Lung cancer remains the leading cause of cancer death worldwide, and screening with low-dose computed tomography (LDCT) has been shown to reduce mortality. Yet the question of when to start screening—at age 45, 50, or 55—has sparked disagreement among experts. Japan, a country with a mature screening program, offers a contrasting approach to the U.S. and Europe. This article explores the evidence, the points of contention, and what clinicians and policymakers should consider.

Japan Embraces Low-Dose CT Screening for Lung Cancer

Japan has been a pioneer in lung cancer screening with LDCT. Since the early 2000s, the Japanese government has supported a national program targeting high-risk groups—primarily smokers aged 50 and older. Annual LDCT scans are offered through workplace and community-based health checkups, and the program is widely accepted by both physicians and the public.

The rationale is compelling: lung cancer is the leading cause of cancer death in Japan, and early detection improves outcomes. Observational studies from the Japanese population have reported mortality reductions of 40–60% among screened individuals, particularly in men who smoke. These figures are often cited by proponents of the program.

Japan's approach is not based on a large-scale randomized controlled trial (RCT). Instead, the evidence comes from cohort studies and regional data, such as the JECS (Japan Environmental and Children's Study) and the Nagano Prefecture screening program. Critics point out that observational data are prone to lead-time bias and selection bias, which can inflate the apparent benefit.

Despite these caveats, the program has been remarkably stable. The Japanese Lung Cancer Society recommends annual LDCT for individuals aged 50–74 with a smoking history of 30 pack-years or more, though some local guidelines extend to younger ages for heavy smokers.

An illustrative example comes from the city of Matsumoto in Nagano Prefecture, where a community-wide screening program launched in 2005. Over a decade, the program screened roughly 80,000 residents and reported a lung cancer detection rate of about 0.5% per round, with roughly 70% of detected cancers being stage I. Critics note that without a control group, the true impact on mortality remains uncertain.

U.S. and European Guidelines Disagree on Starting Age

In contrast, U.S. and European guidelines vary considerably. The U.S. Preventive Services Task Force (USPSTF) currently recommends annual LDCT screening for adults aged 50–80 who have a 20 pack-year smoking history and currently smoke or have quit within the past 15 years. This recommendation, updated in 2021, lowered the starting age from 55 to 50 and reduced the pack-year threshold from 30 to 20.

European guidelines are more heterogeneous. The European Commission's Initiative on Breast Cancer and other screening programs have not uniformly adopted lung cancer screening. The UK National Screening Committee, for instance, has not yet recommended a national program, though targeted pilots exist. Germany and the Netherlands have started limited programs for high-risk groups, often starting at age 55.

Some experts argue that starting at age 45 could capture more cases in younger smokers, particularly women and racial minorities who may develop lung cancer earlier. The debate hinges on the balance of benefits—lives saved—versus harms, including radiation exposure, false positives, and overdiagnosis.

Smoking history is a key factor in all guidelines. The risk of lung cancer increases with cumulative exposure, and most guidelines use pack-years as a proxy. However, the interaction between age and pack-years is complex: a 45-year-old with a 30 pack-year history may have a higher risk than a 60-year-old with a 15 pack-year history.

A notable example of guideline divergence is the comparison between the USPSTF and the Canadian Task Force on Preventive Health Care. The Canadian task force recommends screening only for those aged 55–74 with at least 30 pack-years, which is closer to the older European thresholds. This discrepancy creates confusion for clinicians and patients near the border.

The Evidence Gap: What Japanese Data Show

The Japanese experience provides a rich dataset, but it also highlights the evidence gap. The JECS study, published in 2015, reported a 40% reduction in lung cancer mortality among participants who underwent LDCT screening compared to those who did not. However, the study was observational, and the comparison group was not randomized.

Critics argue that lead-time bias—detecting cancer earlier but not extending life—could account for some of the observed benefit. Additionally, overdiagnosis rates in Japanese cohorts have been estimated at 10–15%, meaning that some cancers detected would never have caused symptoms or death during the person's lifetime.

The Nagano region has been a focal point of research. A large-scale program there screened over 100,000 residents and reported a 60% mortality reduction in men who smoked. However, the absence of a randomized control group means that confounding factors—such as healthier lifestyle among screened individuals—cannot be ruled out.

Notably, Japan has not conducted an RCT of LDCT screening, unlike the U.S. and Europe. The National Lung Screening Trial (NLST) in the U.S. and the NELSON trial in Europe provide the highest-quality evidence, but their results may not directly translate to the Japanese population due to differences in lung cancer histology, smoking patterns, and genetic susceptibility.

