PublishedOctober 7, 2026
JournalThe Oncologist
TrialProspective multicenter observational cohort (UMIN000039761)

Baseline Interstitial Lung Abnormalities Are Associated With Higher Checkpoint Inhibitor Pneumonitis Risk in Thoracic Cancers, in a Prospective Observational Cohort of 614 Patients

How this article was produced

How this article was produced

This article was written about a single named primary publication, with AI assistance for literature discovery, drafting and editing. AI tools are not the author and are not responsible for the conclusions.

How the draft was prepared

  • Discovery: PubMed is searched for oncology publications from the previous 14 days in a fixed list of major journals — JAMA Oncology, The Lancet Oncology, Journal of Clinical Oncology, Annals of Oncology, Nature Reviews Clinical Oncology, New England Journal of Medicine, Nature Medicine, Cancer Discovery and Clinical Cancer Research — falling back to a named set of open-access journals only if that list yields nothing. Candidates are ranked by how completely they report quantitative results, and studies already covered are excluded.
  • Source: the article is drafted from the open full text in PubMed Central where one exists. Where only the abstract is available, it is supplemented with cited external sources — the trial registry record, conference abstracts, sponsor releases — and facts drawn from those sources are labelled as such.
  • Numbers are reported exactly as the source gives them, with confidence intervals and p-values, and are not rounded. Data the source does not report is stated as not specified rather than estimated.
  • Subgroup statistics are reported only where the source states them explicitly, and quotations are reproduced verbatim.
  • Benchmarks from other studies are labelled as cross-study comparisons and never presented as results of the trial itself.
  • Validation: every factual claim in the draft is checked against the source, and key claims are spot-checked against independent sources such as the trial registry. Fixable discrepancies are corrected before the draft is handed on for publication.

These steps are instructions followed by an AI system, not checks enforced by software. The validation step is itself performed by AI rather than a person, and its verdict does not by itself stop an article from being published. This describes the current workflow; some earlier articles predate parts of it.

Checks completed before publication

  • The primary publication is recorded in structured form — journal, publication date, DOI, PubMed ID and PMC ID where they exist — rather than only as prose in a reference list.
  • Every cited DOI is confirmed to resolve.
  • Every cited PubMed ID is confirmed to exist in PubMed.
  • The cited DOI is cross-checked against the DOI that PubMed records for that PubMed ID, so a citation cannot point at a different paper from the one the article describes.
  • Trial registry identifiers are format-checked and linked to their ClinicalTrials.gov record.
  • Every reference must carry a resolvable identifier or link, unless it is explicitly labelled as a funding note, press release or background statistic.
  • The article must contain a limitations discussion specific to the study — what it could not establish — not a generic disclaimer.
  • Absolutist claims such as “proves”, “guarantees”, “will cure” or “completely effective” are rejected.
  • HTML markup and entities that would render as literal text — such as an unsupported tag or " " — are rejected outside the reference list.
  • Structural checks confirm the article has a unique URL and complete metadata.

These run automatically on every article. If any of them fails, the build fails and the article is held rather than published.

What these checks do not cover

These checks confirm that the sources are real and correctly cited. They do not confirm that the analysis drawn from them is sound. This article has not been read against its primary source by a human editor, and should be read as a signpost to the primary publication rather than a substitute for it.

Evidence note

This article summarizes research findings and does not provide medical advice or treatment recommendations. Investigational findings, preliminary results, and preclinical evidence should not be interpreted as established clinical care. Always consult the cited primary source, and a qualified clinician, before acting on anything described here.

Corrections

Report a factual error or a sourcing concern through our corrections policy. Substantive corrections are dated and described on the article page. Our editorial policy explains how these articles are sourced, verified, and corrected.

Article Summary
  • In a prospective, multicenter Japanese cohort of 614 assessable patients with unresectable advanced or recurrent thoracic malignancies starting nivolumab, pembrolizumab or atezolizumab, 78 (12.7%) developed immune checkpoint inhibitor–related pneumonitis (ICI-P) over a median observation period of 14.4 months. Grade ≥3 pneumonitis occurred in 29 patients and 7 patients died owing to ICI-P.
  • In multivariable Fine-Gray analysis, baseline interstitial lung abnormalities (ILAs) on HRCT (SHR 3.42; 95% CI, 2.05-5.71; P < .001), PD-1 inhibitor use (SHR 2.15; 95% CI, 1.28-3.63; P = .004), immunochemotherapy (SHR 2.16; 95% CI, 1.17-3.98; P = .013) and more than 50 pack-years of smoking (SHR 1.90; 95% CI, 1.17-3.10; P = .009) were independently associated with ICI-P.
  • This was an observational study. Regimen selection was at clinician discretion, so selection bias and unmeasured confounding cannot be fully excluded; the cohort was Japanese, there was no external validation, and the findings are associations that do not establish causation.

