Abstract
Purpose
Immune checkpoint inhibitor (ICI)-related pneumonitis (ICI-pneumonitis) is a serious complication, whose clinical course and optimal management strategies remain inadequately elucidated. This study clarified whether post-pneumonitis ICI continuation is safe and feasible, identified risk factors for fatal outcomes or successful continuation, and assessed whether findings from patients with lung cancer can be generalized to those with other malignancies.
Methods
This retrospective study analyzed data from 1,320 patients who received ICI therapy at Kyushu University Hospital between September 2014 and March 2023. Among these, 300 and 1,020 patients had lung and non-lung cancers, respectively. Clinical characteristics, treatment strategies, and outcomes were compared between the two groups. Predictors of fatal pneumonitis and successful ICI continuation were identified using univariable logistic regression.
Results
ICI-pneumonitis occurred in 96 (7.3%) patients, with a higher incidence in those with lung cancer (14.0%, [42/300]) than in those with non-lung cancers (5.3%, [54/1020]). Clinical characteristics remained similar across cancer types post-pneumonitis. Diffuse alveolar damage pattern, grade ≥ 3 pneumonitis at diagnosis, and elevated C-reactive protein level were significant predictors of fatal pneumonitis. Among 34 patients who attempted ICI continuation, 20 (58.8%) maintained ICI treatment until disease progression. Successful ICI continuation showed a trend toward association with normal baseline lung parenchyma.
Conclusions
The similar characteristics of ICI-pneumonitis across malignancies support the generalizability of management strategies. Approximately one-fifth of patients successfully continued ICI therapy after pneumonitis. Systematic radiological surveillance and appropriate severity-based treatment may help optimize outcomes in patients receiving ICI therapy.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00262-026-04375-2.
Keywords: Immune checkpoint inhibitor, Pneumonitis, Immune-related adverse event, Lung cancer
Introduction
Immune checkpoint inhibitors (ICIs) have transformed cancer treatment by blocking immune-regulatory proteins, such as programmed cell death protein 1 (PD-1), programmed death-ligand 1, and cytotoxic T-lymphocyte-associated protein 4, thereby reactivating antitumor immune responses [1]. ICIs have demonstrated remarkable efficacy against various malignancies and significantly improved survival outcomes in patients with advanced cancer [2]. However, enhanced immune activation can lead to immune-related adverse events (irAEs) that affect multiple organ systems [3]. ICI-related pneumonitis (ICI-pneumonitis) is one of the most serious complications of ICIs, with reported incidence rates of 3–5% and accounting for approximately 35% of all ICI-related mortalities [4–6].
Current management guidelines recommend ICI discontinuation for grade ≥ 2 pneumonitis, with corticosteroid therapy serving as the primary treatment [7, 8]. Permanent ICI discontinuation is typically advised for grades 3–4 pneumonitis, potentially hindering effective anticancer therapy [7–9]. The clinical course of ICI-pneumonitis and its optimal management strategies remain inadequately characterized. A key unresolved question is whether continuation or rechallenge of ICIs is safe and feasible in patients who achieve a favorable tumor response but subsequently develop ICI-pneumonitis, and which factors predict successful continuation. Despite the expanding use of ICIs across diverse malignancies, it remains unclear whether the clinical characteristics, prognostic factors, and management strategies established in patients with lung cancer are generalizable to those with non-lung cancers.
To address these knowledge gaps, we conducted a comprehensive analysis of patients with diverse malignancies and ICI-pneumonitis to examine the clinical characteristics, risk factors for fatal outcomes, and feasibility of ICI continuation across different cancer types.
Materials and methods
Study design and participants
This observational, retrospective study was conducted at Kyushu University Hospital in Japan, where 1,320 patients with advanced malignancies received ICI therapy between September 2014 and March 2023. The cohort included 300 and 1,020 patients with lung and non-lung cancers, respectively (Fig. 1). Clinical follow-up data were collected through July 31, 2025 (data cut-off). Supplementary Table 1 shows the distribution of non-lung cancer types. Clinical data, including patient demographics, treatment details, and outcomes, were retrieved from electronic medical records.
Fig. 1.
