Abstract

Lung cancer remains the leading cause of cancer-related mortality worldwide, yet it is now understood to be a complex and biologically diverse disease rather than one driven solely by tobacco use. Although smoking continues to account for a large proportion of cases, a significant and growing number of lung cancers occur in people who have never smoked, emphasising the importance of environmental exposures, inherited susceptibility, and distinct molecular mechanisms. This shift in understanding has reshaped both the clinical approach to lung cancer and the way risk factors are interpreted globally.

Advances in imaging, histopathology, and molecular diagnostics have transformed lung cancer management from a largely uniform, histology based model into one centred on precision oncology. Treatment decisions increasingly depend on tumour stage, histological subtype, genomic alterations, and immune features rather than smoking history alone. Targeted therapies have become central to the treatment of many never-smokers, whose tumours frequently contain actionable driver mutations such as EGFR and ALK, while immunotherapy has improved outcomes for selected patients, particularly those with smoking-related cancers and higher tumour mutational burden. In parallel, emerging technologies, including artificial intelligence (AI) assisted diagnostic tools and radiotherapy planning systems, offer new opportunities for earlier detection, improved accuracy, and more individualised treatment.

Despite these advances, major inequalities persist. Access to molecular testing, targeted therapies, immunotherapy, and AI-supported technologies remains largely confined to high-income healthcare systems. In many low- and middle-income regions, limited infrastructure and cost barriers prevent patients from benefitting from personalised treatment, even when their tumours carry actionable molecular features. Addressing lung cancer therefore requires not only continued scientific and technological progress, but also a focus on equitable access to diagnostics, treatment, and environmental risk reduction. Recognising the biological differences between smoking-related and never-smoker lung cancers is essential for improving outcomes and reducing global disparities in survival.

Introduction

Lung cancer remains the leading cause of cancer-related mortality worldwide, accounting for approximately 1.8 million deaths annually and representing nearly one in five cancer deaths globally (World Health Organization, 2023; Bade & Dela Cruz, 2020). Despite decades of research, public health intervention, and therapeutic innovation, survival outcomes remain poor in many regions, largely due to late-stage diagnosis and persistent exposure to carcinogenic risk factors reducing mortality rates (de Groot et al., 2018). Although historically conceptualised as a disease almost exclusively attributable to tobacco consumption, lung cancer is now understood to comprise a heterogeneous group of malignancies arising through diverse environmental, molecular, and genetic pathways (Herbst et al., 2018). Increasing recognition of lung cancer in never-smokers has challenged traditional assumptions about causation and has reframed the disease as being shaped by many biological influences instead of being explained by a single, uniform behaviour (Sun et al., 2007; Couraud et al., 2012).

The carcinogenic effects of tobacco smoke are among the most studied and clearly defined within oncological research. Cigarette smoke contains more than 60 established carcinogens, including polycyclic aromatic hydrocarbons (PAHs), tobacco-specific nitrosamines such as NNK, benzene derivatives, and formaldehyde  (Bade & Dela Cruz, 2020; Herbst et al., 2018). These compounds induce direct DNA damage through adduct formation, oxidative stress, and chronic inflammatory signalling. Repeated cycles of epithelial injury and regeneration create a permissive environment for the accumulation of somatic mutations (Herbst et al., 2018). Large-scale genomic analyses have identified characteristic mutational signatures associated with tobacco exposure, particularly C>A transversions resulting from PAH-induced DNA damage (Alexandrov et al., 2013). Tobacco-associated tumours frequently exhibit high tumour mutational burden (TMB), widespread chromosomal instability, and recurrent alterations in tumour suppressor genes such as TP53 (Govindan et al., 2012; Herbst et al., 2018). This mutational complexity contributes not only to oncogenesis but also to therapeutic behaviour, particularly in the context of immunogenicity.

Yet tobacco exposure alone does not account for global lung cancer incidence. Between 10 and 25 percent of lung cancer cases occur in individuals who have never smoked, and in some East Asian populations the proportion is substantially higher, particularly among women (Thun et al., 2008; Couraud et al., 2012). In absolute numbers, lung cancer in never-smokers would constitute one of the most common causes of cancer mortality worldwide if considered independently (Sun et al., 2007; Samet et al., 2009). This phenomena has prompted growing recognition that lung cancer in never-smokers is a distinct disease, rather than a residual subset of smoking-related disease (Govindan et al., 2012).

Environmental and occupational exposures beyond tobacco play a critical role in lung carcinogenesis. Radon gas, produced by the radioactive decay of uranium in soil and building materials, is recognised as the second leading cause of lung cancer in several Western countries (Samet et al., 2009; Bade & Dela Cruz, 2020). Chronic exposure to ambient fine particulate matter (PM2.5), especially in highly industrialised or rapidly urbanising regions, has also been associated with increased lung cancer risk, likely mediated through sustained oxidative stress and inflammatory signalling, even among never-smokers (Turner et al., 2011). Occupational carcinogens, including asbestos, silica dust, diesel exhaust, and certain heavy metals, contribute further to cumulative risk (Samet et al., 2009). In many low- and middle-income countries, indoor air pollution from biomass fuel combustion remains a significant yet under-recognised contributor, disproportionately affecting women in domestic settings (Couraud et al., 2012). These exposures highlight the multifactorial nature of lung carcinogenesis and the interaction between environmental context and biological susceptibility.

