A retrospective study of combination therapy with glucocorticoids and pirfenidone for PD-1 inhibitor-related immune pneumonitis

Abstract

Immune checkpoint inhibitor pneumonitis (ICIP) is thought to be a self-limiting disease; however, an effective treatment option does not currently exist. This study aimed to determine the clinical efficacy of combination therapy with glucocorticoids and pirfenidone for ICIP related to programmed cell death protein-1 (PD-1) inhibitors. We conducted a retrospective analysis of 45 patients with advanced non-small cell lung cancer who developed ICIP following PD-1 inhibitor and albumin-bound paclitaxel or carboplatin treatment at our hospital. The PD-1 inhibitor was discontinued, and glucocorticoids were used alone or in combination with pirfenidone to treat ICIP. The relevant clinical data of these patients were collected and analyzed. Compared with the glucocorticoid alone group, the glucocorticoid-pirfenidone group showed significant improvement in forced vital capacity (FVC), carbon monoxide diffusing capacity [%], peripheral capillary oxygen saturation, and 6-minute walk distance (P < .05). There were benefits with respect to the St. George’s Respiratory Questionnaire score and the recurrence rate of ICIP, but there was no significant difference between the 2 groups (P > .05). Adding pirfenidone to glucocorticoid treatment was shown to be safe and may be more beneficial than glucocorticoids alone for improving pulmonary interstitial lesions, reversing ICIP, and preventing its recurrence.

1. Introduction

Lung cancer is currently the most common and deadly malignant tumor worldwide, thus posing a serious threat to human health.^[1] The use of immune checkpoint inhibitors (ICIs) for immune therapy in advanced lung cancer patients has been shown to result in significant long-term survival benefits.^[2] While ICIs block immune checkpoints and enhances tumor-specific immune responses, ICIs also nonspecifically activate the immune system, leading to the disruption of immune homeostasis and the development of immune-related adverse events (irAEs), which can affect almost all organ systems. The incidence of immune checkpoint inhibitor pneumonitis (ICIP) is found to be higher in non-small cell lung cancer (NSCLC) and renal cell carcinoma than other cancers; and is higher during combination therapy than monotherapy.^[3] The incidence of ICIP was 2.92% for all-grade and 1.53% for high-grade pneumonitis according to a meta-analysis.^[4]

ICIP is a potentially fatal irAE from programmed death 1 (PD-1)/programmed death ligand 1 immunotherapy.^[5,6] Although the pathologic and physiologic mechanisms underlying ICIP have not been established, ICIP is thought to be a self-limiting disease.^[7] No prospective trials have been conducted to determine the best treatment option for ICIP. The majority of ICIP cases resolve or improve with corticosteroids^[8]; however, a small percentage of patients may develop recurrent pneumonitis when ICIs are reintroduced.^[5,8] The domestic^[9] and abroad^[10] experts consensus for ICIP recommend corticosteroids as the primary therapy approach. If clinical improvement does not occur after 2 to 7 days of monitoring, the corticosteroid dose should be increased and immunosuppressive drugs added.^[9] The key to lowering the incidence of ICIP and ultimately ensuring the efficacy of immunotherapy for NSCLC is early detection and treatment. To standardize terminology regarding treatment-related adverse events, ICIP is graded according to international guideline on idiopathic pulmonary fibrosis treatment.^[11] Patients with grade 1 or 2 pneumonitis have no or milder symptoms and are typically managed as outpatients, while patients with grade 3 (severe symptoms and an oxygen requirement) or grade 4 (life-threatening respiratory compromise necessitating urgent intervention [intubation or tracheostomy]) require more intensive management.^[12]

