Nivolumab Retreatment in Non–Small Cell Lung Cancer Patients Who Responded to Prior Immune Checkpoint Inhibitors and Had ICI-Free Intervals (WJOG9616L)

Immune checkpoint inhibitors (ICI) have become a key player in the treatment of advanced non–small cell lung cancer (NSCLC). However, recent follow-up of phase III studies demonstrated that ≥60% of patients who initially responded to either ICI alone or combined with chemotherapy ultimately acquired resistance and progressed (1, 2). Therefore, to develop an effective subsequent treatment is a critical unmet need.

For patients whose tumor had progressed with anti–programmed death 1 (anti–PD-1)/anti–programmed death ligand 1 (anti–PD-L1) inhibitors, adding anti-CTLA4 antibody has been investigated in various tumor types, but has shown modest benefit (3–5). Translational approaches have elucidated several mechanisms that may explain acquired resistance to ICIs, including mutational changes in tumor-specific neoantigens, PTEN loss, and/or β-catenin/Wnt signaling alterations (6–8). However, effective agents have not been approved yet.

Retreatment with anticancer drugs that showed efficacy during prior lines of therapy and had certain drug-free interval is commonly accepted in lung cancer. Although the duration of retreatment is typically relatively shorter compared with their initial administration (9, 10), the acceptable adverse events profiles of these agents allow clinicians to consider retreatment even in later line. However, data regarding the effectiveness of retreatment with ICIs have been conflicting. Another issue is the heterogeneity of patient population among studies. Schoenfeld and colleagues recently proposed a definition of acquired resistance to ICI in patients with advanced NSCLC as follows: (i) patients who received anti–PD-1/anti–PD-L1 inhibitor-containing regimen(s), (ii) those who had complete response (CR) or partial response (PR) as their best response, and (iii) those who progressed within 6 months of prior ICI (11). While they encouraged ICI retreatment if patients progressed after ≥ 6 months of ICI-free interval, clinical data regarding the effectiveness of such retreatment is lacking.

On the basis of these backgrounds, we performed this multi-institutional, phase II trial of nivolumab retreatment in patients with advanced NSCLC who responded to prior anti–PD-1/anti–PD-L1 inhibitors and acquired resistance [West Japan Oncology Group (WJOG) 9616L].

We conducted this study as an open-label, multi-institutional, single-arm phase II study. Eligible patients were those with pathologically confirmed NSCLC who previously received at least one systemic anticancer treatment including ICI. Patients should have received clinical benefit from prior ICI-containing regimen preceding progression. We defined a clinical benefit as a CR, PR, or stable disease (SD) ≥ 6 months. To avoid those who recently progressed with prior ICI, patients must have had ICI-free interval ≥ 60 days. Additionally, patients had Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 or 1, and had at least one measurable target lesion according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Patients with brain metastases were allowed unless they were symptomatic. We excluded patients who had active autoimmune disease, however, those with controlled thyroid dysfunction or skin disease that did not require systemic therapy were allowed.

All patients were intravenously administered nivolumab every 2 weeks until progression. The nivolumab dose was in accordance with that approved by the Japanese government. Initially, approved dose was 3 mg/kg but was changed to a 240-mg flat dose in August 2018.

To assess efficacy, chest CT was performed every 6 weeks for up to 40 weeks and then every 12 weeks thereafter. Gadolinium-enhanced MRI of the brain was performed at baseline for all patients prior to registration. Tumor responses were based on the RECIST version 1.1. Adverse events were graded according to the Common Terminology Criteria for Adverse Events, version 4.0.

The primary endpoint of the study was the overall response rate (ORR). After progression of ICI-containing regimen, the standard treatment was cytotoxic chemotherapy. The sample size was calculated on the basis of an assumption that retreatment with nivolumab would improve the ORR from 10% to 20% in the second- or later-line setting. As such, 60 patients were required to ensure a statistical power of 0.79 with a one-sided alpha error of 0.05. The secondary endpoints were progression-free survival (PFS), overall survival (OS), and adverse events.

