
==== Front
Thorac Cancer
Thorac Cancer
10.1111/(ISSN)1759-7714
TCA
Thoracic Cancer
1759-7706
1759-7714
John Wiley & Sons Australia, Ltd Melbourne

39086088
10.1111/1759-7714.15413
TCA15413
Original Article
Original Article
Efficacy of first‐line immune checkpoint inhibitor and anti‐angiogenic agent combination therapy for Kirsten rat sarcoma viral antigen‐mutant advanced non‐small‐cell lung cancer: a systematic review and network meta‐analysis
Tsukada et al.
Tsukada Akinari https://orcid.org/0000-0001-8582-1964
1 atsukada@hosp.ncgm.go.jp

Morita Chie 1
Shimizu Yosuke 2
Uemura Yukari 2
Naka Go 1
Takasaki Jin 1
Nokihara Hiroshi https://orcid.org/0000-0002-1219-6966
1
Izumi Shinyu 1
Hojo Masayuki 1
1 Department of Respiratory Medicine National Center for Global Health and Medicine Tokyo Japan
2 Center for Clinical Sciences National Center for Global Health and Medicine Tokyo Japan
* Correspondence
Akinari Tsukada, Department of Respiratory Medicine, National Center for Global Health and Medicine, 1‐21‐1 Toyama, Shinjuku‐ku, Tokyo 162‐8655, Japan.
Email: atsukada@hosp.ncgm.go.jp

31 7 2024
9 2024
15 25 10.1111/tca.v15.25 18541862
29 6 2024
21 4 2024
18 7 2024
© 2024 The Author(s). Thoracic Cancer published by John Wiley & Sons Australia, Ltd.
https://creativecommons.org/licenses/by-nc/4.0/ This is an open access article under the terms of the http://creativecommons.org/licenses/by-nc/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

Abstract

Background

Recent advancements in advanced non‐small‐cell lung cancer (NSCLC) treatment have significantly improved primary therapy outcomes owing to the emergence of various molecular targeted therapies and immune checkpoint inhibitors (ICIs). However, for Kirsten rat sarcoma viral antigen (KRAS) mutations, molecular targeted drugs, such as sotorasib, are not applicable as first‐line treatments, and the optimal primary treatment remains unclear. Therefore, we aimed to investigate the efficacy of ICI combination therapy as first‐line treatment for KRAS‐mutant NSCLC.

Methods

We conducted a systematic search for phase 3 randomized controlled trials (RCTs) that presented data on KRAS mutation status in advanced NSCLC. The primary endpoints were progression‐free survival (PFS) and overall survival (OS). A random‐effects network meta‐analysis was conducted to perform direct and indirect comparisons among treatment groups.

Results

Six RCTs were eligible for inclusion. In the network meta‐analysis for KRAS‐mutant NSCLC, Chemo + bevacizumab (Bev) + ICI was associated with improved PFS (hazard ratio [HR] 0.38, 95% confidence interval [CI] 0.22–0.64), followed by Chemo + ICI + ICI (HR 0.66, 95% CI 0.47–0.93) and Chemo + ICI (HR 0.67, 95% CI 0.49–0.91). The most beneficial effect on OS was observed with Chemo + Bev + ICI (HR 0.50, 95% CI 0.34–0.73), followed by Chemo + ICI + ICI (HR 0.64, 95% CI 0.48–0.87) and Chemo + ICI (HR 0.72, 95% CI 0.56–0.92). Regarding OS in wild‐type KRAS, ICI + ICI (HR 0.73, 95% CI 0.50–1.07) produced the most favorable effects, followed by Chemo + ICI (HR 0.79, 95% CI 0.63–0.99).

Conclusion

The efficacy of Chemo + Bev + ICI is potentially high for improving PFS and OS in KRAS‐mutant NSCLC. In advanced NSCLC, the presence or absence of KRAS mutations may need to be considered when administering first‐line treatment.

We conducted a systematic search for phase 3 randomized controlled trials that presented data on Kirsten rat sarcoma viral antigen (KRAS) mutation status in advanced non‐small‐cell lung cancer (NSCLC). A random‐effects network meta‐analysis was conducted to perform direct and indirect comparisons among treatment groups. For KRAS‐mutant NSCLC, the most beneficial effect on overall survival (OS) was observed with Chemo + bevacizumab (Bev) + immune checkpoint inhibitors (ICI) (hazard ratio [HR] 0.50, 95% confidence interval [CI] 0.34–0.73), followed by Chemo + ICI + ICI (HR 0.64, 95% CI 0.48–0.87) and Chemo + ICI (HR 0.72, 95% CI 0.56–0.92). Regarding OS in wild‐type KRAS, ICI + ICI (HR 0.73, 95% CI 0.50–1.07) produced the most favorable effects, followed by Chemo + ICI (HR 0.79, 95% CI 0.63–0.99).