Pirfenidone is an orally administered small molecule compound with multifunctional anti-fibrotic, anti-inflammatory, and antioxidant properties. Pirfenidone has been confirmed to have anti-inflammatory and anti-fibrotic effects.^[13,14] Chinese expert consensus on the diagnosis and treatment of idiopathic pulmonary fibrosis^[15] recommend the use of pirfenidone in the treatment of idiopathic pulmonary fibrosis, and it has been widely used in clinical practice. The efficacy pirfenidone is validated in patients with mild-to-moderate impairment of pulmonary functions with adverse events such as photosensitivity and anorexia; the efficacy in patients with severe impairment is not confirmed, but the treatment is demonstrated to be well tolerated.^[16] According to the study by Yu et al,^[17] a patient with grade 3 ICIP was treated with an initial high-dose glucocorticoid pulse and oral pirfenidone (300 mg thrice-daily) for > 11 months, and the patient demonstrated a considerable improvement based on computed tomography imaging and clinical symptoms.

Herein the clinical efficacy of pirfenidone used alone or in combination with glucocorticoids in the treatment of ICIP was retrospectively analyzed, thus providing new clinical ideas for the traditional treatment of immune pneumonia.

2. Materials and methods

2.1. Patients

This retrospective study was approved by the Institutional Ethics Committee of Fujian Medical University Union Hospital (No. 2022KY033).

Once anti-PD(L)1 medication was administered, the diagnosis of ICIP pneumonitis was made based on clinical and radiographic evidence of lung inflammation after other diagnoses, such as infection and malignancy, were ruled out. The following criteria constituted ICIP symptoms: worsening dyspnea; ongoing need for supplemental oxygen; persistent exertional desaturation; and lack of heart failure or anemia. Resolution of ICIP was defined as full weaning from steroids followed by at least 1 month without the appearance of any new lung abnormalities on a chest CT scan or increasing dyspnea.^[5] According to the Chinese Experts Consensus on Immune Checkpoint Inhibitors for Non-small Cell Lung Cancer (2020 Version),^[9] a total of 45 patients with stage III/IV pulmonary cancer who developed ICIP after PD-1 immune checkpoint inhibitor immunotherapy were included in this retrospective study. The study period was from May 2018 to January 2021. The patients who were pathologically-diagnosed with NSCLC, all of whom were first-time patients, were included in this study. Those patients with severe abnormalities in the heart, liver, kidneys, blood, endocrine system, or rheumatic connective tissues were excluded, as well as patients who were uncooperative or had a mental impairment.

2.2. Treatment methods

All patients were diagnosed and treated according to the expert consensus on the diagnosis and treatment of immune checkpoint inhibitor-related pneumonia in lung cancer patients.

Symptomatic supportive treatment was provided for grade 1 ICIP with close follow-up evaluation. Intravenous administration of methylprednisolone (1–2 mg/kg/d) was administered first for grade 2 ICIP, followed by oral administration of methylprednisolone (1–2 mg/kg/d) after 48 to 72 hours. Steroid treatment was gradually tapered (10 mg/week) after the symptoms abated and imaging improved; the treatment course was > 6 weeks. Intravenous administration of methylprednisolone (2–4 mg/kg/d) was administered first for grade 3 and 4 ICIP, followed by oral administration of methylprednisolone (2–4 mg/kg/d) after 48 to 72 hours. Steroid treatment was gradually tapered (10 mg per week) after the symptoms abated and imaging improved; the treatment course was > 8 weeks. There were 30 cases in the glucocorticoid alone group and 15 cases in the glucocorticoid-pirfenidone group, in which pirfenidone (approval number: National Drug Approval H20133375; Beijing Kontine Pharmaceutical Co., Ltd., Beijing) was added to the steroid treatment. The treatment regimen consisted of pirfenidone (100 mg) administered thrice-daily for a duration of 7 days, followed by pirfenidone (200 mg) administered thrice-daily for 14 days. Maintenance pirfenidone therapy was then initiated at a dose of 300 mg administered thrice-daily.

