Lung cancer is a malignant tumor with the highest morbidity and mortality among all types of cancers globally (1). About 85% of patients with lung cancer were diagnosed with non-small cell lung cancer (NSCLC), and more than half of these patients were already in the advanced stage of cancer at the time of diagnosis. Drug therapy including chemotherapy, targeted drugs, and anti-angiogenic drugs can improve the prognosis of patients with NSCLC, but achieving long-term benefits is a challenge (2). The emergence of immune checkpoint inhibitors (ICIs) has been a huge breakthrough in the treatment of NSCLC, including programmed cell death protein 1 (PD-1) and programmed death-ligand 1 (PD-L1) inhibitors. These immune therapies significantly prolong the survival time of patients and have become one of the most essential treatments for advanced NSCLC. However, only a proportion of patients can benefit from ICIs, and others do not respond to immunotherapy, which limits their clinical application. Therefore, it is necessary to screen patients sensitive to immunotherapy in order to carry out the application of ICIs efficiently. PD-L1 expression and tumor mutational burden (TMB) are two popular biomarkers for predicting the efficacy of ICIs. PD-L1 is expressed in tumor cells and the surrounding microenvironment. When PD-1 is expressed in effector T cells binding to PD-L1, the activity of T cells can be suppressed, leading to the escape of tumor cells. Therefore, as the expression of PD-L1 is higher in tumor cells, the immunosuppressive effect of the PD-1/PD-L1 pathway becomes more active and treatment with ICIs becomes more effective. A higher TMB value means more mutated genes. Somatic mutations can lead to the expression of neoantigens, which can be recognized by T cells and can enhance the activity of ICIs. A lot of studies have shown that, in NSCLC, a high expression of PD-L1 is related to the efficacy of PD-1 inhibitors (3, 4). However, patients with negative PD-L1 expression can also benefit from ICIs (5–8). TMB has been recommended as a marker for NSCLC immunotherapy in the 2019 National Comprehensive Cancer Network (NCCN) guidelines, but there is no criterion to define a high level of TMB (9). Tissue biopsy needs to be conducted for the detection of PD-L1 and TMB, which requires high-quality specimens and is of high cost, but there is still a lack of uniform standard for detection platforms and measurement methods. In contrast, the acquisition of peripheral blood biomarkers is less invasive, low cost, and is easy to perform.

Inflammation is an essential component of the tumor microenvironment, promoting tumor proliferation, angiogenesis, metastasis, and destruction of adaptive immunity (10). The level of inflammation is related to the prognosis of malignant tumors. Tissue homeostasis can be perturbed by chronic inflammation, including local and systemic inflammation. Local inflammation includes tumor necrosis, inflammasomes, cytokines, chemokines, transcription factors, and tumor immune cell infiltrates, while systemic inflammation consists of circulating cytokines, small inflammatory proteins, circulating immune cells, and acute-phase proteins. The inflammatory factors coordinate the intracellular communication in the tumor microenvironment, and inflammatory conditions can disrupt the regulation of the innate and adaptive immune systems, leading to excessive tissue remodeling, loss of tissue architecture, protein alterations, and genotoxic DNA damage. These factors subsequently increase the risk of cancer and are crucial to distant tumor metastasis (11, 12). Inflammation-related index in serum, such as neutrophil count/lymphocyte count (NLR), lymphocyte count/monocyte count (LMR), and platelet count/lymphocyte count (PLR), can reflect the level of systemic inflammation. As opposed to a single peripheral blood index, the current trend is to combine multiple indexes to predict prognosis. The Lung Immune Prognostic Index (LIPI) is a predictive score that combines the lactate dehydrogenase (LDH) level and the derived neutrophil-to-lymphocyte ratio (dNLR) (13). Higher activity and expression level of LDH is closely related to the rapid proliferation of cancer cells through anaerobic glycolysis (14). The association between high pretreatment dNLR and poor prognosis was first reported in advanced melanoma patients treated with ICIs (15, 16), followed by studies on the prognostic value of dNLR for patients with NSCLC treated with immunotherapy. However, the association between dNLR and NSCLC prognosis was not significant in all related studies (17), and the role of dNLR in advanced NSCLC immunotherapy remains controversial.

In this study, we explore the potential predictive ability of LDH and dNLR and provide evidence for the prediction of LIPI. Therefore, we conducted this meta-analysis to examine the prognostic value of pretreatment serum dNLR and LDH levels in patients with advanced NSCLC receiving immunotherapy.

More and more studies have shown that systemic inflammation plays a vital role in the occurrence, development, and metastasis of tumors (46). LDH is a classic inflammatory index. The Warburg effect is one of the crucial characteristics of tumor cells: glycolysis occurs in aerobic conditions, and LDH catalyzes pyruvate to produce a large amount of lactic acid, which provides energy for tumor growth (47). The low pH and hypoxic microenvironment promote the proliferation of immunosuppressive cells, and LDH is related to the activation of the vascular endothelial growth factor (VEGF) pathway and the stimulation of tumor angiogenesis (48). Various pieces of evidence have indicated the promotional role of LDH in tumor growth. Neutrophils remodel the extracellular matrix by releasing specific proteases and then migrate into the remodeled extracellular matrix and tissues to form a new vascular supply, thereby promoting the migration of tumor-associated endothelial cells in new blood vessels (49). Lymphocytes, such as tumor-infiltrating lymphocytes (TILs), play an essential role in antitumor immunity and also function as a good prognostic indicator in many cancer types (50). A meta-analysis reported by Zhang et al. evaluated the correlation between NLR and the prognosis of patients with advanced NSCLC treated with immunotherapy (51). Both formulas for NLR and dNLR used the neutrophil count value as the numerator, but their denominators differed. The formula for NLR used the lymphocyte count value as the denominator. However, in terms of dNLR, the value for the white blood cell count minus the neutrophil count in the denominator was similar to that of the lymphocyte count. There may be a decrease in the lymphocyte count and an increase in the monocyte count in patients with cancer. NLR was unable to take the effect of monocytes into account. However, in the formula for dNLR, the denominator mixed two cell types, namely, lymphocytes and monocytes, with possible contradictory effects on the predictive value (18).