immune checkpoint inhibitors
KRAS
network meta‐analysis
non‐small‐cell lung cancer
source-schema-version-number2.0
cover-dateSeptember 2024
details-of-publishers-convertorConverter:WILEY_ML3GV2_TO_JATSPMC version:6.4.8 mode:remove_FC converted:02.09.2024
Tsukada A , Morita C , Shimizu Y , Uemura Y , Naka G , Takasaki J , et al. Efficacy of first‐line immune checkpoint inhibitor and anti‐angiogenic agent combination therapy for Kirsten rat sarcoma viral antigen‐mutant advanced non‐small‐cell lung cancer: a systematic review and network meta‐analysis. Thorac Cancer. 2024;15 (25 ):1854–1862. 10.1111/1759-7714.15413
==== Body
pmcINTRODUCTION

Therapeutic outcomes in advanced non‐small‐cell lung cancer (NSCLC) have improved with the advent of various targeted molecular drugs, such as epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) inhibitors, as well as immune checkpoint inhibitors (ICIs). 1 Although there are diverse reports regarding the types of genetic mutations in NSCLC, approximately 30% of patients with NSCLC in Western countries exhibit Kirsten rat sarcoma viral antigen (KRAS) mutations. 2 In Asian populations, genomic screening projects have identified KRAS mutations in approximately 14% of NSCLC cases. 3 Although various molecular targeted drugs have been approved for the primary treatment of EGFR or ALK mutations, targeted therapy for KRAS mutations, such as sotorasib and adagrasib, is currently approved as second‐line or later treatments for KRAS G12C mutation‐positive advanced NSCLC. The primary first‐line treatment for KRAS mutation‐positive advanced NSCLC currently involves a combination of ICIs and chemotherapy, similar to that for mutation‐negative NSCLC. However, the appropriateness of using a similar primary treatment for KRAS mutation‐positive and mutation‐negative NSCLC remains unclear. In addition to G12C mutations, KRAS mutations, such as G12V and G12D, are known to be frequent, 4 and currently there is no targeted therapy available for these mutations.

Under these circumstances, immune combination therapy is considered a significant option for KRAS mutation‐positive advanced NSCLC. Previous meta‐analyses investigating the efficacy of ICIs for KRAS‐positive lung cancer have shown improvements in overall survival (OS) and progression‐free survival (PFS) compared to when chemotherapy is administered alone. 5 However, at the time this previous meta‐analysis was performed, there were limited options for first‐line immune combination therapy, and combinations with anti‐cytotoxic T‐lymphocyte antigen 4 (CTLA‐4) antibodies were not included in the analysis, therefore the results may differ from those that include current treatment options.

Currently, there are multiple therapeutic options for the primary treatment of advanced NSCLC. However, uncertainties remain regarding the most appropriate first‐line treatment for patients with KRAS mutations. Moreover, there is a lack of direct comparisons of these options performed through randomized controlled trials (RCTs).

In this study, we aimed to conduct a systematic review and network meta‐analysis to investigate the efficacy of ICI combination therapy as a first‐line treatment based on the KRAS status. The objective of this study was to identify the optimal first‐line treatment for this patient population.

METHODS

This network meta‐analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines. 6 The study protocol was registered with the University Hospital Medical Information Network (UMIN) Center, Japan, under the registration ID UMIN000052814.

Literature search

We conducted a literature search using the PubMed, Web of Science, and Cochrane Library databases. The search was designed to include trials targeting patients with advanced NSCLC with KRAS mutations or trials conducted in patients with advanced NSCLC where data on KRAS mutations were available. We focused on studies that evaluated outcomes such as OS and PFS. The systematic data search was conducted until October 30, 2023, using terms such as “lung neoplasm,” “NSCLC,” “immune checkpoint inhibitor,” “PD‐1,” “PD‐L1,” “CTLA‐4,” “KRAS,” “KRAS‐positive,” “KRAS‐mutated,” “KRAS wild‐type,” “clinical trial,” and “randomized.” The detailed search terms are listed in Supporting Information Table S1.

Inclusion and exclusion criteria

In this literature search, we focused on phase 3 RCTs targeting advanced NSCLC. The first‐line treatment in the control group in these trials was conventional platinum‐based chemotherapy, while that in the intervention groups involved the addition of ICIs, such as anti‐PD‐1/PD‐L1 antibodies or antiangiogenic agents, and combinations with anti‐CTLA‐4 antibodies. The inclusion criteria were trials that compared these interventions, provided information on KRAS mutations, and reported efficacy outcomes, such as OS and PFS. Trials without KRAS mutation data, retrospective studies, or those focused on treatments beyond second‐line options were excluded.

Data extraction and quality assessment

Relevant literature was extracted according to the abovementioned criteria by two independent researchers (A.T. and C.M.). In the case of discrepancies regarding data exclusion, a discussion was held before deciding whether to include or exclude any data. Data extracted included the trial name, information source, publication year, overall sample size, median age, histological findings, numbers of patients with KRAS mutations and wild‐type KRAS in the intervention/control groups, treatment details, and PFS/OS details.