2.3. Observation indicators

Changes in clinical indicators before and after treatment were demonstrated based on lung function tests, pulse oximetry, the 6-minute walk test, and symptom assessment 4 weeks later. The major observation indicators included forced vital capacity (FVC) (L) and diffusing capacity of the lung for carbon monoxide (DLCO [%]). The secondary observation indicators included peripheral capillary oxygen saturation (SPO₂ [%]), 6-minute walk distance (6MWD [m]), and St. George’s Respiratory Questionnaire (SGRQ) score. For patients with grade 1 and 2 ICIP, immunotherapy was restarted after ICIP reversal and steroid cessation for 4 and 8 weeks, respectively. No further immunotherapy was given for grade 3 ICIP. The recurrence rate of ICIP was determined in both groups at 4, 8, and 12 weeks after restarting immunotherapy. During treatment, the patients’ conditions were closely monitored, and their oxygen saturation levels were checked regularly. Follow-up evaluations were performed until January 2022.

2.4. Statistical methods

SPSS 24.0 statistical software was used. Normally distributed data are expressed as the mean ± standard deviation, and non-normally distributed data are expressed as median (IQR). Quantitative data were compared using the t-test or the Wilcoxon rank-sum test, while categorical data were compared using the chi-square test or the Wilcoxon nonparametric test. A P < .05 was used to define statistical significance for a difference.

3. Results

3.1. General information

According to the inclusion and exclusion criteria, the retrospective analysis included 45 ICIP patients, consisting of 34 males and 11 females (average age = 64.5 ± 17.64 years) (Fig. 1).

Figure 1.

Flow chart of the patients included and followed up. ICIP = immune checkpoint inhibitor pneumonitis.

There were 40 patients with adenocarcinoma and 5 patients with brain metastases, as well as 25 and 20 patients with stage III and IV squamous cell carcinoma, respectively. There were 9, 30, and 6 patients with ICIP grade 1, 2, and 3, respectively. Thirty-eight patients had respiratory symptoms and 7 had no apparent symptoms. Of those patients with respiratory symptoms, 33 had dyspnea, 13 had coughing, 12 had chest tightness or pain, 8 had fatigue, and 5 had fevers. After treatment, all patients had symptomatic improvement. All grade 1 and 2 ICIP patients resumed immunotherapy, while grade 3 ICIP patients did not continue with immunotherapy. The baseline data for the 2 groups are shown in Table 1. The results showed that the 2 groups were balanced with respect to baseline data, such as gender, age, pathologic type, whether there was or was not brain metastasis, stage, and ICIP grade.

Table 1.

Summary of baseline data for the 2 groups of patients.

General information		Glucocorticoid-pirfenidone group (n = 15)	Glucocorticoid alone group (n = 30)	χ2/t	P
Gender	Male	11	23	0.0596	.8071
	Female	4	7
Pathology	Squamous carcinoma	2	3	0.1125	.7373
	Adenocarcinoma	13	27
Brain metastasis	No	13	27	0.1125	.7373
	Yes	2	3
Staging	Stage III	7	18	0.7200	.3961
	Stage IV	8	12
ICIP classification	1	3	6	0.0000	1.0000
	2	10	20
	3	2	4
Age (year)		65.93 ± 8.43	63.80 ± 7.27	0.88	.3836

Note: Data are presented as mean ± standard deviation or number of patients.

DLCO = diffusing capacity of the lung for carbon monoxide, FEV1 = forced expiratory volume in 1 second, FVC = forced vital capacity, ICIP = immune checkpoint inhibitor pneumonitis, PaCO2 = arterial carbon dioxide partial pressure, PaO2 = arterial oxygen partial pressure, SaO2 = arterial oxygen saturation.

