Background: Lung cancer remains a leading cause of cancer-related mortality worldwide, largely driven by tobacco exposure and other carcinogenic exposures. Low-dose computed tomography (LDCT) has demonstrated superior sensitivity over chest radiography for early detection, leading to reduced lung cancer mortality and favorable cost-effectiveness. However, evidence from low- and middle-income countries, particularly in Latin America, remains limited. In Brazil, the Brazilian Early Lung Cancer Detection Trial (BRELT) was established to address this gap. BRELT1 was the first lung cancer screening study conducted in Latin America, demonstrating the feasibility of LDCT-based screening and providing initial estimates of lung cancer prevalence. BRELT2 expanded the program by incorporating refined risk stratification models and follow-up strategies, generating additional data on nodule management and screening performance. BRELT3 represents a further step by implementing LDCT screening through a mobile unit (ProPulmão), specifically targeting socially vulnerable populations in resource-limited and geographically remote areas.
Objectives: To describe tomographic findings in a high-risk lung cancer population in a resource-limited Brazilian area and to quantify pulmonary nodules and lung cancer frequency in the BRELT3/mobile ProPulmão program.
Methods: This is a prospective, single-arm, longitudinal observational study involving individuals aged 50–80 years, current or former smokers with at least 20 pack-years and up to 15 years since smoking cessation, undergoing LDCT screening in a mobile unit. Results were classified according to Lung-RADS v2022, with categories 3 and 4 referred for further diagnostic evaluation. Ethical approval was obtained (SENAI-CIMATEC/Santa Izabel Hospital; n°67431523.6.0000.9287 / 67431523.6.3001.5520).
Results: A total of 5,223 individuals were identified as potential participants. After applying inclusion criteria, 2,018 underwent LDCT screening; 53% were female, 87% self-identified as non-white, and 64% had primary-level education. Most participants were current smokers (58.8%), and 70.7% had high or very high nicotine dependence by the Fagerström test. The median smoking exposure was 40 pack-years, and 22% reported family history of lung cancer. Using the PLCO2012-No-Race model, 66% had an estimated lung cancer risk >2% over 6 years. Radiological findings suggestive of lung cancer risk were identified in 283 individuals (12%), with 5% classified as Lung-RADS 3 and 7% as Lung-RADS 4. Biopsy was indicated for 46 participants; 34 procedures were performed. Nineteen lung cancer cases were diagnosed (0.94%). Histological types included non-small cell lung cancer (42%), adenocarcinoma (37%), squamous cell carcinoma (11%), neuroendocrine tumor (5%), and round cell/plasmacytoid tumor (5%). Seven patients (37%) were diagnosed at advanced stages (III or IV), and surgical treatment was performed in 7 cases (37%).
Conclusion: Lung cancer screening using LDCT is feasible and effective in socially vulnerable populations within resource-limited setting, enabling early diagnosis and generating critical evidence to support the expansion of screening programs and reduction of lung cancer mortality.

Table 1. Demographic and clinical characteristics of patients