The quality of the RCTs was assessed using the Cochrane Risk of Bias Tool, version 2. This tool evaluates six items: randomization process, deviations from the intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Publication bias in relation to the PFS and OS outcomes was assessed using funnel plots and Egger's test.

Statistical analysis

Hazard ratios (HRs) with 95% confidence intervals (CIs) were used as indices to evaluate effect sizes on PFS and OS. The interventions were categorized into specific groups based on the treatment contents: those where bevacizumab was added to chemotherapy were defined as the Chemo + Bev group, and those where anti‐PD‐1/PD‐L1 antibodies were added were termed the Chemo + Bev + ICI group. Similarly, interventions involving chemotherapy with added anti‐PD‐1/PD‐L1 antibodies were categorized as the Chemo + ICI group, while those with the addition of both anti‐PD‐1/PD‐L1 and anti‐CTLA‐4 antibodies were classified as the Chemo + ICI + ICI group. Treatments involving anti‐PD‐1/PD‐L1 and anti‐CTLA‐4 antibodies without chemotherapy were categorized as the ICI + ICI group.

To examine the efficacy of combination immunotherapy compared to that of chemotherapy, a pairwise meta‐analysis was initially conducted in each integrated group and the overall group using a random‐effects model. Heterogeneity was assessed using I 2 statistics and the Cochran Q test. I 2 statistics <40% were considered low, 40%–75% as moderate, and >75% as considerable. 7 On using Cochran's Q test, p < 0.05 indicated high heterogeneity.

Subsequently, in the network meta‐analysis performed using a frequentist approach, we evaluated PFS/OS outcomes in each group using a random‐effects model and conducted direct and indirect comparisons to explore the optimal first‐line treatment in patients with KRAS‐positive advanced NSCLC. We estimated the treatment effects of traditional chemotherapy in each group, and each treatment group was ranked based on the P‐score. Higher P‐scores were indicative of a higher predicted therapeutic effect. Network plots were generated to visualize the overall network structure. To evaluate the heterogeneity and inconsistency in the network meta‐analysis, Cochran's Q test was conducted, and instances were deemed to have significant heterogeneity or inconsistency if the p value was <0.05.

We used R software, version 4.2.2, meta package, version 6.5.0, and netmeta package, version 2.9.0 to perform the statistical analyses. Statistical significance was set at p < 0.05.

RESULTS

Characteristics of the studies

In total, 790 records were identified during the initial screening. Subsequent processes, including the removal of duplicates, resulted in 145 records being subjected to full screening. Subsequently, seven articles remained under consideration (Figure 1 8 , 9 , 10 , 11 , 12 , 13 , 14 ). The KEYNOTE‐042 trial met the inclusion criteria. However, as this trial exclusively targeted patients with PD‐L1 expression of ≥1%, it was excluded from the analysis to maintain homogeneity. Ultimately, six studies with eight experimental arms were included. The characteristics of the included studies are summarized in Table 1. The NVALT22 trial specifically focused on patients with KRAS mutations, whereas the other five studies included data on the presence of KRAS mutations. Notably, the NVALT22 trial did not provide data on OS, the CheckMate‐227 trial lacked PFS data, and the POSEIDON trial did not present PFS data for wild‐type KRAS.

FIGURE 1 PRISMA flow diagram.

TABLE 1 Characteristics of the included studies.

Study	Author/year	Total patients	Median age (years)	Histology	nSQ (%)	KRAS‐mutant patients (Int/con)	KRAS‐wild type patients (Int/con)	Intervention	Control	
NVALT22	Dingemans/2021	203	65	nSQ	100	203	‐	Chemo + Bev (CB + PTX + Bev)	Chemo (CP + PEM)	
IMpower150 (Arm1)	West/2022	1047	63	nSQ	100	151 (80/71)	461 (235/226)	Chemo+Bev + ICI (CB + PTX + Bev + Ate)	Check + Bev (CB + PTX + Bev)	
IMpower150 (Arm2)	West/2022	1047	63	nSQ	100	145 (74/71)	460 (234/226)	Chemo + ICI (CB + PTX + Ate)	Check + Bev (CB + PTX + Bev)	
KEYNOTE‐189	Gadgeel/2019	616	64	nSQ	100	89 (59/30)	200 (145/55)	Chemo + ICI (CB/CP + PEM + Pemb)	Chemo (CB/CP + PEM)	
CheckMate‐9LA	Paz‐Ares/2022	719	65	NSCLC	68.8	122 (61/61)	191 (105/86)	Chemo + ICI + ICI (CB/CP + PEM + Ipi + Niv)	Chemo (CB/CP + PEM)	
CheckMate‐227	Ramalingam/2021	1166	64	NSCLC	72.1	163 (88/75)	312 (150/162)	ICI + ICI (Ipi + Niv)	Chemo (CB/CD + PEM)	
POSEIDON (Arm1)	Peters/2022	1013	64	NSCLC	62.9	113 (60/53)	296 (148/148)	Chemo + ICI + ICI (CTx a +Tre + Dur)	Chemo (CTx a )	
POSEIDON (Arm2)	Peters/2022	1013	64	NSCLC	62.9	122 (69/53)	282 (134/148)	Chemo + ICI (CTx a +Dur)	Chemo (CTx a )	
Abbreviations: Ate, atezolizumab; Bev, bevacizumab; CB, carboplatin; Chemo, platinum‐based chemotherapy; CP, cisplatin; Dur, durvalumab; ICI, immune checkpoint inhibitors; Int/Con, intervention/control; Ipi, ipilimumab; Niv, nivolumab; nSQ, non‐squamous; NSCLC, non‐small‐cell lung cancer; Pemb, pembrolizumab; PEM, pemetrexed; PTX, paclitaxel; Tre: tremelimumab.