3.2. Clinical changes in relevant indicators

Both groups showed great improvements in each parameter after treatment compared to before treatment (Table 2). The glucocorticoid–pirfenidone group showed significant improvements in FVC, DLCO, 6MWD, and SPO₂ without oxygen supplementation after treatment compared to the glucocorticoid alone group (P < .05, Table 3). There was also an improvement in the SGRQ score, but the difference was not significant (P > .05; Table 3). Among the grade 2 ICIP patients, the glucocorticoid–pirfenidone group had significantly shorter symptom improvement days compared to the glucocorticoid alone group (7.20 ± 0.63 d vs 7.90 ± 0.79 d; P < .05; Table 4). The results were comparable between the 2 groups of grade 3 ICIP patients (P > .05; Table 4). Patients with grade 1 ICIP were not included in this study due to a lack of apparent symptoms.

Table 2.

The data comparison before and after treatment (4 weeks) in 2 groups of patients.

Indicators	Glucocorticoid–pirfenidone group (n = 15)		Glucocorticoid alone group (n = 30)
	Before treatment	After treatment	Before treatment	After treatment
∆FVC (L)	1.75 ± 0.19	1.89 ± 0.17	1.81 ± 0.27	1.92 ± 0.28
∆DLCO (%)	58.05 ± 3.07	69.65 ± 2.86	59.23 ± 3.15	70.03 ± 3.16
∆6MWD (m)	280.47 ± 31.56	327.53 ± 30.99	287.17 ± 20.78	326.27 ± 22.10
∆SPO2 (%)	90.47 ± 0.92	95.93 ± 0.96	90.83 ± 0.75	95.53 ± 0.73
∆SGRQ score	40.87 ± 5.73	36.93 ± 5.38	42.97 ± 6.24	39.97 ± 5.99

DLCO = diffusing capacity of the lung for carbon monoxide, FVC = forced vital capacity, SGRQ = St. George’s Respiratory Questionnaire, SPO2 = saturation of peripheral oxygen.

Table 3.

Changes in data before and after treatment (4 weeks) in 2 groups of patients.

∆Before and after treatment	Glucocorticoid–pirfenidone group (n = 15)	Glucocorticoid alone group (n = 30)	t/Z	P
∆FVC (L)	0.14 ± 0.05	0.12 ± 0.04	2.10	.0416*
∆DLCO (%)	11.60 ± 0.60	10.80 ± 1.23	2.93	.0054*
∆6MWD (m)	47.07 ± 9.58	39.10 ± 12.11	2.22	.0317*
∆SPO2 (%)†	5.47 ± 1.25	4.70 ± 0.99	2.25	.0298
∆SGRQ score†	3.93 ± 1.71	3.00 ± 1.15	2.198	.0349
∆SPO2 (%)†	5 (5,7)	4 (5,5)	2.3145	.0206*
∆SGRQ score†	3 (4,5)	2 (2,4)	1.8151	.0695

DLCO = diffusing capacity of the lung for carbon monoxide, FVC = forced vital capacity, SGRQ = St. George’s Respiratory Questionnaire, SPO2 = saturation of peripheral oxygen.

^†Wilcoxon rank sum test.

^*P<.05.

Table 4.

Time to significant improvement in symptoms after treatment (days).

ICIP grade	Glucocorticoid-pirfenidone group (n = 15)	Glucocorticoid alone group (n = 30)	t	P
2†	7.20 ± 0.63(10)	7.90 ± 0.79(20)	−2.1515	.0314*
3†	13.5 ± 0.71(2)	13.00 ± 0.82(4)	0.5000	.6171

^†Wilcoxon rank sum test.

^*P < .05.

After reversal of ICIP, there were ICIP recurrences in the 8th and 12th weeks after restarting immunotherapy, all of which were initially grade 2 ICIP. The recurrence rate was relatively low in the glucocorticoid–pirfenidone group; there was no significant difference compared to the glucocorticoid alone group (P > .05; Table 5).

Table 5.