Flow chart of this study. Patient enrollment and distribution of malignancies. Flow diagram showing the selection of 96 patients with ICI-pneumonitis from the cohort of 1,320 patients who received ICI therapy for various malignancies. ICI, immune checkpoint inhibitor; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer; HCC, hepatocellular carcinoma
Study assessments
Baseline chest computed tomography (CT) scans obtained before ICI initiation were reviewed for background lung abnormalities. Emphysema was defined as areas of abnormally low attenuation without visible walls on CT images [10, 11]. Interstitial lung abnormalities (ILA) were defined as non-dependent abnormalities, including ground-glass or reticular abnormalities, lung distortion, honeycombing, and traction bronchiectasis-affecting > 5% of any lung zone [12, 13]. Definite interstitial lung disease (ILD) was diagnosed when patients met established diagnostic criteria based on international guidelines [14]. Radiological patterns of ICI-pneumonitis at diagnosis were classified based on chest CT findings using established radiological criteria [5, 11] as organizing pneumonia (OP), nonspecific interstitial pneumonia (NSIP), diffuse alveolar damage (DAD), hypersensitivity pneumonitis (HP), or acute eosinophilic pneumonia (AEP). Two independent pulmonologists (K.M. and K.T.) with > 5 years of experience in thoracic imaging, blinded to clinical outcomes, classified these patterns. Any discordant cases were resolved through consensus discussion. Interobserver agreement for radiological findings was assessed using Cohen’s κ statistic, demonstrating substantial agreement (κ = 0.782, 95% confidence interval [CI] 0.727–0.834).
Diagnosis and grading of immune checkpoint inhibitor (ICI)-related pneumonitis
ICI-pneumonitis was diagnosed based on clinical symptoms, laboratory findings, and radiological evidence of new or worsening pulmonary infiltrates, after excluding other causes such as infection, tumor progression, and pulmonary edema [7, 8, 15]. Initial diagnosis was made by the treating physicians and subsequently confirmed by a multidisciplinary team, comprising specialists in oncology, pulmonology, and radiology, as well as oncology nurses and pharmacists [16]. Pneumonitis severity was graded according to the Common Terminology Criteria for Adverse Events version 5.0 [17, 18].
Fatal pneumonitis (grade 5) was defined as death that was primarily attributable to respiratory failure secondary to ICI-pneumonitis occurring within 30 days of diagnosis, without clear evidence of rapid tumor progression. Independent pulmonologists determined the primary cause of death based on clinical course, imaging findings, and laboratory data.
Ethics statement
This study’s protocol was approved by the Institutional Review Board and Research Ethics Committee of Kyushu University Hospital (approval no. 25125). It was conducted in accordance with the Declaration of Helsinki and the research ethics guidelines at the hospital. Written informed consent was waived due to the retrospective nature of the study and minimal risk to participants. All data were anonymized prior to analysis.
Patient involvement
Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of this study.
Statistical analysis
All statistical analyses were performed using the R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was defined as a two-sided p-value < 0.05. Categorical variables are expressed as frequencies and percentages, while continuous variables are summarized as medians with ranges. Baseline characteristics and clinical outcomes were compared between patients with lung and non-lung cancers using the Student’s t-test, Mann–Whitney U test, and chi-square or Fisher’s exact tests for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectively.
Risk factors for fatal pneumonitis (grade 5) and ICI continuation were identified using univariable logistic regression. Multivariate analysis was not conducted, given the limited number of fatal events, to avoid model overfitting. Odds ratios (ORs) and 95% CIs were calculated for all variables.
Results
Participants and ICI-pneumonitis incidence
Among the 1,320 patients included, 300 and 1,020 had lung and non-lung cancers, respectively (Fig. 1). ICI-pneumonitis occurred in 96 patients (7.3%), including 42 (14.0%, 42/300) and 54 (5.3%, 54/1,020) with lung and non-lung cancers, respectively. All 42 patients with lung cancer who developed ICI-pneumonitis had non-small cell lung cancer, with no cases of small cell lung cancer. Among the patients with non-lung cancers, the highest incidence of ICI-pneumonitis was observed in urothelial carcinoma (11.0%, 8/73), followed by esophageal cancer (7.0%, 8/115), renal cell carcinoma (6.9%, 10/144), and head and neck cancer (5.2%, 12/232) (Supplementary Table 1).
Baseline patient characteristics at ICI initiation
The baseline characteristics of patients at ICI initiation are summarized in Table 1. The median age for patients with lung and non-lung cancers was 71 (range: 40–81) and 68 (range: 45–83) years, respectively. Males were predominant in the lung and non-lung cancer groups (90.5% and 85.2%, respectively). An Eastern Cooperative Oncology Group (ECOG) performance status of 2 was significantly less frequent in patients with lung cancer than in those with non-lung cancers (4.8% vs. 11.1%, p = 0.006). A history of smoking was significantly more prevalent in patients with lung cancer than in those with non-lung cancers (88.1% vs. 70.4%, p = 0.047). Baseline chest CT findings revealed significant differences in background lung parenchyma. Emphysema was significantly more common in patients with lung cancer than in those with non-lung cancers (61.9% vs. 18.5%, p < 0.001), consistent with a higher smoking prevalence. In contrast, the prevalence of interstitial findings, including ILA and ILD, was comparable in both groups (p = 0.155). Anti-PD-1 antibodies (nivolumab and pembrolizumab) were commonly administered in both groups.
