Elsevier

Respiratory Investigation

Volume 57, Issue 5, September 2019, Pages 451-459
Respiratory Investigation

Original article
Pre-existing interstitial lung abnormalities are risk factors for immune checkpoint inhibitor-induced interstitial lung disease in non-small cell lung cancer

https://doi.org/10.1016/j.resinv.2019.05.002Get rights and content

Abstract

Background

Approximately 5% of non-small cell lung cancer (NSCLC) patients develop immune checkpoint inhibitor (ICI)-induced interstitial lung disease (ICI-ILD), 10% of whom die. However, there are no established risk factors for its occurrence. Interstitial lung abnormalities (ILA) are areas of increased lung density on lung computed tomography (CT) in individuals with no known ILD. This study retrospectively investigated whether any patient characteristics, including ILA, were risk factors for ICI-ILD in patients with NSCLC.

Methods

NSCLC patients who received anti-programmed death (PD)-1 antibody treatment at our hospital between September 2015 and December 2017 were enrolled. Information on patient characteristics before anti-PD-1 antibody administration, including chest CT findings and laboratory data, were obtained.

Results

Among 83 enrolled patients, the incidence of ICI-ILD was 16.9% (14/83). All ICI-ILD cases developed by the third line of treatment. The incidence of ICI-ILD was significantly higher in patients with pre-existing ILA than that in those without (p = 0.007). Furthermore, patients with ground glass attenuation (GGA) in ILA had a higher incidence of ICI-ILD than that in those without (p < 0.001). In univariate logistic analysis, ILA were significant risk factors for ICI-ILD (p = 0.005). Multivariate logistic analysis revealed that only GGA in ILA was a significant risk factor for ICI-ILD (p < 0.001).

Conclusions

Pre-existing ILA are risk factors for ICI-ILD and GGA in ILA is an independent risk factor for ICI-ILD. Therefore, we should be more aware of the development of ICI-ILD in patients with ILA, especially those with GGA.

Introduction

Anti-programmed death (PD)-1 antibodies, such as nivolumab and pembrolizumab, belong to the family of immune checkpoint inhibitors (ICIs) and are approved for use for the treatment of multiple advanced cancers, including lung cancer, head and neck cancer, melanoma, and renal cell carcinoma. Anti-PD-1 antibodies exert their antitumor effect by suppressing immune tolerance to cancer cells, and clinical trials have demonstrated their superiority to chemotherapy in patients with non-small cell lung cancer (NSCLC). In addition, the incidence of severe adverse events in patients receiving ICIs is almost one-third that in patients receiving chemotherapy [1], [2], [3], [4], [5].

However, the PD-1 blockade is associated with a unique set of toxicities, termed immune-related adverse events (ir-AEs), due to the mechanisms of action of anti-PD-1 antibodies. Ir-AEs differ from toxicities observed in conventional cytotoxic chemotherapy or targeted therapies. Ir-AEs such as skin rash, thyroiditis/hypothyroidism, diarrhea, and interstitial lung disease (ILD) have been observed with anti-PD-1 treatment [1], [2], [3], [6], [7]. Although ILD is relatively rare, it is a clinically serious and life-threatening toxicity [8]. A systematic review and meta-analysis showed that at approximately 5%, the incidence of ICI-induced ILD (ICI-ILD) is higher in NSCLC than that in melanoma [6], [9]. In addition, 10% of patients who develop ICI-ILD eventually die [2]. Therefore, the identification of risk factors for ICI-ILD is urgently needed.

Previous reports showed that pre-existing ILD was a risk factor for chemotherapy-induced ILD [10], [11]. To our knowledge, only two studies have examined whether pre-existing ILD is a risk factor for ICI-ILD [12], [13]. These studies revealed that the risk of ICI-ILD was associated with the presence or extent of ILD. However, ICI-ILD is frequently observed clinically in patients without pre-existing ILD. Interstitial lung abnormalities (ILA), which have been defined as areas of increased lung density on lung computed tomography (CT) in individuals with no known ILD, are present in 14% of treatment-naïve patients with advanced NSCLC [14]. From these observations, we hypothesized that ILA would be a risk factor of ICI-ILD. This study retrospectively investigated whether any patient characteristics, including ILA, were risk factors for ICI-ILD in patients with NSCLC.

Section snippets

Study design and participants

Patients with advanced NSCLC who received ICIs (nivolumab or pembrolizumab) at the Hiroshima University Hospital between September 2015 and December 2017 were enrolled. Patient characteristics and clinical data before administration of anti-PD-1 antibodies were obtained. Tumor responses were evaluated using the Response Evaluation Criteria for Solid Tumors version 1.1.

Assessment of high-resolution CT (HRCT) findings

We investigated the presence of abnormal findings other than lung cancer including pre-existing ILA on HRCT. ILA were defined as

Patient characteristics

A total of 83 patients were enrolled in the study. The clinical characteristics of these patients are shown in Table 1. The median patient age was 68 years. Male patients (69.9%) and patients with Eastern Cooperative Oncology Group (ECOG) performance statuses of 0 or 1 (85.5%) were predominant. Many patients (79.5%) were current or former smokers. Squamous cell carcinoma histology accounted for 15.7% of cases. Eleven patients (13.2%) had a driver mutation; 10.8% of patients harbored an

Discussion

The results of this study showed that pre-existing ILA were risk factors for ICI-ILD in patients with NSCLC. In addition, GGA in ILA was an independent risk factor. To our knowledge, this is the first study to report these specific CT findings as risk factors for ICI-ILD. In this study, ILA were observed in 15.7% (13/83) of patients. Although the extent of ILA was slight in most patients, their presence was a risk factor for ICI-ILD in this study. Therefore, we should estimate the risk of

Conclusions

In this study, pre-existing ILA were significant risk factors for ICI-ILD. In addition, GGA in ILA was an independent risk factor. Therefore, patients with ILA, especially GGA, in HRCT prior to ICI administration must be carefully monitored to prevent the occurrence of ICI-ILD.

Financial support

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest statement

N.H. and M.O. received research funding from ONO PHARMACEUTICAL CO. LTD., CHUGAI PHARMACEUTICAL CO. LTD., and Astra Zeneca K.K. M.O. received honoraria from ONO PHARMACEUTICAL CO. LTD., CHUGAI PHARMACEUTICAL CO. LTD., and Astra Zeneca K.K.

Ethical approval

This retrospective analysis was approved by the Hiroshima University Institutional Review Board (No. E939). All procedures performed in

References (24)

  • J. Brahmer et al.

    Nivolumab versus Docetaxel in Advanced Squamous-Cell Non-Small-Cell Lung Cancer

    N Eng J Med

    (2015)
  • M. Nishino et al.

    Incidence of Programmed Cell Death 1 Inhibitor-Related Pneumonitis in Patients With Advanced Cancer: A Systematic Review and Meta-analysis

    JAMA Oncol

    (2016)
  • Cited by (0)

    View full text