ICIP recurrence rate after restarting immunotherapy in both groups.

	n	Number of recurrences at 4 weeks (%)	Number of recurrences at 8 weeks (%)	Number of recurrences at 12 weeks (%)
Glucocorticoid–pirfenidone group	13	0 (0)	2 (15.4)	3 (23.1)
Glucocorticoid alone group	26	0 (0)	5 (19.2)	7 (26.9)

3.3. Adverse events and outcomes

Two patients in each group had gastrointestinal reactions, manifested as nausea and upper abdominal discomfort, which improved after treatment with acid-suppressing agents. During the treatment process, routine blood, liver, and kidney function were monitored in all patients. Two patients in the glucocorticoid–pirfenidone group had elevated transaminase levels, but the levels were <2 times the normal value. Hepatic protection treatment was administered and liver function returned to normal without interrupting treatment. The follow up was update as Figure 1 shows. Two patients in the glucocorticoid group with grade 3 ICIP were lost to follow-up and 2 died due to worsening lung infection with respiratory failure. Six patients with grade 2 ICIP were lost to follow-up and 2 died because of worsening lung infection with respiratory failure. The remaining 12 patients were followed (4 with stable disease and 8 with progressive disease). Six patients with grade 1 ICIP were followed (2 with stable disease and 4 with progressive disease). One patient in the glucocorticoid–pirfenidone group with grade 3 ICIP was lost to follow-up and 1 died of lung infection and respiratory failure. Three patients with grade 2 ICIP were lost to follow-up and 1 died due to lung infection and respiratory failure. The remaining 6 patients were followed (2 with stable disease and 4 with progressive disease). Three patients with grade 1 ICIP were followed (1 with stable disease and 2 with progressive disease).

4. Discussion

Due to the widespread use of ICIs in lung cancer patients during recent years, irAEs have become increasingly important as patients experience extended survival. The incidence of ICIP ranges from 3.6% to 4.1% among patients with NSCLC receiving immunotherapy; PD-1 inhibitors have a higher incidence of ICIP than programmed death ligand 1 inhibitors.^[18,19] Currently, glucocorticoids are the first-line treatment for ICIP.^[9,10] In addition, for high-grade and/or steroid-insensitive ICIP patients, immunosuppressive treatments such as tocilizumab and cyclophosphamide may be recommended.^[10] As stated previously, pirfenidone is considered to apply in ICIP because of its anti-fibrotic and anti-inflammatory effect.^[13–17]

This was retrospective study of PD-1 inhibitor-associated ICIP in NSCLC patients. The combination of pirfenidone and traditional corticosteroid therapy in ICIP patients has advantages in improving symptoms and level of activity. Moreover, significant differences were detected for several indicators, such as FVC, DLCO, SPO₂ without oxygen inhalation, and 6MWD (P < .05), in the glucocorticoid-pirfenidone group than the glucocorticoid alone group. A possible reason for these findings is that ICIP often manifests as diffuse ground-glass opacities, which are nonspecific interstitial pneumonias involving multiple lung lobes accompanied by local inflammation and fibrotic changes. Pirfenidone is a drug that can be used to treat idiopathic pulmonary fibrosis.^[13,14] Factors, such as transforming growth factor-beta 1, platelet-derived growth factor, and vascular endothelial growth factor, are all key molecules involved in pulmonary fibrosis. transforming growth factor-beta 1 is an important cytokine that stimulates fibroblast proliferation and collagen synthesis, leading to pulmonary fibrosis. Platelet-derived growth factor and vascular endothelial growth factor also participate in the occurrence and development of pulmonary fibrosis by promoting fibroblast proliferation and extracellular matrix synthesis. Pirfenidone reduces the proliferation of fibroblasts and collagen synthesis by inhibiting the expression and biological activity of these molecules, thus reducing the degree of pulmonary fibrosis. In addition, Pirfenidone inhibits inflammatory and oxidative stress reactions, reduces the production of oxygen-free radicals, and protects lung tissue, thereby reducing lung injury and improving lung function.^[20,21]Therefore, pirfenidone improves lung function, improves oxygenation, reduces the secretion of various inflammatory mediators (interleukin [IL]-4, IL-13, and IL-10), clears free radicals, reduces lipid peroxidation, and inhibits oxidative stress,^[22] thereby improving patient symptoms and quality of life. Pirfenidone also regulates the immune microenvironment and enhances the antitumor activity of ICIs by increasing the expression of cytokines and chemokines,^[23,24] displaying a synergistic effect and improving patient physical function. However, in grade 2 ICIP, patients in the glucocorticoid–pirfenidone group exhibited shorter time to significant symptom improvement and had a statistically significant difference compared to patients in the glucocorticoid alone group (P < .05), while no difference was observed in the grade 3 ICIP patients (P > .05), which may be related to the small sample size.

