association (ADA) “Seal of Acceptance” in 2000.

The ADA launched 2 nat
ional oral cancer awareness campaigns, the first in 2001 and the second in 2003, and the primary message to the public in both campaigns was to emphasize that there was a new reason to see their dentist: “testing is now painless.” The campaign urged patients to see their dentists for oral cancer examinations and to have unexplained white and red spots tested painlessly. The ADA, through press releases, has stated that the success of the campaigns was overwhelmingly positive.

According to CDx Laboratories, the oral brush biopsy has been used to detect over 5,000 pre-cancers and cancers among lesions that would not have aroused sufficient suspicion to biopsy prior to the advent of this test. The brush biopsy has been integrated into the curriculum and clinics at many dental schools in the United States, including ours, Virginia Commonwealth University.

The brush biopsy technique was demonstrated to be accurate in identifying dysplasia and carcinoma with high sensitivity and specificity in a U.S. multicenter trial.12 The study involved nearly 1000 patients and was conducted by academicians who specialized in oral pathology, oral medicine, and oral surgery in 35 dental institutions. The results of scalpel biopsy and brush biopsies were obtained independently from one another and in a blinded fashion.

We subsequently performed additional research on the oral brush biopsy. We compared the results of brush biopsies with results from scalpel biopsies that were triggered as a result of an abnormal brush biopsy to determine the positive predictive value of the test.13 We determined that the positive predictive value of an abnormal brush biopsy, defined as a measure of the likelihood that a patient with an abnormal brush biopsy will have a histologic report of dysplasia or carcinoma on follow-up scalpel biopsy, was 38.3 percent.

This figure is comparable to the value determined in the US multicenter trial and significantly higher than the positive predictive value for other cancer screening tests such as the Pap smear and mammography. Stated simply, if a dentist performs a brush biopsy and the test is abnormal, when the patient undergoes a follow-up scalpel biopsy, the chances that the lesion will prove to be pre-cancer or cancer is about 30 percent.


Indications for a brush biopsy include small or tiny, nonsuspicious, common, unexplained red or white spots.14 These lesions traditionally have been “watched” since they were thought to always be benign and not suspicious enough to subject to biopsy. By definitively determining the significance of these types of oral lesions, dentists no longer have to speculate about which are benign and which are potentially malignant.15 Specifically, benign-appearing lesions that are in fact benign are confirmed as such when they are tested with the brush biopsy.

More importantly, benign appearing oral lesions that are in fact pre-cancerous or cancerous are identified with the test. Since pre-cancers and early stage oral cancers are not easily recognizable by clinical inspection alone, identifying potentially harmful oral lesions that would not have ordinarily been referred for biopsy represents the major breakthrough of the brush biopsy tool in the early detection of cancers and their precursor lesions.

Contraindications for the brush biopsy are lesions that are highly suspicious and require immediate scalpel biopsy. Additionally, lesions with an intact epithelium such as submucosal masses, pigmented lesions, fibromas, and mucoceles should be biopsied with a scalpel when indicated.


Learning to perform the oral brush biopsy does not require extensive training. An extensive free online course is available on the CDx Laboratories web site at In addition, an instructional video or CD-Rom on the technique is supplied to all dentists who request kits from CDx Laboratories (Suffern, NY), providers of OralCDx. Dentists are supplied with kits that contain all of the materials needed to perform the brush biopsy procedure as well as complete instructions for use.

A major advantage of the oral brush biopsy is that it requires no topical or local anesthetic and causes minimal or no bleeding or pain. The brush biopsy instrument has two cutting surfaces, the flat end of the brush and the circular border of the brush (Fig. 6). Either surface may be used to obtain the specimen. For example, the flat surface of the brush may be more suitable for lesions on the buccal mucosa and labial mucosa (Fig. 7), while the circular end of the brush may be utilized for lesions on the lateral tongue and gingiva (Fig. 8).

