Efficacy and safety of camrelizumab plus apatinib in previously treated patients with advanced non-small cell lung cancer harboring EGFR or ALK genetic aberration.

Background: Camrelizumab plus apatinib shows encouraging antitumor activity and acceptable toxicity in chemotherapy-pretreated patients with advanced non-small cell lung cancer (NSCLC); however, clinical benefits from this combination regimen in NSCLC patients with EGFR mutations or ALK rearrangements (EGFR+/ALK+) have not been reported. We assessed the efficacy and safety of this combined regimen in pretreated patients with advanced NSCLC and defined EGFR/ALK status (EGFR+/ALK+) in a phase 1b/2 trial.
Methods: Previously treated patients with advanced EGFR+/ALK+ NSCLC were enrolled and given camrelizumab 200 mg intravenously every 2 weeks plus apatinib at the recommended dose of 250 mg orally once daily. Patients harboring sensitive EGFR mutations or ALK fusion genes had received at least one EGFR/ALK TKI and a platinum-based chemotherapy regimen before the enrollment. The primary endpoint was objective response rate (ORR).
Results: All 43 enrolled patients comprised the efficacy and safety analysis population. The confirmed ORR was 18.6% (95% CI: 8.4-33.4%) and the clinical benefit response rate was 27.9% (95% CI: 15.3-43.7%). Median progression-free survival (PFS) was 2.8 months (95% CI: 1.9-5.5 months) and median overall survival was not reached (95% CI: 7.3 months-not reached), with a median follow-up period of 15.7 months (range, 0.5-24.4 months). The most common grade ≥3 treatment-related adverse events (TRAEs) were hypertension (16.3%), proteinuria (11.6%) and palmar-plantar erythrodysaesthesia syndrome (9.3%). No unexpected adverse events were recorded. Conclusions: Camrelizumab plus apatinib showed moderate antitumor activity and acceptable safety profile in previously treated patients with advanced NSCLC and EGFR or ALK genetic aberrations, which warranted further validation. Trial Registration: ClinicalTrials.gov identifier: NCT03083041. Registered March 17, 2017. 2022 Translational Lung Cancer Research. All rights reserved.

Introduction

Targeted therapies and immunotherapy have expanded therapeutic options in patients with advanced non-small cell lung cancer (NSCLC) over the past decade (1). The therapy landscape for NSCLC patients with epidermal growth factor receptor (EGFR) tyrosine kinase mutations and anaplastic lymphoma kinase (ALK) rearrangements (EGFR+/ALK+) have been radically revolutionized based on genetic alteration in actionable oncogene drivers (2). Despite the prominent activity of tyrosine kinase inhibitors (TKIs) in these patients (3), acquired resistance almost inevitably develops (4,5), and there are no standard options after EGFR/ALK TKIs and platinum-based chemotherapy.

In recent years, immune checkpoint inhibitors (ICIs) have become additional important treatment options for advanced NSCLC (6-8). However, a previous retrospective study indicated that NSCLC harboring EGFR mutations or ALK rearrangements are associated with low objective response rates (ORRs) to programmed death 1 (PD-1)/PD ligand 1 (PD-L1) inhibitors (9). In addition, the only evidence available to date in regard of PD-1 axis blockade (durvalumab monotherapy) as the third- or later-line treatment in patients with EGFR+/ALK+ NSCLC comes from the ATLANTIC phase 2 study (10). So far, the definite clinical benefits of anti-PD-1/PD-L1 monotherapy or combination regimen in pretreated patients with EGFR+/ALK+ NSCLC who have progressed after prior TKI and/or chemotherapy are still unclear.

