Comparative efficacy and safety of first-line neoadjuvant therapy for early-stage non-small cell lung cancer based on immune checkpoint inhibitor therapy: a systematic review and network meta-analysis

Abstract

Introduction

Although there are a number of neoadjuvant immunotherapy combinations that can be applied to the treatment of perioperative non-small cell lung cancer patients, the optimal treatment combination strategy has not yet been determined.

Methods

We searched PubMed, EMBASE, Cochrane Library, ClinicalTrials.go and randomised controlled trials (RCTs) from major international conferences for literature related to neoadjuvant immunotherapy combinations published as first-line treatment options for non-small cell lung cancer from the start of the library to 20 February 2024, and performed a systematic review and network meta-analysis.

Results

We analyzed nine studies involving 3431 patients, including eight perioperative neoadjuvant immunotherapy combinations for non-small cell lung cancer. For patients without programmed death-ligand 1(PD-L1) selection, Toripalimab plus chemotherapy provided the best Pathological complete response (PCR) benefit (OR = 32.89,95% CI:7.88-137.32), best Major Pathological response (MPR) benefit (OR = 10.25, 95% CI: 5.81–18.10) and best Event-free survival (EFS) benefit (HR = 0.40,95% CI: 0.28–0.57). Nivolumab plus chemotherapy provided the best surgical resection rate (OR = 1.71, 95% CI:0.87–3.40) and pembrolizumab plus chemotherapy provided the best R0 surgical resection rate (OR = 2.20, 95% CI:1.28–3.79). In contrast, the combination of ipilimumab, nivolumab and chemotherapy, and the combination of toripalimab and chemotherapy were associated with the lowest incidence of adverse events of grade 3 or above during neoadjuvant therapy.

Conclusions

Our findings suggest that: Toripalimab plus chemotherapy showed better neoadjuvant efficacy and may have an overall survival benefit, but also increased the incidence of serious adverse events during neoadjuvant therapy.

Introduction

Approximately 30% of patients diagnosed with non-small cell lung cancer (NSCLC) are identified as having stage III disease, encompassing stage IIIA, stage IIIB, and stage IIIC. Due to the considerable discrepancies in tumour size (T1-4) and lymph node involvement (N0-3) across different stages of the tumour, clinical and pathological heterogeneity is markedly elevated. Consequently, despite surgical resection, the risk of recurrence or metastasis remains considerable. For instance, the five-year survival rate for patients with stage IIIA NSCLC is only 36%, which may be attributable to the presence of micrometastatic lesions prior to treatment. This renders it a significant staging group with a high rate of recurrence following surgery [1,2,3]. For example, the five-year survival rate of patients with NSCLC stage IIIA is only 36%, which may be related to the presence of micrometastatic lesions prior to treatment, making it a major staging group with a high rate of postoperative recurrence. The optimal treatment regimen for patients with stage III NSCLC remains a topic of debate. Traditional postoperative adjuvant chemotherapy combined with preoperative neoadjuvant chemotherapy has been shown to offer only a modest survival benefit, with a 5% improvement in 5-year survival compared to postoperative chemotherapy alone [4, 5]. The development of new systemic adjuvant therapies is imperative to reduce the risk of recurrence and improve the prognosis of resectable stage III NSCLC.

With the increasing number of scholars’ clinical studies on relevant early-stage non-small cell lung cancers and the continuous research on immune checkpoint inhibitors, neoadjuvant immunotherapy applied to the perioperative period of lung cancer stage III has been found to have a good advantage of the application. The immune checkpoint inhibitors can be used to trigger a stronger adaptive anti-tumour response by promoting the initiation and expansion of tumour-specific T-cells and thus triggering a more powerful adaptive anti-tumour response, which is more advantageous than adjuvant chemotherapy [6]. They have been shown to be effective in reducing tumour load staging, increasing the likelihood of complete surgical resection. They are able to reduce the risk of postoperative tumour metastasis and recurrence by removing circulating tumour cells from the bloodstream [7]. Concurrently, with advancements in detection technologies, the expression of certain biomarkers, such as PD-L1, has emerged as a critical factor in the neoadjuvant immunotherapy for non-small cell lung cancer (NSCLC). Research indicates a strong correlation between PD-L1 expression levels and patient responsiveness to immune checkpoint inhibitors, including anti-PD-1/PD-L1 antibodies. Specifically, patients with PD-L1-positive tumors undergoing anti-PD-1/PD-L1 therapy tend to exhibit a higher objective response rate (ORR) and extended progression-free survival (PFS) [8]. These findings facilitate the identification of patients likely to benefit from immunotherapy, particularly within the intricate tumor microenvironment, where personalized treatment strategies can enhance both the efficacy and safety of neoadjuvant immunotherapy. Therefore, the further use of immunotherapy in the perioperative treatment of NSCLC patients is a promising direction to explore.

New therapeutic strategies have provided new perspectives on the perioperative treatment of NSCLC, and different neoadjuvant immunotherapy combinations in NSCLC have received widespread attention. A meta-analysis study showed that the Major Pathological response (MPR) of patients treated with neoadjuvant immunotherapy in combination with chemotherapy was 11.76%, and the Pathological complete response (PCR) was 32.4% both of which were significantly superior to chemotherapy alone, and had a reliable safety profile [9]. However, for surgically resectable NSCLC stage III patients, how to accurately select neoadjuvant immunotherapy strategies in the perioperative period of lung cancer so as to reduce the risk of postoperative metastatic recurrence and prolong the survival time of patients has become an important research direction in medicine.

