Evaluation of Clinical Outcomes of Icotinib in Patients With Clinically Diagnosed Advanced Lung Cancer With EGFR-Sensitizing Variants Assessed by Circulating Tumor DNA Testing: A Phase 2 Nonrandomized Clinical Trial

Key Points

Question  How is the clinical outcome of icotinib in treatment-naive patients with clinically diagnosed advanced lung cancer with unknown pathological status and positive epidermal growth factor receptor (EGFR) variants assessed by circulating tumor DNA (ctDNA)?

Findings  In this phase 2 nonrandomized clinical trial including 116 patients with clinically diagnosed advanced lung cancer with unknown pathological status and positive plasma EGFR variants, the objective response rate of first-line icotinib was 52.6%, and the median progression-free survival was 10.3 months.

Meaning  For patients with clinically diagnosed advanced lung cancer who were unable to provide tissues for pathological diagnosis, ctDNA-based EGFR genotyping could help decision-making in particular clinical situations.

Importance  The inability to obtain a pathological diagnosis in a certain proportion of patients with clinically diagnosed advanced lung cancer impedes precision treatment in clinical practice.

Objective  To evaluate the clinical outcome of first-line icotinib in patients with clinically diagnosed advanced lung cancer with unknown pathological status and positive epidermal growth factor receptor (EGFR)–sensitizing variants assessed by circulating tumor DNA (ctDNA).

Design, Setting, and Participants  The Efficiency of Icotinib in Plasma ctDNA EGFR Mutation-Positive Patients Diagnosed With Lung Cancer (CHALLENGE) trial is a prospective, multicentered, open-label, single-arm phase 2 nonrandomized clinical trial conducted between July 1, 2017, and July 31, 2019. Patients with systemic treatment-naive, clinically diagnosed advanced peripheral lung cancer, unknown pathological status, and positive pretreatment plasma EGFR-sensitizing variants were eligible. A total of 391 potentially eligible Chinese patients from 19 centers in China were screened for ctDNA EGFR variants by 3 independent detection platforms (Super amplification refractory mutation system [SuperARMS] polymerase chain reaction, droplet digital polymerase chain reaction, and next-generation sequencing), and those with EGFR variants tested by any platform were included. Analyses were conducted from September 9 to December 31, 2021.

Interventions  Enrolled patients were treated with oral icotinib tablets (125 mg 3 times daily) until disease progression, death, or treatment discontinuation due to various reasons, such as toxic effects and withdrawing consent.

Main Outcomes and Measures  The primary end point was objective response rate (ORR). The secondary end points included progression-free survival (PFS), overall survival (OS), disease control rate (DCR), and the concordance among the 3 detection platforms.

Results  Of 116 included patients, 76 (65.5%) were female, and the median (range) age was 64 (37-85) years. The median (IQR) follow-up duration was 36.3 (30.2-40.7) months. The ORR was 52.6% (95% CI, 43.1%-61.9%). The median PFS and OS were 10.3 months (95% CI, 8.3-12.2) and 23.2 months (95% CI, 17.7-28.0), respectively, and the DCR was 84.5% (95% CI, 76.6%-90.5%). The concordance rate among the 3 detection platforms was 80.1% (313 of 391), and the clinical outcomes in patients identified as positive by any platform were comparable.

Conclusions and Relevance  This prospective phase 2 nonrandomized clinical trial suggests that for patients with clinically diagnosed advanced lung cancer with unknown pathological status, ctDNA-based EGFR genotyping could help decision-making in particular clinical situations, while still warranting future larger-scaled real-world exploration.

Trial Registration  ClinicalTrials.gov Identifier: NCT03346811

Histopathological diagnosis and molecular genotyping have been the cornerstones of precision medicine in cancer treatment, especially in non–small-cell lung cancer (NSCLC) with epidermal growth factor receptor (EGFR) variants. However, there always exist a certain proportion of patients who cannot provide tissues for pathological diagnosis, let alone tissue-based molecular testing.^1,2 Even in well-designed prospective trials, approximately one-third of patients failed to provide sufficient tissues for molecular detection.³ Therefore, tissue-based diagnosis and treatment remains challenging in a certain subset of patients.