In this study, the FVC has increased by 0.12 to 0.14 L, which has significant clinical meaning. Physiologic evidence of disease progression of pulmonary fibrosis was judged in the pulmonary function evaluation index as (i) an absolute decrease of ≥5% in FVC as a % of predicted value for 1 year of follow-up, and (ii) an absolute decrease of ≥10% in DLCO as a % of predicted value for 1 year of follow-up.^[25] The FVC in this study improves 0.12 to 0.14 L (an increase of 6.7% to 8%), which means the patients have had significant improvements. SGRQ is originally designed as a health-related quality of life assessment specific to diseases characterized by obstructive ventilatory defect such as COPD. Recently, the use of the SGRQ to assess health-related quality of life during rehabilitation and follow-up of interstitial pneumonia has been reported in the literature.^[26–28] The questions in SGRQ mostly relate with the patient’s perception of respiratory function, respiratory symptoms, which may also be applied in patients with interstitial pneumonia or other lung impairments. Thus, this study used SGRQ as a reference for evaluating the patient’s improvements.

We found that after controlling ICIP, some patients continued to receive pirfenidone maintenance therapy (300 mg 3 times a day), which showed some advantages in controlling the recurrence rate of ICIP after restarting immune therapy, although there was no significant difference compared to steroids, but this may be related to the small number of subjects in this study. It is worth considering and researching whether increasing the dose of pirfenidone to 400 mg 4 times a day has further synergistic effects in preventing ICIP recurrence. This study also found that there was no significant increase in adverse reactions with steroids or combination therapy, and occasional liver function impairment was easy to control, which is consistent with domestic and international research reports.^[29,30]

The next step in research will be to increase the sample size or increase the dosage of pirfenidone to further clarify its role in controlling the recurrence rate of ICIP. For lung cancer patients treated with ICI, especially those with comorbid connective tissue disease, pneumoconiosis, and other conditions that can cause pulmonary interstitial damage, can early use of pirfenidone reduce the incidence of ICIP? Can it enhance the antitumor efficacy of ICI through synergistic effects? These questions require further validation through cellular, molecular biology, and animal experiments in the future.

In conclusion, the combination of steroids and pirfenidone has good safety and may be more beneficial in improving patients’ interstitial lung disease and reversing ICIP. Continuing the use of pirfenidone may be beneficial in preventing ICIP recurrence. Due to the small sample size of the current study, the sample size and/or dosage of pirfenidone will be increased in a corollary study to further clarify the role in controlling the ICIP recurrence rate.

Author contributions

Conceptualization: Yong Li.

Data curation: Yong Li, Huiqin Huang, Xiangli Ye, Bangwei Zeng, Feijian Huang, Limin Chen.

Formal analysis: Yong Li, Huiqin Huang, Bangwei Zeng.

Investigation: Yong Li, Huiqin Huang, Xiangli Ye, Feijian Huang, Limin Chen.

Methodology: Yong Li, Huiqin Huang, Xiangli Ye.

Project administration: Bangwei Zeng, Limin Chen.

Resources: Huiqin Huang.

Software: Feijian Huang.

Supervision: Limin Chen.

Validation: Xiangli Ye, Limin Chen.

Visualization: Feijian Huang.

Writing – original draft: Yong Li, Huiqin Huang, Bangwei Zeng.

Writing – review & editing: Xiangli Ye, Feijian Huang, Limin Chen.