Why This Study Matters

Checkpoint inhibitors (ICIs) are a mainstay of treatment for thoracic malignancies, and pneumonitis is one of their most serious complications. The authors note that in clinical trials, which mostly excluded patients with preexisting lung disease, ICI-P incidence has been reported at approximately 3%, while real-world reports range from 3% to 23%. Most of those data came from retrospective studies, and the few prospective studies mostly involved ICI monotherapy. Whether the picture holds in current practice, where immunochemotherapy is widely used, remained unclear.

The study also focused on interstitial lung abnormalities (ILAs): incidental high-resolution CT (HRCT) findings affecting at least 5% of any lung zone in people without clinical suspicion of interstitial lung disease, which the authors say are often overlooked on conventional CT.

1
Among 614 eligible patients, 78 (12.7%) developed ICI-related pneumonitis over a median observation period of 14.4 months.
2
29 patients had grade ≥3 pneumonitis and 7 patients died owing to ICI-P.
3
Independent risk factors (multivariable Fine-Gray): baseline ILAs (SHR 3.42; 95% CI, 2.05-5.71; P < .001), PD-1 inhibitor use (SHR 2.15; 95% CI, 1.28-3.63; P = .004), immunochemotherapy (SHR 2.16; 95% CI, 1.17-3.98; P = .013), and heavy smoking history, >50 pack-years (SHR 1.90; 95% CI, 1.17-3.10; P = .009).

Study Design

This multicenter prospective observational study ran at 15 hospitals in Japan between May 2019 and July 2023, with a data cutoff of September 30, 2024. It was not a randomized trial. Patients were enrolled consecutively. All underwent HRCT at ICI initiation, and additional HRCT was performed when new opacities were found or ICI-P was suspected. Baseline scans were read by 2 independent radiologists and 1 pulmonologist blinded to clinical information. The primary endpoint was ICI-P incidence; secondary endpoints included risk factors and the characteristics, clinical courses and radiological features of ICI-P. Risk factors were analysed with Fine-Gray competing-risks regression, with death from any cause and discontinuation of ICIs for reasons other than ICI-P as competing risks. Registry ID: UMIN000039761 (UMIN Clinical Trial Registry); no ClinicalTrials.gov ID was reported. Funding: the authors state that the research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Patient Population

Of 632 patients registered, 614 were assessable. Eligible patients were ≥20 years old with unresectable advanced or recurrent thoracic malignancies scheduled for nivolumab, pembrolizumab or atezolizumab; patients with previously diagnosed or clinically suspected ILD, and those receiving durvalumab, were excluded. Median age was 72 years (range 34-90) and 475 (77.4%) were male. Histology was NSCLC in 542 (88.3%), small cell carcinoma in 51 (8.3%) and mesothelioma in 21 (3.4%). PD-1 inhibitors were given to 366 patients (59.6%) and the PD-L1 inhibitor atezolizumab to 248 (40.4%); 417 (67.9%) received immunochemotherapy and 197 (32.1%) monotherapy. Baseline HRCT showed ILAs in 81 patients (13.2%) and emphysema (>10% of lung field) in 301 (49.0%).

Primary Endpoint: Incidence of ICI-P

78 of 614 patients (12.7%) developed ICI-P; in the NSCLC cohort the incidence was 13.1%. Cumulative incidence at 1, 3 and 6 months was 2.1% (95% CI, 1.2%-3.5%), 6.5% (95% CI, 4.5%-8.3%) and 10.1% (95% CI, 7.9%-12.6%). Median time to onset from ICI initiation was 3.0 months. Incidence was 14.9% with immunochemotherapy versus 8.1% with PD-1/PD-L1 monotherapy (P = .019), and cumulative incidence differed significantly between these groups (Gray's test, P = .016).