Inherited genetic susceptibility further modulates lung cancer risk. Genome-wide association studies have identified risk loci on chromosomes including 5p15 (containing TERT) and 15q25 (nicotinic acetylcholine receptor gene cluster). Some variants influence nicotine dependence and therefore indirectly affect exposure, whereas others appear to modify intrinsic cancer susceptibility independent of smoking behaviour (Govindan et al., 2012; Bade & Dela Cruz, 2020).  Molecular profiling has demonstrated that lung cancers in never-smokers are more likely to harbour targetable driver mutations such as EGFR, with marked variation in mutation prevalence across ethnic groups (Midha et al., 2015). The interaction between inherited predisposition, environmental carcinogens, and stochastic somatic mutations (random genetic changes that occur in the body over time) underscores the complexity of tumour initiation and progression (Vogelstein & Kinzler, 2013).

Global patterns of lung cancer incidence are evolving in response to shifting smoking behaviours, environmental change, demographic ageing, and improvements in data collection (Islami et al., 2015; WHO, 2023). In many Western countries, particularly the United States and parts of Northern Europe, age-standardised incidence rates among men have declined steadily over recent decades, reflecting reductions in smoking prevalence following sustained public health interventions (Islami et al., 2015). Female incidence rates, historically lower due to later uptake of smoking, have plateaued or begun to decline in some regions but remain elevated in others. In contrast, several East Asian countries continue to experience high incidence rates, influenced by persistent smoking prevalence among men, severe air pollution in urban centres, and regional genetic susceptibility patterns (Couraud et al., 2012). Notably, a significant proportion of female lung cancer patients in certain Asian populations report no history of tobacco exposure, prompting investigation into alternative risk factors such as indoor pollution, cooking oil fumes, and inherited predisposition (Couraud et al., 2012).

Histological distributions have also shifted over time. Adenocarcinoma has emerged as the dominant subtype globally, surpassing squamous cell carcinoma even in populations with historically heavy smoking exposure (Devesa et al., 2005; International Agency for Research on Cancer, 2020). Changes in cigarette composition, including filter ventilation and reduced tar yields, are hypothesised to have altered smoke deposition patterns within the lung, potentially contributing to this transition toward peripheral tumour development (Devesa et al., 2005). Small cell lung carcinoma, by contrast, remains strongly associated with heavy tobacco exposure and continues to carry a particularly poor prognosis (Herbst et al., 2018).

Despite substantial advances in molecular understanding and therapeutic innovation, survival disparities remain pronounced across geographic and socioeconomic boundaries (Ward et al., 2004). Five-year survival rates exceed 25% in some high-income settings (such as North America, Western Europe, and East Asia) but remain considerably lower in regions where access to early detection, molecular diagnostics, and advanced systemic therapies is limited (de Groot et al., 2018; Bade & Dela Cruz, 2020). As precision oncology advances, the difference between healthcare systems with genomic profiling and those without such setups and infrastructure widens the gap.

Lung cancer therefore represents not a single disease entity but a variety of biologically diverse malignancies shaped by environmental exposure, inherited susceptibility, and regional demographic. Understanding this complexity is critical not only for diagnosis but also for addressing global disparities in outcomes. This review will examine the diversity of lung cancer, exploring the impact of lifestyle and genetic factors on tumour biology, while discussing current and emerging therapeutic approaches and highlighting how these advances align with global health disparities.

Diagnosis and Treatment of Lung Cancer in Smokers and Never-Smokers

For decades, lung cancer management relied almost exclusively on histological classification – the process of classifying tumours based on their appearance under a microscope (Charipidou et al., 2024; Galli & Rossi, 2020). This era of broad generalisation has been replaced by a more nuanced, precision-based framework. Contemporary practice now integrates anatomical staging with a comprehensive analysis of the molecular and immunological profile of the disease (Herbst et al., 2018). While tobacco remains the dominant risk factor globally (Bade & Dela Cruz, 2020; de Groot, Munden & Carter, 2018), it is no longer the sole determinant for treatment. Instead, a patient’s smoking history acts as a biological indicator, often hinting at the tumour’s mutational burden and the likely presence of specific genetic drivers (Govindan et al., 2012).