a In the POSEIDON trial, for non‐squamous cell carcinoma, carboplatin/cisplatin plus pemetrexed or carboplatin/cisplatin plus nab‐paclitaxel were viable options.

Risk of bias and publication bias assessment

The results of the risk of bias assessment are shown in Supporting Information Figure S1. All the studies included in the analysis were phase 3 RCTs. Supporting Information Figure S2 shows a funnel plot of publication bias. For PFS, we analyzed seven trial groups (Table 1): NVALT22, IMpower150 (Arm 1/Arm 2), KEYNOTE‐189, CheckMate‐9LA, and POSEIDON (Arm1/Arm2). For OS, we examined seven trial groups from the IMpower150 (Arm1/Arm2), KEYNOTE‐189, CheckMate‐9LA, CheckMate‐227, and POSEIDON (Arm1/Arm2) studies. The p values in Egger's test were 0.2091 for PFS and 0.7198 for OS, indicating the absence of publication bias for both outcomes.

Pairwise meta‐analysis of immunotherapy combination treatment

The forest plot in Figure 2 illustrates the meta‐analysis of PFS and OS for patients with KRAS‐mutated advanced NSCLC across individual trials. In this pairwise meta‐analysis, the focus was on the efficacy of immunotherapy combination treatment; therefore, the Chemo + Bev group was excluded. Results of one to three trials were integrated for each treatment group. Moderate heterogeneity (I 2 = 44%, p = 0.17) was observed for PFS in the Chemo + ICI group, but heterogeneity was low (I 2 = 0%, p = 0.39) in the Chemo + ICI + ICI group. In the OS analysis, both the Chemo + ICI and Chemo + ICI + ICI groups had low heterogeneity (I 2 = 0%, p = 0.66; I 2 = 0%, p = 0.41). In the population with KRAS mutations, the HR of overall PFS with immunotherapy combination treatment was 0.62 (95% CI 0.49–0.79) and that of OS was 0.67 (95% CI 0.57–0.78).

FIGURE 2 Pairwise meta‐analysis of (a) PFS and (b) OS in patients with KRAS mutations. 95% CI, 95% confidence interval; Ate, atezolizumab; Bev, bevacizumab; Chemo, platinum‐based chemotherapy; Dur, durvalumab; HR, hazard ratio; ICI, immune checkpoint inhibitors; Ipi, ipilimumab; KRAS, Kirsten rat sarcoma viral antigen; Niv, nivolumab; OS, overall survival; Pem, pembrolizumab; PFS, progression‐free survival; Tre, tremelimumab.

In the case of PFS and OS in wild‐type KRAS (Supporting Information Figure S3), the limited number of integrated trials contributed to higher heterogeneity in PFS in the Chemo + ICI group (I 2 = 92%, p < 0.01); conversely, in terms of OS, the Chemo + ICI group exhibited moderate heterogeneity (I 2 = 59%, p = 0.09), while the Chemo + ICI + ICI group exhibited low heterogeneity (I 2 = 8%, p = 0.30). In the population with wild‐type KRAS, the HR of overall OS with immunotherapy combination treatment was 0.85 (95% CI 0.75–0.95).

Network meta‐analysis

Figure 3A presents a network plot illustrating the PFS results in KRAS‐mutant advanced NSCLC. Each circle represents an intervention node in the network. The size of each circle and width of the connecting lines are scaled according to the number of RCTs associated with each intervention and the number of comparisons made. In this analysis, the Chemo + Bev group was included for indirect comparison. The analysis that included data from five trials and seven experimental arms did not reveal any heterogeneity (Q = 3.81, p = 0.28) or inconsistency (Q = 0.41, p = 0.52). The efficacy of the different treatment groups in terms of PFS is detailed in Figures 4a and 5a, and the P‐scores are listed in Table 2. Notably, the Chemo + Bev + ICI group had an HR of 0.38 (95% CI 0.22–0.64), ranking the highest in terms of the P‐score, indicating its superior treatment efficacy. The Chemo + ICI and Chemo + ICI + ICI groups had comparable efficacy.

