Evaluation of gefitinib alone or combined with bevacizumab in patients with EGFR L858R-positive advanced non-squamous non-small cell lung cancer: an open-label, randomized, phase 2 trial (BEVA-FLFX-001) with exploratory analysis of plasma biomarkers

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide (1). Non-small cell lung cancer (NSCLC) is the predominant histological type, with most patients presenting with advanced or metastatic disease at diagnosis. Epidermal growth factor receptor (EGFR) mutations in NSCLC define a special group of individuals, with distinct etiology of cancer, clinicopathological features, prognosis, and treatment strategies compared with other NSCLC patients. Most EGFR mutations occur in exons 18–21 of the tyrosine kinase domain of the receptor; exon 19 deletions (19del) and exon 21 alterations account for 90% of these mutations, with L858R being the predominant subtype of exon 21 mutations (2). Both types of mutations confer sensitivity to small-molecule EGFR tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib. However, almost all patients inevitably develop resistance to EGFR-TKIs, necessitating the development of novel therapeutic strategies for these patients. Although third-generation EGFR-TKIs such as osimertinib have demonstrated superior clinical benefit over first-generation agents and are now recommended as the standard first-line therapy, their accessibility remains limited in certain regions due to cost and reimbursement barriers. Moreover, resistance to osimertinib is ultimately inevitable, and combination strategies with antiangiogenic agents like bevacizumab are still under clinical investigation, with no conclusive survival benefit demonstrated to date (3-6). As such, optimizing alternative strategies with first-generation TKIs, particularly in high-risk subgroups such as patients with L858R-mutant NSCLC, continues to be clinically important.

Bevacizumab is a recombinant antiangiogenic monoclonal antibody that inhibits tumor angiogenesis and tumor growth via the vascular endothelial growth factor (VEGF) signaling pathway (7). The combination treatment of first-generation TKIs and bevacizumab has been assessed in multiple clinical trials, showing the potential to become a standard of care for NSCLC patients with either 19del or L858R mutations (8-12). Of particular interest, the JO25567 trial (9), NEJ026 trial (11), and CTONG1509 trial (12) all demonstrated significant progression-free survival (PFS) benefits for untreated NSCLC patients with common EGFR sensitizing mutations who received erlotinib plus bevacizumab compared with erlotinib [median PFS (mPFS): JO25567 trial, 16.0 vs. 9.7 months, P=0.02; NEJ026 trial, 16.9 vs. 13.3 months, P=0.02; CTONG1509 trial, 17.9 vs. 11.2 months, P<0.001]. Furthermore, subgroup analyses in the JO25567 trial showed a more favorable prognostic outcome of combined targeted and antiangiogenic therapy for 19del-positive NSCLC patients, while the NEJ026 trial and CTONG1509 trial showed superiority of combined therapy for L858R patients. In addition, the BELIEF trial, an international, multicenter, single-arm phase 2 study, evaluated the combination of erlotinib and bevacizumab in treatment-naive patients with EGFR-mutated advanced NSCLC (13). Patients were stratified by baseline T790M mutation status, with median PFS reported as 16.0 months in T790M-positive and 10.5 months in T790M-negative subgroups. These findings support the potential value of antiangiogenic therapy, particularly in molecularly defined populations, and reinforce the rationale for exploring combination strategies in EGFR-mutated NSCLC. However, it is worth noting that most patients in BELIEF harbored exon 19 deletions, and data specific to L858R-mutant cases remain limited. Besides, despite survival benefits observed in advanced EGFR-mutated NSCLC patients following first-line gefitinib plus bevacizumab therapy in the single-arm clinical trial, 19del patients could benefit more from treatment than those with L858R mutation (mPFS: 18.0 vs. 9.4 months, P=0.006) (10). The exploration of combination therapies in EGFR-mutant NSCLC patients is crucial given the varying therapeutic responses observed among patients with different EGFR mutation subtypes. While previous clinical trials have predominantly focused on the combination of bevacizumab and erlotinib, limited reports are available on the efficacy and safety of combining bevacizumab and gefitinib, an EGFR-TKI demonstrating similar effectiveness but a slightly improved safety profile compared to erlotinib.

To fill in the gap, we herein conducted this randomized, phase 2 clinical trial to compare the efficacy and safety of combining bevacizumab with gefitinib versus gefitinib monotherapy in advanced non-squamous NSCLC patients harboring L858R mutations. Through exploratory analysis of plasma biomarkers, this trial offers significant insights into the potential utility of plasma circulating tumor DNA (ctDNA) as a non-invasive biomarker for predicting treatment response and understanding mechanisms of acquired resistance, thereby informing potential second-line treatment options for patients with progressive disease. We present this article in accordance with the CONSORT reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/rc).

Methods

Trial design and oversight

This study was a single-center, open-label, randomized, phase 2 trial. All patients were recruited at Fudan University Shanghai Cancer Center between June 2020 and June 2021. Eligible patients were randomly allocated in a 1:1 ratio to receive gefitinib plus bevacizumab or gefitinib alone, using a balanced block randomization procedure. The randomization was stratified by sex (male vs. female) and clinical stage (stage IIIB vs. IV). All patients and investigators were unblinded to the treatment allocation. Gefitinib was administered orally at 250 mg/d, starting on the first day of the treatment protocol. Bevacizumab was administered intravenously at 7.5 mg/kg on the first day of the treatment protocol and was repeated at 3-week intervals. Vital signs, hematology, serum biochemistry, and urinalysis were performed every three weeks during the treatment period. Therapy continued until disease progression or unacceptable toxicity occurred.

This trial was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, the Good Clinical Practice principles of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use, and the guidelines for Good Clinical Practice of Drugs issued by the China Food and Drug Administration. The study protocol was approved by the Ethical Committee of the Fudan University Shanghai Cancer Center (No. 2004216-19; date of approval: June 2, 2020). All patients signed the written informed consent before any study procedure.