Risk Factor Analysis

Predictor (Fine-Gray, n = 614)Univariable SHR (95% CI)PMultivariable SHR (95% CI)P
Pack-years >502.12 (1.36-3.31)<.0011.90 (1.17-3.10).009
Interstitial lung abnormality3.67 (2.27-5.96)<.0013.42 (2.05-5.71)<.001
Residual normal lung <50%2.79 (1.62-4.78)<.0011.71 (0.91-3.20).09
PD-1 vs PD-L1 inhibitor2.23 (1.33-3.73).0022.15 (1.28-3.63).004
Combined with cytotoxic agents1.89 (1.09-3.28).0242.16 (1.17-3.98).013
Age ≥75 vs <75 years1.00 (0.62-1.60).981.00 (0.60-1.67)1
Performance status ≥2 vs <20.97 (0.41-2.29).951.17 (0.45-3.01).75

The multivariable model was adjusted for age and ECOG-PS. Squamous histology and prior thoracic radiotherapy were not associated with ICI-P in this cohort.

Subgroup Analyses

A subgroup analysis restricted to NSCLC yielded similar results; reduced residual normal lung volume (<50%) was also significantly associated with ICI-P in that subgroup (SHR 1.91; 95% CI, 1.00-3.62; P = .048). ILA subtype data (from supplementary Table S2) showed subpleural fibrotic ILAs in 25.6% of the ICI-P group versus 8.0% of the non-ICI-P group, and cumulative incidence was highest for subpleural fibrotic ILAs (Gray's test, P < .001). Other subgroup-level statistics are in the paper's supplementary material and are not reproduced here.

Baseline characteristicICI-P (n = 78)Non-ICI-P (n = 536)P
>50 pack-years36 (46.2%)149 (27.8%).001
Interstitial lung abnormality25 (32.1%)56 (10.4%)<.001
Immunochemotherapy62 (79.5%)355 (66.2%).019
PD-1 inhibitor59 (75.6%)307 (57.3%).002
Oncogenic driver alteration (n = 475 tested)4 (5.1%)92 (17.2%).002

Safety Profile of ICI-P

Item (patients with ICI-P, n = 78)n (%)
CTCAE grade 1 / 218 (23.1) / 31 (39.7)
CTCAE grade 3 / 4 / 517 (21.8) / 5 (6.4) / 7 (9.0)
Organizing pneumonia pattern42 (53.8)
Nonspecific interstitial pneumonia15 (19.2)
Hypersensitivity pneumonitis13 (16.7)
Simple pulmonary eosinophilia / diffuse alveolar damage5 (6.4) / 3 (3.8)
Prednisolone / methylprednisolone pulse57 (73.1) / 15 (19.2)
Recovered70 (89.7)

Grade ≥3 pneumonitis occurred in 29 patients (37.2% of those with ICI-P). Seven patients (9.0%) died of ICI-P. ICI rechallenge was attempted in 7 patients (9.0%), of whom 2 (2.6%) had recurrence.

Interpretation and Broader Context

In conclusion, this prospective study suggests that baseline ILAs on CT, heavy smoking history, and treatment type (PD-1 inhibitor and immunochemotherapy) are independent risk factors for ICI-P.
— Kitahara et al., The Oncologist, 2026 (Discussion)

The authors write that the findings highlight the importance of evaluating baseline CT for ILAs and smoking history before starting ICI therapy.

Limitations

Regimen selection was at clinician discretion, so selection bias and unmeasured confounding cannot be fully excluded. The cohort was Japanese, which may limit generalizability, and the study was a single cohort without external validation. Few patients received dual ICI regimens (crude incidence 7/37, 18.9%), limiting conclusions there. Patients with oncogenic driver alterations, mostly EGFR, had lower ICI-P incidence, which the authors say may reflect treatment selection rather than a protective effect. Findings are associations from an observational study and do not establish causation.

Primary source: DOI 10.1093/oncolo/oyag377 · PubMed 42776614 · PMC13645426

References

  1. Kitahara Y, Inui N, Koyauchi T, et al. Incidence of and risk factors for immune checkpoint inhibitor–related pneumonitis in thoracic malignancies: a prospective multicenter study. The Oncologist. 2026. DOI: 10.1093/oncolo/oyag377. PMID: 42776614. PMCID: PMC13645426.
  2. UMIN Clinical Trials Registry: UMIN000039761 (no ClinicalTrials.gov identifier reported).