Camrelizumab (SHR-1210) is a humanized anti-PD-1 IgG4 monoclonal antibody. The combination of camrelizumab and chemotherapy (carboplatin and pemetrexed) has been approved for the treatment of patients with chemotherapy-naive, advanced non-squamous NSCLC without EGFR and ALK alterations in China in June 2020 (11). Apatinib, a vascular endothelial growth factor receptor 2 (VEGFR2) TKI, has been approved as a third- or subsequent line treatment for advanced gastric cancer in China (12). Recently, combining anti-PD-1 antibodies with anti-angiogenic agents has been attracting great interest. Preliminary preclinical study has revealed that apatinib alleviates hypoxia through modulating tumor immune microenvironment, enhances tumoral infiltration of CD8+ T cells, and limits immunosuppressive activity of tumor-associated macrophages (13). Additionally, encouraging results from the phase 1b/2 trials also indicated that camrelizumab plus apatinib, administrated at the recommended phase II dose (RP2D; apatinib 250 mg), showed promising antitumor activities and manageable safety profile in patients with advance hepatocellular carcinoma, gastric cancer or non-squamous NSCLC (14,15). Nevertheless, no relevant clinical evidence is available for the antitumor activity of the combination therapy of anti-PD-1 inhibitor plus angiogenesis inhibitors in patients with advanced EGFR+/ALK+ NSCLC. There is an unmet clinical need for more effective subsequent treatment options in this patient population. Therefore, we conducted a phase 1b/2 trial to explore the efficacy and safety of camrelizumab plus apatinib in pretreated patients with advanced EGFR+/ALK+ NSCLC and present the results in accordance with the TREND reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-22-22/rc).

Methods

Study design

This phase 1b/2, open-label, multicenter, multicohort study enrolled patients with advanced NSCLC at 16 medical centers in China. It was registered at Clinical Trials.gov with the identifier NCT03083041. This trial consisted of 2 parts: a phase 1b dose-escalation part and a phase 2 dose-expansion part. Part 1 was designed to evaluate the tolerability, safety, pharmacokinetics, and pharmacodynamics of camrelizumab in combination with apatinib and to determine the RP2D of apatinib. Part 2 was designed to further assess the efficacy and safety of camrelizumab plus apatinib at RP2D (15). Here, we only reported the preliminary efficacy and safety of camrelizumab plus apatinib in previously treated patients with EGFR+/ALK+ NSCLC.

Patients

Eligible patients were aged 18–70 years old, who had a histologically or cytologically confirmed advanced NSCLC (stage IIIB or IV) harboring EGFR mutations within exons 18–21 or ALK fusion gene rearrangements. Patients had disease progression or recurrence occurred after at least one platinum-based doublet chemotherapy and patients harboring sensitive EGFR mutations or ALK fusion genes should also have received at least one TKI(s) targeting EGFR or ALK. Other key eligibility criteria included at least 1 measurable lesion according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1; an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; a life expectancy of ≥12 weeks; and adequate organ functions. Patients were excluded if they had any of the following: active or history of autoimmune disease; use of immunosuppressive agents within 2 weeks before study entry; previous treatment with anti-PD-1/PD-L1 monoclonal antibody or apatinib; untreated central nervous system metastases; evidence of major blood vessel invasion; and intratumor cavitation or necrosis. All patients were required to provide a fresh or archival tissue sample for exploratory analyses. The study protocol and amendments were reviewed and approved by the Ethics Committees of each study site (Table S1). The study was conducted in compliance with the Good Clinical Practice Guidelines, the Declaration of Helsinki and its amendments. All patients provided written informed consent before enrollment.

Procedures

Patients enrolled were administered a fixed dose of camrelizumab (200 mg intravenously once every 2 weeks) plus the RP2D of apatinib (250 mg orally once daily established by the phase 1b study) every 4-week cycle until disease progression, intolerable toxicity, patient withdrawal, investigator withdrawal, or camrelizumab treatment for up to 2 years, whichever occurred first (16). Dose modification of camrelizumab was not allowed and dose interruptions of camrelizumab were permitted for up to 12 weeks. Dose interruption, dose reduction, administration schedule modifications (initial modification: 5 days on, 2 days off; subsequent modification: 1 day on, 1 day off) were allowed. Investigator-assessed tumor response were performed every 2 cycles during the first 6 months, and every 3 cycles thereafter according to the RECIST version 1.1. Initial complete response (CR) or partial response (PR) was required to be confirmed at least 4 weeks later. Patients who had radiologically progressive disease (PD) could continue study treatment at the investigator’s discretion. Patients who discontinued study treatment for reasons other than radiographic disease progression continued tumor assessment every 3 months until documented disease progression, start of a new anticancer therapy, or death. After treatment discontinuation, patients were followed every 2 months to collect survival.