Participants

The eligibility criteria of this study were as follows: age greater than 18 years, pathologically confirmed with non-squamous NSCLC, stage IIIB/IV patients with inoperable locally advanced, recurrent or metastatic disease according to the eighth edition of the American Joint Committee on Cancer (AJCC) staging manual, primary tumors with solely EGFR L858R mutations, no prior systemic therapy, at least one measurable lesion within 28 days before randomization allocation, Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1, expected survival for at least 12 weeks, appropriate organ function, and acquisition of written informed consent. Baseline pretreatment evaluations are mainly composed of a complete history and physical examination, laboratory genetic testing, a chest radiograph, a computed tomography (CT) scan of the chest and abdomen, or a positron emission tomography CT scan. The baseline CT scan must be performed within 4 weeks before registration. Exclusion criteria were as follows: potential for curative treatment, pathologically confirmed mixed adenosquamous carcinoma with squamous cells as the main component (≥10%), history of malignancy other than NSCLC within 5 years before randomization except for those with negligible risk of metastasis or death (i.e., appropriately treated cervical carcinoma in situ, non-melanotic skin cancer, limited prostate cancer, breast ductal carcinoma in situ, stage I uterine cancer), history of hypertensive crisis or hypertensive encephalopathy, uncontrolled hypertension (blood pressure: systolic >150 mmHg and/or diastolic >100 mmHg), imaging evidence of tumor invasion to major blood vessels, and lactation/pregnancy.

Clinical outcomes

The primary endpoint of this study was PFS from randomization (Figure S1A). PFS in this study was defined as the interval between the date of randomization and the date of the first observation of disease progression or death from any cause, whichever comes first. The first patient was enrolled on June 11, 2020, and the data cutoff date was set for January 10, 2023. Secondary endpoints were objective response rate (ORR), disease control rate (DCR), and treatment-related adverse events. Exploratory endpoints were assessing the potential of plasma biomarkers and acquired resistance mechanisms that were explored through targeted next-generation sequencing (NGS).

Patient evaluation and follow-up

Treatment evaluation

Treatment responses were evaluated according to the standard Response Evaluation Criteria in Solid Tumors (RECIST, v1.1). All toxicities were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events v5.0. The safety evaluation includes monitoring and recording all adverse events, regular physical examinations, measurement of vital signs, laboratory tests, regular electrocardiogram (ECG), and echocardiography examinations to assess cardiac function and pregnancy tests.

Treatment adjustment and discontinuation

Treatment was discontinued if patients developed certain complications or comorbidities, including pneumonitis, intracranial hemorrhage, or gastrointestinal perforation (any grades); grade 4 vomiting; or other non-hematologic toxicity of grade ≥3 other than thromboembolic events, hypertension, and proteinuria. If patients experience intolerable diarrhea or skin adverse reactions, a short-term treatment interruption (up to 14 days) can be implemented, followed by a resumption of the daily gefitinib dose of 250 mg. In general, dose adjustment of bevacizumab is not recommended. However, the use of bevacizumab should be discontinued if patients develop gastrointestinal perforation involving visceral fistula formation, wound healing complications, severe bleeding, arterial thrombotic events, hypertensive crisis or encephalopathy, reversible posterior leukoencephalopathy syndrome, nephrotic syndrome. In the following conditions, the use of bevacizumab should be suspended: poorly controlled severe hypertension, moderate to severe proteinuria, and severe infusion reaction.

CtDNA profiling and outcome prediction

The assessment of EGFR mutation status was carried out on baseline formalin-fixed paraffin-embedded (FFPE) tumor tissues using RT-PCR (AmoyDx^® EGFR 29 Mutations Detection Kit, Amoy Diagnostics Co., Ltd., Xiamen, China) or targeted NGS (PULMOCAN^TM, Nanjing Geneseeq Technology Inc., Nanjing, China). Clinical samples were collected at three longitudinal time points for explorative analyses, including baseline, 6 weeks post-treatment (day 43, D43), and post-progression (Figure S1B). Targeted NGS was performed using a hybridization capture-based panel (PULMOCAN^TM, Nanjing Geneseeq Technology Inc., Nanjing, China) covering 139 lung cancer-related genes (14,15). The cutoff thresholds used to detect somatic genomic variants were: variant allele frequency (VAF) of ≥1% for single-nucleotide variants and insertion/deletion mutations; gene ratio of ≥2.0 for copy number gain and ≤0.6 for copy number loss; split-reads ≥2 for structural variants. Details are provided in the Appendix 1.

Statistical analysis

Sample size calculation was performed using the Power Analysis and Sample Size (PASS, version 15.0.5) software (NCSS, Kaysville, Utah). We hypothesized that patients who received gefitinib treatment would reach a median PFS of 9 months, and patients treated with gefitinib plus bevacizumab would reach a hazard ratio (HR) of 0.50 (9,16). To detect the anticipated improvement in PFS within an 18-month follow-up period with 80% power at a two-sided 5% significance level, a total of 104 patients were required. When assuming a cumulative dropout rate of 10%, a total of 116 patients were planned to be enrolled, with an allocation of 58 patients in each treatment group.

PFS was described using the Kaplan-Meier survival curves. If progression or death did not occur, patients were continued on follow-up and censored at the date of last visit, or the date of study closure, whichever comes first. Cox proportional hazard models were fitted to estimate HRs with 95% confidence intervals (CIs), and the proportionality of hazards was assessed using log(−log) survival plots. The differences between categorical variables were assessed using a Chi-squared test or Fisher’s exact test as appropriate. The differences between numerical variables were calculated using a Wilcoxon signed-rank test. A two-sided P value of less than 0.05 was considered significant for all tests unless indicated otherwise.