With an increasing number of studies in a series of randomised controlled trials (RCTs) of neoadjuvant immunotherapy for NSCLC and most of the trials comparing the results with standard neoadjuvant chemotherapy, there is a lack of data to support the optimal neoadjuvant treatment strategy for resectable NSCLC. In this context, this study included randomised controlled trials (RCTs) comparing the efficacy of neoadjuvant immunotherapy for lung cancer and performed a net meta-analysis of various combinations of neoadjuvant immunotherapy to provide some guidance for the clinical application of neoadjuvant immunotherapy for lung cancer and subsequent clinical studies.

Materials and methods

According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), we will perform the meta-analysis according to the PRISMA project list (Supplementary Table 1) and the network meta-analysis (NMA) according to the PRISMA extended version [10]. The study protocol has been registered on the International Registry of Prospective Systematic Reviews (CRD42023477356).

Data sources and search strategy

We systematically searched PubMed, EMBASE, Cochrane Library, and ClinicalTrials.gov databases for relevant studies conducted before 20 February 2024, and in order to include the most recent results, we also collected RCTs from major international conferences. The keywords used for the literature search were “non-small cell lung cancer, randomised clinical trials, neoadjuvant therapy, neoadjuvant immunotherapy, PD-1, PD-L1, CTLA-4” (supplementary Table 2).

Selection criteria

The study included published and unpublished sources that met the specified criteria:

• (1) Clinical studies enrolling patients with histologically or cytologically confirmed early-stage non-small cell lung cancer (stage III and before).

• (2) Including randomized controlled trials of neoadjuvant immunotherapy combination therapy as first-line treatment for NSCLC.

• (3) Randomized controlled trials using neoadjuvant chemotherapy combined with other treatments or placebo as the first-line treatment setting.

• (4) This text reports on RCTs that include at least one of the following clinical outcomes: Pathological complete response (PCR) defined as no residual tumour cells after evaluation of resected tumour tissue and regional lymph; Major Pathological response (MPR) defined as ≤ 10% residual tumour cells by pathological testing of postoperative specimens; Event-free survival (EFS) was defined as the time from randomisation to exclusion of surgery for disease progression, local or distant recurrence, or death from any cause; Surgical resection rate was defined as the proportion of patients amenable to surgery after neoadjuvant therapy; R0 resection rate was defined as; the proportion of patients in whom no cancer cells could be found at the microscopic margin and the lesion was completely resected. Adverse Event (AE) Grade ≥ 3 AEs were defined as grade 3 and higher adverse events that occurred during neoadjuvant therapy, defined and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE).

The exclusion criteria were as follows:

• (1) Repeat patients should not be included in randomized controlled trials (RCTs).

• (2) Clinical outcomes should be clearly reported in RCTs.

• (3) The treatment measures of the experimental and control groups should include only drug interventions; non-drug interventions should be avoided.

• (4) Before conducting clinical trials, consideration should be given to preclinical studies, such as animal studies.

Before the evaluation of the full text for eligibility, we screened the titles and abstracts. Abstracts of all included trials and meetings were double-checked online. Any discrepancies were resolved through discussion with the senior author.

Data extraction and quality assessment

According to the meta-analysis reporting system, the data were collected by three researchers (WLF, HY, and ZGD) for independent extraction in the project, and any differences with the other authors (LY and LR) were discussed and resolved. Information on the first author, publication source, year of publication, trial phase, national clinical trial number, sample size, patient age and sex distribution, smoking status, histologic type, PD-L1 expression, and Eastern Cooperative Oncology Group performance status score were extracted from the articles. Clinical outcomes extracted included hazard ratios (HRs) with 95% confidence intervals (95% CI) for event-free survival (EFS). Incidence of pathologic complete response (PCR), major pathologic response (MPR), Rates of undergoing surgery, R0 resection rate, and grade ≥ 3 adverse events (AEs) were also extracted.

The quality of the studies was assessed using the Cochrane Risk of Bias Tool for Randomized Controlled Trials (2.0). The tool assesses five domains: risk of bias due to the randomization process, risk of bias due to deviation from the intended intervention, risk of bias due to missing outcome data, risk of bias due to outcome measurement, and risk of bias due to the selection of reported outcomes [11]. The studies were categorized into three groups: low risk, high risk, and with some concern.

Statistical analysis

Studies have proved that pathological assessment can more accurately reflect the efficacy of neoadjuvant therapy, is highly correlated with progression-free survival and overall survival, and can avoid the “pseudoprogression” assessed by the traditional imaging-based evaluation of the efficacy of solid tumours [12]. Therefore, PCR and MPR were selected as the primary endpoints, and the secondary endpoints were Rates of undergoing surgery, R0 resection rate, and Grade ≥ 3 AEs. Hazard ratio (HR) and the corresponding 95% confidence interval (95% ci) were used as the effect sizes of EFS, and odds ratio (OR) was used as the effect sizes of Rates of undergoing surgery, R0 resection rate, and Grade ≥ 3 AEs. The quality of the literature for each outcome was evaluated using Rev Man 5.3 software, and a network meta-analysis based on the frequency framework of the included literature was performed using Stata 16. A frequency network meta-analysis was conducted using a random effects model to estimate the treatment effect for each direct pairwise comparison. The outcome measures of each intervention were ranked by surface under the cumulative ranking curve (SUCRA). The network meta-analysis was conducted using the netmeta and meta-packages in R. The treatments were ranked based on the P score, which evaluates the likelihood of a treatment strategy being the best and safest option. Statistical heterogeneity and inconsistencies were assessed using the Q test, inconsistent statistics index (I²), and P values. If P > 0.1 and I² ≤ 50%, there was no statistical heterogeneity among the included studies. If P is 0.1 or less and I² is greater than 50%, investigate the reason for heterogeneity between studies. To analyze the possibility of publication bias, funnel plot tests and symmetric plots indicate the absence of publication bias. Additionally, draw a network diagram where the size of the points on the graph represents the sample size of the intervention, and the thickness of the lines represents the amount of direct evidence between the interventions. An inconsistency test is required to analyze the difference between the results of direct and indirect comparisons when there is a closed loop in the relationship graph. In this study, the consistency model was used to analyze both direct and indirect comparisons because each outcome measure did not form a closed loop.