The brush biopsy instrument is supplied sterile and is designed to obtain a trans-epithelial biopsy specimen, capturing cells from all three layers of the epithelium of the lesion. Unlike a typical cytologic smear, such as a Pap smear, which samples cells only from the superficial layer, the brush biopsy samples the superficial and deep layers of the lesion. The deeper layers of the oral epithelium are often the only layers that contain the pre-cancerous and cancerous cells, which are not sampled by traditional cytology methods. The brush biopsy overcomes this problem by sampling tissue of the entire lesion down to the basement membrane.

Important in the use of the brush, is learning how much pressure to apply and how many rotations are required. The cutting edge of the brush is placed against the lesion and while maintaining firm pressure, rotated in a clockwise fashion. As you bear down on the tissue surface, you should see a small or slight bend in the shaft of the brush prior to beginning your rotational movement. As you continue to press down firmly, bend and rotate all in the same motion. Although the brush biopsy procedure is relatively painless, it is crucial that you apply enough pressure to achieve a complete biopsy sample, capturing cells from the entire thickness of the epithelium.

The brush should be repeatedly rotated in most cases about 5-15 times. Red lesions and ulcerations generally require little pressure and few rotations while white lesions, which are typically covered with keratin, require more pressure and more rotations to reach the basement membrane. Lesions on the palate and gingiva also require more rotations and firmer pressure since these sites are keratinized. If the mucosa at site of the oral lesion turns pink or red, and if pinpoint bleeding is observed, then the brush has penetrated to the basement membrane and a complete transepithelial sample has been achieved. Typically, after just the first or second brush biopsy procedure, dentists feel very confident and adept at performing the technique.

After obtaining the sample, the cellular material on the brush needs to be transferred to the glass slide provided in the kit. The surface of the brush used to sample the lesion should be rotated on the glass slide from one end to another. Transfer as much cellular material from the brush to the slide by using the same rotating motion on the slide as used in the mouth. A thin film of material representing the biopsy specimen should be observed on the glass slide if it is held up to the light. If this is not evident, obtain additional material from the patient’s lesion utilizing the same brush biopsy instrument.

Once the transfer is complete, the glass slide is then flooded with fixative that is supplied in individual packets with each kit. At first, it may look like the fixative washes away the cells from the slide since the fixative often overflows the slide. Excessive fixative that overflows from the slide is not a concern, and it is best to empty the entire contents of the fixative packet onto the slide.

The glass slide is set aside to dry. Usually after 15 or 20 minutes, the fixative has

dried and the slide is ready to place into the slide holder that is supplied with the kit. A one-page form should be completed and submitted with the brush biopsy specimen. The form is easy to complete and includes information such as the doctor’s name and pr
actice number. It has a series of boxes that can be quickly marked to provide patient history, insurance and the clinical description of the lesion.

The test slide is placed back into the same slide holder in which it was shipped. The glass slide in its holder and the test form are enclosed in the provided mailer.


Every brush biopsy specimen is checked to ensure that the dentist sampled cells from all layers of the epithelium. This mechanism assures the dentist that reports are not generated from specimens whose brush biopsies were not performed correctly. Infrequently, cells from the deepest layer, the basal layer, may not have been collected by the brush biopsy. This may occur if the dentist did not apply sufficient pressure while rotating the brush.

If the number of rotations was insufficient, the brush may not have penetrated to the basement membrane. In these instances, the laboratory will detect the inadequacy of the sample, and the dentist will be informed that a complete brush biopsy of the lesion has not been submitted and is therefore inadequate for analysis. In case of an inadequate test result, the patient is not billed by the laboratory and should be rescheduled and retested.

The high accuracy of OralCDx is due to the fact that analyses of oral brush biopsies are accomplished with the assistance of sophisticated computers and advances in image recognition originally developed for the “Star Wars” missile defense program (Figs. 9A & B). The computer systems scan every cell on each specimen searching for potentially abnormal cells.