Adverse events (AEs) were collected and coded according to the Medical Dictionary for Regulatory Activities Version 23.1, and graded in accordance with the Common Terminology Criteria for Adverse Events (CTCAE) version 4.03, and the causality to study drugs was recorded. AEs and serious AEs were monitored until 90 days after the last dose of camrelizumab or 30 days after the last dose of apatinib, whichever occurred later.

Detection of PD-L1 expression

PD-L1 expression was centrally assessed using immunohistochemistry (IHC) method with 22C3 pharmDx kit (Agilent Technologies, Santa Clara, CA, USA). PD-L1 positivity was defined as the tumor proportion score (TPS) ≥1%, which was estimated as percentage of viable tumor cells (TCs) showing partial or complete membrane staining (17).

Endpoints and assessments

The primary endpoint was ORR per RECIST version 1.1, defined as the proportion of patients achieving a confirmed CR or PR based on investigators’ assessment. The secondary endpoints included clinical benefit rate [CBR; defined as the proportion of patients with CR or PR, or stable disease (SD) lasting ≥24 weeks], duration of response (DoR; defined as time from first evidence of CR or PR to disease progression or death, whichever occurred first), progression-free survival (PFS; defined as time from first dose of study treatment to disease progression per RECIST version 1.1 or death, whichever occurred first), OS (defined as time from first dose of study treatment to death due to any cause), 12-month OS rate, and safety. Exploratory analysis on ORR, and survival data by PD-L1 TPS or EGFR subtypes were also performed.

Statistical analysis

Assuming an ORR of 30% and a dropout rate of 20%, a sample size of 38 patients was able to ensure the width of 90% confidence interval (CI) for ORR would be 0.30.

The full analysis set included all patients who received at least one dose of study treatment, and all patients who received at least one dose of study treatment and had safety assessment after treatment initiation were included in the safety analysis set. The ORR and CBR and the corresponding 95% CIs were calculated using the Clopper-Pearson method. Median and range of time to response (TTR) were calculated. Kaplan-Meier method was used to estimate the DoR, PFS and OS, and the corresponding 95% CIs were calculated with the Brookmeyer and Crowley method. The 95% CIs of 12-month OS rate were calculated using the log-log transformation according to the normal approximation with back transformation to CIs on the untransformed scale. Exploratory analyses of response and survival in correlation to tumor PD-L1 expression or EGFR mutation subtypes were also performed. All statistical analyses were conducted with the SAS software version 9.4 (SAS Institute Inc. Cary, NC, USA).

Results

Patient characteristics and disposition

Between November 13, 2017 and January 16, 2019, a total of 43 previously treated patients with advanced EGFR+/ALK+ NSCLC received the combination regimen of intravenous camrelizumab 200 mg every 2 weeks plus oral apatinib 250 mg once daily (). Due to the similarity in gene mutation and treatment settings in the 3 patients from the phase 1b study and 40 patients from Cohort 2 of the dose-expansion phase 2 study, their data were combined for analysis.

Figure 1

Flow chart of patients with advanced EGFR+/ALK+ non-small cell lung cancer (n=43). EGFR, epidermal growth factor receptor; ALK, anaplastic lymphoma kinase.

All enrolled patients were previously treated, and majority of patients harboring sensitive EGFR mutations or ALK fusion genes (n=40; 93.0%) had received at least one EGFR/ALK TKI and a platinum-based doublet chemotherapy regimen. Of them, 40 (93.0%) patients had received first- or second-generation EGFR TKIs (first-generation, n=39; second-generation, n=1) and the remaining 3 (7.0%) patients carrying EGFR 20ins mutation did not receive targeted therapy due to drug-resistant mutation. A total of 10 (23.3%) patients had received third-generation EGFR TKIs (osimertinib, n=9; avitinib, n=1). Furthermore, 9 (20.9%) patients had developed EGFR T790M resistance mutations after the treatment of first- or second-generation TKI, among whom 4 had received osimertinib treatment. The median age of enrolled patients was 55 years (range, 33–69 years). Thirty-nine (90.7%) patients had an ECOG performance status of 1, 19 (44.2%) patients were current or former smokers, and 12 (27.9%) patients had metastases in more than two organs. In addition, 22 (51.2%) patients had PD-L1 TPS ≥1% and 13 (30.2%) patients had PD-L1 TPS <1% tumors. Demographics and baseline characteristics of patients are shown in .