Clinical presentation remains a challenge because symptoms of lung cancer in never-smokers are frequently non-specific. Persistent cough, dyspnoea and unexplained weight loss occur across all patient groups, regardless of their exposure to tobacco (Bade & Dela Cruz, 2020). Despite this, there is a clear divergence in the anatomical distribution where these tumours tend to grow. Smoking-associated malignancies often take root in the central bronchi. These include squamous cell carcinoma and small cell lung cancer (SCLC), which frequently cause airway obstructions that lead to an earlier diagnosis (Herbst et al., 2018). In contrast, lung cancer in never-smokers (LCINS) are predominantly adenocarcinomas and typically develop in the peripheral regions of the lung (Sun, Schiller & Gazdar, 2007). This statistic is increasingly linked to environmental triggers rather than combustion – ranging from radon and second-hand smoke to the long-term inhalation of fine particulate air pollution (Samet et al., 2009; Turner et al., 2011).

Accurate staging is the foundation of any successful management plan. Clinicians use contrast-enhanced CT and PET-CT scans to map the spread of the disease, while brain MRIs are essential in advanced cases where intracranial metastases are common (Herbst et al., 2018). This data allows the TNM classification system to define whether the clinical intent is curative or palliative. While the World Health Organisation (WHO) still categorises these tumours into non-small cell (NSCLC) and small-cell (SCLC) variants (International Agency for Research on Cancer, 2020), the modern diagnostic imperative is the acquisition of tissue for comprehensive genomic profiling.

Genomic data suggests that lung cancer in a smoker and a never-smoker are effectively two different diseases (Couraud et al., 2012). Smoker-associated tumours are typically “noisy”, showing high genomic instability and a heavy mutational load (Couraud et al., 2012). Never-smokers, by contrast, often present with “cleaner” profiles driven by specific oncogenic alterations. EGFR mutations and ALK rearrangements are significantly more common in this group, particularly among East Asian populations (Midha, Dearden & McCormack, 2015; Sun, Schiller & Gazdar, 2007). As these mutations are highly actionable, universal molecular testing is now the gold standard for all advanced NSCLC cases, ensuring that no patient is overlooked based on their smoking status alone (Herbst et al., 2018).

In the surgical setting (Stages I-II), resection with systematic lymph node dissection remains the best chance for a cure. Treatment strategies have moved beyond simple chemotherapy to include adjuvant immunotherapy and targeted EGFR inhibitors for high-risk patients (Herbst et al., 2018). For unresectable Stage III disease, the current benchmark is concurrent chemoradiotherapy followed by consolidation immunotherapy, a regimen that has significantly improved overall survival.

Therapeutic strategies in the metastatic setting are now entirely biomarker-driven. Where no driver mutation exists, immune checkpoint inhibitors (ICIs) are the frontline choice. Interestingly, the high mutational load found in smokers can actually make these tumours more “visible” to the immune system, often resulting in a stronger response to immunotherapy (Govindan et al., 2012). If a specific alteration is present, however, tyrosine kinase inhibitors (TKIs) are preferred. While these drugs often yield dramatic results, the tumour will eventually adapt and develop resistance, requiring repeat biopsies and a shift in the genomic strategy (Govindan et al., 2012). SCLC remains the most difficult variant to treat; it is almost always linked to tobacco (Thun et al., 2008) and, despite some progress with new immunotherapy combinations, the prognosis remains comparatively poor.

Ultimately, the disparity between smokers and never-smokers reflects a fundamental biological divergence. Recognising these complexities is critical to advancing patient-specific treatment plans and tackling the global inequalities that limit access to molecular diagnostics underpinning therapeutic success.

Artificial Intelligence in Lung Oncology

Artificial intelligence (AI) is increasingly being used in oncology for pattern detection, treatment response prediction, and histological imaging assistance rather than manual review and traditional interpretation due to its higher efficiency. For instance, a diagnostic test created by Shin et al. (2023) utilised AI to detect cancerous cells by isolating exosomes from the patient’s plasma via a liquid biopsy, then placing the suspension on a surface-enhanced Raman spectroscopy (SERS) surface. The system amplifies these signals and scans multiple spots before AI is used to analyse the SERS patterns to identify cancer scores. This was done through the following: comparing signals, finding the tissue of origin, and using a heat map to deduce whether the patient has cancer and identify the organ where the cancer originated from. By using artificial intelligence in assisting with Raman spectroscopy, the findings demonstrated roughly 90% sensitivity and 94% specificity (Shin et al., 2023), which highlights how utilising AI in molecular analysis results in better detection of cancer at an earlier stage. As well as this, obtaining a liquid biopsy in the early stages of this test could be viewed as more beneficial for lower-income countries due to the fewer resources required. However, weak SERS signals delay image clarification, leading to delayed treatments and backlogs, and costly AI-assisted imaging biopsies limit their global availability, especially in developing countries with fewer resources for complex imaging techniques. Overall, implementing AI within tissue diagnosis and molecular analysis in lung oncology results in a smoother and more rapid progression from detection to recovery. However, considering the limitations of integrating this method of cancer diagnosis is pivotal due to its intricacy and expenses for system development and maintenance, especially in nations with limited financial capabilities that severely impact their healthcare system.