Results

Systematic review and characteristics of the included studies

In a preliminary literature search, we identified 1350 records from a database and found 48 additional records from online sources. After screening out repetitions and irrelevant studies, we considered 69 studies, of which 9 met our eligibility criteria (Fig. 1) [13,14,15,16,17,18,19,20,21]. A total of 3431 patients were included, all of whom received the following 8 treatments: The study included various treatment options such as chemotherapy(CT), nivolumab plus che-motherapy(Nivo-CT), Toripalimab plus chemotherapy(Tori-CT), duvalumab plus chemotherapy(Durv-CT), pembrolizumab plus che-motherapy(Pemb-CT), Tislelizumab plus chemotherapy (Tis-CT) and ipili-mumab plus nivolumab plus chemotherapy (Ipi + Nivo-CT). Detailed information on all the included studies has been presented in Table 1 and Supplementary Table 4.

According to the risk of bias assessment table, all enrolled trials had a low risk of bias (Supplementary Figure S1).

Fig. 1

Literature search and selection. The study procedure followed the PRISMA guidelines

Table 1 Baseline characteristics of the studies included in the network Meta-analysis

Figure 2 and Supplementary Figure S4 displays the network diagram. Supplementary Figure S2 shows the risk of bias assessment. The funnel plot shows a symmetric distribution, indicating that there was no publication bias in the included studies.

Fig. 2

Network plots comparing efficacy and toxicity of neoadjuvant combination. (A): PCR (B): MPR, each circle represents an intervention as a node in the network. The size of the circle and the width of the line are proportional to the number of RCTS and comparisons, respectively, with chemotherapy serving only as a transitive node for indirect comparisons.CT: chemotherapy; Nivo-CT: nivolumab plus chemotherapy, Tori-CT: Toripalimab plus chemotherapy; Durv-CT: duvalumab plus chemotherapy, Pemb-CT: pembrolizumab plus chemotherapy; Tis-CT: Tislelizumab plus chemotherapy; Ipi + Nivo-CT: ipil-imumab plus nivolumab plus chemotherapy

Comparison of PCR, MPR, EFS, rates of undergoing surgery and R0 resection rate

The primary endpoints were PCR and MPR, and the secondary endpoints were EFS, Rates of undergoing surgery and R0 resection rate. NMA included 8 neoadjuvant therapy combinations for PCR, MPR, EFS, Rates of undergoing surgery and R0 resection rate in NSCLC patients treated with neoadjuvant therapy (Fig. 2).

Regarding PCR and MPR(Figs. 3 and 4), PCR and MPR were reported in nine studies containing a total of eight treatment combination strategies (I² = 0%, P = 0.5369, I² = 0%,P = 0.4841), and the results showed that patients receiving neoadjuvant chemotherapy in combination with immunotherapy were more likely to have greater PCR and MPR benefits compared with neoadjuvant chemotherapy.

In terms of providing PCR benefit, Tori-CT (OR = 32.89, 95% CI: 7.88-137.32) provided a significant benefit, compared to TIS-CT (OR = 11.30, 95% CI: 6.08-21.00), Nivo-CT(OR = 8.32, 95% CI: 4.89–14.17), Ipi + Nivo-CT (OR = 8.32,95% CI:1.64–42.12), Pemb-CT(OR = 5.32,95%CI:3.03–9.32), Camr-CT(OR = 4.95,95% CI:1.48–16.57), Durv-CT (OR = 4.65,95% CI:2.63–8.22) compared to Tori-CT (OR = 32.89, 95% CI: 7.88-137.32) provided a significant benefit.

In terms of providing MPR benefit, among neoadjuvant immunotherapy combinations, Tori-CT provided the best MPR (OR = 10.25, 95% CI: 5.81–18.10), followed by Camr-CT (OR = 10.13,95% CI: 3.65–28.14), Ipi + Nivo -CT (OR = 10.37, 95% CI:2.89–37.16), TIS-CT (OR = 7.28, 95% CI: 4.65–11.41), Durv-CT (OR = 3.57, 95% CI: 2.44–5.20), Pemb-CT (OR = 3.51, 95% CI: 2.40–5.12).

Regarding EFS (Figs. 3 and 4), EFS was reported in a total of seven studies containing six neoadjuvant immunotherapy combinations (I² = 0%, P = 0.7450), and patients receiving immune-combination therapies were more likely to have a greater EFS benefit compared to patients receiving neoadjuvant chemotherapy, and among the neoadjuvant immunotherapy combinations, Tori-CT provided the best EFS (HR 0.40, 95% CI: 0.28–0.58) followed by Camr-CT (HR = 0.52, 95% CI: 0.21–1.29), TIS-CT (HR = 0.56, 95% CI: 0.40–0.79), Pemb-CT (HR = 0.58 95% CI: 0.46–0.73), Durv-CT (HR = 0.68, 95% CI: 0.53–0.88).

Regarding Rates of undergoing surgery(Figs. 3 and 4), Rates of undergoing surgery were reported in nine studies, containing a total of eight treatment combination strategies (I² = 0%, P = 0.5176), with the exception of Ipi + Nivo-CT (OR = 0.31, 95% CI: 0.01–8.56), all neoadjuvant immunotherapy combinations were superior to neoadjuvant chemotherapy alone, with Nivo-CT providing the best Rates of undergoing surgery (OR = 1.71, 95% CI: 0.87–3.40).