FIGURE 3 Network plot of (a) PFS and (b) OS in patients with KRAS mutations. Each circle represents an intervention node in the network. The size of each circle and width of the connecting lines were scaled according to the number of randomized controlled trials associated with each intervention and number of comparisons made. Bev, bevacizumab; Chemo, platinum‐based chemotherapy; ICI, immune checkpoint inhibitors; KRAS, Kirsten rat sarcoma viral antigen; OS, overall survival; PFS, progression‐free survival.

FIGURE 4 Efficacy profiles of the network meta‐analysis in patients with KRAS mutations hazard ratios (HRs) with 95% confidence intervals (CIs) for (a) PFS and (b) OS. Bev, bevacizumab; Chemo, platinum‐based chemotherapy; ICI, immune checkpoint inhibitors; KRAS, Kirsten rat sarcoma viral antigen.

FIGURE 5 Network meta‐analysis‐derived forest plots of (a) PFS and (b) OS in patients with KRAS mutations. 95% CI, 95% confidence interval; Bev, bevacizumab; Chemo, platinum‐based chemotherapy; HR, hazard ratio; ICI, immune checkpoint inhibitors; KRAS, Kirsten rat sarcoma viral antigen.

TABLE 2 P‐scores of the included treatment groups for KRAS‐mutant NSCLC.

Treatment	PFS	OS	
Chemo + Bev + ICI	0.9849	0.9371	
Chemo + ICI + ICI	0.6137	0.6676	
Chemo + ICI	0.6143	0.5017	
ICI + ICI	–	0.3677	
Chemo + Bev	0.2186	–	
Chemo	0.0685	0.0259	
Abbreviations: Bev, bevacizumab; Chemo, platinum‐based chemotherapy; ICI, immune checkpoint inhibitors; KRAS, Kirsten rat sarcoma viral antigen; NSCLC, non‐small‐cell lung cancer; OS, overall survival; PFS, progression‐free survival.

Figure 3B presents the network plot of OS in KRAS‐mutant advanced NSCLC. Similar to the PFS analysis, this analysis included data from five trials and seven experimental arms and did not show heterogeneity (Q = 1.51, p = 0.68). Figures 4b and 5b present the OS in the treatment groups, and Table 2 presents the respective P‐scores. The HR of the Chemo + Bev + ICI group was favorable at 0.50 (95% CI 0.34–0.73), and that of the Chemo + ICI + ICI group was 0.64 (95% CI 0.48–0.87); both HRs indicated significant treatment efficacy. The order of the treatments according to efficacy based on the P‐scores was as follows: Chemo + Bev + ICI, Chemo + ICI + ICI, and Chemo + ICI.

The results the KRAS‐wild type analysis are presented in Supporting Information Figure S4. Regarding PFS, caution is warranted because of the limited number of integrated trials and potential heterogeneity. However, Chemo + ICI showed superior treatment efficacy (HR 0.58, 95% CI 0.29–1.17). In terms of OS, ICI + ICI (HR 0.73, 95% CI 0.50–1.07) and Chemo + ICI (HR 0.79, 95% CI 0.63–0.99) exhibited higher treatment efficacy.

DISCUSSION

We conducted a systematic review and network meta‐analysis of immunotherapy combinations as first‐line treatments for patients with advanced NSCLC harboring KRAS mutations. In the initial pairwise meta‐analysis, it was observed that in patients with KRAS mutations, all the investigated combination immunotherapies produced prolonged PFS and OS compared to chemotherapy alone. In the network meta‐analysis, the Chemo + Bev + ICI group produced favorable treatment effects in terms of both PFS (HR 0.38, 95% CI 0.22–0.64) and OS (HR 0.50, 95% CI 0.34–0.73) in the patient population with KRAS mutations. By contrast, in the population with wild‐type KRAS, the treatment effect of the Chemo + Bev + ICI group on OS was relatively low (HR 0.98, 95% CI 0.69–1.39), while the ICI + ICI (HR 0.73, 95% CI 0.50–1.07) and Chemo + ICI (HR 0.79, 95% CI 0.63–0.99) groups exhibited a higher potential for treatment efficacy. These findings underscore the importance of focusing on KRAS mutations when considering the treatment options for patients with advanced NSCLC.

In the 2023 FDA pooled analysis, it was reported that superior therapeutic effects were observed with Chemo + ICI compared to Chemo alone or ICI alone in KRAS‐mutant NSCLC. 15 Although the specific type of ICI was not explicitly mentioned in this analysis, in the KRAS mutation group the median OS was 22.4 months with Chemo + ICI as opposed to 17.1 months with chemotherapy alone. In the KRAS wild‐type group, the median OS was 18.7 months with Chemo + ICI and 14.9 months with chemotherapy alone. Consequently, it was concluded that Chemo + ICI exhibits higher treatment efficacy in KRAS‐mutant NSCLC, supporting the recommendation of upfront Chemo + ICI treatment in such patients. While our current analysis did not include ICI monotherapy owing to considerations of homogeneity, in our pairwise meta‐analysis, favorable treatment effects were observed in the KRAS mutation group with Chemo + Bev + ICI, Chemo + ICI, and Chemo + ICI + ICI compared to with chemotherapy alone, further supporting the efficacy of combining chemotherapy and immunotherapy in this population.