Results

Patient characteristics

Between June 2020 and June 2021, a total of 993 untreated patients with non-squamous NSCLC were screened for baseline somatic EGFR variants (Figure 1). After excluding EGFR wild-type patients and those harboring non-L858R EGFR mutations, a total of 198 patients were further assessed to determine eligibility for the clinical trial. Among these, 14 patients did not fulfill the inclusion criteria. Additionally, with osimertinib being covered by Chinese medical insurance as of March 1, 2021, and difficulties in hospital admission caused by coronavirus disease 2019 (COVID-19), 103 eligible patients opted not to participate in the clinical trial. As of the data cutoff date (January 10, 2023), the intention-to-treat population comprised 81 patients who were randomly assigned to receive gefitinib alone (N=39) or gefitinib plus bevacizumab (N=42) (Figure 1; Figure S1). The median age of patients was 62 years (range: 32–77 years), and 61.7% of patients (50/81) were female (Table 1). The majority of patients presented stage IV advanced NSCLC (97.5%, 79/81), and were never smokers (87.7 %, 71/81). All patients had a good performance status (ECOG status 0–1) and exhibited adenocarcinoma histology. Brain metastasis was observed in 26 patients (32.1%), while 55 patients (67.9%) had distant organ metastasis upon trial randomization. There were no significant differences in baseline clinical characteristics between the two treatment groups (Table 1).

Figure 1 Study flow. DCR, disease control rate; NSCLC, non-small cell lung cancer; ORR, objective response rate; PFS, progression-free survival.

Efficacy

At data cutoff (January 10, 2023), the median follow-up time for patients in the combination and monotherapy treatment groups was 25.4 months [interquartile range (IQR): 1.1–25.8 months] and 21.0 (IQR: 1.3–23.8 months), respectively. Censored data was observed in 12 patients, with 9 patients (2 in the control group and 7 in the experimental group) showing no progression of their disease, and three patients (2 in the control group and 1 in the experimental group) lost to follow-up. The PFS of patients receiving a combination of gefitinib and bevacizumab was significantly longer compared to the control group that received gefitinib alone [mPFS: 15.1 vs. 8.2 months; HR (95% CI): 0.49 (0.30–0.79); P=0.003] (Figure 2A). The subgroup analysis revealed that the combination therapy in EGFR L858R-mutated non-squamous NSCLC appeared to be a profound factor for a prolonged PFS based on various clinical features (Figure 2B). Notably, the addition of bevacizumab to gefitinib therapy was beneficial for patients with or without distant organ metastases (P=0.01 and P=0.05, respectively) (Figure S2A). Consistently, non-brain metastatic patients treated with combination therapy showed a significantly longer PFS compared to those treated with monotherapy [mPFS: 16.1 vs. 8.2 months, HR (95% CI): 0.45 (0.25–0.81), P=0.01] (Figure S2B). However, the survival difference between treatment groups did not reach statistical significance among patients with brain metastases, presumably due to the limited sample size.

Figure 2 Efficacy of gefitinib plus bevacizumab versus gefitinib alone in NSCLC patients with EGFR L858R mutations. (A) Kaplan-Meier curves of the PFS in patients in strata of type of treatment. (B) The forest plot of the HR for PFS to analyze the effects of gefitinib plus bevacizumab combination therapy versus gefitinib monotherapy after subgrouping of various clinical features in the study cohort. (C) Bar plots show the proportion of patients with various responses to treatment. (D,E) Bar plots showing the proportion of patients in strata of ORR (D) and DCR (E). CI, confidence interval; CR, complete response; DCR, disease control rate; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; mPFS, median PFS; ORR, objective response rate; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease.

One patient in the gefitinib plus bevacizumab group had a change of therapy and was withdrawn from the treatment response analysis due to the presence of interstitial pneumonia. Notably, none of the patients achieved a complete response (CR) following gefitinib treatment, compared to 2 (4.9%) in the gefitinib plus bevacizumab group (Figure 2C). Additionally, the combination therapy resulted in a higher number of patients achieving a partial response (PR) compared to the control group (35/41, 85.4% vs. 23/39, 59.0%). Consistently, the ORR was significantly higher in patients who underwent gefitinib plus bevacizumab treatment compared to those receiving monotherapy (90.2% vs. 59.0%, P<0.01) (Figure 2D). However, no significant difference was observed in the disease control rate (DCR) between the monotherapy and combination therapy groups (92.3% vs. 97.6%, P=0.35) (Figure 2E). Collectively, these results suggest that gefitinib plus bevacizumab might be a promising first-line treatment strategy for Chinese NSCLC patients who carry EGFR L858R mutations.

Toxicity

Treatment-related toxicity was assessed in all 81 patients. As shown in Table 2, rash was the most common adverse event observed in both treatment groups, and a slightly higher frequency was observed in patients who received gefitinib plus bevacizumab therapy (gefitinib + bevacizumab vs. gefitinib: 35.7% vs. 20.5%, P=0.15). The second most frequently reported adverse event was the alanine aminotransferase increase (13.6%, 11/81). Other common adverse events included aspartic transaminase increase, diarrhea, and hypertension. Notably, hypertension was exclusively found in patients treated with gefitinib plus bevacizumab therapy (19%, 8/42). In contrast, the frequency of patients experiencing an increase in red cell distribution width in the gefitinib treatment group was significantly higher compared to the other group (gefitinib vs. gefitinib + bevacizumab: 10.3% vs. 0%, P=0.049). Furthermore, two Grade 3 adverse events were observed in our study. While vein embolism was found in both treatment groups, interstitial pneumonia was found in one patient who received gefitinib plus bevacizumab combination therapy. However, it is important to note that these adverse events occurred at a low frequency, and the difference in incidence between the two treatment groups did not reach statistical significance.

