We retrospectively screened patients with histologically or cytologically confirmed pulmonary adenocarcinoma who were diagnosed and treated at Dongnam Institute of Radiological & Medical Sciences, a secondary referral hospital in Busan, South Korea, from January 2008 through December 2015. Patients with the results of EGFR mutation testing available were included in this study. We excluded patients who had active pulmonary TB along with lung cancer and who were diagnosed with other cancers within 5 years since lung cancer diagnosis, as their clinical course may be influenced by those comorbidities. We reviewed the electronic medical records to obtain demographic and clinical data including age, sex, Eastern Cooperative Oncology Group (ECOG) performance status, smoking status, TNM stage, systemic or pulmonary comorbidities and tumor mutational status. All patients underwent chest computed tomography, brain magnetic resonance imaging, and ¹⁸F-fluorodeoxyglucose positron-emission tomography (PET) for clinical staging determined according to the International Association for the Study of Lung Cancer TNM staging classification of nonsmall cell lung cancer [13]. Tumor response was examined by computed tomography (CT) every two cycles of systemic treatment and evaluated according to the Response Evaluation Criteria in Solid Tumors (RECIST) ver. 1.1 [20]. In case of EGFR-TKIs, a single cycle comprised 4 weeks of treatment.

We reviewed initial chest CT scan at diagnosis to determine concomitant old TB lesions. We divided our lung cancer population into two groups: patients with old TB lesions (TB group) and those without lesions (non-TB group). Old TB lesions were defined as at least one of the following as previously described [21]: fibrotic scar with volume loss, fibronodular lesions, calcified or non-calcified granulomas, or pleural thickening. The existence and location of old TB lesions were determined based on the consensus of two independent reviewers (I.K. Hwang and S.S. Paik). In case of ambiguous findings, consensus was arrived at after discussion with a chest radiologist at our institution. In case of difficult diagnosis of old TB lesions and lung cancer-related or metastatic lesions based on chest CT, the final decision was made by comparison of the corresponding images with those of PET. To exclude patients with active pulmonary TB, we confirmed the negative results of polymerase chain reaction for TB, acid-fast bacilli staining and TB culture using sputum or bronchial washing specimens obtained at the time of lung cancer diagnosis.

Genomic DNA was extracted from formalin-fixed, paraffin-embedded, 5-μm-thick tissue sections using the High Pure polymerase chain reaction (PCR) Template Preparation Kit (Roche Applied Science, Mannheim, Germany) according to the manufacturer’s instructions. The extracted DNA was stored at –20°C until analysis. EGFR Pyro Kit (QIAGEN Korea Ltd., Seoul, Korea) and PyroMark Q24 System (QIAGEN Korea Ltd.) were used to detect EGFR mutations by real-time PCR. The primer sets cover mutations or deletions spanning exons 18 to 21 of the genes encoding tyrosine kinase domain of EGFR. Details of these methods were described in a previous study [22].

Statistical analyses were performed using the SPSS ver. 20.0 software for Windows (IBM Corp., Armonk, NY). PFS was defined as the time from the first day of chemotherapy to disease progression or death from any cause. Overall survival (OS) was defined from the start day of systemic treatment for advanced disease or the day of surgical resection for early-stage disease to death from any cause. Associations between categorical variables were compared using the chisquared test. Associations between clinicopathological parameters and survival were evaluated by univariate analysis using the log-rank test. Subsequently, multivariate Cox proportional hazards regression analysis was conducted after adjusting for parameters with p-values of < 0.2 in the univariate analysis. Survival probability was estimated by the Kaplan-Meier method. A p-value of < 0.05 was considered statistically significant.

Informed consents were obtained from all patients alive and the study protocol was approved by the Institutional Review Board of the Dongnam Institute of Radiological & Medical Sciences (D-1705-002-002).

During the study period, 513 patients were diagnosed as lung adenocarcinoma at our institution. After exclusion of patients with cancers other than lung cancer (n=11), active pulmonary TB at the initial diagnosis (n=2) and without any available data for tumor EGFR mutational status (n=23), 477 patients were found eligible for this study. The clinicopathological characteristics of the patients are summarized in Table 1. All subjects were Korean with a median age of 64 years (range, 38 to 84 years). Old TB lesions on chest CT scan were detected in 100 patients (21%, TB group) while no TB lesions in 377 patients (79%, non-TB group). 280 patients (58%) were male, 173 patients (36%) were above the age of 65 years and 272 patients (57%) were current or former smokers. 395 patients (83%) had ECOG performance status of 0 or 1, and 252 patients (53%) were stage IIIB or IV. Tests for anaplastic lymphoma kinase (ALK) translocation including immunohistochemistry and fluorescence in situ hybridization were performed in 238 patients (50%), while KRAS mutation testing was performed in 199 patients (42%); 10 patients (4%) were positive for ALK translocation and 22 patients (11%) were positive for KRASmutation. 270 patients (57%) had more than one systemic comorbidity including hypertension and diabetes while 64 patients (13%) had benign lung diseases other than TB. One hundred fortyseven patients (30%) received EGFR-TKIs during their clinical course.

Overall, 183 patients (39%) carried EGFR-mutated tumors. The most frequent mutation was deletion in exon 19 (50%) followed by L858R point mutation (43%). The frequency of EGFR mutations was significantly higher in the TB group than in the non-TB group (56/100, 56% vs. 127/377, 34%; p=0.038). In addition, the frequency of exon 19 deletion in the TB group was significantly higher than in the non-TB group (33/100, 33% vs. 60/377, 16%; p=0.011), while the frequency of L858R point mutation was not different between the two groups (20/100, 20% vs. 58/377, 15%; p=0.254) (Table 1).