The ORR estimates and 95% confidence intervals (CI) were calculated, and the 12-month PFS and OS were estimated using Kaplan–Meier curves and are reported with their 95% CIs. Differences in ORR and PFS according to clinical backgrounds were evaluated, with the significance level set to a two-sided alpha of 0.05. Univariate analyses were performed to identify factors predictive of ORR and PFS using a Cox proportional hazards model. The investigated covariates included clinical backgrounds (such as sex, smoking status, and histology) and ICI-related factors (such as PD-L1 expression level, history of immune-related adverse events (irAE), ICI-free interval, and duration of prior ICI). Continuous variables were dichotomized at the median. Covariates that were significantly correlated with PFS in the univariate analyses (defined as P < 0.05) were subjected to multivariate analyses. Statistical analyses were conducted using the JMP software (version 11; SAS Institute, Cary, NC) and Prism version 7.00 for Windows (GraphPad Software, San Diego, CA).

We collected tumor tissue upon progression with prior ICI. PD-L1 and CD8+ lymphocytes were independently assessed using immunohistochemistry. PD-L1 (clone 22c3, Dako) and CD8 (clone SP57, Ventana/Roche) antibodies were used according to the manufacturers’ guidelines. The tumor proportion score was calculated to quantify PD-L1 expression as previously described (12). To evaluate CD8, the density of stained lymphocytes was quantified by averaging the number of all lymphocytes counted in five high-power fields (400× magnification) and then categorizing them as minimal (average density <50), mild (50–75), and moderate (≥75) as described previously (13). According to our previous study (14), we also collected 20 mL of peripheral blood before and after treatment (6–8 weeks), and circulating tumor cells (CTC) were detected using a microcavity array system (Hitachi Chemical Co. Ltd., Chikusei, Japan).

The study was conducted in compliance with the principles of the Declaration of Helsinki, and the institutional review board of each participating institution approved the protocol. Written informed consent was obtained from all patients before any screening or inclusion procedure. This protocol was registered in the University Hospital Medical Information Network, Japan (protocol identification no. UMIN000028561).

The data generated in this study are available upon request from the corresponding authors.

Sixty-one patients were enrolled between October 2017 and February 2020. One dropped out before commencing nivolumab and another was excluded from efficacy analysis because this patient did not meet eligibility criteria after registration (Supplementary Fig. S1). The median follow-up interval for the 59 patients who were evaluated for efficacy was 19.5 months (range, 2.2–30.7 months). At the time of data cut-off, 56 patients had discontinued nivolumab retreatment. The median number of administrations was five (range, 1–31).

The baseline characteristics of 59 patients are described in Table 1. Median age was 70 years (range, 38–81). Most patients were male (n = 43, 73%) and smokers (n = 49, 83%). Median number of prior regimen was three (range, 1–6). Regarding prior ICI, most patients (n = 54, 92%) received monotherapy and 41 (70%) achieved CR or PR. Median duration of prior ICI was 8.1 months (range, 0.8–37.0 months) and median ICI-free interval was 9.2 months (range, 2.4–29.4 months). Prior ICI was discontinued in 20 patients due to irAEs (9 pneumonitis, 4 colitis, 2 thyroid dysfunction, and 5 other AEs).

Tumor shrinkage owing to nivolumab retreatment is shown in Fig. 1. Five patients achieved PR, 25 had SD, 28 had progressive disease, and 1 was not evaluable. Thus, the ORR was 8.5% (95% CI, 2.8%–18.7%), which failed to meet the primary endpoint. Median PFS was 2.6 months (95% CI, 1.6–2.8 months), and PFS rate at 12 months was 4.8% (95% CI, 1.0%–13.6%; Fig. 2A). Median OS was 11.0 months (95% CI, 9.0–14.5 months), and OS rate at 12 months was 46.4% (95% CI, 32.8%–59.0%; Supplementary Fig. S2A).

The clinical backgrounds of the 5 patients who experienced PR are described in Supplementary Table S1. Despite the small number of patients, it was notable that their median PFS was 11.1 months (Fig. 2A), and most of them were alive at the time of analysis (Fig. 2B; Supplementary Fig. S2B).

