Reply to: Pharmacotherapy for lung cancer with comorbid interstitial pneumonia: limited evidence requires appropriate evaluation

From the authors:

We appreciate the interest raised by our recent article [1] and take the opportunity to address the points raised by Y. Ishida and colleagues.

Y. Ishida and colleagues questioned our suggestions for second-line lung cancer therapy in fibrotic interstitial lung disease (ILD) with vinorelbine (squamous cell carcinoma) and pemetrexed (adenocarcinoma). They suggest docetaxel rather than vinorelbine and pemetrexed, but there are opposing arguments. They refer to acute exacerbation rates from one retrospective nationwide surveillance of 396 patients with idiopathic interstitial pneumonia and lung cancer in Japan [2]. However, this study does not reflect experience in other populations and regions. Drug-induced lung toxicities vary greatly between studies and ethnicities. For example, the incidence of adverse events associated with pemetrexed has been as low as 12.5% and 13.3% in idiopathic pulmonary fibrosis (IPF) and ILD, respectively [3]. Drug chemotherapeutic toxicities are reportedly higher in Asians than in Caucasians [4].

Furthermore, docetaxel causes acute exacerbation in 18.4% and 20.8% of patients with ILD and IPF, respectively [3]. In a limited number of patients, the incidence of docetaxel-associated ILD exacerbation was reported to be as high as 50%, with none associated with pemetrexed and vinorelbine [5]. Kenmotsu et al. [6] reported the incidence of acute exacerbation of ILD with use of second-line chemotherapy in non-small cell lung cancer (NSCLC) to be 26% with docetaxel, 25% with pemetrexed and 20% with vinorelbine. Watanabe et al. [7] found that docetaxel monotherapy has a poor activity and substantial risks when used for the treatment of platinum-resistant NSCLC with interstitial pneumonia. Tamiya et al. [8] explicitly recommended against the use of docetaxel in pre-existing ILD patients with NSCLC due to a high incidence of chemotherapy-associated radiological ILD changes. Finally, in the very recently published results of the J-Sonic Phase III trial using nab-paclitaxel plus carboplatin±nintedanib in the first-line setting for lung cancer associated with fibrosing ILD, Japanese patients received, in the second line, most frequently, S-1 and, less frequently but in similar proportions, docetaxel, vinorelbine or pemetrexed [9].

Hayashi and Mitsudomi [10] summarise why docetaxel represents an inferior choice and elaborate reasons for regional differences in S-1 use. S-1 is an effective anticancer drug in patients with ILD in Japan [11] but is not used in Europe. Firstly, the higher incidence of side-effects in Caucasians compared to Asian populations argues against S-1 [12, 13]. Secondly and most importantly, S-1 is authorised for use in advanced gastric cancer and metastatic colorectal cancer [10] but has not been approved in lung cancer patients by the European Medicines Agency.

Thus, the divergence in views on lung cancer treatments in patients with fibrosing ILD is likely to reflect differences in chemotherapy approval and usage, based in part on the variable incidence of side-effects in different countries and among ethnicities.

Lung cancer and fibrosing ILD are both heterogeneous conditions, and treatment plans for both conditions depend on case-by-case considerations and regional variability. The onset of lung cancer in pre-existing fibrosing ILDs greatly complicates management, which must be individualised according to patient tolerance, ethnicity and local resources. This gives rise to considerable variability in reports on anticancer chemotherapy in ILD.

Y. Ishida and colleagues aptly summarise the challenge faced with immune checkpoint inhibitor (ICI) trial data. We have pointed out that alarming data concerning the use of ICIs were mostly retrospective and have described the only two phase II trials published at the time of our review submission. Owing to editorial constraints, we did not cite the pilot study of six patients treated with nivolumab by Fujimoto et al. [14].

We are surprised that Y. Ishida and colleagues argue for the supposed superior efficacy of immunotherapy, compared to chemotherapy, in lung cancer associated with fibrosing ILD based on the recent article by Ikeda et al. [15]. That study was halted due to drug-related toxicity.

We acknowledge that we misleadingly formulated the trial design of the J-Sonic trial as a study result. Unfortunately, the publication became available only after submission of our paper [9]. The primary end-point of reduced acute exacerbation rate was not met. A correction to our publication has now been published [1].

In conclusion, the complexity of lung cancer associated with fibrosing ILD requires international efforts to better define treatment approaches. However, regional data must be generated as ethnicity influences the phenotype, as well as drug efficacy and toxicities.