Table 1.
Baseline characteristics of patients with lung and non-lung cancers
| Characteristics | Lung cancer (n = 42) |
Non-lung cancers (n = 54) |
P value |
|---|---|---|---|
| Age, years, median (range) | 71 (40–81) | 68 (45–83) | – |
| Sex, n (%) | 0.542 | ||
| Male | 38 (90.5) | 46 (85.2) | |
| Female | 4 (9.5) | 8 (14.8) | |
| BMI, kg/m2, median (range) | 21.6 (13.9–31.4) | 21.4 (13.3–35.3) | 0.700 |
| ECOG PS at ICI administration, n (%) | 0.006 | ||
| 0 | 13 (31.0) | 31 (57.4) | |
| 1 | 27 (64.2) | 17 (31.5) | |
| 2 | 2 (4.8) | 6 (11.1) | |
| Smoking, n (%) | 37 (88.1) | 38 (70.4) | 0.047 |
| Emphysema on CT, n (%) | 26 (61.9) | 10 (18.5) | < 0.001 |
| Interstitial findings on CT, n (%) | 0.155 | ||
| Normal | 37 (88.1) | 43 (79.6) | |
| ILD | 3 (7.1) | 2 (3.7) | |
| ILA | 2 (4.8) | 9 (16.7) | |
| Pulmonary metastasis, n (%) | 11 (26.2) | 22 (40.7) | 0.194 |
| ICI regimens, n (%) | 0.315 | ||
| Pembrolizumab | 18 (42.9) | 16 (29.6) | |
| Nivolumab | 14 (33.3) | 28 (51.9) | |
| Atezolizumab | 5 (11.9) | 4 (7.4) | |
| Nivolumab + Ipilimumab | 5 (11.9) | 6 (11.1) | |
| Line of ICI, n (%) | 0.006 | ||
| 1 st line | 23 (54.8) | 15 (27.8) | |
| 2nd line | 15 (35.7) | 21 (38.9) | |
| ≥ 3rd line | 4 (9.5) | 18 (33.3) |
Data are presented as median (range) for continuous variables and n (%) for categorical variables. P-values were calculated using the Mann–Whitney U test for continuous variables and the chi-square or Fisher’s exact test for categorical variables
BMI, body mass index; CT, computed tomography; ECOG, Eastern Cooperative Oncology Group; ICI, immune checkpoint inhibitor; ILA, interstitial lung abnormalities; ILD, interstitial lung disease; PS, performance status
Patient characteristics at ICI-pneumonitis diagnosis
The clinical and laboratory characteristics at the time of ICI-pneumonitis diagnosis are shown in Table 2. The median time from ICI initiation to ICI-pneumonitis onset was 79 (range: 2–1161) and 134.5 (range: 1–1041) days in patients with lung and non-lung cancers, respectively (p = 0.221). Although the overall distribution of radiological patterns on chest CT at diagnosis did not differ significantly between groups (p = 0.085), the OP pattern tended to be more prevalent in patients with lung cancer (78.6%) than in those with non-lung cancers (59.3%). The NSIP pattern was more common in patients with non-lung cancer (22.2%) than in those with lung cancer (7.1%). Other patterns, including DAD (7.1 vs. 3.7%), HP (2.4 vs. 11.1%), and AEP (4.8 vs. 3.7%), were less frequently observed in both groups. Laboratory findings, including C-reactive protein (CRP) and Krebs von den Lungen-6 (KL-6) levels, did not significantly differ between the two groups.
Table 2.