We performed univariate and multivariate analyses to identify the predictive factor of response to nivolumab retreatment (Table 2). None of the clinical backgrounds significantly influenced ORR (Supplementary Fig. S3). However, non-squamous histology, duration of prior ICI (>8.1 months), ICI-free interval (>9.2 months), and history of irAE were associated with significantly longer PFS (Supplementary Fig. S4). Multivariate analysis revealed that ICI-free interval was the only significant predictor of longer PFS (≤9.2 vs. >9.2 months; HR, 2.02; 95% CI, 1.10–3.73; P = 0.02), while prior ICI response or history of irAE was not. As shown in Fig. 3, ICI-free intervals were correlated with longer PFS in a duration-dependent manner. Median PFS values according to ICI-free intervals were 1.4, 1.6, 4.3, 2.8, and 8.4 months among ICI-free intervals with 2–6, 6–12, 12–18, 18–24, and 24–30 months, respectively.

Adverse events related to nivolumab retreatment are summarized in Supplementary Table S2. Nine grade 3 and three grade 4 events were observed, while no treatment-related deaths occurred. Frequent adverse events were skin disorders (n = 14, 23%), malaise (n = 12, 20%), and hypoalbuminemia (n = 9, 15%). Various irAEs were observed, among which pneumonitis and colitis were the most common irAEs. Other severe irAEs included myocarditis (grade 4, n = 1), pancreatitis (grade 3, n = 1), vasculitis (grade 3, n = 1), and adrenal insufficiency (grade 3, n = 1). Of 20 patients who discontinued prior ICI due to irAEs, 4 had severe irAEs by nivolumab retreatment. Thus, the history of irAEs with prior ICI was not related to the occurrence of severe irAE with nivolumab retreatment.

As an exploratory analysis, we planned to collect tissue samples immediately before retreatment. However, a fresh biopsy was challenging in this heavily pretreated patient population. Two specimens were obtained from responders of retreatment: one had high PD-L1 expression (tumor proportion score >90%) with lower CD8⁺ cells, while the other expressed neither PD-L1 or CD8⁺ cells (data not shown). The median number of CTCs at baseline was zero (range, 0–512), with this count being higher among patients who had progressive disease (Supplementary Fig. S5). Changes in CTCs at 6 to 8 weeks were not predictive because most patients had lower CTC counts. PD-L1 expression on CTCs was detected in 11 cases at baseline and in only one case at 6 to 8 weeks.

The rationale of retreatment is from the hypothesis that chemotherapy-sensitive tumor cells ultimately revive through drug-free interval (15) as clinically replicated in patients with small cell lung cancer (16). However, it has not been investigated whether ICI retreatment could show efficacy in patients who responded to prior ICI after ICI-free intervals.

Several studies already demonstrated the safety of ICI retreatment in patients with NSCLC even after the occurrence of irAEs (17, 18), however, its efficacy has not been properly assessed. A retrospective case series of 18 patients with NSCLC that included both ICI responders and nonresponders showed that retreatment with PD-L1 inhibitor was not effective (19). Meanwhile, a post hoc analysis of phase III studies revealed ORRs of 15% to 33% with pembrolizumab retreatment (2, 20). It should be noted that participants in the latter report were highly selective, as they were allowed to receive pembrolizumab retreatment only after they had completed two years of pembrolizumab. Such heterogeneity of patient population made it difficult to interpret these results.

Our key eligibility criteria mostly fulfilled with the aforementioned definition of acquired resistance to ICI (11): 70% of our patients had CR/PR with prior ICI, and the median ICI-free interval was 9.2 months. Given this point, our result suggested that retreatment with ICI is not a mainstay treatment even if prior ICI had shown response for a period of time. Subgroup analysis showed that patients with non-squamous histology, longer duration with prior ICI, longer ICI-free interval, or history of irAE tended to have longer PFS. Among these factors, multivariate analysis found that a longer ICI-free interval was the only predictor of improved PFS. Interestingly, as the ICI-free interval extended, PFS was significantly longer (Fig. 3). This study suggested that sufficient ICI-free interval might provide a clue to achieve efficacy with ICI retreatment, however, the optimal duration is still unknown. Moreover, even though patients with ICI-free interval of 24 to 30 months had 8.4 months of median PFS, we could not deny the possibility that their tumor simply progressed slowly.