Table 1

Baseline characteristics

Characteristics	All patients (n=43)
Age, years, median [range]	55 [33–69]
Male, n (%)	25 (58.1)
ECOG performance status, n (%)
0	4 (9.3)
1	39 (90.7)
Disease stage, n (%)
IV	43 (100.0)
Tumor histology, n (%)
Adenocarcinoma	41 (95.3)
Squamous	2 (4.7)
Smoking status, n (%)
Never smoked	24 (55.8)
Current or former smoker	19 (44.2)
No. of organs with metastasis, n (%)
≤2	31 (72.1)
>2	12 (27.9)
PD-L1 TPS, n (%)
<1%	13 (30.2)
≥1%	22 (51.2)
Unknown	8 (18.6)
EGFR mutation*, n (%)
Positive	40 (93.0)
Negative	1 (2.3)
Unknown	2 (4.7)
ALK rearrangement*, n (%)
Positive	4 (9.3)
Negative	28 (65.1)
Unknown	11 (25.6)
Previous therapy, n (%)
Surgery	11 (25.6)
Chemotherapy	43 (100.0)
Targeted therapy*,†	40 (93.0)
EGFR TKI	37 (86.0)
ALK TKI	5 (11.6)
Radiotherapy	11 (25.6)
Others	3 (7.0)
Median duration from diagnosis (range), years	2.0 (0.4–7.4)

*, one patient had a concomitant EGFR and ALK mutation and had previously received both EGFR-TKI and ALK-TKI; †, there were three patients carrying EGFR 20ins mutation, which were drug-resistant mutation, therefore, no targeted therapy was available. ECOG, Eastern Cooperative Oncology Group; PD-L1, programmed death-ligand 1; TPS, tumor proportion score; EGFR, epidermal growth factor receptor; ALK, anaplastic lymphoma kinase; TKI, tyrosine kinase inhibitor.

As of data cutoff (June 12, 2020), the median follow-up duration was 15.7 months (range, 0.5–24.4 months). Among all 43 patients, 1 (2.3%) patient was still receiving study treatment. The reasons for study treatment discontinuation were disease progression (n=26, 60.5%), AEs (n=7, 16.3%), physician decision (n=5, 11.6%), patient decision (n=2, 4.7%) and others (n=2, 4.7%).

Efficacy

At data cutoff, 43 patients were included in the full analysis set. Best change in sum of diameters of target lesion from baseline is presented in , 20 (46.5%) patients showed decreased total tumor burden. As shown in , no patient achieved CR, 8 (18.6%) patients achieved PR as their best response, 17 (39.5%) patients had SD, 12 (27.9%) patients had PD, and the overall responses of 6 (14.0%) patients were not evaluable. The confirmed ORR assessed by investigators per RECIST version 1.1 was 18.6% (95% CI: 8.4–33.4%). The CBR with this combination regimen was 27.9% (95% CI: 15.3–43.7%). Proportions of patients who achieved an objective response across all subgroups by baseline characteristics are shown in Table S2. None of the 4 patients with ALK fusion genes had an objective response (Table S3). Treatment duration and tumor response in 8 responders are shown in , and the decreased tumor burden sustained over several assessments. Response occurred at a median of 1.8 months (range, 1.8–2.0 months), and the median DoR was 6.5 months (95% CI: 3.5–18.2 months).

