Pathogen surveillance and risk factors for pulmonary infection in patients with lung cancer: A retrospective single-center study

Abstract

Background: Early and accurate diagnosis of pulmonary infection (PI) is crucial for the timely implementation of appropriate treatment strategies in lung cancer patients.

Methods: Metagenomic next-generation sequencing and conventional testing were performed in lung cancer patients with and without PI. The pathogen profiles were analyzed, and risk factors for PI were explored using univariate and multivariate logistic regression models.

Results: A total of 55 lung cancer patients with PI and 59 non-infected lung cancer patients were included. There were 41 underlying pathogens identified by both methods in lung cancer patients with PI. The coexistence of different pathogen types was common, particularly between fungi and viruses, which was observed in 28.57% of cases. The incidence of Streptococcus pneumoniae and Pneumocystis jirovecii is significantly higher in small-cell lung carcinoma patients compared to that in non-small-cell lung carcinoma patients. Besides, cytomegalovirus, P. jirovecii, and Aspergillus were more likely to be found in advanced-stage patients. Risk factor analysis revealed that Karnofsky Performance Status <90 and chemotherapy were strongly associated with PI in lung cancer patients.

Conclusions: This study highlights the complexity of PI in lung cancer patients, emphasizing the need for tailored diagnostic and therapeutic strategies based on cancer type and stage.

Keywords: co-infection; lung cancer; mNGS; pathogens; pulmonary infection.

Figure 1

Potential pathogens detected by mNGS and CTM. Fungi and viruses were the most frequently detected pathogens, followed by G− and G+ bacteria. The bar graph shows the number of cases in which each pathogen was detected out of the 63 infection episodes. Pathogens are categorized into three groups: identical detection (green): pathogens detected by both mNGS and CTM; extra detection (red): pathogens detected exclusively by either mNGS or CTM, with pathogenicity not ruled out; and false-positive detection (blue): microorganisms detected by either mNGS or CTM but clinically diagnosed as non-causative pathogens. CMV: cytomegalovirus; EBV: Epstein-Barr virus; HSV: herpes simplex virus; HHV: human herpes virus.

Figure 2

Infection patterns in PI across different cancer types and stages. (a) Infection patterns in 63 infection episodes, with bars color-coded to represent the cancer stage (I–IV) and histological type (NSCLC or SCLC) of each case. Multiple pathogens were detected in 61.02% of infection episodes, with fungi and viruses being the most common co-detected combination. (b) Coexistence patterns of predominant pathogens identified in the cohort, illustrating the complexity of polymicrobial infections in lung cancer patients. C. albicans was frequently co-detected with EBV, and P. jirovecii was often found alongside CMV. MTBC: M. tuberculosis complex; CMV: cytomegalovirus; EBV: Epstein-Barr virus; HSV: herpes simplex virus.

Figure 3

Pathogen distribution across lung cancer stages and histological types. (a) Proportions of pathogens detected in different histological types of lung cancer, calculated as the number of cases positive for a specific pathogen divided by the total number of patients within each cancer type. (b) Proportions of pathogens detected across different cancer stages, calculated as the number of cases positive for a specific pathogen divided by the total number of patients within each stage group. Notable trends include the increased detection of CMV in late-stage patients and the exclusive presence of P. jirovecii and A. flavus in this group. Statistical significance was determined using the χ² test (*p < 0.05; **p < 0.01).

Figure A1

Multivariate analysis of factors associated with pulmonary infections in lung cancer patients.