Currently, circulating tumor DNA (ctDNA) has been established as an important surrogate of tissues for molecular testing.^4-7 Several studies have exhibited the pivotal role of ctDNA to prospectively predict clinical trajectories,^8,9 and our previous prospective trial also proved the feasibility of using ctDNA-based EGFR variation to guide gefitinib in treatment-naive patients with advanced lung adenocarcinoma.^10,11 However, the patients in these studies all have definitive pathological diagnoses. For patients with unknown pathologic status, the potential role of ctDNA remains unclear.

Therefore, we initiated this prospective multicentered clinical trial to evaluate the potential of ctDNA to instruct coping strategy in this specific subset of real-world patients.

The Efficiency of Icotinib in Plasma ctDNA EGFR Mutation-Positive Patients Diagnosed With Lung Cancer (CHALLENGE) trial is a prospective, multicentered, open-label, single-arm phase 2 nonrandomized clinical trial aiming to evaluate the clinical outcome of icotinib in treatment-naive patients with clinically diagnosed advanced peripheral lung cancer with unknown pathological status and positive plasma EGFR-sensitizing variants (EGFR Exon 19 deletion and/or EGFR L858R). The reasons for unobtained pathological diagnosis were prospectively categorized and documented. All potentially eligible patients were screened by 3 independent platforms: Super amplification refractory mutation system (SuperARMS) polymerase chain reaction (PCR), droplet digital PCR (ddPCR), or next-generation sequencing (NGS), and patients with plasma EGFR variants tested by any platform were eligible. Enrolled patients were treated with icotinib tablets (125 mg 3 times daily; Betta Pharmaceuticals) until disease progression, death, or treatment discontinuation due to various reasons, such as toxic effects and withdrawing consent. This study was approved by the Ethics Committee of Cancer Hospital Chinese Academy of Medical Sciences and independent local ethics committees and was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all participants. The trial protocol can be found in Supplement 1. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

The primary end point was the objective response rate (ORR). The secondary end points included progression-free survival (PFS), overall survival (OS), disease control rate (DCR), and the concordance among the SuperARMS PCR, ddPCR, and NGS platforms. The justification of sample size, the statistical methods we used to analyze the end points, and the calculation of concordance can be found in the eMethods in Supplement 2. All analyses were conducted using SAS version 9.4 (SAS Institute).

Between July 1, 2017, and July 31, 2019, 391 Chinese patients from 19 centers in China were screened; 134 patients were eligible for enrollment, and a total of 116 patients were included in the trial (Figure 1). Of 116 included patients, 76 (65.5%) were female, and the median (range) age was 64 (37-85) years. The median (IQR) follow-up duration was 36.3 (30.2-40.7) months. The reasons for unobtained pathological diagnosis included advanced age (32 [27.6%]), violation of cultural traditions (31 [26.7%]), and failure of repeating previous biopsies (23 [19.8%]) (Table). After enrollment or medication, 18 patients (15.6%) subsequently provided pathological diagnoses, including 16 lung adenocarcinomas, 1 lung squamous cell carcinoma, and 1 lung adenosquamous carcinoma.

By the data cutoff (September 9, 2021), 28 of 116 patients (24.1%) were alive, and 6 (5.2%) were still receiving treatment. A total of 12 patients (10.3%) were lost to follow-up. In the 116 enrolled patients, the ORR was 52.6% (95% CI, 43.1%-61.9%). The median PFS and OS was 10.3 months (95% CI, 8.3-12.2) and 23.2 months (95% CI, 17.7-28.0), respectively, and the DCR was 84.5% (95% CI, 76.6%-90.5%) (Figure 2; eTable 1 in Supplement 2). A total of 89 patients (76.7%) experienced at least 1 treatment-related adverse event (TRAE), and the most common TRAEs were rash (60 [51.7%]) and fatigue (58 [50.0%]). Eight patients (6.9%) experienced grade 3 to 4 TRAEs, and the most common grade 3 to 4 TRAE was fatigue (3 [2.6%]) (eTable 2 in Supplement 2). There were no newly reported TRAEs, and none of the patients discontinued the treatment due to toxic effects.