Figure 2

Clinical activity of camrelizumab plus apatinib in advanced non-small cell lung cancer patients harboring EGFR or ALK genetic aberrations. (A) Best percentage changes in sum of the diameters of target lesion from baseline. (B) Treatment duration and tumor response in 8 responders (0 CR and 8 PR). EGFR, epidermal growth factor receptor; ALK, anaplastic lymphoma kinase; CR, complete response; PR, partial response; PD, progressive disease. Red asterisks indicated confirmed responses.

Table 2

Investigator-assessed best overall tumor response and survival data

Variables	All patients (n=43)
Median follow-up (range), months	15.7 (0.5–24.4)
Best overall response, n (%)
CR (confirmed)	0
PR (confirmed)	8 (18.6)
SD	17 (39.5)
PD	12 (27.9)
NE*	6 (14.0)
Confirmed ORR, n (%; 95% CI)	8 (18.6; 8.4–33.4)
CBR (CR/PR/SD ≥24 weeks), n (%) [95% CI]	12 (27.9) [15.3–43.7]
PFS, months, median (95% CI)	2.8 (1.9–5.5)
OS, months, median (95% CI)	NR (7.3–NR)
12-month OS rate, % (95% CI)	57.2 (41.0–70.5)
TTR, months, median (range)	1.8 (1.8–2.0)
DoR, months, median (95% CI)	6.5 (3.5–18.2)

*, 6 patients were not evaluable due to study discontinuation (adverse events, n=3; withdrawal of consent, n=1; death, n=1; investigator decision, n=1). CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; NE, not evaluable; ORR, objective response rate; CBR, clinical benefit rate; PFS, progression-free survival; OS, overall survival; TTR, time to response; DoR, duration of response; CI, confidence interval; NR, not reached.

A total of 37 (86.0%) patients had PFS events (documented progressive disease or death), and the median PFS was 2.8 months (95% CI: 1.9–5.5 months; ). As of data cutoff, 20 (46.5%) patients had died. The median OS was not reached (NR) (95% CI: 7.3 months–NR months; ), and the estimated 6-, 9- and 12-month OS rate with this combination regimen was 73.9% (95% CI: 57.8–84.6%), 62.0% (95% CI: 45.6–74.7%) and 57.2% (95% CI: 41.0–70.5%), respectively. After the end of the study treatment, 32 (74.4%) patients received at least one subsequent antitumor therapy, and the results are presented in Table S4.

Figure 3

PFS and OS in advanced non-small cell lung cancer patients harboring EGFR or ALK genetic aberrations (n=43). (A) Kaplan-Meier curves for PFS; (B) Kaplan-Meier curves for OS. EGFR, epidermal growth factor receptor; ALK, anaplastic lymphoma kinase; PFS, progression-free survival; OS, overall survival; CI, confidence interval; NR, not reached.

Efficacy by EGFR mutation and PD-L1 expression

Among all 40 patients with EGFR mutation, 22 (55.0%) patients had EGFR 19del, 14 (35.0%) patients had EGFR L858R, and 3 (7.5%) patients had EGFR 20ins. The ORR in patients with EGFR L858R was numerically higher than those with EGFR 19del (21.4% vs. 13.6%) in this phase 1b/2 trial. Similarly, longer median PFS was observed in patients with EGFR L858R than in those with EGFR 19del (5.3 vs. 2.8 months; Table S5).

In exploratory analysis, tumor PD-L1 expression were available in 35 (81.4%) of the 43 enrolled patients, including 22 PD-L1 TPS ≥1% and 13 TPS <1%. The ORR was 27.3% in patients with PD-L1 TPS ≥1% and 7.7% in those with TPS <1%. The median PFS was 2.0 and 5.1 months in patients with PD-L1 TPS ≥1% and TPS <1%, respectively (Figure S1A). The median OS was 8.9 months in patients with tumor PD-L1 TPS ≥1% and NR in those with TPS <1% (Figure S1B). The distribution of EGFR mutation (L858R, 19del or 20ins) in patients with tumor PD-L1 TPS ≥1% and those with tumor PD-L1 TPS <1% is presented in Table S6.