Despite our negative data, it was noteworthy that 5 patients had durable responses with ICI retreatment. These patients exhibited no distinct characteristics, but all of them had achieved PR with prior ICI and most had a relatively long ICI-free interval. To better understand their biological traits, we planned to collect tumor samples just before retreatment. However, it was difficult to obtain tissues from heavily treated patients. Alternatively, we also examined whether the number of CTCs at baseline of retreatment could be a possible predictor of response. Unfortunately, we could not extract useful information regarding CTC counts or PD-L1 expression therein.

This study had several limitations. First, its single-arm nature could not rule out selection bias. For example, prior treatment such as chemotherapy or radiotherapy may affect the result despite setting the appropriate interval before enrollment. However, regarding the 5 responders in this study, 3 did not receive any treatment after prior ICI. Although the other 2 completed chemotherapy 33 and 45 days before enrollment, it may not be unusual for the delayed efficacy of chemotherapy to persist for almost a year. Furthermore, it is more important that this was the first prospectively designed investigation with formal statistical calculation of power to evaluate the utility of ICI retreatment among patients with NSCLC. Second, the lack of independent radiological review could have influenced the efficacy data. However, antitumor activity is rarely underestimated because of this drawback.

In conclusion, even in patients who initially responded to prior ICI and had ICI-free interval, once resistance was acquired, retreatment with nivolumab had limited efficacy. Only a small proportion of patients achieved durable responses, thus identifying biomarkers that are predictive of favorable responses to ICI retreatment is warranted.