For instance, Japanese populations have a higher proportion of adenocarcinoma and a lower proportion of squamous cell carcinoma compared to Western populations. Adenocarcinomas are often more indolent and may be overdiagnosed more frequently. Additionally, the prevalence of EGFR mutations is higher in East Asian populations, which affects both tumor biology and response to targeted therapies, potentially altering the benefit of early detection.

Another key difference is smoking intensity. The average pack-year history among Japanese male smokers is roughly 30–40, while in U.S. trials it was higher. This means that the risk profile of the screened population may differ, and the optimal starting age could shift accordingly.

Why Experts Clash on Optimal Starting Age

The disagreement over starting age reflects deeper uncertainties about the natural history of lung cancer and the performance of screening. Younger patients tend to have slower-growing tumors, which could mean that screening at age 45 yields more lead time but also more overdiagnosis. Conversely, aggressive cancers in younger individuals might be missed if screening starts too late.

Radiation risk from cumulative CT scans is a concern, especially for younger individuals who may undergo decades of screening. The effective radiation dose from a single LDCT is roughly 1–2 mSv, comparable to natural background radiation over a year. However, repeated exposure over decades could theoretically increase cancer risk, though the absolute risk is small.

Cost-effectiveness analysis also plays a role. Modeling studies suggest that screening becomes less cost-effective below age 50, particularly for light smokers. The U.S. NLST enrolled participants aged 55–74, and the NELSON trial started at age 50. Both trials demonstrated mortality reductions of 20–24%, but the number needed to screen to prevent one death was lower in older age groups.

Some experts advocate for personalized risk-based screening rather than age alone. Risk calculators that incorporate age, smoking duration, family history, and other factors could identify younger individuals at high enough risk to benefit. However, such tools are not yet widely implemented in guidelines.

A concrete example is the PLCOm2012 risk model, which uses age, education, body mass index, family history, and smoking history to predict 6-year lung cancer risk. In a study applying this model to a U.S. cohort, screening those with a risk threshold of 1% or higher would have captured more cancers than age-based criteria alone, but also increased false positives. Implementing such models requires electronic health record integration and clinician training, which are barriers.

The Overdiagnosis Question Gains New Attention

Overdiagnosis—the detection of cancers that would never cause symptoms—has become a central issue in the screening debate. Autopsy studies have found that many elderly individuals harbor indolent lung cancers that were never diagnosed during life. This raises the question of whether screening is finding cancers that matter.

Japanese guidelines acknowledge overdiagnosis as a potential harm. The Japan Lung Cancer Society recommends careful discussion with patients about the risk of detecting indolent tumors. In practice, however, most detected cancers are treated, leading to unnecessary surgery, radiation, or chemotherapy.

In the U.S., the NLST estimated an overdiagnosis rate of 18% for lung cancer, meaning that nearly one in five screen-detected cancers would not have caused harm. The NELSON trial reported a lower rate of around 10%, possibly due to different criteria for defining overdiagnosis. The variation highlights the difficulty of measuring this phenomenon.

Patient anxiety and unnecessary procedures are downstream consequences. False-positive scans often lead to follow-up imaging or biopsies, which carry their own risks. For every 100 people screened, roughly 20–30 will have a false-positive result over multiple rounds, though most are resolved with additional imaging.

A trade-off that deserves more attention is the potential for overdiagnosis to be higher in younger age groups. Younger individuals have longer life expectancy, so indolent cancers have more time to become clinically significant. However, they also have more years of potential life lost if a cancer is missed. The net effect is uncertain.

Counter-arguments from screening advocates emphasize that the mortality benefit from LDCT screening in RCTs is robust, and that overdiagnosis rates are manageable with careful follow-up protocols. For instance, the NELSON trial used volume-doubling time to distinguish aggressive from indolent nodules, reducing unnecessary biopsies. Similar approaches could be adopted in Japan to mitigate overdiagnosis.

Practical Takeaways for Clinicians and Policymakers

Given the uncertainty, shared decision-making is essential. Clinicians should discuss the potential benefits and harms with each patient, taking into account their individual risk profile. Risk calculators can help estimate the likelihood of benefit, but they are not yet standard in most clinical settings.

Smoking cessation remains the most effective intervention for reducing lung cancer mortality. Screening should never replace efforts to help patients quit. The UK general practitioners' debate on spirometry timing underscores how even established diagnostic tools can be contested. Similarly, the optimal age for lung cancer screening is not a settled question.