Safety

All 43 patients were evaluable for safety analysis. The median duration of camrelizumab exposure was 3.2 months (range, 0.5–18.3 months), and the median duration of apatinib exposure was 3.3 months (range, 0.4–17.9 months). Among the 43 patients, 10 (23.3%) patients discontinued any study treatment due to treatment-related adverse events (TRAEs), of which 7 (16.3%) discontinued both agents, 32 (74.4%) had dose interruption because of TRAEs, and 8 (18.6%) had apatinib dose reduction caused by TRAEs (Table S7). As illustrated in , all patients had at least 1 TRAE, and the most commonly reported TRAEs of any grade were hypertension (67.4%), proteinuria (65.1%), and increased aspartate aminotransferase (37.2%). The most common grade ≥3 TRAEs were hypertension (16.3%), proteinuria (11.6%), and palmar-plantar erythrodysesthesia (9.3%). Reactive cutaneous capillary endothelial proliferation (RCCEP), a common and self-resolving side effect associated with camrelizumab, occurred in 20.9% of patients, and all were grade 1 or 2. Treatment-related serious adverse events (SAEs) were reported in 12 (27.9%) patients, with hemoptysis [n=2 (4.7%)], immune-mediated pneumonitis [n=2 (4.7%)] and immune-mediated hepatitis [n=2 (4.7%)] as the most common events (Table S8).

Table 3

Treatment-related adverse events that occurred in ≥10% of patients

Adverse events	All patients, n (%)
	Any grade	≥Grade 3
Any TRAE	43 (100.0)	28 (65.1)
Hypertension	29 (67.4)	7 (16.3)
Proteinuria	28 (65.1)	5 (11.6)
Aspartate aminotransferase increased	16 (37.2)	1 (2.3)
Palmar-plantar erythrodysaesthesia syndrome	15 (34.9)	4 (9.3)
Alanine aminotransferase increased	15 (34.9)	0
Asthenia	13 (30.2)	0
Rash	11 (25.6)	2 (4.7)
White blood cell count decreased	10 (23.3)	1 (2.3)
Neutrophil count decreased	9 (20.9)	0
RCCEP	9 (20.9)	0
Blood bilirubin increased	8 (18.6)	1 (2.3)
Platelet count decreased	8 (18.6)	1 (2.3)
Decreased appetite	8 (18.6)	0
Gamma-glutamyltransferase increased	7 (16.3)	2 (4.7)
Hemoptysis	6 (14.0)	1 (2.3)
Headache	6 (14.0)	1 (2.3)
Blood creatinine increased	6 (14.0)	0
Vomiting	6 (14.0)	0
Dysphonia	6 (14.0)	0
Hypertriglyceridemia	5 (11.6)	2 (4.7)
Hypokalemia	5 (11.6)	1 (2.3)
Occult blood positive	5 (11.6)	0
Hypocalcemia	5 (11.6)	0
Pyrexia	5 (11.6)	0
Anemia	5 (11.6)	0

TRAE, treatment-related adverse event; RCCEP, reactive cutaneous capillary endothelial proliferation.

As reported by the investigators, 3 (7.0%) patients died due to TRAEs (pneumonia, n=1; immune-mediate pneumonitis, n=1; hemoptysis, n=1). Other than RCCEP, the most frequently reported AEs of special interest were grade ≥3 immune-mediated hepatitis (4.7%) and grade ≥3 rash (4.7%) (Table S9).

Grade ≥3 immune-related AEs (irAEs) occurred in 10 (23.3%) patients, with rash [n=2 (4.7%)] and immune-mediated hepatitis [n=2 (4.7%)] as the most common events (Table S10). No unexpected AEs were observed.

Discussion

To the best of our knowledge, this was the first reported phase 1b/2 trial investigating the combination strategy of an ICI (camrelizumab) plus a VEGF-TKI (apatinib) in pretreated patients with advanced EGFR+/ALK+ NSCLC. A prior meta-analysis involving Keynote 010, CheckMate 057, OAK and POPLAR trials demonstrated that the application of anti-PD-(L)1 monotherapy, including nivolumab, pembrolizumab, or atezolizumab, did not show more benefits in previously treated EGFR-mutant NSCLC patients than in the wild-type population (18-23). In the phase 2 ATLANTIC trial (10), NSCLC patients with both EGFR/ALK-alterations and high PD-L1 expression (defined as ≥25% of TCs) were administered durvalumab as third- or later-line therapy, despite that NSCLC patients with EGFR/ALK alterations who had PD-L1 expression ≥25% showed a relatively high response rate, the overall response obtained was still lower than that in the wild-type population and could not translated into improved survival outcomes.