H. Akamatsu reports personal fees from AstraZeneca K.K., Boehringer Ingelheim Japan Inc., Bristol Myers Squibb Co. Ltd., Novartis Pharma K.K., Taiho Pharmaceutical Co. Ltd., and Ono Pharmaceutical Co. Ltd., as well as grants and personal fees from Chugai Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., and MSD K.K. outside the submitted work. S. Teraoka reports personal fees from AstraZeneca K.K., Boehringer Ingelheim Japan Inc., Chugai Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., Novartis Pharma K.K., Ono Pharmaceutical Co. Ltd., Pfizer R&D Japan G.K., and Taiho Pharmaceutical Co. Ltd. outside the submitted work. S. Takamori reports grants from AstraZeneca K.K., Chugai Pharmaceutical Co. Ltd., and Pharmaceutical Co. Ltd. outside the submitted work. S. Miura reports personal fees from Ono Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., Boehringer Ingelheim Japan Inc., AstraZeneca K.K., Pfizer Japan Inc., Taiho Pharmaceutical Co. Ltd., Novartis Pharma K.K., Kyowa Hakko Kirin Co. Ltd., Daiichi Sankyo Co. Ltd., AbbVie, Nippon Kayaku Co. Ltd., Chugai Pharmaceutical Co. Ltd., and Bristol Myers Squibb Co. Ltd. during the conduct of the study; S. Miura also reports personal fees from MSD K.K. outside the submitted work. H. Hayashi reports grants from AstraZeneca K.K., Astellas Pharma Inc., MSD K.K., Ono Pharmaceutical Co. Ltd., Nippon Boehringer Ingelheim Co. Ltd., Novartis Pharma K.K., Pfizer Japan Inc., Bristol Myers Squibb Company, Eli Lilly Japan K.K., Chugai Pharmaceutical Co. Ltd., Daiichi Sankyo Co. Ltd., Merck Serono Co. Ltd./Merck Biopharma Co. Ltd., Takeda Pharmaceutical Co. Ltd., Taiho Pharmaceutical Co. Ltd., SymBio Pharmaceuticals Ltd., AbbVie Inc, inVentiv Health Japan, ICON Japan K.K., Gritstone Oncology Inc, Parexel International Corp., Kissei Pharmaceutical Co. Ltd., EPS Corporation, Syneos Health, Pfizer R&D Japan G.K., A2 Healthcare Corp., Quintiles Inc./IQVIA Services JAPAN K.K., EP-CRSUCo. Ltd., Linical Co. Ltd., Eisai Co. Ltd., CMIC Shift Zero K.K., Kyowa Hakko Kirin Co. Ltd., Bayer Yakuhin Ltd., EPS International Co. Ltd., and Otsuka Pharmaceutical Co. Ltd., as well as personal fees from Amgen K.K., AstraZeneca K.K., Boehringer Ingelheim Japan Inc., Bristol Myers Squibb Co. Ltd., Chugai Pharmaceutical Co. Ltd., Daiichi Sankyo Co. Ltd., Eli Lilly Japan K.K., Guardant Healthcare, Kyorin Pharmaceutical Co. Ltd., Merck Biopharma Co. Ltd., MSD K.K., Novartis Pharmaceuticals K.K., Ono Pharmaceutical Co. Ltd., Shanghai Haihe Biopharm, Taiho Pharmaceutical Co. Ltd., Pfizer, and Takeda Pharmaceutical Co. Ltd. outside the submitted work. A. Hata reports personal fees from ONO outside the submitted work. Y. Toi reports personal fees from Bristol Myers Squibb Company, Ono Pharmaceutical Co. Ltd., MSD K.K., AstraZeneca plc, Chugai Pharmaceutical Co. Ltd., TAIHO Pharmaceutical Co. Ltd., Pfizer Inc., and Kyowa Kirin Co. Ltd. outside the submitted work. Y. Shiraishi reports grants and personal fees from Chugai Pharma, as well as personal fees from Eli Lilly, Ono Pharmaceutical, AstraZeneca, Taiho Pharmaceutical, and Bristol Myers Squibb outside the submitted work. N. Mamesaya reports grants and personal fees from Boehringer Ingelheim, as well as personal fees from Chugai Pharmaceutical Co. Ltd., Taiho Pharmaceutical Co. Ltd., MSD K.K., AstraZeneca K.K., and Ono Pharmaceutical Co. Ltd. outside the submitted work. Y. Sato reports personal fees from AstraZeneca K.K., Bristol Myers Squibb Co. Ltd., Chugai Pharmaceutical Co. Ltd., MSD K.K., Ono Pharmaceutical Co. Ltd., Nippon Kayaku Co. Ltd., Novartis Pharma K.K., and Taiho Pharmaceutical Co. Ltd. outside the submitted work. N. Furuya reports personal fees from Eli Lilly Japan, Chugai, AstraZeneca, Bristol Myers Squibb, Boehringer Ingelheim Japan, Taiho, Pfizer Japan, and Novartis outside the submitted work. Y. Koh reports non-financial support from Hitachi Chemical during the conduct of the study. Y. Koh also reports grants and personal fees from Chugai Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., Amgen, Takeda, AstraZeneca, and Tosoh Corporation; personal fees from Guardant Health, Hitachi High Technologies, and Pfizer Japan Inc.; and grants from Boehringer Ingelheim, Daiichi Sankyo, Bristol Myers Squibb, Ono Pharmaceutical, Zeon Corporation, On-chip Biotechnologies, Terumo, and Toppan outside the submitted work. N. Yamamoto reports grants and personal fees from AstraZeneca K.K., Boehringer Ingelheim Japan Inc., Chugai Pharmaceutical Co. Ltd., Eli Lilly Japan K.K., MSD K.K., Ono Pharmaceutical Co. Ltd., Pfizer Japan Inc., Takeda Pharmaceutical, and Taiho Pharmaceutical Co. Ltd.; N. Yamamoto also reports personal fees from Novartis Pharma K.K., as well as grants from Astellas Pharma Inc., A2 Healthcare Corporation, Bristol Myers Squibb Co. Ltd., CMIC Shift Zero K.K., Daiichi Sankyo Co. Ltd., IQVIA Services Japan K.K., and PPD-SNBL K.K. outside the submitted work. K. Nakagawa reports grants and personal fees from AstraZeneca K.K., MSD K.K., Nippon Boehringer Ingelheim Co. Ltd., Novartis Pharma K.K., Bristol Myers Squibb Company, and Chugai Pharmaceutical Co. Ltd.; grants, personal fees, and other support from Ono Pharmaceutical Co. Ltd., Pfizer Japan Inc., and Eli Lilly Japan K.K.; grants from Daiichi Sankyo Co. Ltd.; and personal fees from Merck Biopharma Co. Ltd. during the conduct of the study. K. Nakagawa also reports personal fees from CMIC Co. Ltd., Care Net Inc., Medical Review Co. Ltd., Medical Mobile Communications Co. Ltd, 3H Clinical Trial Inc., YODOSHA Co. Ltd., Nikkei Business Publications Inc., TAIYO Pharma Co. Ltd., Nippon Kayaku Co. Ltd., CMIC Shift Zero K.K., Kyowa Kirin Co. Ltd., Life Technologies Japan Ltd., Neo Communication, and Roche Diagnostics K.K.; personal fees and other support from KYORIN Pharmaceutical Co. Ltd.; grants and personal fees from Takeda Pharmaceutical Co. Ltd., Taiho Pharmaceutical Co. Ltd., AbbVie Inc., Bayer Yakuhin Ltd., Japan Clinical Research Operations, and Amgen Inc.; grants from Parexel International Corp., Kissei Pharmaceutical Co. Ltd., EPS Corporation, Syneos Health Clinical K.K., Pfizer R&D Japan G.K., A2 Healthcare Corp., IQVIA Services Japan K.K., Eisai Co. Ltd., EPS International Co. Ltd., Otsuka Pharmaceutical Co. Ltd., PRA Health Sciences, Covance Japan Inc., Medical Research Support, Sanofi K.K., PPD-SNBL K.K, Sysmex Corporation, Mochida Pharmaceutical Co. Ltd., GlaxoSmithKline K.K., SymBio Pharmaceuticals Ltd., Mebix Inc., Janssen Pharmaceutical K.K., and SRL Inc. outside the submitted work. In addition, K. Nakagawa has a patent for Daiichi Sankyo Co. Ltd. issued. No disclosures were reported by the other authors.

