ABSTRACT

Objective: To describe the performance of a pulmonologist-led lung cancer screening pro-gram using low dose CT (LDCT) in a cohort of outpatients with stable respiratory diseases in the Brazilian public health care system. Methods: This was a retrospective analysis of the first two rounds of lung cancer screening of patients enrolled in the program. Inclusion criteria were being between 55 and 80 years of age, being a current or former smoker (smoking cessation = 15 years), and having a smoking history = 30 pack-years. LDCT results were interpreted in accordance with the Lung CT Screening Reporting and Data System, and those with a score of 3 or 4 were considered posi-tive screening. Incidental pleuropulmonary findings were sought in all reports. Results: LDCTs were requested for 791 patients during the study period, and 712 patients (90%) met the screening criteria. The mean patient age was 63 years, and most participants were current smokers (56%) with emphysema (78.5%) and other pleuropulmonary findings on CT (64%). Screening was positive in 14.0% and 5.6% of the cases in the first and second screening rounds, respectively. Lung cancer was detected in 1.5% of the patients in both first and second rounds (positive predictive value: 11.0% and 26.6%, respectively). The rate of early-stage (TNM I or II) screen-detected non-small cell carcinoma was 64.3%. Of the patients with positive screening, 19% were lost to follow-up before investigation was complete. Conclusions: The results of this screening program suggest its adequate performance in a cohort of patients with significant respiratory morbidity. The loss to follow-up rate highlights the need for constant monitoring and interventions to ensure adherence.

Keywords: Lung neoplasms; Diagnostic screening programs; Early detection of cancer; Brazil; Tuberculosis.

RESUMO

Objetivo: Descrever o desempenho de um programa de rastreamento de câncer de pulmão conduzido por pneu-mologistas usando TC de baixa dose (TCBD) em uma coorte de pacientes ambulatoriais com doença respiratória estável no sistema público de saúde brasileiro. Métodos: Análise retrospectiva das duas primeiras rodadas de rastreamento de câncer de pulmão em pacientes inscritos no programa. Os critérios de inclusão foram ter idade entre 55 e 80 anos, ser fumante atual ou ex-tabagista (cessação do tabagismo ≤ 15 anos) e carga tabágica ≥ 30 anos-maço. Os resultados do TCBD foram interpretados de acordo com o Lung CT Screening Reporting and Data System, e aqueles com pontuação 3 ou 4 foram considerados exames positivos. Achados pleuropulmonares incidentais fo-ram verificados em todos os relatórios. Resultados: TCBD foram solicitadas para 791 pacientes durante o período do estudo, e 712 pacientes (90%) preencheram os critérios de rastreamento. A média de idade dos pacientes foi de 63 anos, e a maioria dos participantes era fumante atual (56%) com enfise-ma (78,5%) e outros achados pleuropulmonares na TC (64%). O rastreamento foi positivo em 14,0% e 5,6% dos casos na primeira e segunda rodada, respectivamente. O câncer de pulmão foi detectado em 1,5% dos pacientes tanto na primeira quanto na segunda rodada (valor preditivo posi-tivo: 11,0% e 26,6%, respectivamente). A taxa de carcinoma de células não pequenas detectado em estágio inicial (TNM I ou II) foi de 64,3%. Dos pacien-tes com rastreamento positivo, 19% foram perdidos no seguimento antes da conclusão da investigação. Conclusões: Os resultados deste programa de rastreamento sugerem um desempenho adequado em uma coorte de pacientes com morbidades respiratórias significativas. A taxa de perda de seguimento destaca a necessidade de monitoramento constante e intervenções para garantir a adesão.

Palavras-chave: Neoplasias pulmonares; Programas de triagem diagnóstica; Detecção precoce de câncer; Brasil; Tuber-culose.