Approximately 30%–40% of KRAS mutations in NSCLC are G12C mutations, while other mutations, such as G12A and G12V, are also known to exist. 4 , 16 , 17 Previous reports have indicated that the efficacy of ICI in NSCLC does not differ significantly between patients with G12C and other KRAS mutations. 18 , 19 In this study, we were unable to obtain data on KRAS subtypes. However, considering the aforementioned reports, our results suggest the potential applicability of our findings across various KRAS subtypes, including G12C.

In our study, the potential superiority of Chemo + Bev + ICI in terms of the effects on PFS and OS was suggested in the population with KRAS mutations. In addition to its antiangiogenic effects, bevacizumab may enhance atezolizumab‐mediated cell death by inhibiting VEGF‐related immunosuppression and promoting T cell tumor infiltration. 20 Altering the tumor microenvironment through the VEGF‐inhibiting effect of bevacizumab has been reported to make tumors more conducive to PD‐L1 inhibition. 21 The involvement of these VEGF‐inhibiting effects could be a contributing factor to our results.

Recently, a phase 3 trial investigated the effect of adding bevacizumab to Chemo + ICI treatment in patients with advanced NSCLC, either without genetic mutations or with EGFR/ALK/ROS‐1/BRAF mutations, who had a history of tyrosine kinase inhibitor use. 22 Although KRAS status was not specified in the inclusion criteria, the trial showed no significant difference in treatment efficacy between the two groups in the overall patient population and in that with driver gene negativity. However, an extended PFS was observed in the population with driver gene mutation positivity. Our analysis also indicated limited benefits of Chemo + Bev + ICI in wild‐type KRAS, suggesting that the effect of adding bevacizumab to Chemo + ICI is limited to a specific subgroup.

In our network meta‐analysis, we were unable to analyze the incidence of adverse events in the population with KRAS mutations. Generally, the addition of anti‐angiogenic agents may increase the occurrence of adverse events. In the previous trial investigating the effect of adding bevacizumab to Chemo + ICI treatment in patients with advanced NSCLC, 22 it was reported that the frequency of hypertension and proteinuria increased by approximately 25%–30% in the group receiving bevacizumab. However, most of these events were mild and manageable with the addition of antihypertensive medication or temporary discontinuation of bevacizumab. Therefore, when adding anti‐angiogenic agents, appropriate follow‐up, including regular blood pressure and urine tests, is considered important.

Our study had several limitations. First, we were unable to conduct an analysis based on the PD‐L1 expression status. This was because of the unavailability of individual PD‐L1 information in the published study data of the population with KRAS mutations. Variation in treatment efficacy based on PD‐L1 expression levels remains a possibility. However, to enhance the similarity between studies, we opted not to include studies that focused only on PD‐L1 expression of ≥1%. Second, we could not consider data on covariates, such as STK‐11/Keap‐1. It is known that STK‐11/Keap‐1 mutations lead to poorer outcomes with ICI than the wild‐type. 23 , 24 Although in our literature search data on the co‐occurrence of KRAS and STK‐11/Keap1 were only available in the Impower150 trial, the ABCP treatment demonstrated efficacy against BCP even in the presence of STK‐11 or Keap1 mutations (14). Third, in the network meta‐analysis of OS in KRAS‐mutant NSCLC, since the included studies did not form loops in the network, inconsistency could not be evaluated. However, in the case of PFS, inconsistency could be assessed, and no significant inconsistency was observed. Finally, except for in the NVALT22 trial, the analyses were based on results from subgroups of KRAS mutations, and the study designs did not include a sufficient sample size to demonstrate efficacy, therefore caution is required while interpreting the results. To address these challenges, prospective trials targeting KRAS‐mutant NSCLC are required.

In conclusion, our study suggests that the Chemo + Bev + ICI group produces favorable treatment effects in terms of PFS/OS as a first‐line treatment for advanced KRAS‐mutant NSCLC. Conversely, in wild‐type KRAS, there was an indication that the Chemo + ICI or ICI + ICI groups may have more favorable treatment effects. These findings highlight the importance of focusing on KRAS mutations when considering primary treatment of advanced NSCLC. However, future prospective trials with a head‐to‐head design are required in the population with KRAS mutations.

