Lung cancer across borders: From molecular epidemiology to precision treatment strategies

Abstract

Lung cancer remains the leading cause of cancer-related mortality worldwide, accounting for approximately 1.8 million deaths annually. Beyond this universal burden, lung cancer is profoundly heterogeneous, shaped by geography, environmental exposures, genetic ancestry, and unequal access to molecular diagnostics and modern therapies. Compelling evidence indicates that non-small cell lung cancer (NSCLC) is not a single entity but a constellation of molecularly distinct subtypes whose distribution reflects regional carcinogens and population-specific susceptibilities. This Special Collection of Therapeutic Advances in Medical Oncology assembles five complementary studies that address the geographical heterogeneity of NSCLC. A comprehensive review by Laguna and colleagues maps the worldwide distribution of risk factors and molecular subtypes, illustrating how tobacco, indoor radon, air pollution, arsenic, biomass smoke, and genetic ancestry converge to produce distinct molecular landscapes across continents. García-Pardo and colleagues extend this framework with the RADON EUROPE study, the first ecological analysis to link estimated indoor radon exposure with ALK fusion prevalence across 21 European countries. Marjanski and colleagues expose critical gaps in real-world perioperative assessment in Poland, where limited preoperative biomarker testing and prolonged surgical intervals may exclude patients from neoadjuvant chemoimmunotherapy. Mo and colleagues describe an emerging consequence of therapeutic progress: a temporal rise in cardiovascular and pulmonary diseasespecific mortality in NSCLC, coinciding with the expanded use of targeted therapies and immune checkpoint inhibitors. Pham and colleagues demonstrate, in a Vietnamese cohort with EGFR-mutant NSCLC, that flexible, individualized afatinib dosing improves outcomes and underscores the persistent underrepresentation of non-European populations in pivotal trials. Taken together, these articles illustrate that the geographic distribution of carcinogens shapes molecular subtypes; molecular subtypes dictate treatment options; treatment options generate toxicities that must be managed within local healthcare systems; and those systems, in turn, determine equitable access to diagnostic and therapeutic innovation. Bridging these gaps will require integration of exposome-informed approaches, more diverse clinical trial populations, cardio-oncology surveillance, and pharmacological individualization. This editorial advocates for a geographically informed and globally equitable approach to thoracic oncology.

Plain Language Summary

Why was this editorial written?

Lung cancer is the leading cause of cancer death in the world, but it is not the same disease everywhere. The risk factors that cause lung cancer, the genetic changes inside tumors, and the treatments that patients receive all depend on where they live. Therapeutic Advances in Medical Oncology invited us to introduce a special collection of five studies that explore these differences and explain why they matter for patients, doctors, and health systems around the world.

What does the collection cover?

The five studies look at lung cancer from a global perspective. The first reviews how risk factors and tumor types differ across continents, including tobacco, indoor radon gas, air pollution, arsenic in drinking water, and smoke from cooking fires. The second examines indoor radon exposure across 21 European countries. The third describes how lung cancer surgery is planned in Poland. The fourth analyses how newer cancer drugs affect heart and lung health in the United States. The fifth studies a targeted lung cancer drug called afatinib in patients in Vietnam.

What did the studies show?

Together, the studies show that lung cancer is shaped by the environment, by people's genetic backgrounds, and by the resources of their local health systems. Indoor radon at home may be linked to certain genetic changes in lung tumors. Important tests are often missed before lung cancer surgery. Newer cancer drugs can cause heart and lung side effects that need closer monitoring. Lower, individualised doses of a targeted drug improved outcomes in some patients.

What do these findings mean?

Lung cancer care should be tailored to where patients live and to who they are. A globally informed approach is needed so that all patients have the best possible chance of long-term survival.

Lung cancer remains the leading cause of cancer-related mortality worldwide, responsible for approximately 1.8 million deaths annually. ¹ However, lung cancer is remarkably heterogeneous, shaped by geography, genetics, environmental exposures, and the therapeutic landscape available to each patient. The recognition that lung cancer is not a single disease but rather a constellation of molecularly distinct subtypes, ² each influenced by regional carcinogens and population-specific susceptibilities, demands a shift toward more context-aware and globally informed approaches. This Special Collection brings together five studies that collectively address the geographical heterogeneity of non-small cell lung cancer (NSCLC), including the molecular epidemiology of carcinogen-driven alterations and the real-world challenges of perioperative assessment, treatment-related toxicities, and dose optimization in resource-diverse settings.