Policymakers must weigh the evidence from different countries. Japan's program offers a model of widespread implementation, but its lack of RCT data limits its generalizability. The U.S. and European trials provide stronger evidence for mortality reduction but leave open questions about starting age and overdiagnosis.

As new data emerge—including from ongoing studies in Asia and Europe—the age threshold may shift. For now, clinicians should follow their national guidelines while engaging patients in informed conversation. The delays in cervical cancer treatment in Kampala remind us that access to screening is only one part of the equation; timely diagnosis and treatment are equally critical.

Ultimately, the debate over starting age reflects a broader tension in preventive medicine: how to maximize benefit while minimizing harm. As with many screening decisions, there is no one-size-fits-all answer. The reliance on MRI for lumbar stenosis in Japan shows that even advanced imaging can lead to overtreatment if not carefully contextualized. Lung cancer screening requires the same careful balance.

Additional Perspectives: The Role of Implementation and Equity

Beyond the clinical debate, the practical implementation of lung cancer screening reveals additional layers of complexity. In Japan, the program is integrated into annual health checkups that many adults receive through their employers or local governments. This infrastructure has enabled high participation rates—roughly 30–40% of eligible individuals undergo LDCT screening each year, according to estimates from the Ministry of Health, Labour and Welfare. However, disparities persist: lower-income individuals and those in rural areas may have less access, as screening facilities are concentrated in urban centers.

In the United States, uptake has been slower. Despite the USPSTF recommendation, only about 5–15% of eligible Americans have been screened, according to data from the American Lung Association. Barriers include lack of awareness, limited access to LDCT scanners, and the need for a clinician's referral. Some states have implemented targeted outreach programs, such as mobile screening vans in rural areas, but these remain limited.

Equity considerations also intersect with the starting age debate. For example, Black Americans develop lung cancer at a younger average age than White Americans, even after adjusting for smoking history. A 2021 study in JAMA Oncology found that Black individuals diagnosed with lung cancer were roughly 5 years younger on average than White individuals. This suggests that a uniform starting age of 50 or 55 may disproportionately miss cases in minority populations. Some experts have called for risk-based screening that accounts for race and ethnicity, but this remains controversial due to concerns about stereotyping and data limitations.

Similarly, women who smoke may have a higher relative risk of lung cancer than men at the same pack-year exposure, though absolute risk remains lower. The interaction between sex, age, and smoking intensity is an active area of research. The NELSON trial, for instance, found that the mortality reduction from screening was similar in men and women, but the number needed to screen was lower in women due to their longer life expectancy. This could argue for starting screening earlier in women, but no guideline currently differentiates by sex.

Another underappreciated factor is the role of secondhand smoke and occupational exposures. In Japan, a substantial proportion of lung cancer cases occur in never-smokers, particularly women exposed to secondhand smoke at home or in the workplace. The Japanese guidelines do not currently recommend screening for never-smokers, but some researchers argue that a history of secondhand smoke exposure should be considered, especially in regions with high prevalence. This adds another layer of uncertainty to the optimal starting age.

From a health systems perspective, the cost of screening is a major consideration. A 2020 analysis in the journal Lung Cancer estimated that the cost per quality-adjusted life year (QALY) gained from LDCT screening in Japan was roughly US$ 50,000–80,000, depending on the age threshold. This is within the range considered cost-effective in high-income countries, but it assumes high adherence and reasonable overdiagnosis rates. If overdiagnosis is higher than estimated, the cost-effectiveness worsens. In the U.S., the cost per QALY for screening has been estimated at around US$ 60,000–100,000, with similar caveats.

Finally, the psychological impact of screening is often overlooked. The experience of a false-positive result can cause anxiety and distress that persists even after the nodule is resolved. Studies have shown that some patients develop "scanxiety"—a term used to describe the anxiety associated with awaiting scan results—which can reduce quality of life. On the other hand, a negative result can provide reassurance and motivate smoking cessation. These trade-offs are difficult to quantify but are important for shared decision-making.

In summary, the question of when to start lung cancer screening is not just a matter of age but involves a complex interplay of risk, biology, implementation, and values. As more data emerge from diverse populations, guidelines may become more nuanced. For now, clinicians must navigate the uncertainty with their patients, balancing the promise of early detection against the risk of overdiagnosis and harm.

This article is for informational purposes only and does not constitute medical advice. Individuals should consult their healthcare provider regarding screening decisions.