Stereotactic body radiotherapy (SBRT) is the standard therapy for stage 1 non-small cell lung cancer (Kubo et al., 2024), which is localised and usually takes place in 1-4 sessions when surgery cannot be performed safely. SBRT initially starts with CT imaging so radiographers can formulate a treatment plan before inserting skin markings that are used as reference points for the treatment. In the session, the patient is inserted into a radiotherapy couch and a linear accelerator or CyberKnife System uses radiation to kill the cancerous cells; it can run from 15 minutes to ~2 hours. Some advantages of utilising this method of radiotherapy include the short hospitalisation period after treatment, which is beneficial for hospitals that do not have high bedspace availability, as well as its non-invasive nature, reducing the use of anaesthetics that would be used in pulmonary resections. However, the side effects resulting from this treatment can be severe and potentially damage organs near the lungs, which can have a more deleterious impact in a lower-income country as the additional complications from this treatment further strain a depleted healthcare system to carry out radiotherapy. Since this method of external beam radiation therapy requires high levels of precision and accuracy, Al could assist with SBRT by reducing the time spent on decision-making in a planner’s treatment system (TPS) (Wang et al., 2019). This could be done through controlling parts of the imaging guidance, mapping the coordinates of the tumour through the body’s movements, and generating high-resolution imaging within the 4D-CT scans. Through AI managing parts of the radiotherapy, oncologists and radiographers can focus more on drug dosage and optimisation that can be carefully managed through knowledge-based treatment planning and deep learning techniques (Kawamura et al., 2024). Despite having a higher efficiency and minimising the time spent planning radiotherapy treatments, these models need to be reviewed regularly by healthcare professionals to ensure that they are meeting standards regarding confidentiality, performance, and security. 

A recent clinical trial was carried out on patients with EGFR mutations in advanced NSCLC, where they were given savolitinib on top of their baseline treatment: osimertinib with varying dosages (De Marinis et al., 2025). This was because many people who were on osimertinib used to treat EGFR-mutated NSCLC developed resistance to the drug, and savolitinib inhibits the MET changes, which reduces the gene amplification from occurring. Resulting from this trial, more than half of the patients displayed tumour shrinkage with an average response of 7.1 months (De Marinis et al., 2025), suggesting potential longevity in patients due to a median progression-free survival of 7.4 months, resulting in an improvement in the quality-adjusted life years in patients. On the other hand, the data indicated that 46% of patients experienced swelling, and around 40% had nausea, which could be a bigger concern for a country with an ageing population as the impacts of these side effects would be more detrimental to them. A key reason behind the conduction of this trial was to display how precision oncology could be used more in clinical settings, which focuses on choosing treatments based on a patient’s genetic profile of cancer with individualised care, shifting the focus away from a one-size-fits-all model. In the growing prevalence of AI, Mohammed et al. (2025) highlight the benefits of AI within forecasting drug efficacy and analysing the pharmacodynamics of chemotherapy drugs, resulting in a better quality of drug interactions on a molecular level. Despite AI exhibiting superior standards compared to manual reviews and diagnoses, progression into advanced testing and planning would be slower in developing countries due to high costs. Another potential limitation of utilising AI within pharmacokinetic and pharmacodynamic interactions in oncology would be the regional financial disparities within countries that rely on a private healthcare system. This is due to pharmaceutical corporations having greater control of their marketing, which could lead to false advertisements or drug shortages, such as the ones during the coronavirus pandemic (Socal & Sharfstein, 2024). In summary, the findings demonstrate that the combined effects of both savolitinib and osimertinib could improve patient outcomes regarding chemotherapy in NSCLC with better results when synthesing molecular and respiratory histological analysis with AI. It’s important to consider that long-term studies will be essential in evaluating and adapting these advanced models into countries with varying healthcare demands and financial structures to ensure that drug distribution across countries remains fairly proportional. 

Akhtar & Bansal (2017) note that with lung cancer being one of the most prominent forms of cancer, utilising AI to assist in diagnosis, treatment, and rehabilitation is essential in mitigating the genetic and environmental effects of this form of this condition. The clinical significance of this study lies in the fundamental dynamics of global diversity, such as age, gender, ethnicity, and socioeconomic differences, that could be explored further when analysing the varying differences within non-smokers with lung cancer, thus reducing any underlying healthcare bias and formulating new methods of modifying current treatments aided with AI exploration.