Exploratory endpoints

Plasma ctDNA as a measurement of treatment outcome

A biomarker exploration for prognostic outcomes was conducted using targeted NGS (Figure S1). Somatic genomic alterations were detected from 66 plasma samples collected from patients at baseline, and the mutational landscape of these patients was delineated based on the type of treatment received (Figure S3A). The top frequently mutated genes were TP53 (60.6%), PIK3CA (9.1%), and RB1 (6.1%). Concomitant EGFR copy-number variants (CNVs) were identified in 5 patients (7.6%). Besides, we observed concurrent EGFR mutations in 9.1% of patients (6/66), including those in codons V834L, V689L, D1012V, P992R, L747S, and V1010E. Notably, we observed a significantly longer PFS in patients who carried co-existing mutations in addition to EGFR L858R who were treated with gefitinib plus bevacizumab compared to monotherapy [mPFS: 15.1 vs. 6.2 months; HR (95% CI): 0.40 (0.19–0.82); P=0.01] (Figure S3B). In contrast, a comparable survival benefit was observed among patients without L858R co-mutations treated with different modalities (P=0.28). Furthermore, we conducted a univariate analysis of PFS for genetic alterations with a mutational count ≥3 (Figure S3C). For EGFR L858R-mutated patients undergoing combination therapy, those with co-existing TP53 mutations had significantly shorter PFS compared to those without the mutation [mPFS: 15.1 vs. 18.8 months; HR (95% CI): 3.11 (1.21–7.97); P=0.01] (Figure S3D). However, there was no significant difference in PFS for those treated with monotherapy. Additionally, patients with baseline TP53 mutations who received combination therapy showed a better PFS than those in the monotherapy group, though the difference in survival benefit was not statistically significant (Figure S3E).

Next, we analyzed the association between plasma ctDNA and the patient’s prognosis in two treatment groups (Figure S1B). As shown in Figure 3A, gefitinib-treated patients who tested positive for ctDNA (ctDNA+) at baseline exhibited a nearly significant reduction in PFS compared to those with a negative ctDNA status (ctDNA-) [mPFS: 7.6 vs. 15.9 months, HR (95% CI): 2.66 (0.91–7.79); P=0.07]. In contrast, baseline plasma ctDNA status showed minimal value for predicting the PFS in patients treated with gefitinib plus bevacizumab combination therapy [mPFS: 16.2 vs. 19.1 months, HR (95% CI): 1.01 (0.35–2.97); P=0.99]. On the other hand, positive plasma ctDNA at day 43 (D43) post-treatment indicated a worse prognosis in patients regardless of the type of treatment (Figure 3B). These findings highlighted the clinical significance of plasma ctDNA as a promising noninvasive biomarker for patient prognosis.

Figure 3 Plasma ctDNA is a potential biomarker for prognosis estimation. (A) Kaplan-Meier curves of PFS in patients stratified by baseline plasma ctDNA status in two treatment groups. (B) Kaplan-Meier curves of PFS in patients stratified by plasma ctDNA status at day 43 (D43) post-treatment within the two patient subgroups. (C) Kaplan-Meier curves of PFS in patients stratified by ctDNA dynamics based on plasma ctDNA status in paired baseline and D43 samples within the two patient subgroups. (D) Bar plots showing the ORR among all patients with paired plasma samples before and after treatment, and the ORR of subgroup patients stratified by ctDNA dynamics. (E) Paired sample comparison of the maxVAF at baseline and post-treatment at 43 days. (F) Bar plots showing the proportion of patients with different treatment responses after subgrouping based on the maxVAF decline using a cutoff of 98% within the study cohort. (G) Subgroup analysis of the ORR of patients receiving gefitinib monotherapy versus gefitinib plus bevacizumab combination therapy. (H) Kaplan-Meier curves of the PFS in patients stratified by maxVAF decline using a cutoff of 98% within the two treatment groups. (I) Kaplan-Meier curves of the PFS in persistent ctDNA positive patients (N=21) stratified by maxVAF decline using a cutoff of 98% within the two treatment groups. CI, confidence interval; CR, complete response; ctDNA, circulating tumor DNA; HR, hazard ratio; maxVAF, maximal somatic variant allelic frequency; mPFS, median PFS; NEG, negative at baseline and D43; ORR, objective response rate; PD, progressive disease; PFS, progression-free survival; POS, positive at baseline and D43; PR, partial response; PTN, positive at baseline to negative at D43; SD, stable disease.

Next, we analyzed the association between ctDNA dynamics and prognosis using paired plasma samples obtained at both baseline and post-treatment from 67 patients (Figure S1B). Patients were categorized into three subgroups: ctDNA remained negative at baseline and D43 (NEG, N=11), ctDNA status changed from positive at baseline to negative at D43 (PTN, N=35), and ctDNA remained positive at baseline and D43 (POS, N=21). In patients treated with gefitinib alone, PTN patients showed a notably more prolonged PFS compared to POS patients [mPFS: 10.2 vs. 4.7 months, HR (95% CI): 0.16 (0.07–0.40), P<0.001] (Figure 3C). Similarly, for patients who underwent gefitinib plus bevacizumab combination therapy, PTN patients exhibited a significantly longer PFS than those with consistently positive ctDNA status [mPFS: 18.8 vs. 11.1 months, HR (95% CI): 0.27 (0.10–0.70), P=0.01]. Through assessment of patients’ treatment response, we found a significantly higher proportion of patients achieving CR/PR in the combination therapy group compared to the monotherapy treatment group, for both total paired plasma samples (93.5% vs. 55.6%, P<0.001) and PTN patients (100% vs. 64.7%, P=0.01) (Figure 3D). Collectively, these results indicate that post-treatment ctDNA obtained from patients’ plasma samples, and the dynamic change of ctDNA status could effectively estimate the recurrence risk of patients regardless of gefitinib monotherapy or combination therapy.

An indicative role of maximal somatic variant allelic frequency (maxVAF) decline in post-treatment plasma ctDNA for treatment efficacy

We further analyzed the ctDNA data of patients who had detectable baseline ctDNA and paired D43 samples and found a significant decrease in the maxVAF of D43 plasma ctDNA compared to paired baseline samples in both treatment groups (P<0.001) (Figure 3E). Among patients with a maxVAF decline >98%, the ORR was significantly higher in those who were treated with gefitinib plus bevacizumab combination therapy compared to those receiving only gefitinib (100% vs. 68.2%, P=0.01) (Figure 3F,3G). In contrast, no significant association between the combination therapy and an increased ORR was observed in patients who had a maxVAF decline ≤98%. In both treatment cohorts, patients displaying a maxVAF decline of >98% exhibited a significantly longer PFS compared to those with a maxVAF decline of ≤98% (P<0.001 and P=0.008, respectively) (Figure 3H). Furthermore, among the 21 patients with persistent positive DNA (POS), a similar trend was observed, indicating that a higher percentage of maxVAF decline was likely associated with a better survival benefit in both gefitinib-treated patients and those receiving gefitinib plus bevacizumab combination therapy (Figure 3I).