With respect to combination regimen, the IMpower150 trial indicated that the addition of bevacizumab to atezolizumab and chemotherapy as first-line therapy confers the clinical efficacy observed in the subgroup of patients with sensitive EGFR mutations (24). Notably, in the IMpower 150 trial, the better outcomes described in the atezolizumab + bevacizumab + carboplatin + paclitaxel group for EGFR+/ALK+ patients were not found in those who received atezolizumab plus chemotherapy without bevacizumab, emphasizing the crucial role of VEGF inhibition and angiogenesis in the process of EGFR/ALK cancer immunity.

The enrollment of an independent patient cohort defined by advanced EGFR+/ALK+ NSCLC in our trial permitted prospective evaluation of the combination regimen with camrelizumab and apatinib in a patient population with the different clinical treatment outcome compared to those with EGFR-/ALK- NSCLC. Our study provided a crucial contribution to the evidence on the efficacy and safety of this combination regimen in previously treated patients with advanced EGFR+/ALK+ NSCLC. The ORR and CBR of pretreated patients with advanced EGFR+/ALK+ NSCLC who received this combination regimen were 18.6% and 27.9%, the median PFS was 2.8 months and the median OS had not been reached at the time of data cutoff with a median follow-up period of 15.7 months, suggesting that the improved ORR, durable DoR and PFS with camrelizumab plus apatinib might be translated into prominent OS benefit in previously treated patients with advanced EGFR+/ALK+ NSCLC. In recent years, the ATLANTIC phase 2 trial provided the largest prospective cohort with anti-PD-1/PD-L1 monotherapy in advanced EGFR+/ALK+ NSCLC patients, the median OS in Cohort 1 (pretreated patients with advanced EGFR+/ALK+ NSCLC) was 13.3 months (95% CI: 6.3–24.5 months) in patients with increased PD-L1 expression (defined as ≥25% of TCs) and 9.9 months (95% CI: 4.2–13.3 months) in those with TCs <25% (25). Although cross-trial comparisons can be challenging, in our phase 1b/2 trial, the OS was comparable and competitive to that observed in the similar patient population from ATLANTIC phase 2 trial (10). Our results also supported the point of view that combination of anti-PD-1/PD-L1 agent and low-dose antiangiogenic inhibitor could elicit T cell activation, improve tumor immune microenvironment and drive the responses of TCs to immune checkpoint blockade, which brings about more sufficient anti-tumor immunity than anti-PD-1 monotherapy (26).

Not all EGFR mutations are identical, which could partly interpret the reason some EGFR mutation subtypes respond variously to ICIs. A retrospective study involving multicenters in U.S. revealed that among 171 patients with EGFR mutation who received ICI treatment, patients with EGFR exon 19del tumors showed lower tumor mutation burden (TMB) than those with EGFR L858R. Furthermore, NSCLC patients with EGFR exon 19del tumors showed worse treatment response or prognosis compared to those with EGFR L858R. However, EGFR T790M status or PD-L1 expression was not found to influence responses or survival outcomes (27). The results from the IMMUNOTARGET registry indicated that 125 NSCLC patients with EGFR mutations receiving ICIs achieved an ORR of 12.2%, with the median PFS of 2.1 months. Among them, patients with exon 21 EGFR mutations indicated prolonged PFS than those with exon 19 deletions or T790M mutations (28). In our study, improved ORR and prolonged PFS and OS were also observed in patients with EGFR L858R compared with those with EGFR 19del, highlighting that robust genetic background might exert important influence on the antitumor activity of this combination regimen in advanced EGFR+/ALK+ NSCLC. In addition, a recent study discovered that L858R-mutant tumors were infiltrated with more T cells marked with CD8+ PD-1+ compared with 19del-mutant tumors, which implied that tumor immune microenvironment (TIME) might be different among EGFR mutation subtypes (29). This might partly explain the better clinical benefits in patients with EGFR L858R subgroup than those with EGFR 19del in our trial. However, these findings warranted further validation and confirmation in the further trials with larger sample size.