Laguna and colleagues ³ provide the foundation for this collection with a comprehensive review charting the worldwide distribution of lung cancer risk factors, genetic predisposition, and molecular subtypes across continents. The authors demonstrate that while tobacco consumption accounts for the majority of cases of lung cancer in Western societies, other carcinogens such as indoor radon, air pollution, arsenic in drinking water, occupational exposures, and biomass combustion, may play pivotal roles in specific regions. In Latin America, arsenic contamination in the Antofagasta region of Chile is associated with squamous cell histology, while wood smoke exposure in Mexico correlates with high EGFR mutation rates. In Asia, approximately one-third of lung cancer patients have never smoked, and the prevalence of EGFR mutations is significantly higher compared to Caucasian populations. The review further highlights how genetic ancestry shapes somatic mutation profiles: Native American ancestry in Latin American populations is positively correlated with EGFR mutations, mirroring patterns observed in Asian populations. Taken together, this work highlights that the molecular profile of NSCLC is inseparable from the environment in which it develops.

Building on this carcinogen-molecular interface, García-Pardo and colleagues ⁴ present the RADON EUROPE study, the first ecological analysis to systematically correlate estimated indoor radon concentrations with the frequency of EGFR mutations and ALK fusions across 21 European countries. Drawing on radon mapping data from the European Commission and published molecular profiling studies encompassing over 150,000 NSCLC cases, they identified a statistically significant positive correlation between ALK fusion frequency and the proportion of dwellings exceeding 400 Bq/m³ (r = 0.72, p = 0.001); no such correlation was observed for EGFR mutations at the country level. While the ecological design precludes causal inference, these findings are consistent with prior observations in Spanish and French cohorts linking radon exposure to oncogene-driven NSCLC,^5,6 and with mechanistic evidence that radon-induced alpha-particle radiation causes DNA double-strand breaks and chromosomal rearrangements ⁷ : genomic events that may contribute to oncogenic rearrangements such as ALK fusions. These findings lend support to the biological plausibility of an exposure-related contribution to molecular patterns in lung carcinogenesis. This study advances the hypothesis that environmental exposures may act not merely as initiators of carcinogenesis but as selective promoters of specific molecular subtypes, potentially through mechanisms complementary to those described for air pollution. ⁸ The ongoing EORTC 1920 BIORADON prospective study is expected to provide patient-level data to further interrogate this association and to advance our understanding of the underlying biological mechanisms from a clinical perspective. ⁹

As heterogeneous as the molecular characteristics and environmental risk factors of the disease are the clinical pathways and standard of care practice for early-stage NSCLC across different healthcare systems. Marjanski and colleagues ¹⁰ explored this dimension through a multicenter observational survey of 459 surgically treated NSCLC patients across seven Polish thoracic surgery centers, revealing significant gaps between recommended perioperative assessment standards and real-world practice. Brain imaging was performed in a negligible proportion of patients across all stages, including stage III (where approximately 10% of patients harbor occult brain metastases). The median interval from CT to surgery was 61 days, more than double the 30-day benchmark recommended by ESMO guidelines. Preoperative PD-L1 and next-generation sequencing (NGS) assessments were virtually absent, with molecular profiling conducted almost exclusively on postoperative specimens. These findings carry important implications in an era where neoadjuvant chemoimmunotherapy has demonstrated significant improvements in pathological complete response rates and overall survival.^11,12 Without preoperative biomarker assessment, patients who would benefit from neoadjuvant strategies are excluded from optimal treatment. This gap highlights a critical disconnect between rapidly evolving evidence and its real-world implementation and serves as a reminder that molecular advances are only as impactful as the systems that operationalize them.

