Prognostic value of lung immune prognostic index in non-small cell lung cancer patients receiving immune checkpoint inhibitors: a meta-analysis

Background and Purpose Until now, it has been difficult to accurately predict the efficacy of immunotherapy in patients with non-small cell lung cancer (NSCLC). A novel indicator, the lung immune prognostic index (LIPI), has shown relatively high prognostic value in patients with solid cancer. Therefore, this study aimed to further identify the association between LIPI and the survival of patients with NSCLC who receive immune checkpoint inhibitors (ICIs). Methods Several electronic databases were searched for available publications up to April 23, 2023. Immunotherapy outcomes included overall survival (OS), progression-free survival (PFS), and hazard ratios (HRs) with 95% confidence intervals (CIs). Subgroup analysis based on the study design and comparison of the LIPI was conducted. Results In this meta-analysis, 21 studies with 9,010 patients were included in this meta-analysis. The pooled results demonstrated that elevated LIPI was significantly associated with poor OS (HR = 2.50, 95% CI:2.09–2.99, p < 0.001) and PFS (HR = 1.77, 95% CI:1.64–1.91, p < 0.001). Subgroup analyses stratified by study design (retrospective vs. prospective) and comparison of LIPI (1 vs. 0, 2 vs. 0, 1–2 vs. 0, 2 vs. 1 vs. 0, 2 vs. 0–1 and 2 vs. 1) showed similar results. Conclusion LIPI could serve as a novel and reliable prognostic factor in NSCLC treated with ICIs, and elevated LIPI predicts worse prognosis.


Introduction
Lung cancer remains the most common malignancy and leading cause of tumor-related deaths worldwide [1,2].Nonsmall cell lung cancer (NSCLC) accounts for approximately 85% of all lung cancer cases [3].Despite great advances in early screening, surgical techniques, and adjuvant therapies for NSCLC, the overall prognosis remains poor, representing a relatively high risk of recurrence and therapeutic resistance [4,5].In the last few years, immunotherapy has become an important treatment option for NSCLC, especially for patients with advanced-stage and metastatic NSCLC.Unfortunately, less than 20% of patients could benefit from immunotherapy [6].
In clinics, immune checkpoint inhibitors (ICIs), particularly anti-programmed death ligand 1 (PD-L1)/ programmed death 1 (PD-1) antibodies, are widely used as firstor second-line treatments for metastatic/advanced NSCLC alone or in combination with chemotherapy.However, as mentioned above, the number of patients who benefit from ICIs is fairly limited [6].Thus, accurate and effective indicators to predict the efficacy of ICIs are urgently needed to help select potential beneficiaries of ICIs.Overall, PD-L1 expression and tumor mutation burden (TMB) are the most commonly used biomarkers to select ICI-advantaged populations and predict prognosis.Nevertheless, the predictive effect of these two biomarkers on ICI efficacy is not satisfactory in clinical practice [7,8].Patients with high PD-L1 expression are more likely to experience better survival, but a subset of patients do not benefit from immunotherapy [9,10].Therefore, further exploration of effective predictive indicators for the prognosis of ICIs treated NSCLC is required.
Since the immune checkpoint pathway includes an important circulatory phase, changes in some parameters based on peripheral blood may be associated with the response to immunotherapy.Increasing evidence suggests that inflammatory responses play an essential role in the development and progression of cancers [11,12].The inflammatory process in the body is considered to be the immune resistance mechanism in cancer patients, which promotes the growth and spread of tumor cells and activates the carcinogenic signaling pathway [11,12].Some biomarkers, such as the neutrophil-to-lymphocyte ratio (NLR), derived neutrophil-to-lymphocyte ratio (dNLR), and platelet-to-lymphocyte ratio (PLR) have been used to detect inflammatory status and predict prognosis in various cancers, including NSCLC [11][12][13][14].The lung immune prognostic index (LIPI) is a novel indicator based on a dNLR >3 and lactate dehydrogenase (LDH) > upper limit of normal range (ULN), which was first reported by Mezquita et al. [15].Patients were divided into three groups based on the number of risk factors from the LIPI: low-risk, intermediate-risk, and high-risk groups with 0, 1, and 2 risk factors, respectively.Previous studies have revealed that pretreatment LIPI play a role in predicting the therapeutic outcomes of ICIs in patients with solid cancers.However, whether it can predict the prognosis of ICIs ICI-treated NSCLC remains unclear.
Therefore, this meta-analysis aimed to further identify the association between LIPI and survival of NSCLC patients receiving ICIs, which might contribute to the selection of an advantaged population and improvement of the therapeutic efficacy of immunotherapy among NSCLC patients.

