Does Oral Cancer Screening Have a Clear Benefit in High-Risk Groups, and When Does Screening Become Routine Without Hard Outcome Evidence?

What does "Targeted Screening in High-Risk Groups" mean for me as a patient?

One of the most common questions patients ask about this topic concerns targeted screening in high-risk groups. The answer is not as straightforward as one might hope — but research now provides clear guidance.

The strongest and still singular intervention evidence for oral cancer screening comes from the Trivandrum Oral Cancer Screening Trial, a large cluster-randomized controlled trial from the Kerala region in southern India (Sankaranarayanan et al. 2005). In this study, 13 clusters with approximately 96,000 participants were assigned to a screening group in which trained health workers performed a visual oral examination (VOE). An additional 13 clusters with approximately 95,000 participants served as a control group without systematic screening. The primary endpoint was oral cancer mortality. The study is unique because it is the only randomized trial worldwide providing hard mortality data on oral cancer screening.

In the initial report after 9 years of follow-up (Sankaranarayanan et al. 2005), the intent-to-screen analysis for the overall population showed no statistically significant mortality benefit (rate ratio 0.79; 95% CI 0.51–1.22). In the pre-specified subgroup analysis for tobacco or alcohol users, the effect was more pronounced but did not consistently reach the conventional significance threshold. Only the extended 15-year follow-up (Sankaranarayanan et al. 2013) provided a statistically significant mortality reduction in the high-risk subgroup (tobacco and/or alcohol users) with a rate ratio of 0.76. For the overall cohort, the effect remained non-significant (rate ratio 0.88; 95% CI 0.69–1.12). These results are used as a reference point in the entire international guideline discussion.

The clinical significance of these results must be evaluated in the context of the study population. Kerala is a region with a very high prevalence of chewing tobacco use and betel nut consumption. The background incidence of oral carcinomas there is considerably higher than in European or North American settings. The 24% mortality reduction therefore refers to a population with an a priori substantially higher disease risk. A direct transfer to European low-prevalence populations is not methodologically straightforward. Nevertheless, biological plausibility is high: when pre-test probability for oral carcinomas is elevated by risk factors, the positive predictive value of a screening finding increases and the number needed to screen per prevented death decreases substantially. This is precisely the mechanism reflected in the Kerala subgroup analysis.

Two additional aspects limit transferability: First, the visual examination in Kerala was performed by health workers, not dentists. The sensitivity and specificity of this VOE implementation may differ in the context of specialized Western dental practices. Second, there was no systematic follow-up for screening-negative participants. This means the study primarily measures the effect of a single or repeated screening pass on mortality, not the effect of a continuous surveillance program. Despite these limitations, the Kerala trial remains the only controlled study with hard mortality data and thus represents the most robust single piece of evidence for a mortality benefit of targeted screening in high-risk groups.

The systematic review by Pedroso et al. (2024) specifically examined screening outcomes in the Latin American region, including twelve studies from Brazil and Cuba. A total of 13,277,608 individuals were screened, with 1,516 oral carcinomas detected — a detection rate of 0.01%. Only two of the included studies explicitly targeted high-risk populations (smokers and alcohol users). VOE sensitivity in the two studies with diagnostic data was 100%, with specificity between 75% and 90%. The positive predictive value (PPV) ranged between 0.36 and 0.43 — values considered acceptable but not high for a screening instrument in a low-prevalence population. However, these accuracy data came from small studies (Simonato et al. 2019: 283 and 1,765 participants; Pivovar et al. 2017: 202 participants), substantially limiting generalizability.

The Cuban screening program (Santana et al. 1997), which examined over 10 million individuals by VOE from 1983 to 1990, demonstrated a shift toward earlier tumor stages (Stage I and II from 1985 onward compared with Stage III previously) over several years. However, the program was unable to demonstrate a measurable reduction in overall incidence or mortality. The Argentine intervention campaign, which combined dental continuing education with public health outreach, reported a reduction in diagnostic delay and — after ten years of continuous campaigning — a decrease in mortality rates (Pedroso et al. 2024). This suggests that combining screening with health education may be more effective than screening alone.