Treatment-associated acquired resistance mechanisms

We then compared the mutational profiles of 30 patients who had detectable genetic alterations in baseline and post-progression samples (Figure S4A). In post-progression samples, EGFR T790M mutations showed a higher frequency in patients treated with gefitinib monotherapy compared to patients who underwent gefitinib plus bevacizumab combination therapy (58.8% vs. 30.8%, P=0.16) (Figure S4B). Other less frequently detected acquired resistance mechanisms in the monotherapy group include concurrent T790M plus KRAS G12D, T790M plus BRAF V600M, and KRAS K117R. On the other hand, EGFR amplification was found to be co-mutated with T790M, presenting a novel underlying resistance mechanism in patients who developed resistance to combination therapy.

Discussion

We present this open-label, randomized, phase 2 clinical trial to evaluate the efficacy and safety of the combination therapy using gefitinib plus bevacizumab versus gefitinib monotherapy in untreated, advanced NSCLC patients with sensitizing EGFR L858R mutations, thus providing valuable clinical data from Chinese patients. The rationale for combining EGFR-TKIs with VEGF pathway inhibition lies in their potential synergistic effects on tumor growth and angiogenesis. EGFR activation has been shown to promote VEGF expression through the PI3K/AKT and RAS/RAF/MEK/ERK signaling pathways (17-19). Furthermore, EGFR-mutant NSCLC tumors are often characterized by enhanced VEGF expression and angiogenesis (20). Hypoxia-inducible factor 1-alpha (HIF-1α), a key transcriptional regulator of VEGF, may also be upregulated in EGFR-mutant tumors, either through hypoxic stress or EGFR signaling-mediated stabilization (21,22). These molecular interactions suggest that VEGF blockade may counteract pro-angiogenic signals induced by EGFR activation, potentially improving treatment efficacy. Nonetheless, heterogeneity in VEGF expression and tumor microenvironment may contribute to differential responses to anti-VEGF therapy, warranting further mechanistic exploration and biomarker-driven patient selection.

Frontline treatment using first-generation EGFR-TKIs, such as erlotinib and gefitinib, has demonstrated efficacy in treating EGFR-mutant NSCLC patients. However, resistance mostly occurs within approximately 10 months of EGFR-TKI monotherapy, emphasizing the urgent clinical need to develop new treatment strategies. Moreover, prior studies have demonstrated controversial prognostic outcomes of EGFR-TKI monotherapy treatment or combination therapy with antiangiogenic therapy in patients harboring EGFR subtype mutations (23). Indeed, subgroup analyses from different clinical studies showed distinct survival outcomes for 19del and L858R patients, highlighting the importance of specific treatment strategies for each mutation. In general, 19del patients could benefit more from EGFR-TKI monotherapy than patients with L858R mutations. Conversely, both the CTONG1509 trial and the NEJ026 trial demonstrated the efficacy of erlotinib plus bevacizumab combination therapy as a promising therapeutic approach for patients bearing L858R mutations compared with 19del patients. Hence, we carried out this phase 2 clinical trial to further assess the prognostic benefits and safety of combined targeted and antiangiogenic therapy exclusively in NSCLC patients with L858R mutations.

The dose of bevacizumab administered in this study was 7.5 mg/kg, representing half of the dose utilized in previous studies (9-12). Nonetheless, the median PFS of 15.1 months observed in the combination group in our study aligns closely with the range of 14.9 to 17.9 months reported in previous clinical trials exploring the efficacy of erlotinib plus bevacizumab (9,11,12,24). Indeed, it has been reported that there was no significant difference in efficacy for NSCLC patients who received first-line combination treatment with chemotherapy and bevacizumab, irrespective of whether administered at a dosage of 7.5 or 15 mg/kg (25). This finding highlights the importance of dose selection while taking both therapeutic efficacy and cost-effectiveness into account. On the other hand, a slightly shorter median PFS of 8.2 months was observed in the gefitinib monotherapy group. This observation could be attributed to the presence of baseline TP53 mutations, as co-mutations with TP53 have been associated with a decreased PFS in EGFR-mutated NSCLC patients treated with gefitinib based on the BENEFIT study (26). Comparing the treatment groups, we showed that the median PFS was 6.9 months longer with gefitinib plus bevacizumab treatment compared with the gefitinib monotherapy control (15.1 vs. 8.2 months, P<0.01). This finding was consistent with the NEJ206 trial, which demonstrated that the addition of bevacizumab to erlotinib therapy significantly improved PFS in L858R-mutated patients from 13.7 to 17.4 months (11). Furthermore, our findings recapitulated those of the CTONG1509 trial conducted in Chinese populations, which showed that the erlotinib plus bevacizumab group had doubled the PFS compared to the erlotinib control group in L858R-positive NSCLC patients (19.5 and 9.7 months, respectively) (12).

