Interstitial lung diseases with concomitant lung cancer: a data mining approach revealing a complex condition with gender and immune-associated specific implications

Abstract Background. Interstitial lung diseases (ILDs) identify a family of heterogeneous entities mainly characterized by chronic scarring of the lung parenchyma. Among ILDs, idiopathic pulmonary fibrosis (IPF) represents the most frequent idiopathic interstitial pneumonitis associated with progressive functional decline leading to respiratory failure, high symptoms burden and mortality. Notably, the incidence of lung cancer (LC) in patients already affected by ILDs - mainly IPF- is significantly higher than in general population. Moreover, these cases are often neglected and deprived of active oncologic treatments. Methods.We here aim at identifying variables predictive of outcome in a multicentre retrospective cohort of ILD associated to lung cancer collected from 2018 till the end of 2023. Overall, 73 cases were identified and for 55 patients exhaustive clinico-pathologic data were available. Among them 42 carried IPF. The entire dataset was then analysed by using the JMP partition algorithm (JMP-Statistical Discoveries. From SAS) which is able to choose the optimum splits from many possible trees, making it a powerful modelling, and data discovery tool. Results. The average age of lung cancer diagnosis was 71.4 years whereas that of IPF was 69.5 years, the average Charlson comorbidity index was 4.6. Female patients were the 28.3 % (15) of the evaluated cases. The most frequent tumor hystotype was adenocarcinoma (45.2%) and in more than 60% of the cases (67.9%) cancer was diagnosed at early stage (TNM I-II-IIIA). A significant gender difference emerges regarding the overall patients’ survival and quite unexpectedly surgical approach to IPF-associated LC and the detection of serum autoantibodies are among the strongest outcome predictors. Conclusions. IPF-associated LC emerges as a unique malignant disease defined by specific gender, histopathologic clinical and molecular parameters which might benefit from active treatments.