Patient characteristics at diagnosis of ICI-pneumonitis
| Lung cancer (n = 42) |
Non-lung cancers (n = 54) |
P value | |
|---|---|---|---|
| Time to onset of ICI-pneumonitis, days, median (range) | 79 (2–1161) | 134.5 (1–1041) | 0.221 |
| Radiological patterns, n (%) | 0.085 | ||
| OP | 33 (78.6) | 32 (59.3) | |
| NSIP | 3 (7.1) | 12 (22.2) | |
| DAD | 3 (7.1) | 2 (3.7) | |
| HP | 1 (2.4) | 6 (11.1) | |
| AEP | 2 (4.8) | 2 (3.7) | |
| Laboratory data, median (IQR) | |||
| KL-6, U/L | 476.5 (293.3–700.3) | 324.5 (182.0–577.0) | 0.228 |
| CRP, mg/dL | 1.6 (0.2–5.5) | 1.1 (0.2–3.2) | 0.447 |
Data are presented as median (range) for continuous variables, median (IQR) for laboratory values, and n (%) for categorical variables. P-values were calculated using the Mann–Whitney U test for continuous variables and the chi-square or Fisher’s exact tests for categorical variables
AEP, acute eosinophilic pneumonia; CRP, C-reactive protein; DAD, diffuse alveolar damage; HP, hypersensitivity pneumonitis; ICI, immune checkpoint inhibitor; IQR, interquartile range; KL-6, Krebs von den Lungen-6; NSIP, nonspecific interstitial pneumonia; OP, organizing pneumonia; SD, standard deviation
ICI-pneumonitis grade distribution and treatment
The distribution of ICI-pneumonitis grades at diagnosis and the worst grade are shown in Fig. 2. Most patients presented with grade 1 pneumonitis at diagnosis, followed by grades 3, 2, and 4. No significant difference was observed in grade distribution at diagnosis between patients with lung and non-lung cancers (p = 0.580). However, grade 1 pneumonitis at diagnosis was more common in patients with lung cancer (61.9%) than in those with non-lung cancers (50.0%). During its clinical course, ICI-pneumonitis severity worsened in a substantial proportion of patients. The worst-grade distribution showed that 27.1% (26/96) and 7.3% (7/96) of patients experienced grades 3 and 5 (fatal) pneumonitis, respectively. Specifically, the proportion of patients with worst-grade ≥ 3 pneumonitis was 35.8% and 33.3% in patients with lung and non-lung cancers, respectively, without any significant between-group difference. Grade 5 pneumonitis occurred in 4.8% and 9.2% of patients with lung and non-lung cancers, respectively. The treatment strategies for ICI-pneumonitis were determined based on the severity, as summarized in Supplementary Table 2. High-dose methylprednisolone pulse therapy (1,000 mg/day) was required in 26.2% and 33.3% of patients with lung and non-lung cancers, respectively, primarily for those with grade ≥ 3 pneumonitis. Corticosteroid therapy was not required in 31.0% and 37.0% of patients with lung and non-lung cancers, respectively. No significant between-group difference was found in the distribution of treatment regimens (p = 0.422), indicating that treatment decisions were primarily guided by pneumonitis severity rather than cancer type.
Fig. 2.
Number of patients treated with ICI and ICI-pneumonitis incidence by cancer type. Distribution of ICI-pneumonitis grades at diagnosis and at the worst point. Pneumonitis severity grades at the time of diagnosis (left) and worst point during the clinical course (right) according to CTCAE version 5.0
Risk factors for fatal ICI-pneumonitis
To identify risk factors associated with fatal ICI-pneumonitis (grade 5), we performed univariable logistic regression analysis using the patient characteristics listed in Supplementary Tables 3 and 4. The results are presented as forest plots in Fig. 3. Several factors showed associations with fatal pneumonitis. The DAD pattern on chest CT (OR: 11.47, 95% CI 1.55–85.03, p = 0.017), grade ≥ 3 pneumonitis at diagnosis (OR: 8.62, 95% CI 1.55–47.87, p = 0.014), and elevated CRP levels at diagnosis (OR: 2.09, 95% CI 1.10–3.95, p = 0.024) were associated with increased odds of fatal outcomes. An ECOG performance status of ≥ 2 (OR: 5.53, 95% CI 0.88–34.75, p = 0.068) and shorter time to ICI-pneumonitis onset (OR 0.62, 95% CI 0.37–1.02, p = 0.061) showed a trend toward higher risk. Other factors, including age, sex, smoking history, cancer type, pulmonary metastasis, emphysema, and KL-6 levels, were not significantly associated with fatal outcomes.
Fig. 3.