Among the 391 screened patients, 140 (35.8%) had plasma EGFR-sensitizing variants tested by any of the 3 platforms (Figure 1). The EGFR positive rate was the highest through NGS (114 of 140 [81.4%]), followed by ddPCR (103 of 140 [73.5%]) and SuperARMS PCR (100 of 140 [71.4%]), and the concordance rate among these 3 platforms was 80.1% (313 of 391). Notably, the clinical outcomes in patients detected positive by any platform were comparable (eFigure and eTable 3 in Supplement 2).

The inability to obtain histopathological diagnosis in patients with lung cancer has deprived some patients of the opportunity to benefit from precision treatment. In this study, we unveiled the potential of ctDNA-based EGFR genotyping to prospectively guide first-line icotinib treatment in patients with clinically diagnosed advanced lung cancer and unknown pathological status. To the best of our knowledge, this is the first prospective study focusing on this specific population to provide evidence for possible coping strategies.

Icotinib, a first-generation EGFR tyrosine kinase inhibitor independently developed in China, exhibited an ORR of 64.8% in treatment-naive patients with NSCLC harboring EGFR variants derived from tissue-based testing from the Icotinib Versus First-line Chemotherapy Plus Maintenance Treatment in EGFR Positive Lung Adenocarcinoma Patients (CONVINCE) trial.¹² In a phase IV trial containing 6087 patients with advanced NSCLC treated with icotinib in the real world, the ORR in treatment-naive patients with EGFR-variant NSCLC was 56.3%.¹³ Our study demonstrated an ORR of 52.6% (95% CI, 43.1%-61.9%), inferior than that in CONVINCE trial,¹² which could be mainly explained by the different inclusion criteria between the rigorously designed phase III trial and our study for this specific population. Compared with the CONVINCE trial, we enrolled a higher proportion of patients with older age (median [range] age, 64 [37-85] years vs 56 [35-74] years), Eastern Cooperative Oncology Group performance status 2-4 (11 of 116 [9.5%] vs 7 of 148 [4.7%]), brain metastasis (37 of 116 [31.9%] vs 41 of 148 [27.7%]), and EGFR L858R (65 of 116 [56.0%] vs 68 of 148 [45.9%]). Moreover, the EGFR variants occurred in nonadenocarcinoma¹⁴ and the clonal hematopoiesis,¹⁵ although very rare in EGFR-sensitizing variants (Exon 19 deletion and L858R) could also lead to an inferior ORR. However, the ORR in our study was similar with the real-world study, further indicating the potential of ctDNA-based EGFR genotyping to guide first-line icotinib in patients with clinically diagnosed lung cancer in the real world.

In addition, the median PFS in our study was comparable to the CONVINCE trial (10.3 months [95% CI, 8.3-12.2] vs 11.2 months). The shorter median OS (23.2 months [95% CI, 17.7-28.0] vs 30.5 months) could be explained by the higher proportion of patients with worse conditions and the lack of following precise therapeutic strategy owing to the uncertain pathological status. Moreover, although NGS could uncover more EGFR-positive patients, the clinical outcomes guided by 3 platforms were similar, suggesting that for this specific subset of patients, the assessment of ctDNA-based EGFR variants via any platform was feasible.

This study has limitations. First, this trial was initiated and conducted before osimertinib was approved as first-line treatment for EGFR-variant NSCLC in China. Further studies are needed to explore whether the same findings could be observed with third-generation tyrosine kinase inhibitors. Next, histopathological diagnosis still remains the criterion standard in current oncology practice, and the reasons for unobtained pathological diagnosis in this study might be limited to regional issues in China. However, we indeed provided a promising future solution for this specific subset of patients who do exist worldwide. Additionally, although historical controls were used, the interpretation of our findings should still be cautious because of its single-arm setting.