Variation in Treatment of Lung Cancer in Non-Smokers

Lung cancer in non-smokers is distinguished by the biological features that differentiate it from lung cancer in smokers, creating new treatment requirements that diverge from that used by smokers. Unlike tumours caused by tobacco exposure, which typically carry a high number of mutations, cancer in non-smokers are often driven by genetic alterations and environmental exposures that shape the behaviour of the disease as well as the effectiveness and availability of certain therapies. As a result, the factors that influence treatment outcomes in non-smokers are closely related to the underlying molecular characteristics of the tumours, including the environmental circumstances, such as exposure to radon gas and secondhand smoke, which help the cancers develop. Understanding these influences is essential to reducing the likelihood of lung cancer in non-smokers because the global variation in mutation patterns and environmental risks means that treatment strategies for non-smokers differ widely across regions. 

One of the largest driving factors affecting treatment in non-smokers is the presence of driver mutations, which are genetic alterations that directly impact tumour growth. Mutations in genes like EGFR, ALK, and ROS1 are far more common in non-smokers than in smokers (Curley, 2024; Galzerano, 2026). These mutations are not necessarily random, and they ultimately determine how the cancer behaves, which determines what treatments will be most effective. For instance, EGFR mutations, which are present in a large population of non-smoker lung cancer cases, create tumours that are responsive to EGFR targeted therapies, which are designed to block the abnormal signalling caused by the mutation. In contrast, the people who have cancer and are smokers, often respond better to chemotherapy or immunotherapy because they carry more mutations that the immune system is able to recognise. The dependence on targeted therapy in non-smokers highlights how genetic factors directly shape treatment decisions. 

Despite this, even when targeted therapies are effective initially, they still have their respective challenges. Multiple sources note that tumours containing an EGFR mutation often develop drug resistance over an extended period of time, indicating that the cancer adapts and begins to grow despite treatment. This resistance hinders the long term success of targeted therapies and forces doctors to shift strategies, and in some cases, turn to newer generations of targeted drugs or combination treatments. The development of resistance is a major factor that influences the treatment outcomes for non-smokers, which emphasises the urgency for personalised treatment adjustments. As cancers are driven by specific mutations rather than widespread DNA damage, the treatment landscape is more dynamic and requires continual adaptation (Hrustanovic et al, 2013.)

Another major factor influencing the effectiveness of treatment in non-smokers is the low mutation burden, which is a tumour with a low number of genetic mutations. According to the Lung Cancer Foundation of America and the National Cancer Institute, non-smoker tumours tend to have far fewer genetic abnormalities than those caused by smoking. This has a direct impact on the success of immunotherapy, a treatment that relies on the immune system recognising mutated cancer cells as foreign. With these fewer mutations, the tumours of non-smokers are less visible to the immune system, hence, reducing the effectiveness rate of immunotherapy. The low mutation burden is therefore a critical factor that shapes which therapies are viable and which are unlikely to succeed (Murphy et al., 2025).

Aside from genetics, environmental experiences also play a significant role in changing the types of lung cancer that develop in non-smokers, which in turn affects treatment. Across multiple sources, radon exposure is consistently identified as one of the leading causes of lung cancer in individuals who do not smoke. Radon levels may vary widely based on the region, meaning that the environmental risk is not standardised. In areas with a high radon exposure, such as certain regions in the United States and Europe, non-smoker lung cancers may arise from different pathways than those in regions where radon exposure is low. Similarly, air pollution and exposure to secondhand smoke are also major contributors to non-smoker lung cancer, especially those in densely populated regions. These environmental factors influence the mutation patterns found in tumours, which then determine which treatments are most effective. Environmental exposures are not only risk factors, but also indirect determinants of treatment strategy (Urrutia-Pereira et al., 2023).

The global variation in mutation prevalence is one of the clearest examples of how lung cancer treatment for non-smokers changes around the world. EGFR mutations are significantly more common in East Asia. The geographical difference between the mutation rates of East Asia to the rest of the world has major implications for treatment. In East Asia, EGFR targeted therapies are a central component of lung cancer care for non-smokers because the mutation is so prevalent. In contrast, regions where EGFR mutations are less common, the treatment strategies may depend more on other forms of targeted therapy. Environmental exposures also vary globally, further shaping variants of cancers that can develop. For example, countries with a high level of air pollution may experience higher rates of lung cancer in non-smokers, influencing public health strategies, leading to new treatments. These global disparities demonstrate that lung cancer treatment is not uniform and changes based on location. Treatment must be adapted to the genetic and environmental context of each region (Melosky et al., 2021; Abu-Hijeh, 2018; NCI, 2021).

Another factor that may contribute to global variation in treatment is the availability of genetic testing. Since treatment for non-smokers depends heavily on identifying specific mutations, regions with limited access to molecular testing may struggle to provide the most effective therapies. In certain areas, patients may receive treatments that are used for the majority of patients, such as chemotherapy. However, every person is different, and not all the generalised treatment plans will be as effective, meaning that targeted therapies could be more effective. This disparity showcases how healthcare across the world influences treatment outcomes and contributes to the global differences in lung cancer care (Salloum et al., 2018; Dubin & Griffin, 2020).