Univariable logistic regression analysis of risk factors for fatal ICI-pneumonitis. Forest plot showing ORs and 95% CIs for potential risk factors associated with fatal pneumonitis. CI, confidence interval; CRP, C-reactive protein; CT, computed tomography; DAD, diffuse alveolar damage; ICI, immune checkpoint inhibitor; KL-6, Krebs von den Lungen-6; OR, odds ratio; PS, performance status
Anticancer treatment after ICI-pneumonitis diagnosis
Subsequent treatment strategies for the underlying malignancy following ICI-pneumonitis are summarized in Table 3. Most patients discontinued ICI therapy after pneumonitis was diagnosed, with the best supportive care provided to 40.5% and 42.6% of patients with lung and non-lung cancers, respectively. However, a notable proportion of patients continued ICI therapy despite prior pneumonitis (35.7% and 35.2% of patients with lung and non-lung cancers, respectively). This decision was typically made in patients with grades 1–2 pneumonitis who demonstrated a favorable tumor response to ICI and experienced resolution of ICI-pneumonitis with treatment. Next-line chemotherapy with non-ICI regimens was initiated in 21.4% and 18.5% of patients with lung and non-lung cancers, respectively. A small proportion of patients (2.4% and 3.7% of patients with lung and non-lung cancers, respectively) achieved durable disease control without further anticancer treatment post-ICI discontinuation.
Table 3.
Subsequent anticancer treatment after ICI-pneumonitis
| Lung cancer (n = 42) |
Non-lung cancers (n = 54) |
P value | |
|---|---|---|---|
| Subsequent treatment, n (%) | 0.967 | ||
| No recurrence without treatment | 1 (2.4) | 2 (3.7) | |
| ICI continuation | 15 (35.7) | 19 (35.2) | |
| Next-line chemotherapy | 9 (21.4) | 10 (18.5) | |
| Best supportive care | 17 (40.5) | 23 (42.6) |
The ICI continuation or rechallenge group included patients who continued ICI treatment without interruption (grade 1 pneumonitis) or resumed ICI after pneumonitis resolution (rechallenge). Next-line chemotherapy is a non-ICI systemic therapy that is initiated after permanent ICI discontinuation. P-values were calculated using the chi-square test
ICI, immune checkpoint inhibitor
ICI treatment continuation after an ICI-pneumonitis diagnosis
Figure 4 illustrates the clinical course after an ICI-pneumonitis diagnosis. All 25 patients with grade ≥ 3 pneumonitis at diagnosis discontinued ICI therapy immediately. Among the 71 patients with grades 1–2 ICI-pneumonitis at diagnosis (53 with grade 1 and 18 with grade 2), ICI treatment was continued in 34 (47.9%) but discontinued immediately in 37 (52.1%). Of the 34 patients who continued ICI therapy, the objective response rate (ORR) was 64.7% (22/34), and all patients achieved disease control. Among these, 20 successfully maintained treatment until progressive disease without worsening ICI-pneumonitis. In this subgroup, the ORR was 60.0% (12/20), all patients achieved disease control, and the median duration of ICI continuation was 114 (interquartile range: 62.5–418.0) days. The remaining 14 patients experienced worsening ICI-pneumonitis, requiring ICI discontinuation.
Fig. 4.
Clinical course following the diagnosis of ICI-pneumonitis. Flow diagram showing treatment decisions and outcomes after ICI-pneumonitis diagnosis, stratified by the initial severity grade. ICI, immune checkpoint inhibitor; PD, progressive disease
To explore potential factors associated with successful ICI continuation, univariable logistic regression was performed comparing baseline characteristics between patients who experienced worsening pneumonitis and those who did not (Supplementary Table 5). Normal baseline lung parenchyma (absence of ILA or ILD) showed a trend of association with successful ICI continuation (OR: 0.13, 95% CI 0.01–1.34, p = 0.087). Although not statistically significant, this finding suggests that the absence of preexisting interstitial lung findings on CT may be associated with a low risk of worsening pneumonitis post-ICI continuation. Other baseline characteristics, including age, sex, smoking history, emphysema, pulmonary metastasis, ICI regimen, grade, and radiological pattern, showed no significant association with worsening pneumonitis.
Discussion
In this study, we here analyzed the clinical characteristics and outcomes of patients with diverse malignancies who experienced ICI-pneumonitis. Among the 1,320 patients who received ICI therapy, 96 (7.3%) experienced ICI-pneumonitis, with a higher incidence in those with lung cancer (14.0%, 42/300) than in those with non-lung cancers (5.3%, 54/1,020). However, once ICI-pneumonitis developed, its clinical characteristics―including severity distribution, radiological patterns, treatment requirements, and outcomes―were remarkably similar across cancer types. These findings suggest that the pathophysiology of ICI-pneumonitis is largely independent of the underlying malignancy and is primarily driven by immune-mediated mechanisms rather than tumor-specific factors. Consequently, clinical data on ICI-pneumonitis characteristics, risk factors, and prognostic factors reported in lung cancer populations may reasonably inform management decisions in patients with diverse malignancies.