Another relevant factor contributing to the heterogeneity of outcomes in patients with lung cancer is comorbidity and treatment-related toxicity. ¹³ As targeted therapies and immunotherapies have transformed lung cancer survival, Mo and colleagues ¹⁴ draw attention to an emerging consequence: a temporal rise in cardiovascular and pulmonary disease-specific mortality among NSCLC patients. Analyzing SEER registry data from 2001 to 2019 alongside the FDA Adverse Event Reporting System pharmacovigilance data, the authors demonstrate that while cancer-specific mortality declined steadily, particularly after 2012, cardiovascular disease (CVD)-specific mortality increased significantly. In patients with lung adenocarcinoma, CVD-specific mortality rose by 3.82% annually between 2012 and 2016, coinciding with the expanded adoption of EGFR-targeted therapies; pulmonary disease-specific mortality followed a similar trajectory, with a 6.54% annual increase during the same period. Multivariate analysis in the validation cohort identified immune checkpoint blockade as an independent risk factor for CVD-specific death (HR = 4.602, p = 0.031). Importantly, large-cell lung cancer, a subtype with limited utilization of targeted and immune therapies, did not exhibit comparable increases in cardiopulmonary mortality, supporting a treatment-associated rather than disease-inherent explanation. These data highlight the need to integrate cardio-oncology surveillance into routine NSCLC management, particularly as these therapies are increasingly used in earlier disease settings.

Beyond environmental carcinogens and patient comorbidity, genomic ancestry shapes both the molecular landscape and drug tolerance. Ancestry-linked pharmacogenomic variation can also affect drug metabolism and toxicity, while the persistent underrepresentation of non-European populations in landmark clinical trials limits the generalizability of current evidence and hinders equitable precision oncology. Pham and colleagues ¹⁵ explored the practical challenge of optimizing targeted therapy in a specific population. In a multicenter retrospective analysis of 343 Vietnamese patients with advanced EGFR-mutant NSCLC treated with first-line afatinib, the authors demonstrate that flexible, individualized dosing strategies are both feasible and beneficial. Only 39.9% of patients-initiated treatment at the recommended 40 mg dose, with the majority starting at 30 mg. Patients who began at 40 mg but de-escalated to less than 40 mg due to toxicity achieved the longest median time-to-treatment failure (21.5 months) and overall survival (37.9 months), significantly outperforming those who maintained 40 mg throughout. These findings are particularly relevant for Asian populations, where EGFR mutation prevalence is high but body weight, pharmacogenomic factors, and healthcare infrastructure may necessitate adapted dosing protocols. Multivariate analysis confirmed smoking status, ECOG performance status, and EGFR mutation subtype as independent prognostic factors, while dose reductions due to toxicity were associated with improved rather than compromised outcomes. In line with prior research, this study illustrates that precision medicine extends beyond molecular diagnosis to encompass pharmacological and clinical individualization. ¹⁶

Taken together, these five articles illustrate that the geographical heterogeneity of lung cancer extends well beyond epidemiology, permeating molecular profiling, diagnostic infrastructure, treatment toxicity, and drug dosing. The link between indoor radon and ALK rearrangements across Europe, while hypothesis-generating, awaits confirmation from prospective studies with individual-level exposure data such as the EORTC 1920 BIORADON trial. ⁹ The Polish experience reveals that even in healthcare systems where perioperative therapies are reimbursed, inadequate preoperative biomarker assessment can exclude patients from treatments proven to improve survival, ¹⁷ a gap that is unlikely to be unique to Poland. Meanwhile, the temporal rise in cardiovascular mortality among NSCLC patients treated with targeted agents and immunotherapy underscores the need for integrated cardio-oncology surveillance,^18,19 especially with these therapies moving into earlier disease settings. And the Vietnamese afatinib data remind us that the populations who carry the highest burden of oncogene-driven NSCLC remain underrepresented in the trials that define standard dosing. ¹⁶ Bridging these gaps will require not only larger and more diverse clinical studies, but also the integration of exposome-informed approaches alongside genomics and real-world data into prevention and treatment strategies that account for where patients live. The exposome, defined as the totality of environmental exposures across the lifespan and their biological effects, provides a comprehensive framework to capture the complex interplay between external risk factors and host susceptibility. ²⁰ In summary, this Special Collection illustrates that lung cancer is a disease defined as much by place as by pathology. The geographic distribution of carcinogens shapes molecular subtypes; molecular subtypes dictate treatment options; treatment options generate toxicities that must be managed within local healthcare systems; and those systems, in turn, determine whether patients access the diagnostic and therapeutic innovations that the field has produced. Understanding this continuum (from environment to genome to clinic, within an exposome-informed perspective) is essential to reducing the global burden of lung cancer and to further improving patient outcomes. The articles assembled here provide the evidence for a more integrated, geographically informed approach to thoracic oncology.

ORCID iDs

Miguel García-Pardo https://orcid.org/0000-0001-6339-8501

Herbert H. Loong https://orcid.org/0000-0002-6607-1106

Laura Mezquita https://orcid.org/0000-0003-0936-7338