Why focus on symptomatic lung cancer?
Lung cancer is the largest cause of cancer death, with 1.8 million deaths worldwide per year.1 Despite important improvements in treatment, outcomes remain poor compared with other common cancers. In England and Wales 1-year survival is only 48%.2 Stage of cancer and performance status (PS) are the strongest independent predictors of survival. The adjusted hazard ratio for death was 4.39 for PS 4 (the most impaired PS category) compared with PS 0 (the least impaired category) and 4.58 for stage IV compared with IA–IB in the English National Lung Cancer Audit.3 Unfortunately, almost half of patients with lung cancer are diagnosed with stage IV disease in England4 and 47% of patients present at PS 2–45 when they are not eligible for most systemic treatments. Furthermore, systemic treatments are more effective in patients who are fitter. Achieving diagnosis of lung cancer in earlier stages of the disease is therefore crucial to improving survival.
Targeted screening using low-dose computed tomography (LDCT) is an important way to reduce lung cancer mortality, and implementing national programmes across the UK following the recommendation of the National Screening Committee is vital.6 However, the majority of patients will continue to present via the symptomatic route. More than half of lung cancers arise in people who would not meet the eligibility criteria for screening7 and, of those eligible, only around half choose to participate.8
Evidence suggests there is an opportunity to bring forward the diagnosis in people with symptoms and, given that many patients deteriorate prior to diagnosis, this could have important impacts on treatment eligibility and efficacy. The median symptom lead time to diagnosis of lung cancer is 3 months,9 and 33% of those with lung cancer who attend general practice before diagnosis have three or more consultations prior to referral.10 An Australian study has shown that, in retrospect, numbers of blood tests are increased in those with lung cancer 6 months prior to diagnosis and that these patients frequently have abnormal inflammatory markers and full blood counts.11 Similarly, a Danish population cohort study has demonstrated that 93% of those with lung cancer consult with their GP in the year prior to diagnosis and that those subsequently diagnosed with lung cancer have more frequent contacts with their GP and increased activity, such as investigations and antibiotic prescriptions, than comparator patients who do not have lung cancer.12 However, most patients who present with possible lung cancer symptoms do not receive imaging with chest X-ray (CXR) within 2 weeks as recommended by guidelines.13 Along with the summary recommendations presented in this Analysis article, a comprehensive version of this review is available (see Supplementary Information S1).
Promoting awareness of lung cancer symptoms and help seeking
Concern about not wishing to take up doctors’ time has been identified as a barrier to symptomatic presentation, particularly in the UK. Patients most affected by such concerns also estimate that they would wait longer before consulting with a GP (odds ratio of anticipated interval ≥4 weeks 2.50, 95% confidence interval [CI] = 2.27 to 2.75).14 Therefore, in addition to raising awareness of symptoms, efforts are required to address three aspects of behaviour change important to earlier presentation — motivation, capability, and opportunity.15 A validation–endorsement–motivation–action framework has been proposed to help patients who are deterred from presenting to health services by fear that medical professionals will dismiss their symptoms as not worthy of concern or investigation.16
Campaigns should feature the recognisable branding of the health service prominently to convey a sense of official endorsement and to assure patients that their health service encourages them to seek help for symptoms. Inclusive approaches are required to maximise reach to those with less fluency in English, who have lower levels of health literacy, and/or learning disabilities. Focusing on a positive message that early presentation improves the chances of effective treatment is important to confront nihilism about the disease and to motivate patients to present.
Messaging should also communicate that lung cancer also occurs in those who have never smoked, and thus encourage prompt presentation for persistent symptoms.17 To maintain their profile and public engagement, awareness campaigns should be funded as a long-term commitment, with continuous evaluation to enable content to be refined and updated. Campaigns have tended to focus on the most common symptoms (cough) and there is some evidence that this has achieved greater public recognition of this symptom. The expert group highlighted the need to strike a balance between communicating a clear message, for example, focusing on cough, while not neglecting the range of symptoms that could represent lung cancer. A campaign using a range of media, with links to further information or even the capacity to check cancer symptoms, for example, on the NHS app, may be warranted. Messaging should also address factors that influence the capability to seek help, such as how to request an assessment. It is important to recognise that many patients currently struggle to arrange general practice appointments. Similarly, the adoption of triaging systems and the increased use of remote consultations may add barriers to prompt diagnosis since face-to-face discussion and examination may now be less readily available in many practices.