The higher ICI-pneumonitis incidence in patients with lung cancer than in those with non-lung cancers (14.0 vs 5.3%) may reflect several contributing factors. First, patients with lung cancer had significantly higher rates of emphysema on baseline CT (61.9 vs 18.5%, p < 0.001) and smoking history (88.1 vs 70.4%, p = 0.047) than those with non-lung cancers, both of which have been reported as risk factors for ICI-pneumonitis in prior studies [19, 20]. Second, a numerical trend toward a higher prevalence of asymptomatic grade 1 pneumonitis at diagnosis was observed in patients with lung cancer than in those with non-lung cancers (61.9% vs 50.0%), possibly reflecting surveillance intensity variations rather than biological differences, as individuals with lung cancer typically undergo more frequent chest CT imaging for tumor response assessment. This finding highlights the importance of systematic radiological surveillance in all patients receiving ICI therapy, regardless of cancer type, as earlier detection of pneumonitis may allow timely intervention before progression to more severe grades.
The ability to safely continue ICI therapy after ICI-pneumonitis has important clinical implications, particularly for patients with a favorable tumor response and limited alternative treatment options. Despite the risk of recurrence, patients who underwent ICI rechallenge after pneumonitis had improved survival outcomes compared with those with permanent ICI discontinuation, underscoring the potential clinical benefit of this approach [21]. In the present study, ICI continuation was successfully achieved in a considerable proportion of patients who received ICI rechallenge therapy. Although rechallenge was feasible after irAEs, pneumonitis showed a relatively higher recurrence rate than other toxicities [21, 22]. The initial adverse event severity is also a key predictor of recurrence after rechallenge [6, 21]. Wu et al. reported that 50% of patients with lung cancer who attempted rechallenge after pneumonitis experienced recurrence or worsening [23]. The worsening ICI-pneumonitis observed in 41.2% (14/34) of patients in this study is consistent with this report. Baseline lung parenchymal status may also be associated with outcomes. Normal baseline lung parenchyma (absence of ILA or ILD) showed a trend toward association with successful continuation (OR: 0.13, 95% CI 0.01–1.34, p = 0.087), extending prior observations indicating that baseline ILD increases the risk of initial pneumonitis. Although this finding did not reach statistical significance, it suggests that baseline lung status can inform decisions regarding ICI continuation. Particularly, patients with low-grade pneumonitis, a favorable tumor response, and normal baseline lung parenchyma may represent a subgroup in whom cautious continuation of ICI treatment could be attempted with close monitoring.
Several risk factors for fatal ICI-pneumonitis have been previously reported. Radiologically, the DAD pattern on chest CT and baseline ILD have been consistently associated with poor outcomes [24, 25]. Patients with preexisting ILA have shown significantly higher rates of severe and fatal pneumonitis [26]. Early-onset pneumonitis has also been linked to poor survival [6, 27]. Our findings were generally consistent with these observations. The DAD pattern on chest CT was associated with fatal outcomes (OR: 11.47, 95% CI 1.55–85.03, p = 0.017). Similarly, grade ≥ 3 pneumonitis (OR: 8.62, 95% CI 1.55–47.87, p = 0.014) and elevated CRP (OR: 2.09, 95% CI 1.10–3.95, p = 0.024) at diagnosis were linked to fatal outcomes. A shorter time to pneumonitis onset showed a trend toward worse outcomes (OR 0.62 per 30-day increase, 95% CI 0.37–1.02, p = 0.061). Baseline ILD or ILA was not associated with fatal outcomes in our cohort, possibly due to the limited number of fatal events since none of the seven patients who died had preexisting ILD or ILA. Although these findings should be interpreted with caution, integrating these clinical, laboratory, and radiological factors may help guide risk stratification and decisions on treatment intensity.
This study has some limitations. First, the retrospective, single-center design introduces potential selection bias since treatment decisions were made at the physicians’ discretion rather than according to standardized protocols. However, this design also confers important strengths. Our institution has substantial experience managing ICI-pneumonitis across diverse malignancies over 9 years, and clinical practice remained relatively consistent under a dedicated multidisciplinary team throughout the study period. These factors contribute to the internal consistency of our observations. Second, the limited number of fatal events (n = 7) precluded multivariable analysis, resulting in wide CIs for several risk estimates. Therefore, the identified predictors of fatal pneumonitis should be viewed as hypothesis-generating rather than validated and require confirmation in larger cohorts. Third, the analysis of ICI continuation was limited by a small sample size (n = 34 attempts) and inherent selection bias, as clinicians likely selected patients with more favorable tumor response and performance status for continuation. Corticosteroid administration at the time of ICI continuation was not standardized. Notably, several patients in the successfully maintained group were receiving low-dose corticosteroids, whereas none in the discontinuation group received such treatment. Such differences in management strategies may have influenced the tolerability of ICI continuation. Despite these limitations, our findings provide practical guidance for real-world clinical decision-making and warrant validation in larger, multi-center prospective studies.