The OralCDx computer system has been demonstrated to detect one abnormal cell among one hundred thousand similarly appearing and overlapping normal cells. It is this advance in computer science that makes the OralCDx test extremely sensitive and reliable. The computer does not make the final diagnosis but presents potentially abnormal cells to the pathologist on a high-resolution computer monitor, to aid in diagnosis of the specimen. With the aid of this highly specialized system, the pathologist can detect as few as one or two abnormal individual cells in a brush biopsy specimen.


If no cellular abnormalities are detected, then the dentist receives a “negative” report indicating there is no evidence of pre-cancerous or cancerous activity in the specimen submitted. Negative OralCDx results should be expected since harmless appearing lesions are being tested, and most oral lesions that appear benign will prove to be negative. It is important to emphasize that all patients with “negative” oral brush biopsy reports must be followed. Lesions that persist and remain unchanged should be retested in 6 months. Lesions that change require retesting as soon as either the patient or the dentist notes the change.

If cellular abnormalities are detected, then the dentist receives a report of either “positive”, which invariably means that the lesion is pre-cancerous or cancerous, or “atypical”, which means that there are abnormal cells that require further investigation to identify their significance. An “atypical” brush biopsy specimen contains abnormal cells, and this is important since in approximately one third of these cases, it indicates a pre-cancerous or cancerous lesion.13 Therefore, it is recommended that all patients with an “atypical” report be evaluated for a scalpel biopsy. A “positive” specimen contains dysplastic cells indicating an oral pre-cancerous or cancerous lesion, and therefore, all patients with a “positive” report should receive a scalpel biopsy.

The use of the oral brush biopsy assists the dentist by confirming the benign nature of a lesion (a “negative” brush biopsy result) and identifies which oral lesions may be potentially pre-cancerous and cancerous (an “atypical” or “positive” report). If the brush biopsy is atypical or positive, the dentist also receives a printout of color images of cellular abnormalities with an accompanying explanation. These are useful in demonstrating the abnormalities to the patient and in persuading patients with abnormal results that they must follow-up with the oral surgeon for scalpel biopsy.


With the increased incidence in oral cancer in young patients under age 40 and in women, coupled with the fact that 25 percent of all oral cancer patients have no risk factors, it is incumbent upon dentists to screen all patients for oral cancer. At least 5 percent of your patients will display an oral lesion, and almost all will appear harmless.

Since the majority of lesions are proven benign with brush biopsy, prior to performing one, your patient should be informed that 1) oral lesions like the one you discovered are commonly detected; 2) most prove to be totally benign; 3) you would like to be certain that the lesion is harmless and therefore you are going to test it; 4) the test does not require anesthetic, is painless and accurate; and 5) if the test comes back abnormal, then at least it will be identified early, when it can easily be treated.

If the brush biopsy test does in fact return as “atypical”, your patient should be reassured that there is no cause for alarm. Based upon the positive predictive value of the test, you can explain that approximately two thirds of patients with an “atypical” brush biopsy report will prove to have inflammation or other non-cancerous conditions when they undergo a scalpel biopsy. Since one-third will be pre-cancerous or cancerous, all patients with an “atypical” report should be evaluated for a scalpel biopsy.


The dentist performing the brush biopsy charges the patient for a biopsy procedure. CDx Laboratories, providing computer-assisted analysis of the oral brush biopsy specimen, separately bills the patient for its services. Billing instructions are supplied in each kit.


Prior to 1955, cervical cancer was once one of the most common causes of cancer death for American women. The increased use of the Pap test between 1955 and 1992 resulted in a reduction of cervical cancer deaths in the United States by 74 percent.16 Dentists now have a momentous opportunity to have the same effect on oral cancer as the Pap smear had on cervical cancer.

Adjuncts for early cancer detection, such as the Pap smear for cervical cancer, mammography for breast cancer, and the PSA test for prostate cancer, have been conspicuously lacking in the field of oral diagnosis. By using the oral brush biopsy to test harmless looking lesions, this minimally invasive, easily learned, patient accepted biopsy procedure fills the void in the early detection of oral cancer that will ultimately impact the poor morbidity and mortality of this disease.