AUTHOR CONTRIBUTIONS

Protocol and study design development was performed by Akinari Tsukada. Data analysis and manuscript writing were performed by Akinari Tsukada and Yosuke Shimizu. Literature search and study selection were performed by Akinari Tsukada and Chie Morita. Statistical analysis was conducted by Yosuke Shimizu and Yukari Uemura. Go Naka, Jin Takasaki, Hiroshi Nokihara, Shinyu Izumi, and Masayuki Hojo performed the review and provided advice on the manuscript. All authors have read and agreed to the published version of the manuscript.

FUNDING INFORMATION

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

CONFLICT OF INTEREST STATEMENT

Dr. Tsukada reports receiving honoraria for lectures from AstraZeneca and Chugai Pharmaceutical outside of the submitted work. Dr. Naka reports receiving honoraria for lectures from AstraZeneca, MSD, Ono Pharmaceutical, Chugai Pharmaceutical, and Bristol‐Myers Squibb outside of the submitted work. Dr. Nokihara reports receiving grants from AstraZeneca, MSD, Ono Pharmaceutical, and Chugai Pharmaceutical, and honoraria for lectures from AstraZeneca, MSD, Ono Pharmaceutical, Chugai Pharmaceutical, and Bristol‐Myers Squibb outside of the submitted work. The other authors declare no conflicts of interest.

Supporting information

SUPPORTING INFORMATION DATA S1 Supporting Information

ACKNOWLEDGMENTS

Not applicable.

DATA AVAILABILITY STATEMENT

The authors confirm that all data relevant to the study were included.
==== Refs
REFERENCES