INTRODUCTION
 
Lung cancer causes more deaths than any other neoplasms worldwide, and a substantial and growing proportion of cases occur in regions of middle and low socioeconomic development.(1) In Brazil, advanced-stage lung cancer is identified in about 70% of diagnosed cases,(2) and the estimated number of deaths is 28,000 every year.(3) Low-dose CT (LDCT) screening in high-risk patients, followed by an appropriate diagnostic and therapeutic approach, reduces mortality from this disease by 20% or more, as demonstrated by large clinical trials in the United States and in Europe.(4,5) However, factors associated with the clinical-epidemiological profile of the screened population, as well as the local health care system, can potentially alter the benefits of screening.(6) Current international guidelines recommend continuing to study this strategy in different scenarios, as well as collecting data with a view to improving local programs.(6,7)
 
This study describes the results of a screening program developed for patients at high risk of lung cancer who were being followed up for lung diseases in a large public hospital in southern Brazil. Because the hospital is located in an area with a high incidence of granulomatous diseases, especially tuberculosis (89.9/100,000 population in the city of Porto Alegre and 46.6/100,000 population in the State of Rio Grande do Sul)(8) but also paracoccidioidomycosis(9) and silicosis,(10) there is specific interest in the potential large number of positive screenings associated with inflammatory nodules. In addition, little is known about the feasibility of a lung cancer screening program in the Brazilian public health care system.
 
METHODS
 
This was a retrospective analysis of all patients who underwent LDCT at the institution between the implementation of the lung cancer screening program (June of 2014) and December of 2019. The centralized registry of all LDCTs allowed the location of all patient records for review. In the program’s care routine, pulmonologists requested LDCTs during outpatient visits of patients with lung diseases and smokers who were already being followed at the hospital. Data on demographics and clinical history, in addition to spirometry test results and imaging controls, were available in the electronic medical records and were collected using a structured form. Clinical data from the examination request and previous consultations were reviewed to confirm the intent to screen. The main follow-up diagnosis was recorded as reported by the pulmonologist in the medical records. COPD was confirmed by spirometry, but patients with evidence of chronic bronchitis were also considered COPD cases. Data were recorded anonymously and the project’s ethical and methodological aspects were approved by the institution’s research ethics committee (CAAE 73309317.5.0000.5530). Partial results of the present study have been previously presented as a poster at a conference.(11)
 
LDCT protocol
 
The acquisition and processing of images followed the American College of Radiology recommendations.(12) In summary, LDCT was performed with a 16-channel scanner (BrightSpeed; GE Healthcare, Waukesha, WI, USA) without the use of intravenous contrast, according to the following parameters: 120 kVp; 60 mA; gantry rotation time, 0.5 s; and pitch, 1.375. A single acquisition was performed during inspiration, and subsequent reconstructions were performed with 20-mm collimation, 5-mm increment, and 1.25-mm thickness. Effective radiation doses ranged between 0.8 and 1.3 mSv, with a dose-length product between 69 and 86 mGy•cm. The results were interpreted in accordance with the Lung CT Screening Reporting and Data System (Lung-RADS) standards,(13) and the revised assessment categories (version 1.1; 2019) were used whenever relevant. The reports were completed by radiologists from the institution under the supervision of a certified radiologist, who had developed the program and had specific training in thoracic radiology. The reports included the Lung-RADS classification score, as well as information on the presence of emphysema and other incidental pleuropulmonary findings. These findings included all acute or chronic interstitial, parenchymal, and pleural abnormalities that were described in the report but not used for the Lung-RADS classification. (13) Findings about other thoracic and extrathoracic organs, although described in the report, were not recorded in the present study.
 
Inclusion criteria were being between 55 and 80 years of age; having a smoking history of at least 30 pack-years; and being a current smoker or a former smoker (cessation ≤ 15 years). Exclusion criteria were having a pulmonary or systemic disease that would limit the diagnostic investigation or a possible surgical treatment for lung cancer (defined by the attending physician at the time of requesting the exam); having symptoms or signs compatible with clinical suspicion of lung cancer at the time of LDCT request; and having had lung cancer previously.
 
The procedures for investigation after positive screening (including control CT, biopsy, or referral for surgery) were at the discretion of the attending pulmonologist, although suggestions on the LDCT report in accordance with the Lung-RADS standards(13) were also considered. As part of the program’s routine, most control CTs were also LDCTs, and their reports also followed the Lung-RADS standards.(13) Regular multidisciplinary sessions were not a formal part of the screening program, and difficult cases were individually discussed between the radiologist, the thoracic surgeon, or both, as the routine practice at the institution.
 
The analysis of the present study refers to the outcomes in the first (T0) and second (T1) rounds of screening. Clinical and radiological outcomes were evaluated for every patient after a positive screening, including control CT results and final results of additional diagnostic workup (cancer or benign disease). The Lung-RADS standards(13) were used in order to determine the stability or regression of the lesion in control CTs. The medical records of patients with positive screening were reviewed until diagnostic definition or follow-up loss/closure. Additional data from patients diagnosed with cancer by screening were collected, including histological type and details on staging and treatment.
 