For dental practice, the high-risk evidence means that targeted attention to patients with known tobacco and/or alcohol use is clinically justified. The Kerala data suggest that in high-risk patients, a systematic visual inspection of the oral mucosa — including the tongue, floor of the mouth, inner cheeks, and soft palate — can produce a mortality benefit. This is particularly true for patients with additional risk factors such as chronic betel nut use, HPV-16 positivity, or a family history of head and neck carcinomas. The threshold for further workup (biopsy, referral) should be set low for these patients.

Transferring the Kerala results to European practice populations requires calibration: while the biological plausibility for benefit in high-risk patients also exists in European populations, specific European intervention evidence is lacking. The number needed to screen will likely be considerably higher in populations with lower background prevalence. For clinical decision-making, it is therefore relevant that in Germany, Austria, and Switzerland, the age-standardized incidence of oral carcinomas in men is between 5 and 10 per 100,000 — significantly lower than in Kerala. Blanket screening of all adults cannot be derived from these data, but systematic vigilance in high-risk individuals is well-supported.

In everyday practice, this means: the evidence does not provide a one-size-fits-all answer, but rather a framework for individualized decisions. Patient-specific factors such as overall health, compliance, individual risk profiles, and treatment preferences must be incorporated into the decision.

What does this mean for you? High-risk criteria must be explicitly identified.

As a patient, it is important to know: no examination method is perfect. Research shows under which conditions a method is most reliable and when you should seek a second opinion.

Science has intensively studied this topic in recent years. For this article, more than 11 scientific papers were evaluated. It is important to understand: not every study carries the same weight. Large, well-controlled investigations yield more reliable results than small observational studies. The overall picture from these various studies is what we present to you here.

What is better: routine examination or hard added value?

When it comes to routine examination vs. hard added value, the research picture is clearer than many think. Here you will find out what current studies actually show.

The central question of whether systematic visual oral cancer screening for the entire adult population — regardless of individual risk profile — produces a mortality benefit cannot be answered based on current evidence. The Cochrane review by Brocklehurst et al. (2013) systematically examined the evidence for population-based oral cancer screening and identified only one randomized controlled trial (the Kerala trial), whose overall population analysis showed no significant mortality benefit. The authors concluded that the available evidence is insufficient to make a clear recommendation for or against population-based screening, and emphasized the need for further large randomized studies in various settings.

The US Preventive Services Task Force (USPSTF) published an assessment in 2013 (Moyer et al. 2014) that resulted in an I-rating (Insufficient Evidence). The panel stated that the current evidence is insufficient to weigh the benefits of screening for oral carcinomas in asymptomatic adults by primary care physicians or dentists against potential harms. This assessment explicitly referred to population-wide routine screening, not to clinical examination within the context of risk-adapted assessment. The USPSTF rating was controversially discussed in the professional community, as representatives of the oral medicine community (including Downer et al. 2004; Speight et al. 2006) argued that the absence of evidence should not be equated with evidence of ineffectiveness.

The Taiwanese screening program provides supplementary, albeit non-experimental, evidence. Taiwan implemented a nationwide screening program for betel nut chewers and smokers beginning in 2004, conducted by dentists and physicians through visual inspection. Observational data (Chuang et al. 2017; Chen et al. 2021) show an association between screening participation and earlier tumor diagnosis as well as reduced mortality. In a large long-term observational study (Chen et al. 2021) with more than 4 million participants, screening was associated with a relative mortality reduction of approximately 26% (adjusted hazard ratio 0.74; 95% CI 0.65–0.83). However, these data are observational and may be subject to healthy-screening bias and selection effects. The Taiwanese program also specifically targets a high-risk population and is therefore not evidence of benefit from universal routine screening.