Although our study population is characterized by a higher representation of women and frequently detected brain/distant organ metastases at baseline, the PFS benefit of gefitinib combined with bevacizumab was consistently observed across most patient subgroups categorized based on the patient’s clinical characteristics. In patients with brain metastases at baseline, those who received combination therapy showed a longer PFS than those treated with monotherapy (14.1 vs. 8.1 months), although the difference was not statistically significant, presumably due to the limited cohort size. The combination therapy also provided significant improvement in PFS regardless of whether the patient had already developed distant metastases or not. Furthermore, we found that baseline mutational profiles of patients may also affect the therapeutic outcome of patients receiving bevacizumab plus gefitinib combination therapy. In our study, a subset of patients harboring L858R co-mutations who underwent combination therapy were found to exhibit a significantly longer PFS than those treated with monotherapy. Additionally, our findings suggest that baseline TP53 mutations are likely associated with an unfavorable prognosis in L858R-mutated NSCLC patients, and patients harboring concurrent TP53 mutations at baseline may benefit more from gefitinib plus bevacizumab combination therapy. These findings are in line with prior studies, such as the work by Chua et al. (27), which provides an important molecular characterization of EGFR-mutant NSCLC, particularly in T790M-negative, L858R-mutant tumors. TP53 co-mutations have been associated with a more aggressive tumor phenotype, reduced differentiation as reflected by low TTF-1 expression, and increased immune cell infiltration, suggesting a complex tumor biology that may influence treatment response (27). While our study did not include an assessment of TTF-1 expression or immune microenvironment features, these observations highlight that TP53 co-mutation may contribute to genomic instability and modulate resistance to EGFR-TKI therapy. Consequently, the presence of TP53 co-mutations could represent a biomarker for identifying high-risk patients who may benefit from alternative combination strategies or closer monitoring.

The toxicity of the combination therapy was manageable and tolerable. For the single patient who experienced vein embolism, a stent was successfully placed, and there was no adjustment made to the gefitinib dosage. Although the proportion of patients experiencing rash was higher than expected in the gefitinib plus bevacizumab group, the frequency was comparable to that observed in patients treated with gefitinib alone, and the difference did not reach statistical significance. Additionally, the frequency of hypertension and proteinuria was found to be higher in patients who received combination therapy, which aligns with findings from other trials reported for Chinese and Asian patients (9,11,12). However, it is worth noting that in the NEJ026 trial (11), 5 patients (4%) treated with erlotinib monotherapy were found to have interstitial lung diseases, while in our study, no patients in the gefitinib alone therapy group exhibited such conditions. We reasoned that these disparities among studies could potentially be attributed to ethnic differences among patient populations as well as the relatively small sample size in our study. Besides, we observed only two Grade 3 adverse events in the experimental group, which could be attributed to the lower dosage of bevacizumab administered in our study. Nevertheless, it is imperative to undertake additional research to assess the effectiveness and safety of diverse combinations of gefitinib and bevacizumab across various dosage levels.

Plasma ctDNA has been emerging as a non-invasive biomarker for cancer prognosis and monitoring. In this study, we employed a surveillance timepoint of 6 weeks post-treatment to monitor ctDNA levels, aiming to facilitate early detection of tumor progression. In both treatment groups, positive D43 ctDNA status and dynamic ctDNA levels between paired baseline and post-treatment samples were associated with the PFS and ORR of patients. However, the optimal interval time to assess ctDNA dynamics has been a subject of active investigation, yet a definitive conclusion has not been reached (26,28-31). Indeed, the timing of blood sampling intended for ctDNA analysis should be carefully selected depending on clinical questions, such as treatment received, concurrent inflammatory processes, and surgery, as different factors can affect the release of ctDNA. Hence, further research is warranted to enhance our understanding of the molecular mechanisms underlying ctDNA shedding into the blood and refine the timing of blood sampling for more accurate assessment.

In this present study, it was found that 30.8% of patients who received gefitinib plus bevacizumab combination therapy had EGFR T790M mutations upon disease progression, which was lower than that observed in the control group (58.8%). These results were consistent with the findings from the CTONG1509 trial (12). As osimertinib has shown efficacy in patients with pretreated, T790M-positive, advanced NSCLC patients (32,33), osimertinib monotherapy might serve as a promising approach for those who have developed resistance to gefitinib plus bevacizumab therapy. On the other hand, previously reported acquired mechanisms of resistance to first-/second-generation EGFR-TKIs (i.e., MET amplifications) were not observed in our study (34). This could be due to the limited power of analysis attributable to the small number of patients.

There are limitations to this study. Firstly, the overall survival data remained immature, which limited the scope of analyses. Secondly, we present a prospective single-center, real-world study investigating the first-generation EGFR-TKI gefitinib with or without the addition of bevacizumab, which may carry a considerable risk of selection bias. Thirdly, our study was limited by the sample size for subgroup analysis investigating the differences between molecular mechanisms of acquired resistance between treatment groups upon disease progression. Additionally, the low prevalence of co-occurring genomic alterations other than TP53 mutants in our cohort limited a comprehensive assessment of their impact on clinical outcomes. For example, LRP1B mutations were detected in only 3% (2/66) of patients. Furthermore, our targeted sequencing panel did not include RBM10, precluding analysis of its potential impact as reported in a previous study, which has shown that RBM10 co-mutations are associated with efficacy of EGFR-TKI therapy and may underlie primary resistance mechanisms in EGFR-mutant NSCLC (35). The absence of these data highlights the need for broader genomic profiling in future studies to better elucidate the influence of such co-mutations on therapeutic response.

Conclusions

In conclusion, bevacizumab combined with gefitinib showed significantly improved PFS and ORR in advanced NSCLC patients harboring EGFR L858R mutations with well-tolerated toxicity. Hence, combination therapy with bevacizumab plus gefitinib might be a promising first-line treatment modality for L858R-positive NSCLC patients.

Acknowledgments

We thank all the patients and family members who gave their consent to present the data in this study, as well as the investigators and research staff involved. A portion of the data from this study was previously presented as a poster (Abstract 1332P) at the 2023 European Society for Medical Oncology (ESMO) Congress.