The parameters adopted to evaluate the program’s performance were defined as follows: rate of positive screens—number of patients with a Lung-RADS score of 3 or 4 divided by the number of patients screened, the rate being calculated for T0 and T1 separately; prevalence of lung cancer—number of patients with confirmed lung cancer in T0 divided by the number of screened patients in T0; incidence of lung cancer—number of patients with confirmed lung cancer in T1 divided by the number of screened patients in T1; and positive predictive value—number of patients with confirmed lung cancer divided by the number of patients with positive screening.
 
As an additional element of investigation, non-small cell carcinoma cases detected by screening were compared with cases diagnosed outside the program at the same institution (patients whose investigation was initiated due to symptoms or incidental findings). This comparative sample consisted of all cases diagnosed outside the screening program in 2017, which was the midpoint of the study period.
 
Statistical analysis
 
Descriptive statistics (absolute and relative frequencies; means and standard deviations; and medians and interquartile ranges) were used for reporting data on the prevalence of positive screenings and neoplasms, as well as clinical-epidemiological variables. The chi-square test was used for comparison of frequencies of positive screening between patients with and without additional CT findings, as well as for comparison of early-stage lung cancer between screened and unscreened patients (comparative sample). The significance level was set at 0.05 for all results.
 
RESULTS
 
During the study period, LDCT was performed in 791 patients. In 79 of these patients (10%), LDCT was not requested for screening purposes or the patient did not meet the inclusion criteria of the program. The reasons for excluding these patients are detailed in Figure 1. Of the 712 patients who underwent the first round of screening (T0), 266 (37.3%) underwent the second round (T1) by the end of the study period. Clinical and demographic data of the patients are shown in Table 1. Briefly, the mean age was 63 years, and there was a slight predominance of men (51.5%) and current smokers (56%). The most common diagnosis was COPD, which was the main diagnosis in 69.3% of the patients. The mean FEV1 was 64.9% of the predicted value.
 

 
The rate of positive screenings in T0 was 14%, with a similar distribution between Lung-RADS scores of 3 and 4 (Table 2). In T1, among 266 patients, 15 (5.6%) of the screenings were positive. Overall, 16 cases of cancer were identified in the study: 15 were cases of primary lung cancer and 1 was a case of metastatic breast cancer (the primary tumor had not been diagnosed prior to screening). Of the 15 primary lung malignancies, 11 were identified in T0 (n = 721; cancer prevalence of 1.5%), as were 4 in T1 (n = 266; cancer incidence of 1.5%). The positive predictive value for positive screening in T0 was 11% (11 confirmed neoplasm cases/99 positive screenings). Considering only a Lung-RADS score of 4, the positive predictive value was 23.9% (11/46). In T1, the positive predictive value was 26.6% (4/15) for positive screening and 50% (4/8) for Lung-RADS 4.

 
Details on the patients diagnosed with lung cancer, including staging and treatment, are shown in Table 3. The most common histological type was adenocarcinoma (13/15 patients), and treatment with curative intent (surgery or ablative radiotherapy) was offered to all stage I or II patients. A comparison of staging of non-small cell carcinoma detected in the screening program with those detected outside the screening program in 2017 (n = 134) is shown in Figure 2. TNM staging I-II was found in 64.3% and 22.4% of screened and unscreened patients, respectively (percentage point difference = 41.9%; 95% CI: 15.2-62.2; p = 0.0007).

 

 
Table 4 shows the outcome of the 114 positive screenings (T0 and T1), including the proportion of cases in which the Lung-RADS score regressed on subsequent CTs. One important finding was that 19.3% (n = 22/114, T0 and T1 combined) of the patients with positive screening were lost to follow-up without completing the investigation: 18.3% with a Lung-RADS score of 3 and 20.3% with a Lung-RADS of 4 (T0 and T1 combined).

 
Incidental pleuropulmonary findings (in addition to emphysema) were described in 64% of the CT scans, including parenchymal bands/cicatricial atelectasis in 37.9% of the cases. The frequencies of positive screening between patients with and without incidental LDCT findings were similar (16.4% and 12.5%, respectively; percentage point difference = 3.9; 95% CI: −1.3 to 9.6; p = 0.15). There was no statistically significant difference between patients with and without emphysema on LDCTs (15% and 9.8%, respectively; percentage point difference = 5.2%; 95%CI: −1.15 to 10.1; p = 0.09).
 