In Europe, there are no population-based screening programs for oral carcinomas as a public health measure. The European Commission has not included oral carcinomas as a target pathology for organized screening in its cancer screening recommendations (Council Recommendation 2003, Update 2022). In Germany, the early detection guidelines of the Federal Joint Committee (G-BA) do not contain a specific position on oral cancer screening. The routine dental examination includes inspection of the oral mucosa; however, this is not structured as a formal screening program with defined quality assurance, invitation system, and outcome monitoring. It therefore represents opportunistic clinical vigilance, not screening in the epidemiological sense.

The evidence on harms from population-wide oral cancer screening is also limited, but is considered relevant in the USPSTF analysis. Potential harms include false-positive findings leading to unnecessary biopsies and psychological distress, lead-time and length-time bias in the interpretation of stage shifts, and overdiagnosis of indolent lesions — particularly oral leukoplakias, whose malignant transformation rate varies between 0.13% and 34% per year depending on the cohort (Warnakulasuriya and Ariyawardana 2016). Blanket screening could therefore identify a substantial number of individuals with oral potentially malignant disorders (OPMDs), the vast majority of whom will never develop carcinoma but may be subjected to invasive diagnostics and surveillance.

Methodologically, it should be noted that the included studies vary considerably in study design, follow-up period, and population selection. This heterogeneity limits the comparability of results and explains why pooled effect estimates must be interpreted with caution. Nevertheless, the direction of the effect is consistent across different study types.

For applicability to the German-speaking healthcare context, it is additionally relevant that a substantial proportion of the evidence comes from Anglo-American or Scandinavian healthcare systems. Differences in reimbursement structures, treatment culture, and patient access can influence effect sizes, without invalidating the core conclusion.

For daily dental practice, the state of research means that mucosal inspection as an integral part of the routine examination is clinically reasonable and professional standard — but cannot be considered an evidence-based screening program in the public health sense. The distinction is relevant: attentive clinical examination that identifies suspicious findings and refers them for evaluation is of high value. The claim that this examination demonstrably reduces oral cancer mortality at the population level is not supported by available data.

Dental professional societies (including ADA, FDI, DGZMK) recommend systematic oral mucosal inspection at every recall appointment. This recommendation is based on clinical plausibility and expert consensus, not on high-quality intervention evidence. It is to be regarded as a good-practice statement, not as an evidence-based screening recommendation in the GRADE sense. For practice, this means performing clinical inspection consistently, without communicating to patients that it is a definitively mortality-reducing measure.

In everyday practice, this means: the evidence does not provide a one-size-fits-all answer, but rather a framework for individualized decisions. Patient-specific factors such as overall health, compliance, individual risk profiles, and treatment preferences must be incorporated into the decision.

What does this mean for you? Routine care and screening must be linguistically distinguished.

As a patient, it is important to know: no examination method is perfect. Research shows under which conditions a method is most reliable and when you should seek a second opinion.

How do scientists arrive at these conclusions? They do not evaluate just one study, but examine many investigations simultaneously. This allows them to determine whether a result was coincidental or consistently reproduced. In this case, the findings are based on 11 scientific papers from various countries and research groups.

What does "Adjunctive Tools" mean for me as a patient?

One point that often causes confusion is adjunctive tools. However, science has made important advances in recent years.

In addition to conventional visual examination (VOE), various adjunctive technologies have been investigated for the past two decades to improve oral cancer detection. The most important methods include autofluorescence imaging (e.g., VELscope, Identafi), vital staining with toluidine blue, chemiluminescence (e.g., ViziLite), brush biopsy with cytological analysis (e.g., OralCDx), and more recently saliva-based biomarkers and artificial intelligence approaches. For none of these methods does a randomized controlled trial exist demonstrating mortality reduction as a primary endpoint.