Footnote

Reporting Checklist: The authors have completed the CONSORT reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/rc

Trial Protocol: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/tp

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/prf

Funding: This work was supported by Shanghai’s 2022 “Science and Technology Innovation Action Plan” Special Project in Medical Innovation Research (No. 22Y31920405 to J.W.); the Collaborative Innovation Cluster Program of the Shanghai Municipal Health Commission (No. 2020CXJQ02 to J.W.); the Soaring Program, Shanghai Anti-Cancer Association (No. SACA-AX202210 to J.W.); the Beijing Life Oasis Public Service Center (No. CPHCF-ZLKY-2023016 to J.W.); the China Aging Development Foundation (No. 96 to X.Z.); the China Health and Medical Development Foundation (No. CHMDF2025-XRKY02-12 to X.Z.); and the Pan-Tumor Therapeutic Research Fund of the China Primary Health Care Foundation (No. QD-SH0008 to X.Z.). Additional support was provided by the Beijing Health Alliance Charitable Foundation and the Guangzhou Life Oasis Public Service Center (to X.Z.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-545/coif). X.Z. reports funding from the China Aging Development Foundation (No. 96), the China Health and Medical Development Foundation (No. CHMDF2025-XRKY02-12), and the Pan-Tumor Therapeutic Research Fund of the China Primary Health Care Foundation (No. QD-SH0008), and the Beijing Health Alliance Charitable Foundation and the Guangzhou Life Oasis Public Service Center. K.X., C.L., S.W., H.B., and Q.O. are employees of Nanjing Geneseeq Technology Inc. J.W. reports funding from the Shanghai’s 2022 “Science and Technology Innovation Action Plan” Special Project in Medical Innovation Research (No. 22Y31920405); the Collaborative Innovation Cluster Program of the Shanghai Municipal Health Commission (No. 2020CXJQ02); the Soaring Program, Shanghai Anti-Cancer Association (No. SACA-AX202210); the Beijing Life Oasis Public Service Center (No. CPHCF-ZLKY-2023016). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This trial was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, the Good Clinical Practice principles of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use, and the guidelines for Good Clinical Practice of Drugs issued by the China Food and Drug Administration. The study protocol was approved by the Ethical Committee of the Fudan University Shanghai Cancer Center (No. 2004216-19; date of approval: June 2, 2020). All patients signed the written informed consent before any study procedure.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Miller KD, Wagle NS, et al. Cancer statistics, 2023. CA Cancer J Clin 2023;73:17-48. [Crossref] [PubMed]
2. Liu Y, Chen L, Zhu X, et al. Real-world prognostic factors for first-line EGFR-TKI efficacy in advanced NSCLC patients harboring EGFR 21 L858R mutation. Glob Med Genet 2025;12:100051. [Crossref] [PubMed]
3. Akamatsu H, Toi Y, Hayashi H, et al. Efficacy of Osimertinib Plus Bevacizumab vs Osimertinib in Patients With EGFR T790M-Mutated Non-Small Cell Lung Cancer Previously Treated With Epidermal Growth Factor Receptor-Tyrosine Kinase Inhibitor: West Japan Oncology Group 8715L Phase 2 Randomized Clinical Trial. JAMA Oncol 2021;7:386-94. [Crossref] [PubMed]
4. Soo RA, Han JY, Dafni U, et al. A randomised phase II study of osimertinib and bevacizumab versus osimertinib alone as second-line targeted treatment in advanced NSCLC with confirmed EGFR and acquired T790M mutations: the European Thoracic Oncology Platform (ETOP 10-16) BOOSTER trial. Ann Oncol 2022;33:181-92. [Crossref] [PubMed]
5. Yi Y, Cai J, Xu P, et al. Potential benefit of osismertinib plus bevacizumab in leptomeningeal metastasis with EGFR mutant non-small-cell lung cancer. J Transl Med 2022;20:122. [Crossref] [PubMed]
6. Zhou G, Guo L, Xu J, et al. Comparison of osimertinib plus bevacizumab against osimertinib alone in NSCLC harboring EGFR mutations: a systematic review and meta-analysis. Ther Adv Med Oncol 2024;16:17588359241227677. [Crossref] [PubMed]
7. Chitoran E, Rotaru V, Stefan DC, et al. Blocking Tumoral Angiogenesis VEGF/VEGFR Pathway: Bevacizumab-20 Years of Therapeutic Success and Controversy. Cancers (Basel) 2025;17:1126. [Crossref] [PubMed]
8. Herbst RS, Ansari R, Bustin F, et al. Efficacy of bevacizumab plus erlotinib versus erlotinib alone in advanced non-small-cell lung cancer after failure of standard first-line chemotherapy (BeTa): a double-blind, placebo-controlled, phase 3 trial. Lancet 2011;377:1846-54. [Crossref] [PubMed]
9. Seto T, Kato T, Nishio M, et al. Erlotinib alone or with bevacizumab as first-line therapy in patients with advanced non-squamous non-small-cell lung cancer harbouring EGFR mutations (JO25567): an open-label, randomised, multicentre, phase 2 study. Lancet Oncol 2014;15:1236-44. [Crossref] [PubMed]
10. Ichihara E, Hotta K, Nogami N, et al. Phase II trial of gefitinib in combination with bevacizumab as first-line therapy for advanced non-small cell lung cancer with activating EGFR gene mutations: the Okayama Lung Cancer Study Group Trial 1001. J Thorac Oncol 2015;10:486-91. [Crossref] [PubMed]
11. Saito H, Fukuhara T, Furuya N, et al. Erlotinib plus bevacizumab versus erlotinib alone in patients with EGFR-positive advanced non-squamous non-small-cell lung cancer (NEJ026): interim analysis of an open-label, randomised, multicentre, phase 3 trial. Lancet Oncol 2019;20:625-35. [Crossref] [PubMed]