Regarding R0 resection rate(Figs. 3 and 4), reported in 8 studies comprising a total of 7 treatment combination strategies (I² = 70.4%, P = 0.0340), all neoadjuvant immunotherapy combinations were superior to neoadjuvant chemotherapy alone. Pemb-CT compared with other immunotherapy combinations demonstrated the best (OR = 2.20, 95% CI: 1.28–3.79).

Safety and txicity

Safety and toxicity were determined according to Grade ≥ 3 AEs (Figs. 3 and 4). Of the 9 studies that were analyzed, 8 different immunotherapy combinations in the neoadjuvant therapy portfolio were found to be associated with AEs of grade 3 or higher. (I² = 43.6%, P = 0.1700). Among all treatments, immunodual and chemotherapy combinations had the lowest toxicity and the fewest Grade ≥ 3 AEs, the Ipi + Nivo-CT(OR = 0.32, 95% CI:0.07–1.45) combination had the lowest toxicity, whereas the others increased toxicity by the addition of immunotherapeutic agents during neoadjuvant chemotherapy compared to standard neoadjuvant chemotherapy. Tori-CT (OR = 2.75, 95% CI:1.42–5.33) was associated with relatively more gradations greater than or equal to 3 AEs. Frequently reported treatment-related adverse events of grade 3 or higher Treatment-related adverse events include anaemia, vomiting, fatigue, diarrhoea, neutropenia, leukopenia, and reduced platelet count. Immune-mediated serious adverse events include pneumonia, rash, colitis and endocrine system disorders, and it can be seen that pembrolizumab causes more immune-mediated serious adverse events(Supplementary Table 6).

Fig. 3

Neoadjuvant therapy in patients with lung cancer network meta-analysis of efficacy and safety. (a) : PCR (b) : MPR (c) : EFS (d) : Rates of undergoing surgery (e) : Grade ≥ 3 AEs. (f) : R0 resection rate. (A) Hazard ratios and 95% CIs for Event-free survival(EFS), and a hazard ratio < 1.00 provides better survival benefits. (B) ORs and 95% CIs for Pathological complete response(PCR), Major Pathological response(MPR), Rates of undergoing surgery and R0 resection rate, and an OR>1.00 indicates a better efficacy. ORs and 95% CIs for Grade ≥ 3 AEs, and an OR < 1.00 indicates a better efficacy. CT: chemotherapy; Nivo-CT: nivolumab plus chemotherapy, Tori-CT: Toripalimab plus chemotherapy; Durv-CT: duvalumab plus chemotherapy, Pemb-CT: pembrolizumab plus chemotherapy; Tis-CT: Tislelizumab plus chemotherapy; Ipi + Nivo-CT: ipilimumab plus nivolumab plus chemotherapy

Fig. 4

Efficacy of network meta-analysis in early-stage NSCLC patients without PD-L1 selection. OR > 1.00 indicates better efficacy or poorer safety. CT: chemotherapy; Nivo-CT: niv-olumab plus chemotherapy, Tori-CT: Toripalimab plus chemotherapy; Durv-CT: duvalumab plus chemotherapy, Pemb-CT: pembrolizumab plus chemotherapy; Tis-CT: Tislelizumab plus chemotherapy; Ipi + Nivo-CT: ipilimumab plus nivolumab plus chemotherapy

Rankings

P-scores and area under the curve were used to rank the indicators: each intervention was ranked according to its estimated effect, and the proportion of cycles in which a drug was ranked first was the probability that it was rankedfirst, and the probability of ranking was calculated accordingly. P-score values with the area under the curve showed that Tori-CT treatment yielded the best PCR, MPR, and the best EFS, with the safety rate located at the worst. Ipi + Nivo-CT had the best safety position. Nivo-CT provided the best Rates of undergoing surgery and Pemb-CT demonstrated the best R0 resection rate (Table 2; Fig. 5 and Supplementary Fig S3).

Table 2 Ranking profiles in the bayesian network Meta-analysis

Fig. 5

Radar Chart comparing efficacy and toxicity of neoadjuvant combination

Subgroup analysis

PCR and MPR subgroup analysis based on PD-L1 expression level. Patients were divided into the following 2 subgroups based on PD-L1 expression level less than 1%, greater than or equal to 1%(Fig. 6).

In patients with PD-L1 less than 1%, 5 neoadjuvant immunotherapy combinations were included in the subgroup analysis. Until data from the Tori-CT PD-L1 subgroup analysis were published, Camr-CT provided the best PCR benefit compared to CT, Durv-CT, Nivo-CT, and TIS-CT (OR = 9.33, 95% CI: 0.71-122.57). As for MPR, Camr-CT (OR = 10.0, 95% CI: 0.78-128.77), TIS-CT (OR = 4.67, 95% CI: 2.30–9.49), Nivo-CT (OR = 2.51, 95% CI: 1.12–5.60), Durv-CT (OR = 1.84, 95% CI: 0.90–3.77) were then more effective than neoadjuvant chemotherapy alone.

For patients with PD-L1 greater than or equal to 1%, NMA included 5 neoadjuvant immunotherapy combinations in the subgroup analysis. All IO combinations significantly improved PCR and MPR compared to neoadjuvant chemotherapy alone, regarding PCR, Nivo-CT (OR = 19.27, 95% CI: 5.73–64.89), TIS-CT (OR = 16.27, 95% CI: 6.68–39.59), Camr-CT (OR = 5.79, 95% CI: 0.93–35.81), and Durv-CT (OR = 5.35, 95% CI: 2.78–10.31) were associated with significantly increased PCR. Regarding MPR, Camr-CT (OR = 40.00, 95% CI: 3.56–450.00), Nivo-CT (OR = 13.71, 95% CI: 5.07–37.06), TIS-CT (OR = 9.74, 95% CI: 5.34–17.79), Durv-CT (OR = 4.63, 95% CI: 2.95–7.27) significantly improved MPR.

