Lung cancer screening in clinical practice: identification of high-risk chronic obstructive pulmonary disease patients

• Autor(a) principal: Sousa,Sofia Rodrigues
• Data de Publicação: 2022
• Outros Autores: Caldeira,João Nunes, Rodrigues,Cidália, Figueiredo,Ana, Barata,Fernando
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Revista da Associação Médica Brasileira (Online)
• Texto Completo: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0104-42302022000400502
• topic: Lung cancer; Screening; Chronic obstructive pulmonary disease

SUMMARY OBJECTIVE: The NELSON study demonstrated a positive association between computed tomography scanning and reduced mortality associated with lung cancer. The COPD-LUCSS-DLCO is a tool designed to improve screening selection criteria of lung cancer for chronic obstructive pulmonary disease patients. The aim of this study was to examine and compare the discriminating value of both scores in a community-based cohort of chronic obstructive pulmonary disease patients. METHODS: A retrospective study of chronic obstructive pulmonary disease patients followed in pulmonology consultation for a period of 10 years (2009–2019) was conducted. The NELSON criteria and COPD-LUCSS-DLCO score were calculated for each patient at the time of the study inclusion. The lung cancer incidence was calculated for each of the subgroups during the follow-up period. RESULTS: A total of 103 patients were included in the study (mean age 64.7±9.2 years, 88.3% male). Applying the COPD-LUCSS-DLCO score, high-risk patients have a 5.9-fold greater risk of developing lung cancer versus the low risk. In contrast, there was no significant association between NELSON selection criteria and lung cancer incidence. The area under the curve was 0.69 for COPD-LUCSS-DLCO and 0.59 for NELSON criteria. Comparing test results showed no differences. CONCLUSIONS: The use of the COPD-LUCSS-DLCO score in clinical practice can help to detect chronic obstructive pulmonary disease patients in greater risk of developing lung cancer with better performance than NELSON criteria. Therefore, models that include a risk biomarker strategy can improve selection criteria and consequently can enhance a better lung cancer prediction.