Materials and methods
The current meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 [16].

Literature search
The PubMed, EMBASE, Web of Science, and CNKI databases were searched from inception to April 23, 2023.The following terms were used for the search: PD-1, PD-L1, CTLA-4, ICIs, immune checkpoint inhibitor, lung, pulmonary, cancer, tumor, carcinoma, neoplasm, LIPI, lung immune prognostic index, survival, prognosis, and prognostic.The detailed search strategies were as follows: (PD-1 OR PD-L1 OR CTLA-4 OR ICIs OR immune checkpoint inhibitor) AND (lung OR pulmonary) AND (cancer OR tumor OR carcinoma OR neoplasm) AND (LIPI OR lung immune prognostic index) AND (survival OR prognosis OR prognostic).Free text and Medical Subject Headings terms were also applied.All the references cited in the included studies were reviewed.

Inclusion criteria
Studies that met the following criteria were included:1) patients were pathologically diagnosed with primary NSCLC; 2) patients who received ICIs with or without other combined therapies such as chemotherapy; 3) LIPI score was assessed according to the dNLR values and LDH level before immunotherapy, and the association between LIPI and efficacy of immunotherapy was evaluated; 4) the overall survival (OS) and (or) progression-free survival (PFS) were defined as outcomes of immunotherapy; 5) hazard ratios (HRs) with 95% confidence intervals (CIs) for OS and PFS were directly reported in articles.

Exclusion criteria
Studies that met any of the following criteria were excluded:1) low-quality studies; 2) letters, editorials, reviews, case reports, or animal trials; and 3) studies with insufficient or duplicated data.

Data extraction
The following information was collected from the included studies: name of first author, publication year, country, study design (retrospective or prospective), sample size, TNM stage, pathological type, detailed drugs of ICIs, threshold and comparison of LIPI, endpoint, HR, and 95% CI.

Quality assessment
Owing to the nature of the included studies, the Newcastle-Ottawa Scale (NOS) score system was used to evaluate the quality of the included studies.As mentioned above, only high-quality studies with an NOS score ≥6 were included.
The literature search, selection, information collection, and quality assessment were conducted by two authors independently and any disagreement was resolved by team discussion.

Statistical analysis
All statistical analyses were performed using STATA 12.0.Heterogeneity between studies was evaluated using I 2 statistics and Q test.If significant heterogeneity was observed (I 2 > 50% and/or p < 0.1), the random-effects model was applied; otherwise, the fixed-effects model was used.HRs and 95% CIs were combined to evaluate the association between the LIPI, OS, and PFS.Subgroup analysis based on study design (retrospective vs. prospective) and comparison of LIPI (1 vs. 0, 2 vs. 0, 1-2 vs. 0, 2 vs. 1 vs. 0, 2 vs. 0-1 and 2 vs. 1) were conducted.Sensitivity analysis was conducted to detect the sources of heterogeneity and assess the stability of the overall results.Furthermore, Begg's funnel plot and Egger's test were conducted to detect publication bias, and significant publication bias was defined as p < 0.05 [17][18][19].

Basic characteristics of included studies
A total of 9,010 participants were enrolled in 21 studies published between 2018 and 2023.Most of the included studies were retrospective and focused on patients with advanced NSCLC.The sample size ranged from 51 to 1,489, and all studies applied the same definition of LIPI risk grading: LIPI 0, dNLR≤3 and LDH ≤ ULN; LIPI 1, dNLR >3 or LDH > ULN; and LIPI 2, dNLR >3 and LDH > ULN.Specific data are presented in Table 1.

Association between LIPI and OS and PFS
Seventeen studies explored the relationship between LIPI and OS in NSCLC patients receiving ICIs.The pooled results showed that elevated LIPI predicted poorer OS (HR = 2.50, 95% CI: 2.09-2.99,p < 0.001; I 2 = 68.6%,p < 0.001), and subgroup analysis based on the study design showed the same results (Figure 2A).

Subgroup analysis for PFS
Similarly, subgroup analysis for PFS based on the comparison of LIPI and the study design was performed.2.

Sensitivity analysis
Sensitivity analysis for the association between LIPI and OS and PFS was performed, which demonstrated that the pooled results of this meta-analysis were stable, and none of the included studies had an obvious impact on the relationship between LIPI and OS (Figure 5A) and PFS (Figure 5B) among immunotherapy-treated NSCLC patients.