Fig. 6

Forest plots of the subgroup analysis in patients with advanced. NSCLC according to the PD-L1 expression. CT: chemotherapy; Nivo-CT: nivolumab plus chemotherapy,; Durv-CT: duvalumab plus chemotherapy; Tis-CT: Tislelizumab plus chemotherapy

Discussion

Stage III of non-small cell lung cancer represents the most challenging stage in terms of diagnosis and treatment. Patients with inoperable stage III disease are predominantly concentrated in some stage IIIA, stage IIIB and all stage IIIC patients. Previous treatments were primarily based on synchronous radiotherapy and chemotherapy. While these methods can control tumour progression to a certain extent, the issues of side effects and drug resistance frequently restrict the enhancement of efficacy. In recent years, there has been a growing interest in the use of neoadjuvant immunotherapy as a novel therapeutic approach. A number of studies have demonstrated that immune checkpoint inhibitors have superior survival and progression-free survival rates compared to conventional chemotherapy and radiotherapy in patients with stage III inoperable non-small cell lung cancer [22]. Consequently, the integration of conventional therapy with emerging neoadjuvant immunotherapy may provide novel insights and avenues for enhancing patient prognosis.

Compared with published meta-analyses of neoadjuvant immunotherapy in NSCLC, our study is the first to compare the efficacy and safety of all current combinations of first-line neoadjuvant immunotherapy in NSCLC, which includes a more comprehensive set of neoadjuvant immunotherapy interventions [23, 24]. In this study, we comprehensively pooled and analysed the efficacy and safety of different neoadjuvant immunotherapy combinations in patients with early-stage lung cancer. Because the overall survival of patients with early-stage NSCLC is longer than that of patients with advanced NSCLC, trials for early-stage lung cancer that use overall survival as an indicator of survival benefit need to be longer and fewer trials are relevant, so in this paper, we used pathological complete response (PCR) and major pathological remission (MPR) as the main efficacy indicators of survival benefit. The results of related studies showed that MPR and PCR were significantly correlated with the survival benefit of neoadjuvant therapy, so PCR and MPR were used as the main clinical benefit indicators in this paper. With regard to safety, the majority of the included studies employed a ‘sandwich’ model, combining preoperative neoadjuvant therapy with postoperative adjuvant therapy. Consequently, only severe adverse events occurring during preoperative neoadjuvant therapy were included in this study, which differs from the findings of previous studies. Concurrently, neoadjuvant immunotherapy typically necessitates the utilisation of costly pharmaceuticals, in addition to the implementation of sophisticated diagnostic and monitoring procedures. The financial implications of these costs can be considerable, placing a significant financial burden on patients, hospitals and health insurance systems. The necessity for lengthy courses of treatment, coupled with the frequency of follow-up visits and testing, may result in an overall increase in healthcare expenditures. This may have a significant impact on the financial situation of the patient and the allocation of healthcare resources. Furthermore, neoadjuvant immunotherapy typically prolongs the interval between diagnosis and surgical intervention, which in some instances may influence the timing of surgery and the ultimate outcome. It is necessary to rationalise the treatment process in clinical practice and for clinicians to give further consideration to the balance between efficacy and cost.

Nevertheless, despite the demonstrated efficacy of neoadjuvant immunotherapy in certain cases, conventional concurrent chemoradiotherapy remain the preferred treatment option for some tumour types, particularly in patients with high tumour loads. Accordingly, the selection of an appropriate treatment regimen necessitates a comprehensive assessment of the patient’s specific circumstances and tumour characteristics. Furthermore, the potential adverse effects of concurrent chemoradiotherapy and neoadjuvant immunotherapy, and their influence on the patient’s quality of life, are crucial factors in the decision-making process regarding the most appropriate treatment. Concurrent chemoradiotherapy is typically associated with considerable adverse effects, including nausea, vomiting, alopecia and immunosuppression, which can have a profound impact on patients’ quality of life. In contrast, neoadjuvant immunotherapy is typically associated with less severe adverse effects and is well tolerated by the majority of patients. Immunotherapy has the potential to confer benefits in terms of improved quality of life for patients, particularly in those with a favourable pathological response. It is therefore essential to conduct a comprehensive assessment of the tolerance of side effects and their impact on patients’ quality of life when developing treatment regimens. Furthermore, future studies should aim to identify the optimal combination of the two, with a view to improving survival and quality of life.

In recent years, the rapid development of immune checkpoint inhibitors (ICIs) has led to a significant increase in the number of studies on NSCLC immunotherapy-related drugs. This has enabled immunotherapy-focused researchers to explore novel neoadjuvant treatment options, with an increasing number of drugs being applied to the neoadjuvant treatment of lung cancer, resulting in notable outcomes. In the context of driver-negative NSCLC, immunotherapy has demonstrated novel advantages in the neoadjuvant treatment of lung cancer. It has been demonstrated that initiating immune checkpoint blockade at the early ‘in situ’ stage of the tumour can utilise higher levels of endogenous tumour-associated antigens, which can enhance T-cell activation and eliminate micrometastatic tumour deposits. This approach is more effective than adjuvant therapy for resected micrometastatic disease and provides patients with more possibilities for surgical resection and cure.

However, most of the clinical studies of neoadjuvant immunisation reported to date are mostly single-arm studies and lack head-to-head comparative studies, reporting only pathological response rates with no high level of evidence, and the study in this paper provides a reference for subsequent development in this field. In this study, we found that Tori-CT treatment obtained the best PCR, MPR and best EFS, and there have been studies confirming that both PCR and EFS have been shown to correlate with survival benefits, so this study suggests that Tori-CT may be the best combination strategy for neoadjuvant therapy.