H. Akamatsu: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, investigation, visualization, writing–original draft, project administration, writing–review and editing. S. Teraoka: Resources. S. Takamori: Resources, data curation. S. Miura: Resources, data curation. H. Hayashi: Resources, data curation. A. Hata: Resources, data curation. Y. Toi: Resources. Y. Shiraishi: Resources. N. Mamesaya: Resources. Y. Sato: Resources. N. Furuya: Resources, data curation. J. Oyanagi: Resources, formal analysis. Y. Koh: Data curation, formal analysis, supervision, writing–review and editing. T. Misumi: Formal analysis. N. Yamamoto: Conceptualization, data curation, supervision. K. Nakagawa: Supervision.

We wish to thank Ms. Ryoko Sakagami and Yasuko Matsumoto who assisted our translational analyses. We are also grateful to the data managers and other support staff of the West Japan Oncology Group, particularly Drs. Koji Takeda, Shinichiro Nakamura, and Ms. Seiko Tanaka. The current study was conducted with support from the West Japan Oncology Group Data Center, Osaka, Japan. We thank Editage (www.editage.com) for English proofreading.

Hiroaki Akamatsu, the principal investigator had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

This work was supported by Ono Pharmaceutical Co. Ltd. and Bristol Myers Squibb Co. Ltd. Also, this work was supported by the Project for Development of Innovative Research on Cancer Therapeutics (P-DIRECT) and the Project for Cancer Research and Therapeutic Evolution (P-CREATE) of the Japan Agency for Medical Research and Development, AMED (grant numbers 15cm0106130h0002 and 16cm0106418h0001, respectively).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.