Treating lung cancer in non-smokers raises unique challenges that stem from many biological and global factors such as radon gas and random mutations. Targeted therapies, while often effective at first, frequently lose their effectiveness as the tumours develop resistance. Immunotherapy, which has transformed treatments for many smokers, is less effective in non-smokers due to their tumours’ low mutation burden. Effective treatment therefore requires precise genetic testing to identify the mutations driving each patient’s cancer, but access to such testing varies widely around the world. These challenges highlight the need for personalised treatment strategies and demonstrate the importance of understanding the distinct nature of lung cancer in non-smokers. 

Discussion

Our article investigated the disparities within diagnosis and treatments for lung cancer in non-smokers through the exploration and analysis of genetic and environmental influences on non-smokers with lung cancer worldwide. Within our research, many of our articles highlight the presentation of lung cancer being more common in women and East Asian populations, as well as radon gas and carcinogenic exposure, air pollution and gene mutations being key factors in increasing the likelihood of lung cancer presenting in non-smokers. These findings could be explained by the presence of carcinogens within tobacco, radiation and atmospheric pollutants, mutating DNA within lung cells to cause uncontrolled cell division.

These findings are consistent with earlier research by Dubin and Griffin (2020), who highlight that more EGFR mutations are present in non-smokers, as well as more females presenting as non-smokers with lung cancer. Despite advancements in imaging and detection, a key limitation of these studies was that underlying bias remained an issue when analysing current treatments and experimenting with potential testing in underrepresented groups, which would greatly affect general procedures for dealing with lung cancer. Since it is not representative of the whole population, it could be a bigger issue with non-smokers with this diagnosis, since the most affected demographics are women and East Asians. Another limitation of this study was the confounding variables affecting the diagnosis of lung cancer within varying populations, as diverse global physical environments could merge or segregate common factors, therefore implicitly influencing the results without being directly accounted for. Therefore, additional research is needed to examine molecular diversity within highly affected groups, evaluate AI within lung cancer diagnosis and treatment for wider global use, and investigate aspects that influence a lung cancer diagnosis to see which has a bigger influence.

Conclusion

Lung cancer continues to be the leading cause of cancer-related death worldwide, but it is now clear that it is not a single disease with a single pathway. The differences between smoking-related tumours and those arising in never-smokers reflect distinct molecular drivers, environmental influences, and patterns of immune interaction. These biological differences have real clinical consequences. Modern management increasingly depends on tumour stage, histology, genomic alterations, and immune markers rather than smoking history alone. Targeted therapies, immune checkpoint inhibitors, and advances in radiotherapy have significantly improved outcomes for selected patients, particularly where comprehensive molecular testing is available.

At the same time, progress has been uneven. Access to next-generation sequencing, immunotherapy, and emerging technologies such as AI-supported diagnostics remains concentrated in high-income healthcare systems. In many parts of the world, limited infrastructure and cost barriers mean that patients may not receive molecular testing or personalised therapies, even when their tumours harbour actionable mutations. As a result, survival disparities persist despite major scientific advances. In addition, access to low-dose CT screening remains highly unequal across and within countries, with many low- and middle-income regions lacking organised screening programmes. As lung cancer survival is strongly linked to stage at detection, later-stage diagnosis in under-resourced settings further widens global outcome disparities. The recognition of lung cancer in never-smokers as a distinct clinical entity reinforces the importance of universal molecular profiling in non-small cell lung cancer, regardless of tobacco exposure. It also highlights the need for continued research into mechanisms of drug resistance and improved biomarkers for predicting immunotherapy response, particularly in tumours with low mutational burden. Alongside therapeutic innovation, investment in equitable screening programmes, environmental risk reduction, and scalable diagnostic technologies is essential.

Fundamentally, advances in lung cancer control do not solely depend on developing more effective treatments. In lieu, ameliorating the accessibility of these treatments across wider healthcare settings needs to be emphasised. The right balance between advances in precision oncology, as well as fair and equal access to them, are imperative for a global reduction in mortality.

Abu-Hijeh, M. (2015) ‘Why do nonsmokers get lung cancer?’, UT Southwestern Medical Center [online]. <https://utswmed.org/medblog/nonsmoker-lung-cancer/>

Akhtar, N. and Bansal, J.G. (2017) ‘Risk factors of Lung Cancer in nonsmoker’, Current Problems in Cancer, 41(5), pp. 328-339.

Alexandrov, L.B., Nik-Zainal, S., Wedge, D.C., Aparicio, S.A.J.R., Behjati, S., Biankin, A.V., Bignell, G.R., Bolli, N., Borg, A., Børresen-Dale, A.L. and Boyault, S. (2013) ‘Signatures of mutational processes in human cancer’, Nature, 500(7463), pp. 415–421.

Bade, B.C. and Dela Cruz, C.S. (2020) ‘Lung cancer 2020: epidemiology, etiology, and prevention’, Clinics in Chest Medicine, 41(1), pp. 1–24.