Publication bias
The Begg's funnel plots for OS (Figure 6A) and PFS (Figure 6B) were both symmetrical, and the p-values of Egger's test for OS and PFS were 0.208 and 0.992, respectively.Thus, no significant publication bias was observed in this meta-analysis.

Discussion
This meta-analysis explored the predictive role of LIPI for the long-term survival of NSCLC patients who received ICIs based on current evidence, and the pooled results showed that LIPI was significantly associated with OS and PFS in this group of patients.Patients with elevated LIPIs were more likely to have a worse prognosis than patients with good LIPIs.Therefore, the LIPI could serve as a novel and reliable prognostic indicator in patients with NSCLC receiving ICIs.
The invasiveness of malignant tumors depends on the nature of tumor cells and their microenvironment.Previous studies have indicated that inflammation is a recognized feature of cancer, and inflammatory reactions play a crucial role in the process of carcinogenesis [40].On the one hand, in malignant solid tumors, inflammatory stimulation leads to immune cell infiltration, angiogenesis, and fibroblast proliferation [41,42].In contrast, it is one of the mechanisms of immune tolerance, promoting tumor growth and dissemination, and activating oncogenic signaling pathways in cancer patients [43].The dNLR was calculated using the neutrophil and lymphocyte counts.Neutrophils are key participants in tumor inflammation and immunity, and participate in tumor progression.Studies have found that neutrophils can produce vascular endothelial growth factor (VEGF), which plays an important role in mediating tumor angiogenesis and is a powerful immunosuppressive factor of natural and adaptive anti-tumor immunity [44].In addition, neutrophil-derived proteases can degrade cytokines and chemokines and reshape the extracellular matrix, and neutrophil elastase in tumor cells can overactivate the PI3K pathway, further accelerating uncontrolled tumor proliferation [45].It has been reported that T cells producing interleukin (IL)-17 can release CXC chemokines to supplement neutrophils, and IL17a is involved in resistance to ICIs [46].Therefore, higher dNLR levels may reflect negative inflammation and contribute to resistance to ICIs.Peripheral blood lymphocyte count is considered a predictive factor for the prognosis of various cancers [47].Lymphocytes play an important role in tumor-related immunity, have potential anti-tumor immune functions to inhibit tumor development, participate in cytotoxic cell death and cytokine production, and inhibit tumor cell proliferation and metastasis through the immune response to cancer [48].
LDH is widely distributed in major human organs and catalyzes the conversion of lactate and pyruvate.It is an indicator of tumor burden, cell damage, and necrosis.Studies have shown that elevated LDH levels are an adverse prognostic factor for tumors [49,50].Elevated LDH levels are a product of enhanced tumor glycolysis and hypoxia-induced tumor necrosis [51].On one hand, in tumors with increased glycolytic activity, both aerobic and anaerobic glycolysis under hypoxia can affect immune cell function due to glucose deficiency or tumor acidity [52].Furthermore, hypoxia itself or the excessive expression of hypoxia-   This meta-analysis had several limitations that should be noted.First, all included studies were observational, and most of them were retrospectively conducted.Second, some of the included studies had relatively small sample sizes, which might have caused bias.Third, we were unable to conduct more subgroup analyses based on other parameters such as the pathological subtype, drugs of ICIs, and combinations of other therapies due to the lack of original data and sufficient information reported in the included studies.

Conclusion
Overall, LIPI could serve as a novel and reliable prognostic factor in NSCLC treated with ICIs, and intermediate LIPIs predict a worse prognosis.However, further high-quality studies are required to verify our findings.

FIGURE 1
FIGURE 1Prisma flow diagram of this meta-analysis.

FIGURE 2
FIGURE 2Subgroup analysis based on study design for the association between LIPI and overall survival (A) and progression-free survival (B) of non-small cell lung cancer patients receiving immune checkpoint inhibitors.

FIGURE 5
FIGURE 5Sensitivity analysis about the association between LIPI and overall survival (A) and progression-free survival (B) of non-small cell lung cancer patients receiving immune checkpoint inhibitors.

TABLE 1
Basic characteristics of included studies.

TABLE 1 (
Continued) Basic characteristics of included studies.

TABLE 1 (
Continued) Basic characteristics of included studies.

TABLE 2
Results of meta-analysis.