Interstitial lung disease in patients enrolled in early-phase clinical trials: the ILDE study

•About 5% of all ILDs are related to drugs, mostly cancer medicines; risk of ILD with novel cancer drugs is unknown.•ILDE is a real-world series of patients treated with novel compounds in phase I clinical trials.•In ILDE, 5.3% of patients had ILD, usually not severe, with treatment resumed in two-thirds of cases.•Possible risk factors of ILD in the ILDE were: older age, prior radiotherapy to chest wall, antibody–drug conjugates.•A risk-based monitoring, diagnostic and treatment approach is recommended. BackgroundInterstitial lung disease (ILD) encompasses a heterogeneous group of disorders sharing pathophysiological inflammatory mechanisms, leading to parenchymal distortions. The prevalence of ILD with new cancer drugs is underreported: the identification of potential determinants is priority.Materials and methodsILDE is a retrospective study aimed at describing the clinical course and potential determinants of ILD in patients receiving experimental treatments.ResultsWe identified 226 eligible patients, of whom 5.3% (n = 12) had ILD. In five patients, the diagnosis was radiological, while seven patients had initial cough, dyspnea, fatigue or fever. ILD was graded as grade 1 (G1) in four, G2 in five and G3 in three patients. The first occurrence of ILD resolved completely in 50% of patients (n = 6/12). No patient had fatal ILD. Eight patients (66.7%) resumed the treatment after the first episode of ILD, while four patients (33.3%) had to discontinue the therapy. Five out of six patients had resolved the first ILD episode and then resumed treatment, experiencing a second ILD episode (n = 5/6; 83.3%). The second ILD event was G1 in three patients and G2 in two patients, resulting in three patients who eventually discontinued the treatment (n = 3/5; 60%). Correlation analysis showed a higher risk of ILD in older patients (P = 0.051), those who had received previous chest radiation therapy (P = 0.047) or those receiving antibody–drug conjugates (P = 0.006). In a survival analysis adjusted for immortal time bias, ILD was not independently prognostic (hazard ratio 0.50, 95% confidence interval 0.23-1.09, P = 0.082).ConclusionsIn ILDE, patients experiencing ILD had generally good outcomes, and many could resume the cancer treatments. Implementing best practices to prompt diagnosis and management of ILD is critical to treat a potentially severe adverse effect of new drugs, while not affecting patients' outcomes. Research efforts to identify risk factors is warranted, to implement risk-based monitoring schedules and develop ad hoc strategies to improve the cure rates of ILD. Interstitial lung disease (ILD) encompasses a heterogeneous group of disorders sharing pathophysiological inflammatory mechanisms, leading to parenchymal distortions. The prevalence of ILD with new cancer drugs is underreported: the identification of potential determinants is priority. ILDE is a retrospective study aimed at describing the clinical course and potential determinants of ILD in patients receiving experimental treatments. We identified 226 eligible patients, of whom 5.3% (n = 12) had ILD. In five patients, the diagnosis was radiological, while seven patients had initial cough, dyspnea, fatigue or fever. ILD was graded as grade 1 (G1) in four, G2 in five and G3 in three patients. The first occurrence of ILD resolved completely in 50% of patients (n = 6/12). No patient had fatal ILD. Eight patients (66.7%) resumed the treatment after the first episode of ILD, while four patients (33.3%) had to discontinue the therapy. Five out of six patients had resolved the first ILD episode and then resumed treatment, experiencing a second ILD episode (n = 5/6; 83.3%). The second ILD event was G1 in three patients and G2 in two patients, resulting in three patients who eventually discontinued the treatment (n = 3/5; 60%). Correlation analysis showed a higher risk of ILD in older patients (P = 0.051), those who had received previous chest radiation therapy (P = 0.047) or those receiving antibody–drug conjugates (P = 0.006). In a survival analysis adjusted for immortal time bias, ILD was not independently prognostic (hazard ratio 0.50, 95% confidence interval 0.23-1.09, P = 0.082). In ILDE, patients experiencing ILD had generally good outcomes, and many could resume the cancer treatments. Implementing best practices to prompt diagnosis and management of ILD is critical to treat a potentially severe adverse effect of new drugs, while not affecting patients' outcomes. Research efforts to identify risk factors is warranted, to implement risk-based monitoring schedules and develop ad hoc strategies to improve the cure rates of ILD.