Drs. Svirsky, Abbey and Burns are Professors of Oral & Maxillofacial Pathology and Dr. Page is Associate Professor of Oral & Maxillofacial Pathology at the Virginia Commonwealth University, Richmond, VA.

Oral Health welcomes this original article.


1.Christian DC. Computer-assisted analysis of oral brush biopsies at an oral cancer screening program. J Am Dent Assoc. Mar 2002;133(3):357-362.

2.Silverman S, American Cancer Society. Oral cancer. 4th ed. Hamilton, ON; Lewiston, NY St. Louis, MO: B.C. Decker ; Mosby-Year Book distributor; 1998.

3.Silverman S, Jr. Early diagnosis of oral cancer. Cancer. 1988;62(8 Suppl):1796-1799.

4.Bouquot JE. Common oral lesions found during a mass screening examination. J Am Dent Assoc. 1986;112(1):50-57.

5.Wildt J, Bundgaard T, Bentzen SM. Delay in the diagnosis of oral squamous cell carcinoma. Clin Otolaryngol. 1995;20(1):21-25.

6.Drinnan AJ. Screening for oral cancer and pre-cancer — A valuable new technique. Gen Dent. 2000;48:656 – 660.

7.Zunt SL. Transepithelial Brush Biopsy: an adjunctive diagnostic procedure. J Indiana Dent Assoc. 2001; 80(2):6-8.

8. Felefli S, Flaitz CM. The oral brush biopsy: it’s as easy as 1, 2, 3. Tex Dent J. Jun 2000;117(6):20-24.

9.Schantz SP, Yu GP. Head and neck cancer incidence trends in young Americans, 1973-1997,
with a special analysis for tongue cancer. Arch Otolaryngol Head Neck Surg. Mar 2002;128(3):268-274.

10.Llewellyn CD, Johnson NW, Warnakulasuriya KA. Risk factors for squamous cell carcinoma of the oral cavity in young people–a comprehensive literature review. Oral Oncol. Jul 2001;37(5):401-418.

11.Lingen M, Sturgis EM, Kies MS. Squamous cell carcinoma of the head and neck in nonsmokers: clinical and biologic characteristics and implications for management. Curr Opin Oncol. May 2001;13(3):176-182.

12.Sciubba JJ. Improving detection of pre-cancerous and cancerous oral lesions. Computer-assisted analysis of the oral brush biopsy. U.S. Collaborative OralCDx Study Group. J Am Dent Assoc. 1999;130(10):1445-1457.

13.Svirsky JA, Burns JC, Carpenter WM, et al. Comparison of computer-assisted brush biopsy results with follow up scalpel biopsy and histology. Gen Dent. Nov-Dec 2002;50(6):500-503.

14.Svirsky JA, Burns JC, Page DG, Abbey LM. Computer-assisted analysis of the oral brush biopsy. Compendium. 2001;22:99-106.

15.Tauberg JA. An oral surgeon’s view of the computer-assisted oral brush biopsy. Alpha Omegan. Aug 2002;95(2):9-11.

16.Sawaya GF, Brown AD, Washington AE, Garber AM. Clinical practice. Current approaches to cervical-cancer screening. N Engl J Med. May 24 2001;344(21): 1603-1607.



“Your dentist has recent good news about progress against cancer. It is now easier than ever to detect oral cancer early, when the opportunity for a cure is great.”

“In about 10 percent of patients, the dentist may notice a flat, painless, white or red spot or a small sore.”

“Although most of these are harmless, some are not. Harmful oral spots or sores often look identical to those that are harmless – testing can tell them apart.”

“To ensure that a spot or sore is not dangerous, your dentist may choose to perform a simple test, such as a brush biopsy, which usually is painless and can detect potentially dangerous cells when the disease is still at an early stage.”

“If your dentist notices something that looks very suspicious and dangerous, a scalpel biopsy may be recommended. This usually requires local anesthesia. Your general dentist may perform this procedure or refer you to a specialist for it.”