DISCUSSION
 
Our study reports the initial results of a lung cancer screening program in a cohort of patients with specific characteristics that differ from others: it was developed in a setting of high prevalence of granulomatous diseases and conducted by pulmonologists for patients being already followed up for chronic stable respiratory diseases in the context of the Brazilian public health care system.
 
Although our inclusion criteria were practically the same as were those in the National Lung Screening Trial (NLST),(4) the patients included in our study had a different clinical profile, as expected in a cohort of patients with previous lung diseases in a different epidemiological context. In fact, in our study, emphysema was reported in 78.5% of the patients during the first round of LDCTs, which was much higher than that reported in the NLST (30.7%).(14) Similarly, only 10.6% had a history of COPD/emphysema in that study,(14) whereas, in our study, the main reason for respiratory follow-up was COPD (69.3%). Parenchymal bands/cicatricial atelectasis were also very frequent in our cohort, probably reflecting previous infections, including locally prevalent granulomatous diseases.
 
Despite these differences, the positive screening rates were quite similar according to the Lung-RADS classification: The proportion of Lung-RADS 3 or 4 was 13.6% and 14%, respectively, in the NLST reanalysis(15) and in our study. The same occurred with the prevalence of cancer: 1% and 1.5%, respectively.
 
In a context similar to that in this study, Grover et al.(16) evaluated a screening program in a population previously followed up for COPD in the United Kingdom public health care system. The prevalence of cancer was 2% and, more importantly, 66.7% of these cases were diagnosed at stage I or II. In our study, the proportion of early cases was similar (64.3%), and, of note, that was significantly higher than was the proportion of cases detected outside the screening program in our institution (22.4%) and nationwide. (2) In fact, our results might have underestimated the potential benefit of screening, because a careful review of data revealed that in 2 of the stage IV cases, there was an unintentional delay in performing a control CT and diagnostic workup. Indeed, we cannot be sure whether avoiding these delays would have resulted in more favorable staging, and we understand that these situations may reflect real-life difficulties of a screening program.
 
One important concern is that patients with lung diseases and impaired lung function may present with a limited potential to treatment with curative intent. In our study, with the exception of 1 patient who refused treatment, all patients at TNM I or II received treatment with curative intent (surgery or ablative radiotherapy). Despite the significant number of patients with impaired lung function, including more than a quarter of the participants with FEV1 < 50% of the predicted value, we believe that patients carefully selected on the basis of their overall clinical context may be suitable candidates for screening even at such levels of lung function impairment.
 
Screening with LDCT in developing countries is challenging, and efforts to study and implement it are still incipient.(17,18) Nevertheless, studies such as that by dos Santos et al.(19) demonstrate that cancer detection rates and the need for invasive investigation may be similar to those in developed countries. In fact, in a recent study by Hochhegger et al.,(20) who retrospectively evaluated the screening results of 3,470 patients in Brazil (88% from the private health care system), the results were quite encouraging: the prevalence of cancer was 2.1% and, more importantly, early staging was identified in 70.3% of these cases. These results are similar to those in international studies, and the authors concluded that the local prevalence of granulomatous diseases did not increase the number of lung biopsies. The results in our study are in the same direction and significantly increased the number of patients screened in the public health care system (401 in Hochhegger et al.(20) vs. 721 in this study). Even with the limitations inherent to the public health care system context, our early staging rates were similar to those of Hochhegger et al.(20) (64.3% vs. 70.3%), which represents a very important advance in relation to the usual rates without screening. We believe that the expertise of large-volume or academic centers in a multidisciplinary context with specialists familiar with the local epidemiology and management of granulomatous diseases can contribute to satisfactory results, such as those obtained in the present study, without unnecessary investigations.
 
