Immune checkpoint inhibitors (ICI) have the potential to trigger unpredictable immune-related adverse events (irAEs), which can be severe. The underlying mechanisms of these events are not fully understood. As PD-L1 is upregulated by IFN, the heightened immune activation resulting from PD-1/PD-L1 inhibition may enhance the IFN response, triggering the expression of IFN-inducible genes and contributing to irAE development and its severity. In this study, we investigated the interplay between irAEs and the expression of IFN-inducible chemokines and cytokines in 134 consecutive patients with solid tumours treated with PD-(L)1 inhibitors as monotherapy or in combination with chemotherapy or other immunotherapy agents. We compared the plasma levels of IFN-associated cytokines (CXCL9/10/11, IL-18, IL-10, IL-6 and TGFβ) at various time points (at baseline, at the onset of irAE and previous to irAE onset) in three patient groups categorized by irAE development and severity: patients with serious irAEs, mild irAEs and without irAEs after PD-(L)1 inhibitors. No differences were observed between groups at baseline. However, patients with serious irAEs exhibited significant increases in CXCL9/10/11, IL-18 and IL-10 levels at the onset of the irAE compared to baseline. A network analysis and correlation patterns highlighted a robust relationship among these chemokines and cytokines at serious-irAE onset. Combining all of the analysed proteins in a cluster analysis, we identified a subgroup of patients with a higher incidence of serious irAEs affecting different organs or systems. Finally, an ROC analysis and a decision tree model proposed IL-18 levels ≥ 807 pg/mL and TGFβ levels ≤ 114 pg/mL as predictors for serious irAEs in 90% of cases. In conclusion, our study elucidates the dynamic changes in cytokine profiles associated with serious irAE development during treatment with PD-(L)1 inhibitors. The study’s findings offer valuable insights into the intricate IFN-induced immune responses associated with irAEs and propose potential predictive markers for their severity.

Immune-related adverse events (irAEs) are unpredictable autoimmune-like toxicities induced by immune checkpoint inhibitors (ICI). irAEs are a consequence of a breakdown in self-tolerance. ICIs can induce autoantibody formation, and the presence of antinuclear autoantibodies (ANAs) has been reported in patients who developed irAEs. Our goal was to compare ANA patterns by indirect immunofluorescence at different timepoints before (baseline) and after the initiation of ICI treatment and to analyze the role of ANA pattern changes as predictors of irAEs. This is a 2-year-follow-up prospective study of 152 consecutive patients with solid tumors treated with anti-PD-(L)1 blockade agents. They were included from September 2018 until March 2020 in the Hospital de la Santa Creu I Sant Pau (Barcelona, Spain). We grouped patients into three groups: ANA de novo (patients who showed new ANA patterns at any time after ICI initiation), ANA (ANA positive at baseline without changes in the ANA patterns after initiation of treatment) and non-ANA (ANA negative at baseline and after ICI initiation). We did not find any association between the appearance of ANAs and irAE rates or the number and types of irAEs. However, patients in the ANA de novo group showed higher severe irAE rates (grade ≥ 3) than the other groups. Additionally, in most of the patients with severe irAEs (83.3%), changes in ANA patterns preceded irAE onset. In conclusion, we found ANA induction during ICI therapies in 22 patients and our results suggest that the appearance of ANAs may predict the severity of the irAE.

PD-(L)1 inhibitors are part of the treatment strategy for non-small cell lung cancer (NSCLC) although its efficacy is limited to certain patients. Our study aimed to identify patients who might benefit from anti-PD-(L)1 inhibitors by analyzing the PD-L1 expression on circulating leukocytes and its evolution during treatment. One hundred thirteen NSCLC patients, according to their radiological response after 10–12 weeks of treatment, were classified into responders, stable, and progressive disease. Percentages of circulating PD-L1⁺ leukocytes, PD-L1⁺ platelets (PLTs), and leukocyte-PLT complexes were assessed using flow cytometry, and plasma concentrations of soluble immunomodulatory factors were quantified by ELISA. Responders exhibited significantly higher pre-treatment percentages of PD-L1⁺ neutrophils, PD-L1⁺ CD14⁺ cells, and PD-L1⁺ PLTs than progressors. The percentages of these populations decreased in responders post-treatment, contrasting with stables and progressors. PLTs notably contributed to PD-L1 expression in CD14⁺ cells and neutrophils. Plasma cytokine analysis revealed baseline differences only in IL-17 concentration among groups, whereas network analyses highlighted distinct association patterns between plasma molecules and PD-L1⁺ leukocytes after 10–12 weeks of treatment. Our findings suggest that pre-treatment assessment of circulating PD-L1⁺ neutrophils, PD-L1⁺ CD14⁺ cells, and PD-L1⁺ PLTs may be helpful in identifying NSCLC patients who are potential candidates for anti-PD-(L)1 therapy.

Introduction: Despite improved efficacy with first-line immune checkpoint inhibitors plus platinum-based chemotherapy for extensive-stage small cell lung cancer (ES-SCLC), long-term survival remains limited. There is currently no available predictive biomarker to identify which patients would benefit most from this treatment. We hypothesized that pre-treatment PD-L1 expression on circulating immune cells might predict survival outcomes and toxicity. Material and methods: This prospective, multi-center observational study included patients with untreated ES-SCLC treated with first-line durvalumab plus platinum-based chemotherapy. The percentages of circulating PD-L1 ⁺ immune cells at baseline were analysed by flow cytometry to assess their association with survival outcomes and the development of immune-related adverse events (irAEs). Results: Among 41 patients with ES-SCLC, 65.9% were male, 73.2% had an ECOG-PS 1, 9.8% had central nervous system (CNS) metastases and 31.7% had liver metastases. Sixteen patients (39%) experienced irAEs. Median PFS was longer in patients with high percentages of circulating PD-L1 ⁺ monocytes compared to those with low percentages: 8.97 months (95% CI NR to NR) vs. 5.97 months (95% CI 4.65 to 7.28), p = 0.007. There was a trend toward longer median OS in patients with ES-SCLC and high percentages of circulating PD-L1 ⁺ monocytes versus low percentages: NR (95% CI NR-NR) vs. 9.13 months (95% CI 6.34 to 11.92), p = 0.092. Patients with higher circulating PD-L1 ⁺ neutrophils correlated with the development of irAES ( p = 0.007). Conclusions: Our results showed a statistically significant longer PFS in patients with ES-SCLC and high percentages of circulating PD-L1 ⁺ monocytes. This suggests PD-L1 expression on monocytes might be established as a predictive biomarker for patients with ES-SCLC treated with upfront chemo-immunotherapy. Trial registration: NCT04712903 Trial. Last registered 1 December 2025, https://www.clinicaltrials.gov/study/NCT04712903 .