In conclusion, this study demonstrated that the characteristics of ICI-pneumonitis were remarkably similar between patients with lung and non-lung cancers, supporting the generalizability of management strategies across malignancies. Approximately one-fifth of patients successfully continued ICI therapy until disease progression, indicating that selective continuation of ICIs is feasible. Our findings suggest that systematic radiological surveillance for early detection, severity-guided treatment, and individualized decisions regarding ICI continuation may contribute to improved outcomes in patients receiving ICI therapy.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
None.
Abbreviations
- AEP
Acute eosinophilic pneumonia
- BMI
Body mass index
- BSC
Best supportive care
- DAD
Diffuse alveolar damage
- ECOG
Eastern cooperative oncology group
- HP
Hypersensitivity pneumonitis
- ILA
Interstitial lung abnormalities
- ILD
Interstitial lung disease
- ICI
Immune checkpoint inhibitor
- irAE
Immune-related adverse events
- KL-6
Krebs von den Lungen-6
- NSIP
Nonspecific interstitial pneumonia
- OP
Organizing pneumonia
- PD-1
Programmed cell death protein 1
- PS
Performance status
Author contributions
K.M.: Conceptualization, Formal analysis, Investigation, Data curation, Resources, Visualization, Writing the original draft, Writing the review and editing. K.T.: Conceptualization, Formal analysis, Investigation, Data curation, Resources, Project administration, Visualization, Writing the original draft, Writing the review and editing. H.O.: Formal analysis, Visualization, Writing the review and editing. R.M., S.Y., T.T.: Investigation, Data curation, Writing the review and editing. M.U.: Supervision, Writing the review and editing. I.O.: Supervision, Visualization, Writing the review and editing.
Funding
None.
Data Availability
Statement: Data may be obtained from a third party and are not publicly available.
Declarations
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Pardoll DM (2012) The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12:252–264. 10.1038/nrc3239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Robert C (2020) A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun 11:3801. 10.1038/s41467-020-17670-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Postow MA, Sidlow R, Hellmann MD (2018) Immune-related adverse events associated with immune checkpoint blockade. N Engl J Med 378:158–168. 10.1056/NEJMra1703481 [DOI] [PubMed] [Google Scholar]
- 4.Nishino M, Giobbie-Hurder A, Hatabu H et al (2016) Incidence of programmed cell death 1 inhibitor-related pneumonitis in patients with advanced cancer: a systematic review and meta-analysis. JAMA Oncol 2:1607–1616. 10.1001/jamaoncol.2016.2453 [DOI] [PubMed] [Google Scholar]
- 5.Naidoo J, Wang X, Woo KM et al (2017) Pneumonitis in patients treated with anti–programmed death-1/programmed death ligand 1 therapy. J Clin Oncol 35:709–717. 10.1200/JCO.2016.68.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang DY, Salem J-E, Cohen JV et al (2018) Fatal toxic effects associated with immune checkpoint inhibitors: a systematic review and meta-analysis. JAMA Oncol 4:1721–1728. 10.1001/jamaoncol.2018.3923 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Haanen J, Obeid M, Spain L et al (2022) Management of toxicities from immunotherapy: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 33:1217–1238. 10.1016/j.annonc.2022.10.001 [DOI] [PubMed] [Google Scholar]
- 8.Schneider BJ, Naidoo J, Santomasso BD et al (2021) Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO Guideline Update. J Clin Oncol 39:4073–4126. 10.1200/JCO.21.01440 [DOI] [PubMed] [Google Scholar]
- 9.Puzanov I, Diab A, Abdallah K et al (2017) Managing toxicities associated with immune checkpoint inhibitors: consensus recommendations from the Society for Immunotherapy of Cancer (SITC) Toxicity Management Working Group. J Immunother Cancer 5:95. 10.1186/s40425-017-0300-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lynch DA, Austin JHM, Hogg JC et al (2015) CT-definable subtypes of chronic obstructive pulmonary disease: a statement of the Fleischner Society. Radiology 277:192–205. 10.1148/radiol.2015141579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Johkoh T, Lee KS, Nishino M et al (2021) Chest CT diagnosis and clinical management of drug-related pneumonitis in patients receiving molecular targeting agents and immune checkpoint inhibitors: a position paper from the Fleischner Society. Radiology 298:550–566. 10.1148/radiol.2021203427 [DOI] [PubMed] [Google Scholar]