Finally, the rate of loss to follow-up was a significant limitation. This is a constant concern in clinical screening practice in real-life situations.(21) A recent meta-analysis by Lopez-Olivo et al.(22) included 15 American studies (16,863 patients) and found an overall adherence rate of only 55%. The authors(22) found that the following factors had important associations with low adherence: current smoking, ethnic minorities, age < 65 years, low educational level, and decentralized screening programs. In our study, the reasons for the low rate of T1 screenings were not evaluated and may have been due to either the physician’s or the patient’s decision, which were beyond the scope of this study. However, all positive screenings were carefully reviewed, and our loss to follow-up was approximately 20%. Unfortunately, our retrospective study could not identify causes of nonadherence to the visits or of the failure to carry out the investigations requested after a positive screening. In addition to the usual causes of poor adherence, one possible factor is that some patients with positive screening at the end of 2019 may have had difficulties in scheduling control CTs or medical visits due to restrictions caused by the COVID-19 pandemic (from March of 2020). Although the high rate of patients lost to follow-up is worrisome, we understand that the adherence issues in our study are similar to those reported in real-life studies(22,23) and that detecting such limitations may help improve the program, including strategies to contact missing patients and improve “navigation” after a positive result. We also believe that patients who are already linked to outpatient care may have better adherence to subsequent rounds of screening.
 
In conclusion, a lung cancer screening program for patients undergoing respiratory follow-up in the Brazilian public health care system in an area with a high incidence of granulomatous diseases and with a high rate of residual inflammatory findings on CT obtained satisfactory results that are comparable to results in other cohorts in different contexts. The high rate of early staging is encouraging and suggests a beneficial impact on the number of treatments with curative intent. The frequency of incomplete investigations after positive screening points to the need for constant monitoring and interventions to ensure adherence to screening.
 
AUTHOR CONTRIBUTIONS
 
FMS, MMRL, APGS, RSG, and CFA: conceptualization. FMS, MMRL, and CFA: data curation. FMS and CFA: formal analysis. CTMO, RUB, FMS, RSG, ACC, APGS, and MMRL: investigation. FMS: drafting of the manuscript. FMS, CFA, RSG, MMRL, and APGS: editing and review of the manuscript. All authors approved the final version of the manuscript.
 
CONFLICTS OF INtEREST
 
None declared.
 