We observed that the Tori-CT study enrolled predominantly stage III patients (67.3% and 32.2% of the stage IIIA and IIIB populations, respectively) compared with studies of other interventions and that even with the inclusion of more advanced patients, the combination treatment strategy of Tori-CT has been demonstrated to be both effective, the main reasons may be as follows:

• 1. The unique anti-tumour mechanism of Trepolizumab, similar to other anti-PD-1, Trepolizumab can bind to PD-1, selectively block the interaction between PD-1 and PD-L1, reactivate T-cells and restore the tumour-killing effect, and it has a high affinity for PD-1. Clinical studies have shown that Trepolizumab has a 12-fold higher affinity for PD-1 than Pabolizumab, and it has a higher affinity for PD-1, and it has a more potent T-cell activation effect [25]. In addition, after binding to PD-1, Trepolizumab can induce endocytosis of PD-1 protein on the surface of T-cell membrane and reduce its expression on the cell membrane, so as to better restore T-cell function and promote immune normalisation, thus realising highly effective and long-lasting anti-tumour effects. Clinical studies have shown that when Toripalimab is used in combination with chemotherapy, it significantly improves the overall survival of patients regardless of their PD-L1 status [26].

In previous studies of melanoma, the median OS and PFS of triptorelinumab combined with axitinib in the first-line treatment of mucosal melanoma were significantly improved compared with those of previous studies, with the 1–3 year OS rates of 62.1%, 44.8%, and 31%; and the 1–3 year PFS rates of 41.4%, 13.8%, and 10.3%, respectively, and is expected to become a new standard for first-line treatment of mucosal melanoma [27]. The domestic PD-1 inhibitor approved in China has shown remarkable efficacy in the treatment of advanced NSCLC, small cell lung cancer, melanoma, and other tumor types, and its efficacy is comparable to similar drugs in other countries, such as camrelizumab and nivolumab [28]. Therefore, even in the case of worse baseline characteristics of patients, perioperative treatment with toripalimab is still able to reduce the size of the tumour and the number of cancer cells, which improves the success rate of surgery, and then reduces the risk of postoperative recurrence and provides patients with more survival benefits.

• 2. Original “3 + 1 + 13” perioperative treatment mode: 3 cycles of neoadjuvant therapy with toripalimab combined with chemotherapy, 1 cycle of adjuvant therapy with toripalimab combined with chemotherapy, and 13 cycles of consolidation therapy with toripalimab. This treatment modality has provided significant benefits to patients by providing continuous treatment before, after, and for a period of time after surgery, and by eliminating residual lesions through comprehensive management.

• 3. This study included patients of Asian descent, and studies have shown that Asian populations may benefit more from immunotherapy than non-Asian populations due to genetic differences, which may be one of the reasons for the best outcomes in this study[29].

Overall, triplizumab has demonstrated substantial efficacy in the treatment of non-small cell lung cancer. Particularly in perioperative settings, the “sandwich cake” regimen combining triplizumab with chemotherapy has delivered unprecedented survival benefits to patients. This approach not only markedly prolongs event-free survival but also substantially diminishes the risk of disease recurrence, metastasis, progression, and other adverse events, thereby significantly influencing the current treatment guidelines for NSCLC. The latest edition of the “China Clinical Oncology Society (CSCO) Guidelines for the Diagnosis and Treatment of Non-Small Cell Lung Cancer” lists triplizumab in conjunction with chemotherapy as a Grade I recommendation for stage III resectable NSCLC. This represents the first and sole perioperative immunotherapy regimen in China to receive a Grade I recommendation, offering clinicians and patients an enhanced therapeutic option. Triplizumab has not only achieved notable success in the Chinese market but has also garnered widespread recognition internationally. For instance, it is the first and only approved drug in Europe for the treatment of nasopharyngeal carcinoma and the sole first-line therapy in Europe for advanced or metastatic esophageal squamous cell carcinoma regardless of PD-L1 expression. Extensive clinical trials encompassing multiple indications have been conducted across numerous countries and regions, leading to successful market launches in key areas such as the United States and the European Union, thus laying a robust foundation for its broad application worldwide. As globalization progresses, oncologists beyond China can draw upon China’s successful experiences, enhance international collaboration and exchange, collectively advance the diagnostic and therapeutic standards for NSCLC and other malignancies, and foster the shared development of global lung cancer prevention and treatment through the sharing of research findings and clinical insights.

The U.S. Food and Drug Administration has formally approved the use of anti-PD-L1 and anti-CTLA-4 immunodoublet chemotherapy for the treatment of advanced non-small cell lung cancer [30]. The highest MPR and optimal safety profile were observed in patients undergoing neoadjuvant therapy with Ipi + Nivo-CT. Nivo-Ipi is an interventional agent that targets both PD-1 and CTLA-4, potentially exhibiting internal synergistic effects. Ipilimumab is a monoclonal antibody that promotes T-cell activation and proliferation, while nivolumab is a checkpoint inhibitor that supports existing T-cell recognition and targeting of tumour cells. Furthermore, T cells that can be activated by ibritumab are capable of differentiating into memory T cells. Such memory T cells are more likely to generate long-term immune responses, thereby facilitating the early control of the disease. In regard to safety, our findings indicate that the dual immunisation combination may not markedly elevate the incidence of adverse events (AEs), which is contrary to the findings of previous studies. However, this may be attributed to a potential selection bias associated with the limited sample size of the included studies. In the Nivo + CT group, 12 patients (55%) experienced grade 1–2 TRAEs, while 10 patients (45%) experienced grade 3–4 TRAEs. In the Ipi + Nivo + CT group, 16 patients (80%) experienced grade 1–2 TRAEs, while 4 patients (20%) experienced grade 3–4 TRAEs. As only grade 3–4 TRAEs were included as a safety assessment criterion, it is possible that false-positive findings of a higher safety profile of the dual immunisation combination may have occurred. Further comparative studies are needed to confirm these results. Furthermore, due to the limited patient sample size and potential patient selection bias, it is plausible that the dual immunisation combination group exhibited superior baseline health status and fewer comorbidities. This study revealed that the dual immunisation combination group had a higher proportion of patients below the age of 65, which could have contributed to a relatively lower toxicity profile.