- 12.Podolanczuk AJ, Hunninghake GM, Wilson KC et al (2025) Approach to the evaluation and management of interstitial lung abnormalities: an Official American Thoracic Society Clinical Statement. Am J Respir Crit Care Med 211:1132–1155. 10.1164/rccm.202505-1054ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hatabu H, Hunninghake GM, Richeldi L et al (2020) Interstitial lung abnormalities detected incidentally on CT: a Position Paper from the Fleischner Society. Lancet Respir Med 8:726–737. 10.1016/S2213-2600(20)30168-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Travis WD, Costabel U, Hansell DM et al (2013) An Official American Thoracic Society/European Respiratory Society Statement: update of the International Multidisciplinary Classification of the Idiopathic Interstitial Pneumonias. Am J Respir Crit Care Med 188:733–748. 10.1164/rccm.201308-1483ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang H, Guo X, Zhou J et al (2020) Clinical diagnosis and treatment of immune checkpoint inhibitor-associated pneumonitis. Thorac Cancer 11:191–197. 10.1111/1759-7714.13240 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Matsukane R, Nakamura S, Minami H et al (2024) Krebs von den Lungen-6 surveillance in immune checkpoint inhibitor-induced pneumonitis. J Immunother Cancer 12:e010114. 10.1136/jitc-2024-010114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gosangi B, McIntosh LJ, Keraliya A et al (2022) Imaging features of toxicities associated with immune checkpoint inhibitors. Eur J Radiol Open 9:100434. 10.1016/j.ejro.2022.100434 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Naidoo J, Murphy C, Atkins MB et al (2023) Society for Immunotherapy of Cancer (SITC) consensus definitions for immune checkpoint inhibitor-associated immune-related adverse events (irAEs) terminology. J Immunother Cancer 11:e006398. 10.1136/jitc-2022-006398 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Stahlbaum D, Jablonski R, Strek ME, Bestvina CM, Polley MY, Reid P (2023) Abnormalities on baseline chest imaging are risk factors for immune checkpoint inhibitor associated pneumonitis. Respir Med 217:107330. 10.1016/j.rmed.2023.107330 [DOI] [PubMed] [Google Scholar]
- 20.Uchida Y, Kinose D, Nagatani Y et al (2022) Risk factors for pneumonitis in advanced extrapulmonary cancer patients treated with immune checkpoint inhibitors. BMC Cancer 22(1):551. 10.1186/s12885-022-09642-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dolladille C, Ederhy S, Sassier M et al (2020) Immune checkpoint inhibitor rechallenge after immune-related adverse events in patients with cancer. JAMA Oncol 6(6):865–871. 10.1001/jamaoncol.2020.0726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Simonaggio A, Michot J-M, Voisin A-L et al (2019) Evaluation of readministration of immune checkpoint inhibitors after immune-related adverse events in patients with cancer. JAMA Oncol 5(9):1310–1317. 10.1001/jamaoncol.2019.1022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wu G, Qu J, Zheng J et al (2025) Immunotherapy rechallenge after ICI-related pneumonitis in lung cancer patients: a retrospective cohort study. Front Oncol 15:1527690. 10.3389/fonc.2025.1527690 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Fujimoto D, Miura S, Tomii K et al (2023) Pneumonitis associated with pembrolizumab plus chemotherapy for non-squamous non-small cell lung cancer. Sci Rep 13(1):3698. 10.1038/s41598-023-30676-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yamaguchi T, Shimizu J, Hasegawa T et al (2018) Pre-existing pulmonary fibrosis is a risk factor for anti-PD-1-related pneumonitis in patients with non-small cell lung cancer: a retrospective analysis. Lung Cancer 125:212–217. 10.1016/j.lungcan.2018.10.001 [DOI] [PubMed] [Google Scholar]
- 26.Nakanishi Y, Masuda T, Yamaguchi K et al (2019) Pre-existing interstitial lung abnormalities are risk factors for immune checkpoint inhibitor-induced interstitial lung disease in non-small cell lung cancer. Respir Investig 57:451–459. 10.1016/j.resinv.2019.05.002 [DOI] [PubMed] [Google Scholar]
- 27.Suresh K, Psoter KJ, Voong KR et al (2019) Impact of checkpoint inhibitor pneumonitis on survival in NSCLC patients receiving immune checkpoint immunotherapy. J Thorac Oncol 14(3):494–502. 10.1016/j.jtho.2018.11.016 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Statement: Data may be obtained from a third party and are not publicly available.