To determine the factors associated with the rate of EGFR mutation, we performed a multivariate analysis for possible clinical variables. As presented in Table 2, after adjusting for possible confounders, female sex (odds ratio [OR], 1.21; 95% confidence interval [CI], 1.05 to 1.86), never smoking (OR, 1.27; 95% CI, 1.12 to 1.98), and old TB lesions (OR, 1.43; 95% CI, 1.08 to 2.47) were independently associated with an increased rate of EGFR mutation.

To investigate the relationship between EGFR mutations and location of old TB lesions, we analyzed the frequency of EGFR mutations according to the spatial relationship between the primary lung mass and old TB lesions. As summarized in Table 3, among 100 patients in the TB group, 38 (38%) and 36 patients (36%) had the main tumor mass in the same lobe and different lobe of the ipsilateral lung containing the main TB lesion, respectively, while 26 patients (26%) had main tumor mass in the contralateral lung. The frequency of EGFR mutation in patients who had the main mass in the same lobe and different lobe of the ipsilateral lung with the main TB lesion were 61% (23 among 38 patients) and 72% (26 among 36 patients), respectively, while that of those who had lung cancer in the contralateral lung is 27% (7 among 26 patients). The difference in EGFR mutational frequencies according to the location of lung cancer and TB lesions was statistically significant (p=0.028). In case of exon 19 deletion, the frequency in patients who had the main mass in the same lobe and different lobe of the ipsilateral lung with the main TB lesion were 57% (17 among 30 patients) and 52% (14 among 27 patients), respectively, whereas that of those who had lung cancer in the contralateral lung is 10% (2 among 20 patients). The difference in the frequencies of exon 19 deletion according to the location of lung cancer and TB lesions was also statistically significant (p=0.013).

To evaluate whether patients with TB-related lung adenocarcinoma showed different survival rates compared with those without TB lesions, we performed a survival analysis according to clinicopathological parameters of all study subjects. The results of survival analyses are summarized in Table 4. The median OS in all study subjects was 40.5 months (range, 3.3 to 65.7 months). Univariate analysis revealed that male gender, old age, ever smoking, poor performance status, advanced stage and EGFR wild-type were significantly associated with shorter OS. The presence of old TB lesions was not associated with OS (p=0.154). Multivariate analysis showed that poor performance status (hazard ratio [HR], 1.67; 95% CI, 1.14 to 2.43), advanced stage (HR, 3.61; 95% CI, 2.31 to 5.64) and EGFR wild-type (HR, 1.91; 95% CI, 1.29 to 3.58) were independently associated with shorter OS. We further performed subgroup analyses according to EGFR mutational status. However, OS was similar between the two groups regardless of EGFR mutational status.

To investigate whether the pulmonary TB affects the response to TKI treatment, we compared the PFS and OS in patients who were treated with EGFR-TKIs as a first-line treatment in each group. Among all the patients, 88 (18.5%) were treated with EGFR-TKIs such as gefitinib (n=39), erlotinib (n=28), and afatinib (n=21) for locally advanced or metastatic EGFR-mutated lung adenocarcinoma. Twenty-one and 67 patients were included in TB group and non-TB group, respectively. As first-line EGFR-TKI was approved by Korea Food & Drug Administration at April 2010, the enrolment period for this subgroup was 2010-2015. First-line gefitinib and erlotinib were covered by National Health Insurance System (NHIS) of Korea since April 2011 and March 2013, respectively. Ten patients (4 for gefitinib, 6 for erlotinib) paid for their own first-line EGFR-TKI treatment before the reimbursement of each TKI. First-line afatinib was available since October 2014 and the proportion of patients who were treated with this drug as a first-line therapy was similar in both groups (23% [5/21] in TB group and 24% [16/67] in non-TB group). The number of patients who received firstline EGFR-TKIs and EGFR-TKIs used was distributed throughout the study period, and they are summarized in S1 Table. The survival analyses results are summarized in Table 5. The median PFS of the 88 patients treated with EGFR-TKIs were 10.5 months (range, 4.6 to 21.2 months). In the univariate analysis, poor ECOG performance and advanced stage were significantly associated with shorter PFS. In addition, pulmonary TB lesions was significantly associated with shorter PFS (9.1 months vs. 11.6 months, p=0.020). Multivariate analysis showed that advanced stage (HR, 1.76; 95% CI, 1.10 to 2.83) and old TB lesions (HR, 1.85; 95% CI, 1.16 to 3.56) were independently associated with shorter PFS. The Kaplan-Meier survival curves also showed that patients with old TB lesions were likely to have shorter PFS (Fig. 1A). The median OS of the 88 patients were 21.5 months (range, 5.1 to 39.6 months). In the univariate analysis, ever smoking and advanced stage were associated with shorter OS. In addition, TB lesions was also associated with shorter OS (19.4 months vs. 24.5 months, p=0.014). Multivariate analysis showed that ever smoking (HR, 2.02; 95% CI, 1.47 to 3.84), advanced stage (HR, 1.84; 95% CI, 1.09 to 3.10) and old TB lesions (HR, 1.60; 95% CI, 1.16 to 3.74) were independently associated with shorter OS. The Kaplan-Meier survival curves also showed that patients with old TB lesions were likely to have shorter OS (Fig. 1B).