PD-L1 expression has been widely applied as a biomarker for prediction of response to immunotherapy in NSCLC patients. Nevertheless, PD-L1 expression in tumors harboring actionable driver mutations might not be necessarily associated with clinical responses to ICIs. In our phase 1b/2 trial, the proportion of patients achieving a response was higher in patients with PD-L1 TPS ≥1% than in those with PD-L1 TPS <1%; however, the survival benefit from this combination regimen did not indicate the trend in favor of patients with PD-L1 TPS ≥1%. This discordance between ORR and survival benefits observed in our study might be partially attributed to the multiple confounding factors, such as the relatively small sample size in each PD-L1 TPS category, influences of subsequent anti-tumor therapies and so on, which requires further investigation.

The incidence and severity of TRAEs with camrelizumab plus apatinib were consistent with previously reported toxic effects for camrelizumab or apatinib (11,30). The most common grade 3 or higher TRAEs were hypertension (16.3%), proteinuria (11.6%) and palmar-plantar erythrodysaesthesia (9.3%), which were slightly higher than the incidences reported for apatinib monotherapy (30). No new safety signals were identified. Remarkably, RCCEP, a clinically controllable and self-limiting TRAE induced by camrelizumab, only occurred in 9 (20.9%) patients receiving this combination regimen, with no grade 3 or higher events reported; the incidence was similar to that observed with the identical combination regimen for pretreated advanced non-squamous NSCLC without driver mutations (21.9%) and obviously decreased compared with that reported for camrelizumab monotherapy (74.0%) (31) or camrelizumab plus chemotherapy (77.6%) (32) in advanced NSCLC. Our results verified that addition of apatinib to camrelizumab could reduce the incidence of RCCEP, further suggesting that the mechanism of RCCEP may be correlated with the VEGFA/VEGFR-2 signaling pathway (14,15). Overall, TRAEs that occurred with this combination regimen were unsurprising and well-tolerated. Increased aspartate aminotransferase and alanine aminotransferase might be associated with camrelizumab (11), while the occurrence of hypertension, proteinuria and palmar-plantar erythrodysaesthesia might be related with apatinib (30). This combination regimen also resulted in slightly increased occurrences of asthenia and common hematological toxicity, including decreased white blood cell count, decreased neutrophil count, and decreased platelet count, which might be attributed to the overlapping AE profiles of camrelizumab plus apatinib.

The current study had several limitations. First, the trial was a single-arm phase 1b/2 study design, with no standard-of-care or PD-1/PD-L1 monotherapy arm as a control. Second, due to the small sample size, the proportion of patients with PD-L1 TPS ≥1% was relatively low, nevertheless, improved ORR and prolonged survival data had been derived regardless of PD-L1 expression. Third, the goal of an ORR of 30% was not reached, partially due to the relatively high proportion of patients who discontinued treatment due to TRAE (16.3%) and those who were not evaluable for tumor response (7.0% for reasons other than TRAEs); on the other hand, despite the modest ORR, OS was remarkable with a 12-month rate of 57.2%, suggesting potential benefits beyond initial treatment period. Fourth, the upper age limit for the present study was set at 70 years based on the compliance and tolerance of Chinese patients. Considering the generally lower treatment tolerance in patients with more advanced age, the findings of this study may not extrapolate to patients over 70 years and further research in this population is needed. Moreover, the median OS was still not reached, and long-term follow-up survival data are required to be reported in near future.

In conclusion, our phase 1b/2 trial demonstrated that camrelizumab plus apatinib showed moderate antitumor activity and acceptable safety profile in previously treated patients with advanced NSCLC harboring EGFR or ALK genetic aberrations, which warranted further validation.

The article’s supplementary files as