For patients with clinically diagnosed advanced lung cancer with unknown pathological status, ctDNA-based EGFR genotyping could help decision-making under particular clinical situations, while still warranting future larger-scaled real-world exploration.

Accepted for Publication: May 10, 2022.

Published Online: July 21, 2022. doi:10.1001/jamaoncol.2022.2719

Corresponding Author: Jie Wang, MD, PhD (zlhuxi@163.com), and Zhijie Wang, MD (jie_969@163.com), State Key Laboratory of Molecular Oncology, Department of Medical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No. 17, Panjiayuannanli, Beijing 100021, China.

Author Contributions: Dr J. Wang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Xu, Zheng Liu, Bai, and Dong contributed equally as co–first authors. Drs Duan, Z. Wang, and J. Wang contributed equally as co–senior authors.

Study concept and design: Xu, Zheng Liu, Wan, Zang, Q. Li, Guo, Huang, Lv, J. Zhang, Y. Zhao, Gao, L. Li, Z. Wang, J. Wang.

Acquisition, analysis, or interpretation of data: Xu, Zheng Liu, Bai, Dong, J. Zhong, Wan, X. Li, Q. Li, Guo, Du, D. Zhong, Lv, J. Zhang, Y. Zhao, Gao, L. Li, C. Zhang, J. Zhao, B. Li, Zhe Liu, Yang, Ji, T. Wang, Duan, Z. Wang, J. Wang.

Drafting of the manuscript: Xu, Zheng Liu, Q. Li, Guo, Du, Lv, Y. Zhao, Gao, L. Li, C. Zhang, J. Zhao, Zhe Liu, Yang, Z. Wang, J. Wang.

Critical revision of the manuscript for important intellectual content: Xu, Zheng Liu, Bai, Dong, J. Zhong, Wan, Zang, X. Li, Q. Li, Guo, D. Zhong, Huang, Lv, J. Zhang, Y. Zhao, Gao, C. Zhang, B. Li, Ji, T. Wang, Duan, Z. Wang, J. Wang.

Statistical analysis: Xu, Zheng Liu, Dong, Q. Li, Guo, Huang, Lv, J. Zhang, Y. Zhao, Gao, C. Zhang, B. Li, Yang, Ji, J. Wang.

Obtained funding: Xu, Zheng Liu, Q. Li, Guo, Huang, Lv, Y. Zhao, Gao, Ji, Z. Wang, J. Wang.

Administrative, technical, or material support: Bai, J. Zhong, Wan, Zang, X. Li, Huang, J. Zhang, Y. Zhao, L. Li, C. Zhang, Zhe Liu, Ji, T. Wang, Duan, Z. Wang, J. Wang.

Study supervision: X. Li, Du, Huang, J. Zhang, Y. Zhao, Duan, J. Wang.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by the National Key Research and Development Project (grant 2019YFC1315700 to Dr Jie Wang); the National Natural Sciences Foundation Key Program (grant 81630071 to Dr Jie Wang); Aiyou Foundation (grant KY201701 to Dr Jie Wang); the National Natural Sciences Foundation (grants 81871889 and 82072586 to Dr Zhijie Wang; grant 81972905 to Dr Duan; grant 82102886 to Dr Xu); CAMS Innovation Fund for Medical Sciences (grants 2021-I2M-1-012 to Dr Zhijie Wang); Beijing Natural Science Foundation (grant 7212084 to Dr Zhijie Wang); Beijing Hope Run Special Fund of Cancer Foundation of China (grant LC2020B09 to Dr Xu); and the Special Research Fund for Central Universities, Peking Union Medical College (grant 3332021029 to Dr Xu).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We thank Chunxue Guo, MPH (Hengpu Yinuo [Beijing] Technology Co, Beijing, China), for the statistical analyses. She was not compensated for her work.