In terms of rates of undergoing surgery, we observed that among all interventions, patients who received Nivo + CT had the highest OR, whereas those who received IPI + Nivo had the lowest OR, which is inconsistent with our conventional understanding. In studies involving related interventions, we further analysed the reasons for failure to undergo surgery in both interventions and found that the high acceptance rate of Nivo + CT was mainly due to the cancellation or delay of surgery because no severe AE was detected, and in clinical studies involving Nivo + CT as an intervention, we could observe that compared with neoadjuvant chemotherapy alone, disease progression during neoadjuvant chemotherapy resulted in failure to undergo surgery. In addition, we can observe that compared with neoadjuvant chemotherapy alone, during neoadjuvant therapy, the number of patients who were unable to undergo surgery due to disease progression was significantly less than that of neoadjuvant chemotherapy alone, further proving that Nivo + CT can effectively inhibit disease progression during neoadjuvant therapy and that its effectiveness and safety ensures a high rate of surgery for lung cancer patients. In the literature involving IPI + Nivo, we can find that the reason why the IPI + Nivo + CT group was unable to undergo surgery was mainly unrelated to immunotherapy (complications related to SARS-CoV-2 infection), and the sample size of the included population was small (n = 22), and the individual differences in the samples led to too large an overall difference, which could not accurately reflect their therapeutic effects, and in the subsequent studies single-cell sequencing found that the Ipi + Nivo + CT group generally enhanced the tumour immune infiltration and reduced the immunosuppressive phenotype compared with the Nivo + CT group, further proving the superiority of the combination of dual-immunity drugs, which is in line with the conventional understanding, so we speculate that the reason for the lowest Ipi + Nivo + CT is mainly due to the individual differences of the patients, and that it is necessary to further expand the sample size for an in-depth study.

In terms of R0 resection rate, compared to other studies, the study of Pemb-CT had the best R0 surgical resection rate despite patients having lower ORs for PCR and MPR and compared to other studies, which included a population of patients with earlier-stage stage II lung cancers (approximately 30% of patients), it was still possible to have a higher R0 surgical resection rate in patients with stage II lung cancers even if they did not have a significant remission for PCR and MPR. High tumour R0 surgical resection rates. In addition, neoadjuvant therapy also reduces the oxidation of the resectable tumour bed, so that in some patients after preoperative neoadjuvant therapy, although the PCR and MPR remission is not obvious, the gap between the tumour and the blood vessels has been more clearly defined than before the treatment, which makes it more effective than direct surgery, and ultimately achieves the goal of increasing the R0 surgical resection rate of patients. Secondly, due to the small number of included studies, it is difficult to exclude the factor of individual differences in patients. For tissue sectioning, the pathologist’s professional knowledge and experience can also have a greater impact, such as the cutting of sections can be axial, bilateral or multilateral, etc., and it is difficult to accurately differentiate between exact specimen cutting edges and so on is also an important factor influencing the R0 resection of the tumour.

Nevertheless, the existing neoadjuvant clinical studies predominantly comprise populations of Asian origin, and notable discrepancies exist in the prevalence of mutations in genes such as STK11, TP53 and EGFR between Asian and non-Asian patients. It has been demonstrated that STK11 mutations are associated with a poor response to immunotherapy. Furthermore, there is a significant difference in the incidence of STK11 mutations between Asian and non-Asian populations. This significant difference may account for the superior benefit of immunotherapy in some Asian populations compared to non-Asian populations [31]. Furthermore, some studies have demonstrated disparities in the clearance rates of PD-1 monoclonal antibodies between Asian and non-Asian populations. This could be a potential explanation for the observed discrepancies in the efficacy of immunotherapy across different populations. It is imperative that broader inclusion criteria be explored in order to ensure that the results of neoadjuvant immunotherapy studies are more representative of the global population.

Nevertheless, a significant number of clinical trials of neoadjuvant immunotherapy may fail to adequately include participants of non-Asian ethnicity. Furthermore, the question of how to effectively apply findings from Asian populations to non-Asian populations remains a complex topic. For example, different ethnic groups may exhibit disparate tumour biology, including tumour type, molecular subtype and sensitivity to immunotherapy. Consequently, the response of certain immunotherapeutic agents in Asian patients may not be applicable to other ethnic groups. Accordingly, when contemplating potential therapeutic avenues, comprehensive investigations into the genetic profiles of diverse populations are imperative to ascertain the efficacy and safety of proposed treatments.

To guarantee that the outcomes of neoadjuvant immunotherapy studies are more accurately reflective of global populations, it is imperative to investigate more inclusive inclusion criteria. It is imperative to encourage participants from diverse geographical regions, racial and ethnic backgrounds, and to increase the involvement of non-Asian populations in clinical trials. This will facilitate the validation and dissemination of the results of these studies. One strategy for achieving this is to establish clinical trial centres in different regions, which can provide a more comprehensive understanding of the effects and side-effects of neoadjuvant immunotherapy in different populations. Furthermore, the impact of these factors on efficacy and side effects can be studied by including patients of different age groups and genders. It is of particular importance to consider the response of elderly and female patients to immunotherapy. Furthermore, studies should encompass patients with disparate stages and types of lung cancer, particularly those tumour types that exhibit a high incidence in specific populations, to guarantee the general applicability of treatment effects.