Supporting early investigation for possible lung cancer in primary care
CXR in diagnosis of symptomatic lung cancer
Guidelines worldwide recommend CXR as the first-line test for the majority of lung cancer presentations.18 CXRs are less costly and can be reported in around a fifth of the time taken for a CT scan.19 Therefore, CXR is suitable as a high-volume investigation for large numbers of patients at relatively low levels of risk, with English GPs requesting 2.1 million CXRs in 2022–2023, compared with less than 100 000 abdominal or chest CTs.20 However, the sensitivity of CXR for lung cancer in patients who are symptomatic in primary care is approximately 80%. Therefore, both the use and interpretation of negative results requires consideration of both pre-test probability of lung cancer, persistence of symptomatology, and patient concern. For example, patients with haemoptysis have a risk of having lung cancer of 3% even after a negative CXR and may warrant CT as the first-line test.21
Research on rates of referral for cancer and disease outcomes in the UK has demonstrated a reduced risk of death from lung cancer in patients from GP practices in the highest tertile for referrals on the urgent suspected cancer pathway compared with the lowest tertile (hazard ratio 0.95, 95% CI = 0.94 to 0.97).22 Variation in the use of CXR may be a crucial underlying cause of differences in referral rates for suspected cancer. UK GPs appear to be less likely to arrange investigation for patients who are symptomatic, compared with their peers in similar health systems. A vignette study has suggested that UK GPs were less likely to arrange cancer investigation including CXR.23 There is scope for interventions targeted at primary care providers who use CXR infrequently to encourage greater testing of patients who are symptomatic.24 In particular, we recommended that existing information on annual numbers of CXRs performed per practice should be provided to practices alongside other cancer metrics, such as urgent suspected cancer referral rates.
Improving access through patient-initiated investigation and referral
Given the present difficulties patients face in accessing primary care in several healthcare systems, widespread adoption of services that enable patients to arrange investigation without a GP consultation warrants urgent consideration. Self-request CXR services have been established in English regions including Leeds and Greater Manchester. Patients who fulfil symptom criteria, based on those specified by the National Institute for Health and Care Excellence, are enabled to access radiology services directly, with the report of their X-ray sent to their GP. Patients are advised that, because the test does not identify all lung cancers and because symptoms may be caused by other serious problems, they should still consult with their GP if these persist. Patients also need to be informed that haemoptysis requires assessment regardless of CXR findings.21 The prevalence of lung cancer among patients investigated via self-referral and those who have CXR arranged in the conventional way by a GP are similar, and these services have been targeted effectively at deprived population groups, with higher levels of uptake from these communities.21 An alternative approach, currently under evaluation in some areas including Nottingham, is for patients with symptoms to access a lung cancer concern telephone hotline that applies a risk stratification algorithm to determine the imaging modality (either CXR or CT) along with ‘safety-netting’ advice for patients at lower risk who are investigated with CXR only.
Direct access to CT from general practice
Despite high degrees of concordance between guidelines worldwide, the extent to which GPs in different healthcare systems request CT is highly variable. The low levels of investigation with CT from general practice in the UK25 compared with other high-income countries may reflect greater capacity constraints in delivering CT and uncertainty about the indications.
Since 2022, policy in the English NHS has been that GPs should have direct access to urgent CT for circumstances in which lung cancer is suspected but criteria for urgent cancer referral are not met. Barriers to this include radiology capacity, implementing pathways locally, and supporting GPs with guidance on modality of imaging according to risk. Without appropriate and clear guidance on risk stratification, this could lead to a detrimental impact on CT capacity with a suboptimal return in earlier diagnosis. Thus, the primary imaging modality should be determined according to risk stratification based on basic patient characteristics, such as age, smoking status, and symptoms, along with modelling of likely impact on capacity to ensure that services are not overwhelmed. In the absence of such evidence, it remains very challenging for GPs who have access to direct-access CT to decide which patients should be investigated using direct-access CT, either as a first-line test instead of CXR or following an unremarkable CXR if symptoms persist. Formulation of clear guidance, informed by consideration of radiology capacity, should be prioritised to support GPs to decide when to use direct-access CT. Even where robust evidence is not currently available there may be a role for consensus-based guidelines formulated by a multidisciplinary group.