1 Howlader N , Forjaz G , Mooradian MJ , Meza R , Kong CY , Cronin KA , et al. The effect of advances in lung‐cancer treatment on population mortality. N Engl J Med. 2020;383 :640–649. 10.1056/NEJMoa1916623 32786189
2 Travis WD , Brambilla E , Nicholson AG , Yatabe Y , Austin JHM , Beasley MB , et al. The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol. 2015;10 :1243–1260. 10.1097/JTO.0000000000000630 26291008
3 Tamiya Y , Matsumoto S , Zenke Y , Yoh K , Ikeda T , Shibata Y , et al. Large‐scale clinico‐genomic profile of non‐small cell lung cancer with KRAS G12C: results from LC‐SCRUM‐Asia study. Lung Cancer. 2023;176 :103–111. 10.1016/j.lungcan.2022.12.019 36634571
4 Liu SY , Jie GL , Xie Z , Shao Y , Zhang X , Ye JY , et al. Clinical characteristics and prognostic value of the KRAS G12C mutation in Chinese non‐small cell lung cancer patients. Biomark Res. 2020;8 :22. 10.1186/s40364-020-00199-z 32607238
5 Landre T , Justeau G , Assié JB , Chouahnia K , Davoine C , Taleb C , et al. Anti‐PD‐(L)1 for KRAS‐mutant advanced non‐small‐cell lung cancers: a meta‐analysis of randomized‐controlled trials. Cancer Immunol Immunother. 2022;71 :719–726. 10.1007/s00262-021-03031-1 34378081
6 Hutton B , Salanti G , Caldwell DM , Chaimani A , Schmid CH , Cameron C , et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta‐analyses of health care interventions: checklist and explanations. Ann Intern Med. 2015;162 :777–784. 10.7326/M14-2385 26030634
7 Higgins JPT , Thompson SG , Deeks JJ , Altman DG . Measuring inconsistency in meta‐analyses. BMJ. 2003;327 :557–560. 10.1136/bmj.327.7414.557 12958120
8 Dingemans AMC , Ernst S , Mellema W , Burgers S , Staal‐van den Brekel J , Hendriks LEL , et al. LBA50 a randomized phase III study comparing cisplatin‐pemetrexed (cis‐pem) with carboplatin (C)‐paclitaxel (P)‐bevacizumab (B) in chemotherapy naïve patients (pts) with advanced KRAS mutated non‐small cell lung cancer (NSCLC): NVALT22. Ann Oncol. 2021;32 :S1327. 10.1016/j.annonc.2021.08.2129
9 Gadgeel S , Rodriguez‐Abreu D , Felip E , Esteban E , Speranza G , Reck M , et al. KRAS mutational status and efficacy in KEYNOTE‐189: pembrolizumab (pembro) plus chemotherapy (chemo) vs placebo plus chemo as first‐line therapy for metastatic non‐squamous NSCLC. Ann Oncol. 2019;30 :XI64–XI65. 10.1093/annonc/mdz453.002
10 Herbst RS , Lopes G , Kowalski DM , Kasahara K , Wu YL , De Castro G Jr , et al. LBA4 association of KRAS mutational status with response to pembrolizumab monotherapy given as first‐line therapy for PD‐L1‐positive advanced non‐squamous NSCLC in Keynote‐042. Ann Oncol. 2019;30 :XI63–XI64. 10.1093/annonc/mdz453.001
11 Paz‐Ares LG , Ciuleanu TE , Cobo M , Bennouna J , Schenker M , Cheng Y , et al. First‐line nivolumab plus ipilimumab with chemotherapy versus chemotherapy alone for metastatic NSCLC in CheckMate 9LA: 3‐year clinical update and outcomes in patients with brain metastases or select somatic mutations. J Thorac Oncol. 2023;18 :204–222. 10.1016/j.jtho.2022.10.014
12 Peters S , Cho BC , Luft A , Alatorre‐Alexander J , Geater SL , Kim SW , et al. OA15.04 association between KRAS/STK11/KEAP1 mutations and outcomes in POSEIDON: durvalumab ± tremelimumab + chemotherapy in mNSCLC. J Thorac Oncol. 2022;17 :S39–S41. 10.1016/j.jtho.2022.07.073
13 Ramalingam SS , Balli D , Ciuleanu TE . 4O Nivolumab (NIVO) + ipilimumab (IPI) versus chemotherapy (chemo) as first‐line (1L) treatment for advanced NSCLC (aNSCLC) in CheckMate 227 part 1: efficacy by KRAS, STK11, and KEAP1 mutation status. Ann Oncol. 2021;32 :S1375–S1376. 10.1016/j.annonc.2021.10.020
14 West HJ , McCleland M , Cappuzzo F , Reck M , Mok TS , Jotte RM , et al. Clinical efficacy of atezolizumab plus bevacizumab and chemotherapy in KRAS‐mutated non‐small cell lung cancer with STK11, KEAP1, or TP53 comutations: subgroup results from the phase III IMpower150 trial. J Immunother Cancer. 2022;10 :e003027. 10.1136/jitc-2021-003027 35190375
15 Nakajima EC , Ren Y , Vallejo JJ , Akinboro O , Mishra‐Kalyani PS , Larkins EA , et al. Outcomes of first‐line immune checkpoint inhibitors with or without chemotherapy according to KRAS mutational status and PD‐L1 expression in patients with advanced NSCLC: FDA pooled analysis. J Clin Oncol. 2022;40 (16_suppl ):9001. 10.1200/JCO.2022.40.16_suppl.900
16 Ruppert AM , Beau‐Faller M , Debieuvre D , Ouafik L , Westeel V , Rouquette I , et al. Outcomes of patients with advanced NSCLC from the Intergroupe francophone de Cancérologie Thoracique biomarkers France study by KRAS mutation subtypes. JTO Clin Res Rep. 2020;1 :100052. 10.1016/j.jtocrr.2020.100052 34589947
17 Arbour KC , Rizvi H , Plodkowski AJ , Hellmann MD , Knezevic A , Heller G , et al. Treatment outcomes and clinical characteristics of patients with KRAS‐G12C‐mutant non‐small cell lung cancer. Clin Cancer Res. 2021;27 :2209–2215. 10.1158/1078-0432.CCR-20-4023 33558425
18 Noordhof AL , Swart EM , Damhuis RAM , Hendriks LEL , Kunst PWA , Aarts MJ , et al. Prognostic implication of KRAS G12C mutation in a real‐world KRAS‐mutated stage IV NSCLC cohort treated with immunotherapy in The Netherlands. JTO Clin Res Rep. 2023;4 :100543. 10.1016/j.jtocrr.2023.100543 37674812
19 Jeanson A , Tomasini P , Souquet‐Bressand M , Brandone N , Boucekine M , Grangeon M , et al. Efficacy of immune checkpoint inhibitors in KRAS‐mutant non‐small cell lung cancer (NSCLC). J Thorac Oncol. 2019;14 :1095–1101. 10.1016/j.jtho.2019.01.011 30738221
20 Chen DS , Hurwitz H . Combinations of bevacizumab with cancer immunotherapy. Cancer J. 2018;24 :193–204. 10.1097/PPO.0000000000000327 30119083
21 Kudo M . Scientific rationale for combined immunotherapy with PD‐1/PD‐L1 antibodies and VEGF inhibitors in advanced hepatocellular carcinoma. Cancers (Basel). 2020;12 :1089. 10.3390/cancers12051089
22 Shiraishi Y , Kishimoto J , Sugawara S , Mizutani H , Daga H , Azuma K , et al. Atezolizumab and platinum plus pemetrexed with or without bevacizumab for metastatic nonsquamous non‐small cell lung cancer: A phase 3 randomized clinical trial. JAMA Oncol. 2024;10 :315–324. 10.1001/jamaoncol.2023.5258
23 Skoulidis F , Heymach JV . Co‐occurring genomic alterations in non‐small‐cell lung cancer biology and therapy. Nat Rev Cancer. 2019;19 :495–509. 10.1038/s41568-019-0179-8 31406302
24 Aggarwal C , Thompson JC , Chien AL , Quinn KJ , Hwang WT , Black TA , et al. Baseline plasma tumor mutation burden predicts response to pembrolizumab‐based therapy in patients with metastatic non‐small cell lung cancer. Clin Cancer Res. 2020;26 :2354–2361. 10.1158/1078-0432.CCR-19-3663 32102950