12. Zhou Q, Xu CR, Cheng Y, et al. Bevacizumab plus erlotinib in Chinese patients with untreated, EGFR-mutated, advanced NSCLC (ARTEMIS-CTONG1509): A multicenter phase 3 study. Cancer Cell 2021;39:1279-1291.e3. [Crossref] [PubMed]
13. Rosell R, Dafni U, Felip E, et al. Erlotinib and bevacizumab in patients with advanced non-small-cell lung cancer and activating EGFR mutations (BELIEF): an international, multicentre, single-arm, phase 2 trial. Lancet Respir Med 2017;5:435-44. [Crossref] [PubMed]
14. Zhang C, Zhang J, Xu FP, et al. Genomic Landscape and Immune Microenvironment Features of Preinvasive and Early Invasive Lung Adenocarcinoma. J Thorac Oncol 2019;14:1912-23. [Crossref] [PubMed]
15. Zhao W, Song A, Xu Y, et al. Rare mutation-dominant compound EGFR-positive NSCLC is associated with enriched kinase domain-resided variants of uncertain significance and poor clinical outcomes. BMC Med 2023;21:73. [Crossref] [PubMed]
16. Wu YL, Saijo N, Thongprasert S, et al. Efficacy according to blind independent central review: Post-hoc analyses from the phase III, randomized, multicenter, IPASS study of first-line gefitinib versus carboplatin/paclitaxel in Asian patients with EGFR mutation-positive advanced NSCLC. Lung Cancer 2017;104:119-25. [Crossref] [PubMed]
17. Wang L, Liu WQ, Broussy S, et al. Recent advances of anti-angiogenic inhibitors targeting VEGF/VEGFR axis. Front Pharmacol 2023;14:1307860. [Crossref] [PubMed]
18. Wang Q, Zeng A, Zhu M, et al. Dual inhibition of EGFR‑VEGF: An effective approach to the treatment of advanced non‑small cell lung cancer with EGFR mutation Int J Oncol 2023;62:26. (Review). [Crossref] [PubMed]
19. Osude C, Lin L, Patel M, et al. Mediating EGFR-TKI Resistance by VEGF/VEGFR Autocrine Pathway in Non-Small Cell Lung Cancer. Cells 2022;11:1694. [Crossref] [PubMed]
20. Le X, Nilsson M, Goldman J, et al. Dual EGFR-VEGF Pathway Inhibition: A Promising Strategy for Patients With EGFR-Mutant NSCLC. J Thorac Oncol 2021;16:205-15. [Crossref] [PubMed]
21. Dan A, Burtavel LM, Coman MC, et al. Genetic Blueprints in Lung Cancer: Foundations for Targeted Therapies. Cancers (Basel) 2024;16:4048. [Crossref] [PubMed]
22. Zhang Y, Zheng Y, Zhang J, et al. Apoptotic signaling pathways in bone metastatic lung cancer: a comprehensive analysis. Discov Oncol 2024;15:310. [Crossref] [PubMed]
23. Li WQ, Cui JW. Non-small cell lung cancer patients with ex19del or exon 21 L858R mutation: distinct mechanisms, different efficacies to treatments. J Cancer Res Clin Oncol 2020;146:2329-38. [Crossref] [PubMed]
24. Piccirillo MC, Bonanno L, Garassino MC, et al. Addition of Bevacizumab to Erlotinib as First-Line Treatment of Patients With EGFR-Mutated Advanced Nonsquamous NSCLC: The BEVERLY Multicenter Randomized Phase 3 Trial. J Thorac Oncol 2022;17:1086-97. [Crossref] [PubMed]
25. Zhou CH, Yang F, Jiang WJ, et al. Efficacy and Safety of Different Doses of Bevacizumab Combined With Pemetrexed and Platinum in First-Line Treatment of Advanced NSCLC: A Retrospective-Real World Study. Front Pharmacol 2021;12:727102. [Crossref] [PubMed]
26. Wang Z, Cheng Y, An T, et al. Detection of EGFR mutations in plasma circulating tumour DNA as a selection criterion for first-line gefitinib treatment in patients with advanced lung adenocarcinoma (BENEFIT): a phase 2, single-arm, multicentre clinical trial. Lancet Respir Med 2018;6:681-90. [Crossref] [PubMed]
27. Chua KP, Teng YHF, Tan AC, et al. Integrative Profiling of T790M-Negative EGFR-Mutated NSCLC Reveals Pervasive Lineage Transition and Therapeutic Opportunities. Clin Cancer Res 2021;27:5939-50. [Crossref] [PubMed]
28. Fukuhara T, Saito H, Furuya N, et al. Evaluation of plasma EGFR mutation as an early predictor of response of erlotinib plus bevacizumab treatment in the NEJ026 study. EBioMedicine 2020;57:102861. [Crossref] [PubMed]
29. Ramalingam SS, Vansteenkiste J, Planchard D, et al. Overall Survival with Osimertinib in Untreated, EGFR-Mutated Advanced NSCLC. N Engl J Med 2020;382:41-50. [Crossref] [PubMed]
30. Kwon M, Ku BM, Olsen S, et al. Longitudinal monitoring by next-generation sequencing of plasma cell-free DNA in ALK rearranged NSCLC patients treated with ALK tyrosine kinase inhibitors. Cancer Med 2022;11:2944-56. [Crossref] [PubMed]
31. Yang Y, Zhang T, Wang J, et al. The clinical utility of dynamic ctDNA monitoring in inoperable localized NSCLC patients. Mol Cancer 2022;21:117. [Crossref] [PubMed]
32. Ahn MJ, Tsai CM, Shepherd FA, et al. Osimertinib in patients with T790M mutation-positive, advanced non-small cell lung cancer: Long-term follow-up from a pooled analysis of 2 phase 2 studies. Cancer 2019;125:892-901. [Crossref] [PubMed]
33. Mok TS, Wu Y-L, Ahn M-J, et al. Osimertinib or Platinum-Pemetrexed in EGFR T790M-Positive Lung Cancer. N Engl J Med 2017;376:629-40. [Crossref] [PubMed]
34. Qin K, Hong L, Zhang J, et al. MET Amplification as a Resistance Driver to TKI Therapies in Lung Cancer: Clinical Challenges and Opportunities. Cancers (Basel) 2023;15:612. [Crossref] [PubMed]
35. Nanjo S, Wu W, Karachaliou N, et al. Deficiency of the splicing factor RBM10 limits EGFR inhibitor response in EGFR-mutant lung cancer. J Clin Invest 2022;132:e145099. [Crossref] [PubMed]