Charpidou, A., Harvadelle, G., Boutsikou, E., Penagiotou, E., Simsek, G. Ö., Verbeke, K., Xhemalaj, D. and Domagał-Kulawik, J. (2024) ‘Unravelling the diagnostic pathology and molecular biomarkers in lung cancer’, Breathe, 20(2), p. 230192.

Couraud, S., Zalcman, G., Milleron, B., Morin, F. and Souquet, P.J. (2012) ‘Lung cancer in never smokers – a review’, European Journal of Cancer, 48(9), pp. 1299–1311. 

Curley, B. (2024) ‘Why many people with lung cancer who have never smoked don’t respond well to treatments’, Lung Cancer Foundation of America [online]. <https://lcfamerica.org/research-news/why-many-people-with-lung-cancer-who-have-never-smoked-dont-respond-well-to-treatments/>

de Groot, P., Munden, R.F. and Carter, B.W. (2018) ‘The epidemiology of lung cancer’, Translational Lung Cancer Research, 7(3), pp. 220–233. 

de Marinis, F., Kim, T., Bonanno, L., Cheng, S., Kim, S., Tiseo, M., Chu, Q., Proto, C., Sacher, A., Luo, Y., Novello, S., Hao, D., Baik, C., Bazhenova, L., Lee, J., Cho, B., Cadranel, J., Diep, T., Metro, G., . . . Ahn, M. (2025). ‘Savolitinib plus osimertinib in epidermal growth factor receptor (EGFR)-mutated advanced non-small cell lung cancer with MET overexpression and/or amplification following disease progression on osimertinib: Primary results from the phase II SAVANNAH study’, Annals of Oncology, 36(8), pp. 920-933. 

Devesa, S.S., Bray, F., Vizcaino, A.P. and Parkin, D.M. (2005) ‘International lung cancer trends by histologic type: male: female differences diminishing and adenocarcinoma rates rising’, International Journal of Cancer, 117(2), pp. 294–299. 

Dubin, S. and Griffin, D. (2020) ‘Lung Cancer in Non-Smokers’, Missouri Medicine, 117(4), p. 375.

Galli, G. and Rossi, G. (2020) ‘Lung cancer histology-driven strategic therapeutic approaches’, Shanghai Chest, 4, p. 29.

Galzareno, S. (2023) ‘How Lung Cancer Can Affect People Who Have Never Smoked’, Jefferson Health [online]. <https://www.jeffersonhealth.org/your-health/living-well/how-lung-cancer-can-affect-people-who-have-never-smoked>

Govindan, R., Ding, L., Griffith, M., Subramanian, J., Dees, N.D., Kanchi, K.L., Maher, C.A., Fulton, R., Fulton, L., Wallis, J. and Chen, K. (2012) ‘Genomic landscape of non-small cell lung cancer in smokers and never-smokers’, Cell, 150(6), pp. 1121–1134.

Herbst, R.S., Morgensztern, D. and Boshoff, C. (2018) ‘The biology and management of non-small cell lung cancer’, Nature, 553(7689), pp. 446–454.

Hrustanovic, G., Lee, B.J. and Bivona, T.G. (2013). ‘Mechanisms of resistance to EGFR targeted therapies’, Cancer Biology & Therapy, 14(4), pp. 304–314.

International Agency for Research on Cancer (2020) WHO Classification of Tumours: Thoracic Tumours. 5th edn. Lyon: IARC Press. Islami, F., Torre, L.A. and Jemal, A. (2015) ‘Global trends of lung cancer mortality and smoking prevalence’, Translational Lung Cancer Research, 4(4), pp. 327–338. 

Kawamura, M., Kamomae, T., Yanagawa, M., Kamagata, K., Fujita, S., Ueda, D., Matsui, Y., Fushimi, Y., Fujioka, T., Nozaki, T., Yamada, A., Hirata, K., Ito, R., Fujima, N., Tatsugami, F., Nakaura, T., Tsuboyama, T. and Naganawa, S. (2024) ‘Revolutionizing radiation therapy: The role of AI in clinical practice’, Journal of Radiation Research, 65(1), pp. 1-9.

Kubo, N., Suefuji, H., Nakajima, M., Tokumaru, S., Okano, N., Yoshida, D., Suzuki, O., Ishikawa, H., Satouchi, M., Nakayama, H., Shimizu, K. and Shioyama, Y. (2024) ‘Five-Year Survival Outcomes After Carbon-Ion Radiotherapy for Operable Stage I NSCLC: A Japanese National Registry Study (J-CROS-LUNG)’, Journal of Thoracic Oncology, 19(3), pp. 491-499. 