Mitigating risk of diagnostic delay resulting from lung cancer not identified on chest X-ray
Patients who are deemed to have a high pre-test probability of cancer or other serious illness warrant planned follow-up with a clinician who is aware of the degree of concern to ensure appropriate further management is in place, including immediate referral for CT. For those with lower risk, alternative strategies are required to ensure that the possibility of lung cancer is not overlooked. The development of automated trigger-based interventions or other monitoring systems should be prioritised to reduce risk of diagnostic delay resulting from failure to reassess patients who have symptoms that persist or evolve. Proactive safety-netting interventions, implemented at a system level rather than on individual clinician initiative, could have a role in encouraging patients to re-present if symptoms have persisted or worsened. Such measures could include an automated text message following an interval after CXR investigation prompting patients to re-consult if symptoms have persisted. It is also important that general practices have robust systems in place to ensure that planned repeat imaging or assessment occurs for those who have had abnormal imaging for which repeat imaging after an interval (for example, 6–8 weeks) is recommended.
Expediting lung cancer diagnosis in people who have never smoked
Worldwide over 10% of those diagnosed with lung cancer have never smoked,21 a proportion that will increase as smoking rates decline. People who have never smoked experience barriers to investigation. A qualitative study of the experiences of patients with lung cancer found that smoking history was instrumental to how individuals perceived and responded to early symptoms of lung cancer, resulting in risk of diagnostic delay for those without a smoking history.26 GPs should be encouraged to have a low threshold to use CXR in patients who have never smoked but who have persistent symptoms. Interventions to achieve this include improving GP awareness, which should include information regarding the extremely low radiation dose from a CXR.
Conclusion
Substantial progress has been made in lung cancer care over the last two decades, but much more improvement is required in achieving earlier diagnosis of symptomatic lung cancer. Early identification of lung cancer in people who are symptomatic before they become unfit for treatment is now even more important because of the marked improvement in outcomes seen with modern treatment. To achieve this, the working group recommends enhancing messaging to increase awareness of symptoms, and motivate and empower people to act should symptoms arise. Health systems need to provide improved access for those with symptoms along with timely imaging with an appropriate modality and safety-netting processes, so that those with normal initial imaging receive further assessment should symptoms persist or worsen. A summary of recommendations is presented in Box 1.
Promoting symptom awareness and help seeking
Campaigns should be endorsed by health authorities and assure patients that help seeking is encouraged. Messaging needs to be clear and designed to reach people who have never smoked, as well as those who have. Awareness campaigns should be funded as a long-term commitment, with continuous evaluation and regular refinement and updating of content.
Facilitating prompt investigation for possible symptoms of lung cancer
Provide accessible data on imaging utilisation (for example, annual number of CXRs requested by general practices) to identify services where patients may benefit from increased opportunistic investigation. Facilitate access to investigation through patient-initiated pathways (for example, self-request CXR or lung cancer hotline). Formulate clear guidance for GPs on when to use direct access to CT using expert consensus where robust evidence is not available.
Mitigating diagnostic delays in those at low risk and who have had negative initial investigation
Promote consideration of imaging for those at low risk (including never smokers) who have symptoms, particularly if persistent and/or patient/clinician concern. Provide guidance for GPs, through expert consensus if insufficient evidence available, on patients for whom CT should be considered following a negative CXR. Support GPs to provide specific ‘safety-netting’ advice to encourage patients to re-present if symptoms persist or evolve.
|
Box 1. Summary of recommendations
Notes
Funding
The Roy Castle Lung Cancer Foundation engaged a medical writer (Susan Mayor) to attend the meeting in order to record the presentations and discussion, and to prepare an initial draft summary of the meeting for further development by a group of meeting participants.
Provenance
Freely submitted; externally peer reviewed.
Competing interests
Jesme Fox is the Medical Director of the Roy Castle Lung Cancer Foundation. Samuel WD Merriel is Associate Editor and Editorial Board Member of the British Journal of General Practice. Michael Richards reports that he is Chair of the UK National Screening Committee and is an advisor to NHS England on diagnostic services. Nick Woznitza is a member of the College of Radiographers AI and Diagnostic Imaging advisory groups, the Royal College of Radiologists group for validation standards for thoracic AI, and College of Radiographers and Royal College of Radiologists Chest X-Ray reporting standards joint working group. The other authors have declared no competing interests.
- © British Journal of General Practice 2025
