Association of immune-related pneumonitis with the phenotypic appearance of concurrent ILD in patients treated with anti-PD-1 antibody

Date 24 November 2018
Event ESMO Asia 2018 Congress
Session Poster display - Cocktail
Topics Immunotherapy
Complications/Toxicities of Treatment
Thoracic Malignancies
Presenter Ryota Shibaki
Citation Annals of Oncology (2018) 29 (suppl_9): ix150-ix169. 10.1093/annonc/mdy425
Authors R. Shibaki1, S. Murakami2, Y. Matsumoto2, Y. Goto2, S. Kanda2, H. Horinouchi2, Y. Fujiwara2, N. Yamamoto2, Y. Ohe3
  • 1Department Of Thoracic Oncology, National Cancer Center Hospital, Tokyo, Japan, 1040045 - Tokyo/JP
  • 2Department Of Thoracic Oncology, National Cancer Center Hospital, Tokyo, Japan, 104-0045 - Tokyo/JP
  • 3Department Of Thoracic Oncology, National Cancer Center Hospital, Tokyo, Japan, Tokyo/JP



Concurrent ILD is a risk factor for anti-tumor drug-related pneumonitis. The performance of a risk assessment for concurrent ILD prior to anti-PD-1 antibody (Ab) treatment is important for preventing immune-related (ir) pneumonitis, although this precaution is not well known.


NSCLC patients treated with anti-PD-1 Ab between December 2015 and May 2018 were retrospectively reviewed. The CT appearance of concurrent ILD was classified as either non-ILD, inconsistent with UIP, or UIP/possible UIP. The CT appearance of ir-pneumonitis was classified as having a cryptogenic organizing pneumonia-like (COP) pattern, a ground glass opacities (GGO) pattern, an interstitial pattern, a hypersensitivity (HP) pattern, or a pneumonitis not otherwise specified (Not-specified) pattern. The aim of this study was to evaluate the dependency of concurrent ILD for the development of ir-pneumonitis and the association between the CT appearance of concurrent ILD and that of ir-pneumonitis.


As of the data cutoff on July 1, 2018, a total of 332 patients were included in this study. No difference in PFS was seen between ILD and non-ILD patients (median, 4.4 vs. 2.7 months, P = 0.14). The overall incidence of ir-pneumonitis was 11.4% (38/332 patients). The incidence of ir-pneumonitis was higher in patients with concurrent ILD than in those with non-ILD (35.1% vs. 8.5%, P < 0.0001). The distributions of CT appearances were as follows: for the non-ILD group,: 56.0% COP, 20.0% GGO, 4.0% interstitial, 16.0% HP, and 4.0% Not-specified; for the inconsistent with UIP group,: 57.1% COP, 42.9% GGO, 0.0% interstitial, 0.0% HP, and 0.0% Not-specified; for the UIP/possible UIP group,: 0.0% COP, 16.7% GGO, 83.3% interstitial, 0.0% HP, and 0.0% Not-specified.


Concurrent ILD is a risk factor for ir-pneumonitis, but anti-PD-1 Ab has a similar efficacy in concurrent ILD patients and non-ILD patients. The CT appearances of ir-pneumonitis differ among concurrent ILD groups, with non-ILD and inconsistent groups exhibiting COP and GGO patterns and the possible UIP/UIP group exhibiting a mainly interstitial pattern. Careful monitoring of patients, especially those with a UIP/possible UIP pattern on pretreatment CT, is needed.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.


Has not received any funding.


Y. Goto: Consulting roles: Boehringer Ingelheim; Speakers’ bureaus: Taiho, Boehringer Ingelheim, Ono, Bristol-Myers Squibb, MSD; Research Funding Taiho, Bristol-Myers Squibb, Ono. S. Kanda: Research Funding Ono Pharmaceutical; Honoraria: Ono Pharmaceutical, Bristol-Myers Squibb. H. Horinouchi: Research Funding MSD, Bristol-Myers Squibb, Ono Pharmaceutical,Taiho Pharmaceutical. Y. Fujiwara: Research Funding MSD, Bristol-Myers Squibb; Speakers’ bureaus: MSD, Taiho Pharmaceutical, Bristol-Myers Squibb, Ono Pharmaceutical. N. Yamamoto: Research Funding Taiho Pharmaceutical, Bristol-Myers Squibb, Ono Pharmaceutical; Speakers’ bureaus: Bristol-Myers Squibb, Ono Pharmaceutical. Y. Ohe: Research Funding Taiho Pharmaceutical, MSD; Honoraria: Taiho Pharmaceutical, MSD. All other authors have declared no conflicts of interest.