Lung Cancer Group (2026) ‘Lung Cancer in Nonsmokers’, Lung Cancer Group [online]. <https://www.lungcancergroup.com/lung-cancer/non-smokers-and-lung-cancer/>

Melosky, B., Kambartel, K., Häntschel, M., Bennetts, M., Nickens, D.J., Brinkmann, J., Kayser, A., Moran, M. and Cappuzzo, F. (2021). ‘Worldwide Prevalence of Epidermal Growth Factor Receptor Mutations in Non-Small Cell Lung Cancer: A Meta-Analysis’, Molecular Diagnosis & Therapy, 26(1), pp. 7–18.

Midha, A., Dearden, S. and McCormack, R. (2015) ‘EGFR mutation incidence in non-small-cell lung cancer of adenocarcinoma histology: a systematic review and global map by ethnicity’, American Journal of Cancer Research, 5(9), pp. 2892–2911.

Mohammed, A. H., Hallit, S., Malaeb, D. and Hosseini, H. (2025) ‘Exploring the role of artificial intelligence in chemotherapy development, cancer diagnosis, and treatment: Present achievements and future outlook’, Frontiers in Oncology, 15, p. 1475893.

Murphy, C., Pandya, T., Swanton, C. and Solomon, B.J. (2025). ‘Lung Cancer in Nonsmoking Individuals’, JAMA, 334(20), pp. 1836-1845.

NIH (2021) ‘NIH study illuminates origins of lung cancer in never smokers’, National Institutes of Health [online]. <https://www.nih.gov/news-events/news-releases/nih-study-illuminates-origins-lung-cancer-never-smokers>

Salloum, N.C., Buchalter, E.L.F., Chanani, S., Espejo, G., Ismail, M.S., Laine, R.O., Nageeb, M., Srivastava, A.B., Trapp, N., Trillo, L., Vance, E., Wenzinger, M., Hartz, S.M., David, S.P. and Chen, L.-S. (2018) ‘From genes to treatments: a systematic review of the pharmacogenetics in smoking cessation’, Pharmacogenomics, 19(10), pp. 861–871.

Samet, J.M., Avila-Tang, E., Boffetta, P., Hannan, L.M., Olivo-Marston, S., Thun, M.J. and Rudin, C.M. (2009) ‘Lung cancer in never smokers: clinical epidemiology and environmental risk factors’, Clinical Cancer Research, 15(18), pp. 5626–5645. 

Shin, H., Choi, B.H., Shim, O., Kim, J., Park, Y., Cho, S. K., Kim, H. K., & Choi, Y. (2023) ‘Single test-based diagnosis of multiple cancer types using Exosome-SERS-AI for early-stage cancers’, Nature Communications, 14(1), p. 1644. 

Socal, M.P.&  Sharfstein, J.M. (2024) ‘Drug Shortages—A Study in Complexity’, JAMA Netw Open, 7(4), e244168.

Sun, S., Schiller, J.H. and Gazdar, A.F. (2007) ‘Lung cancer in never smokers — a different disease’, Nature Reviews Cancer, 7(10), pp. 778–790. 

Thun, M.J., Hannan, L.M., Adams-Campbell, L.L., Boffetta, P., Buring, J.E., Feskanich, D., Flanders, W.D., Jee, S.H., Katanoda, K., Kolonel, L.N. and Lee, I.M. (2008) ‘Lung cancer occurrence in never-smokers: an analysis of 13 cohorts and 22 cancer registry studies’, Journal of the National Cancer Institute, 100(10), pp. 691–701. 

Turner, M.C., Krewski, D., Pope, C.A. III, Chen, Y., Gapstur, S.M. and Thun, M.J. (2011) ‘Long-term ambient fine particulate matter air pollution and lung cancer in a large cohort of never-smokers’, American Journal of Respiratory and Critical Care Medicine, 184(12), pp. 1374–1381.

UChicago Medicine (n.d.) ‘Targeted Therapy for Lung Cancer’, UChicago Medicine [online]. <https://www.uchicagomedicine.org/cancer/types-treatments/lung-cancer/targeted-therapy>

Urrutia-Pereira, M., Chatkin, J.M., Chong-Neto, H.J. and Solé, D. (2023) ‘Radon exposure: a major cause of lung cancer in nonsmokers’, Jornal Brasileiro De Pneumologia, 49(6), e20230210.

Wang, C., Zhu, X., Hong, J. C. and Zheng, D. (2019) ‘Artificial Intelligence in Radiotherapy Treatment Planning: Present and Future’, Technology in Cancer Research & Treatment, 18.

Ward, E., Jemal, A., Cokkinides, V., Singh, G.K., Cardinez, C., Ghafoor, A. and Thun,M., 2004. Cancer disparities by race/ethnicity and socioeconomic status.    CA: A Cancer Journal for Clinicians, 54(2), pp.78–93.

World Health Organization (2023) (WHO) Global Cancer Observatory: Cancer Today. Lyon: International Agency for Research on Cancer.

Vogelstein, B. and Kinzler, K.W., 2013. Cancer genes and the pathways they control. Nature Medicine, 10(8), pp. 789–799.