As neoadjuvant immunotherapy is often embodied as a pre-surgical intervention, long-term follow-up is required to assess its long-term efficacy and safety. Consequently, most of the relevant findings use pathological data such as PCR and MPR, and lack long-term survival data. The long-term efficacy of immunotherapy may be affected by a number of factors, including the biological behaviour of the tumour in question, individual patient differences and other therapeutic interventions. This adds to the complexity of data analysis. To address this limitation, a multicentre clinical trial network could be established in the future. By integrating the resources of multiple healthcare institutions and recruiting more patients to participate in the study, the sample size could be increased, the follow-up time extended, and the reliability of the data improved. Following neoadjuvant immunotherapy, long-term follow-up of patients will be incorporated into the research plan. A comprehensive electronic health record will be established, and data on survival rate, recurrence rate and quality of life will be collected on a regular basis. This will provide a theoretical basis for individualised treatment. The aforementioned strategies will facilitate a more comprehensive and profound understanding of the long-term effects of neoadjuvant immunotherapy, thereby providing a more reliable basis for patients’ treatment decisions.

A detailed examination of the original study data in the subgroup analysis of this study revealed that the included clinical studies included only a limited number of individuals and lacked sufficient statistical power. It is acknowledged that there may be considerable variation in the baseline characteristics of patients, treatment regimens and follow-up durations across the studies in question, which may present a challenge to the comparability of results in subgroup analyses. In particular, the diversity of geographic regions, encompassing differences in patient populations, genetic characteristics, treatment acceptance, and follow-up patterns across urban and rural areas and countries, may influence the response to immunotherapy. Furthermore, socioeconomic factors are also a contributing element in the efficacy of treatment. The socioeconomic background of the study patients, including income level, education level, and health insurance status, may influence both access to treatment and patient adherence. Consequently, the current sample size limitations may result in the drawing of conclusions that are subject to greater bias, which may affect the scientific validity and rigour of the study. It is recommended that future subgroup analyses increase the sample size of patients with specific variables, such as PD-L1 expression level, tumour type and stage, genomic features, and immune status before and after treatment. It is recommended that the above variables be integrated into the design of future clinical trials, and that the corresponding enrollment criteria and evaluation indicators be developed to more comprehensively evaluate the efficacy and safety of neoadjuvant immunotherapy. This will facilitate a more comprehensive and in-depth understanding of the effect of neoadjuvant immunotherapy, and enhance the applicability and practical value of research. This will contribute to relevant research in academia, and provide stronger support for clinical practice.

The application of neoadjuvant immunotherapy in the treatment of lung cancer has garnered increasing attention, yet several challenges remain. Future research endeavors should concentrate on the following areas: Firstly, assessing the long-term survival rates and quality of life outcomes associated with neoadjuvant immunotherapy is essential for elucidating its sustained efficacy. This involves a detailed examination of variations among different tumor types, stages, and individual patient profiles to ascertain which patients stand to gain the most from this therapeutic approach. Secondly, identifying robust biomarkers predictive of patient response to neoadjuvant immunotherapy is critical for advancing personalized medicine and enhancing both the effectiveness and safety of the treatment regimen. Additionally, exploring synergistic combinations of neoadjuvant immunotherapy with other modalities, including chemotherapy, radiotherapy, and targeted therapies, is imperative to optimize treatment outcomes while minimizing adverse effects. Lastly, as neoadjuvant immunotherapy gains traction, evaluating its economic and societal impact becomes increasingly pertinent. Comparative analyses of the overall costs and patient burdens associated with neoadjuvant immunotherapy versus traditional treatments across diverse national and healthcare settings will facilitate a more nuanced understanding of its applicability.

Neoadjuvant immunotherapy provides a new perspective on the perioperative treatment of NSCLC, and the efficacy of different therapeutic regimens is still in the exploratory stage. The present study further clarifies the advantages of these regimens, and it is worthwhile to explore the “precision” selection of suitable patients and more effective and safe treatment combination strategies in the future, thus bringing hope for the survival and cure of more patients with NSCLC. In the future, it is worthwhile to further explore a “precise” selection of suitable patients and more effective and safe treatment combination strategies, so as to bring hope to more NSCLC patients for survival and cure. For early-stage operable NSCLC, especially considering the highly heterogeneous nature of stage III patients, further individualised and prospective biomarker testing is needed to identify the precise beneficiaries of neoadjuvant immunotherapy, to expand the chances of curing lung cancer and to reduce the risk of death from lung cancer.

Conclusion

Based on the results, neoadjuvant chemotherapy combined with toripalimab appears to be a more effective neoadjuvant treatment for non-small cell lung cancer, but also increased the incidence of serious adverse events during neoadjuvant therapy. However, further high-quality head-to-head clinical trials are necessary to confirm this conclusion.

Limitations

The study still has many limitations, First of all, for neoadjuvant immunotherapy, there are limited randomised controlled clinical trials with small sample sizes included in the study, and a proportion of the studies have not yet published complete analyses with incomplete data in further subgroup analyses.

Second, NSCLC patients with EGFR mutations are unlikely to derive a survival benefit from neoadjuvant immunotherapy. In the present study, the EGFR of the population involved in most of the included studies is unknown, and more well-designed randomised controlled trials are needed to document and report histological information relevant to patients with NSCLC in order to improve clinical decision-making and develop appropriate treatment strategies.

Thirdly, the efficacy and safety of neoadjuvant immunotherapy in patients with NSCLC may vary depending on the ethnic and geographic differences of the included study populations.

Fourthly, subgroup analyses of tumour stage and pathology type were not performed due to limited information on the included studies.

Finally, the different chemotherapy combinations included were not compared in further subgroup analyses. Differences in the neoadjuvant chemotherapy drugs and doses included in the study may have had some impact on the results.