REFERENCES
 
1.            Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209-249. https://doi.org/10.3322/caac.21660
2.            Araujo LH, Baldotto C, Castro G Jr, Katz A, Ferreira CG, Mathias C, et al. Lung cancer in Brazil. J Bras Pneumol. 2018;44(1):55-64. https://doi.org/10.1590/s1806-37562017000000135
3.            Brasil. Ministério da Saúde. Instituto Nacional de Câncer José Alencar Gomes da Silva [homepage on the Internet]. Rio de Janeiro: INCA; c2019 [cited 2022 Jan 16]. Estimativa 2020: incidência de câncer no Brasil. Available from: https://www.inca.gov.br/sites/ufu.sti.inca.local/files//media/document//estimativa-2020-incidencia-de-cancer-no-brasil.pdf
4.            National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395-409. https://doi.org/10.1056/NEJMoa1102873
5.            de Koning HJ, van der Aalst CM, de Jong PA, Scholten ET, Nackaerts K, Heuvelmans MA, et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. N Engl J Med. 2020;382(6):503-513. https://doi.org/10.1056/NEJMoa1911793
6.            US Preventive Services Task Force, Krist AH, Davidson KW, Mangione CM, Barry MJ, Cabana M, et al. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2021;325(10):962-970. https://doi.org/10.1001/jama.2021.1117
7.            Mazzone PJ, Silvestri GA, Souter LH, Caverly TJ, Kanne JP, Katki HA, et al. Screening for Lung Cancer: CHEST Guideline and Expert Panel Report. Chest. 2021;160(5):e427-e494. https://doi.org/10.1016/j.chest.2021.06.063
8.            Governo do Estado do Rio Grande do Sul. Secretaria da Saúde [homepage on the Internet]. Porto Alegre: Governo do Estado do Rio Grande do Sul [cited 2022 Jan 16]. Informe Epidemiológico: Tuberculose 2020. Available from: https://www.cevs.rs.gov.br/a-tuberculose-no-rio-grande-do-sul
9.            Shikanai-Yasuda MA, Mendes RP, Colombo AL, Queiroz-Telles F, Kono ASG, Paniago AMM, et al. Brazilian guidelines for the clinical management of paracoccidioidomycosis [published correction appears in Rev Soc Bras Med Trop. 2017 Oct 16;:0] [published correction appears in Rev Soc Bras Med Trop. 2017 Nov-Dec;50(6):879-880]. Rev Soc Bras Med Trop. 2017;50(5):715-740. https://doi.org/10.1590/0037-8682-0230-2017
10.          Algranti E, Saito CA, Carneiro APS, Bussacos MA. Mortality from silicosis in Brazil: Temporal trends in the period 1980-2017. Am J Ind Med. 2021;64(3):178-184. https://doi.org/10.1002/ajim.23215
11.          Svartman FM, Sartori APG, Gutierrez RS, Brito RU, Oliveira CTM, Andrade CF. Lung cancer screening with low-dose CT in Brazil: results of the initial CT in the setting of clinical practice at a public hospital. Amer J Resp Crit Care Med. 2018;197:A7358. https://doi.org/10.1164/ajrccm-conference.2018.197.1_MeetingAbstracts.A7358
12.          Kazerooni EA, Austin JH, Black WC, Dyer DS, Hazelton TR, Leung AN, et al. ACR-STR practice parameter for the performance and reporting of lung cancer screening thoracic computed tomography (CT): 2014 (Resolution 4). J Thorac Imaging. 2014;29(5):310-316. https://doi.org/10.1097/RTI.0000000000000097
13.          American College of Radiology (ACR) [homepage on the Internet]. Reston, VA: ACR; [cited 2022 Jan 16]. Lung CT Screening Reporting and Data System (Lung-RADS). Available from: https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/Lung-Rads
14.          Pinsky PF, Lynch DA, Gierada DS. Incidental Findings on Low-Dose CT Scan Lung Cancer Screenings and Deaths From Respiratory Diseases. Chest. 2022;161(4):1092-1100. https://doi.org/10.1016/j.chest.2021.11.015
15.          Pinsky PF, Gierada DS, Black W, Munden R, Nath H, Aberle D, et al. Performance of Lung-RADS in the National Lung Screening Trial: a retrospective assessment. Ann Intern Med. 2015;162(7):485-491. https://doi.org/10.7326/M14-2086
16.          Grover H, Ross T, Fuller E. Implementation of targeted screening for lung cancer in a high-risk population within routine NHS practice using low-dose computed tomography. Thorax. 2020;75(4):348-350. https://doi.org/10.1136/thoraxjnl-2019-214303
17.          Raez LE, Nogueira A, Santos ES, Dos Santos RS, Franceschini J, Ron DA, et al. Challenges in Lung Cancer Screening in Latin America. J Glob Oncol. 2018;4:1-10. https://doi.org/10.1200/JGO.17.00040
18.          Shankar A, Saini D, Dubey A, Roy S, Bharati SJ, Singh N, et al. Feasibility of lung cancer screening in developing countries: challenges, opportunities and way forward. Transl Lung Cancer Res. 2019;8(Suppl 1):S106-S121. https://doi.org/10.21037/tlcr.2019.03.03
19.          dos Santos RS, Franceschini JP, Chate RC, Ghefter MC, Kay F, Trajano AL, et al. Do Current Lung Cancer Screening Guidelines Apply for Populations With High Prevalence of Granulomatous Disease? Results From the First Brazilian Lung Cancer Screening Trial (BRELT1). Ann Thorac Surg. 2016;101(2):481-488. https://doi.org/10.1016/j.athoracsur.2015.07.013
20.          Hochhegger B, Camargo S, da Silva Teles GB, Chate RC, Szarf G, Guimarães MD, et al. Challenges of Implementing Lung Cancer Screening in a Developing Country: Results of the Second Brazilian Early Lung Cancer Screening Trial (BRELT2). JCO Glob Oncol. 2022;8:e2100257. https://doi.org/10.1200/GO.21.00257
21.          Sakoda LC, Henderson LM, Rivera MP. Adherence to Lung Cancer Screening: What Exactly Are We Talking About?. Ann Am Thorac Soc. 2021;18(12):1951-1952. https://doi.org/10.1513/AnnalsATS.202106-724VP
22.          Lopez-Olivo MA, Maki KG, Choi NJ, Hoffman RM, Shih YT, Lowenstein LM, et al. Patient Adherence to Screening for Lung Cancer in the US: A Systematic Review and Meta-analysis. JAMA Netw Open. 2020;3(11):e2025102. https://doi.org/10.1001/jamanetworkopen.2020.25102
23.          Pinsky PF, Miller E. Use and Outcomes of Low-Dose CT Scan Lung Cancer Screening in the Medicare Population. Chest. 2022;S0012-3692(22)00575-X. https://doi.org/10.1016/j.chest.2022.03.031