Autofluorescence technology exploits the differential fluorescence of healthy and pathologically altered oral mucosa under blue-light excitation. The systematic review by Lingen et al. (2017) analyzed the diagnostic accuracy of autofluorescence-based methods and found a pooled sensitivity of 80–100% with a specificity of only 15–58%. The high sensitivity combined with low specificity means the method rarely misses suspicious lesions but produces a large number of false-positive findings. Inflammatory changes, ulcers, melanotic pigmentations, and even amalgam tattoos can produce fluorescence loss indistinguishable from malignant changes. The study by Simonato et al. (2019), reported within the Latin American review (Pedroso et al. 2024), found a sensitivity of 100% and specificity of 90% for the combined VOE-plus-fluorescence approach — values based on a study population of only 283 individuals and therefore not generalizable.

Toluidine blue (tolonium chloride) is a vital stain that binds selectively to dysplastic or neoplastic tissue. The systematic review by Macey et al. (2015) reported a pooled sensitivity of 86% and specificity of 67% for the detection of oral carcinomas and high-grade dysplasias. Diagnostic accuracy varies considerably by setting: false-positive rates are high in high-risk primary care, while specificity is better in specialized clinics. Toluidine blue is used in some specialized centers as an adjunct to clinical decision-making but has not been incorporated into universal screening protocols.

Brush biopsy systems (e.g., OralCDx) enable minimally invasive transepithelial cell sampling with subsequent computer-assisted cytological analysis. The initial study by Sciubba (1999) reported a sensitivity of 100% and specificity of 93%, values that could not be reproduced in subsequent independent validation studies. The Cochrane review (Macey et al. 2015) found high variability in diagnostic accuracy for brush cytology depending on the threshold and reference standard. The method cannot replace incisional biopsy but may serve as a triage instrument before deciding on biopsy.

Saliva-based biomarkers represent the most recent research avenue. Various candidate markers — including IL-8, IL-6, VEGF, p53 autoantibodies, microRNAs (miR-31, miR-21), and cell-free DNA — have been identified in pilot studies as potentially discriminating between healthy mucosa and oral carcinomas (Elashoff et al. 2012; Rapado-González et al. 2020). The diagnostic performance of individual markers is currently insufficient for a screening instrument: AUC values typically range between 0.68 and 0.85, which would generate an unacceptably high false-positive rate for population-based screening with low prevalence. Multi-marker panels show better performance in derivation cohorts (AUC > 0.90) but require validation in prospective screening settings.

Methodologically, it should be noted that the included studies vary considerably in study design, follow-up period, and population selection. This heterogeneity limits the comparability of results and explains why pooled effect estimates must be interpreted with caution. Nevertheless, the direction of the effect is consistent across different study types.

For applicability to the German-speaking healthcare context, it is additionally relevant that a substantial proportion of the evidence comes from Anglo-American or Scandinavian healthcare systems. Differences in reimbursement structures, treatment culture, and patient access can influence effect sizes, without invalidating the core conclusion.

For dental practice, current evidence indicates that no adjunctive method can replace visual inspection as the primary screening instrument or reliably enhance the clinical value of VOE. Autofluorescence, toluidine blue, and brush biopsy can be used in specialized settings as supplementary decision aids for clinically unclear lesions, but are not screening instruments for general practice. The marketing of these methods as practice-ready screening tools substantially overstates the available evidence.

Saliva-based biomarkers and AI-assisted image analysis are currently in the research and development stage. While individual studies show promising results, each of these technologies lacks prospective validation in an unselected screening population. Until such studies are available, these methods should not be communicated or applied as clinically validated screening instruments.

In everyday practice, this means: the evidence does not provide a one-size-fits-all answer, but rather a framework for individualized decisions. Patient-specific factors such as overall health, compliance, individual risk profiles, and treatment preferences must be incorporated into the decision.

What does this mean for you? Adjunctive procedures must not be overstated.

As a patient, it is important to know: no examination method is perfect. Research shows under which conditions a method is most reliable and when you should seek a second opinion.

What makes these results reliable? In medical research, the principle is: the more independent studies reach the same conclusion, the more reliable the statement. The type of study and number of participants also play an important role. Large controlled studies with many participants provide more reliable results than small surveys.