This meta-analysis included 26 studies involving a total of 3,429 patients with advanced lung cancer treated with ICIs. A systematic analysis was conducted to reveal that elevated pretreatment dNLR is associated with poorer PFS and OS. Considering that the first PD-1 inhibitors went on the market in China in 2018, many new studies have investigated the predictive role of LDH in recent years. This study also further verified that the high pretreatment LDH level is related to poor OS, which is consistent with the results of a previous study (52). The pooled results of the multivariate analysis showed that LDH is not an independent prognostic indicator of PFS, which is inconsistent with the results of a previous study (52). However, in the LDH subgroup analysis based on region, patient cohorts from Australia did not show a correlation with PFS or OS, which may be due to there being only one article included and its small sample size. Patient cohorts from Asia showed an association with OS and PFS, while the studies from Europe did not show any correlation to PFS, but were correlated with OS. The differences in the effect of LDH on the prognosis of immunotherapy among races need to be explored with more clinical research. There is still no uniform standard to define the cutoff value of LDH, and the definition of the cutoff in each study differed. Although LIPI clearly defines LDH ≥  ULN as one of the risk factors, the definition of ULN in each research center still differed (21). In this study, the values “3” and “240” were defined as the cutoff for dNLR and LDH, respectively, to perform subgroup analysis. The results showed that both dNLR ≥  3 and dNLR < 3 were associated with poor PFS and OS, and the risk of LDH ≥  240 U/L and LDH < 240 U/L was significantly higher in both PFS and OS. However, more high-quality research is still needed to determine the critical standard for the cutoff values. In this meta-analysis, only four studies involved patients receiving combination treatments. The subgroup analysis conducted based on treatment showed that baseline dNLR did not correlate with poor PFS and that baseline LDH was not associated with poor OS in studies on combination therapy. Current studies have mainly focused on the prognosis of immune monotherapy for NSCLC. However, there are multiple clinical trials of immune combination therapy (7, 53, 54), ICI combination chemotherapy, anti-angiogenic drugs, or double immunotherapy, which are the clinical practice trends. Two retrospective studies explored the predictive value of LIPI for patients treated with ICIs. Wang et al. hypothesized that LIPI is the only prognostic indicator of ICI monotherapy, which was shown to have no correlation with the prognosis of ICIs combined with chemotherapy (55). In contrast, Blanc-Durand et al. found that pretreatment LIPI correlated with ICI survival in monotherapy and in combination with chemotherapy (56). The predictive value of peripheral blood inflammation indicators for immune combination therapy is still unclear.

Our study had some limitations. Firstly, all included articles were retrospective studies with selection bias, and the sample size of each study comparatively differed. Secondly, the combined results for LDH were highly heterogeneous and showed significant publication bias. Thirdly, some of the ICIs were approved only for patients with a positive PD-L1 expression, such as pembrolizumab, atezolizumab, and cemiplimab. In these cases, detection of the PD-L1 tumor proportion score (TPS) is unavoidable, and PD-L1 tumor staining combined with the blood values may underline the essential role of tumor inflammation and its microenvironment. In a multicenter retrospective exploratory analysis, patients with NLR <5 + PD-L1 >80% obtained better 2-year OS and PFS compared with patients with NLR >5 (OS: HR = 0.20, p = 0.006; PFS: HR = 0.44, p = 0.03). This result revealed that the TPS of PD-L1 combined with the blood values showed a stronger prognostic role (57). Fourthly, our meta-analysis did not include studies conducted in the United States and Africa, which leads to the lack of comprehensiveness. We did not find studies published in these countries that met our study inclusion criteria, which may be caused by search bias. Finally, this study only considered the impact of the levels of dNLR and LDH before treatment on the prognosis and their dynamic changes were not taken into consideration. A retrospective analysis investigated the early discrepancies in the NLR and dNLR in patients with NSCLC treated with ICI therapy. The results showed the PFS to be significantly shorter in the group with increased NLR than in the non-increased NLR group (2.6 vs. 9.5 months, p < 0.001), and the same result was found in the dNLR group (4.2 vs. 9.2 months, p = 0.001) (58). Another team explored changes in the Gustave Roussy immune score (GRIm-score) in 45 days, which comprised the NLR, LDH, and serum albumin concentration. Their results suggested that a low GRIm-score in the 45-day group achieved significantly longer PFS and OS than the high-score group. Patients with stable/positive GRIm-score changes had better OS, PFS, and overall response rate (ORR) than did patients with negative changes in their GRIm-score (59). These studies indicated that early changes in the inflammatory blood values might be more reliable predictors of prognosis than pretreatment blood biomarkers. It could be the basis for treatment modifications by monitoring dynamic changes in inflammatory blood markers.

In summary, this study demonstrated that a high baseline LDH is significantly associated with poor OS and that a high baseline dNLR is significantly associated with both poor OS and PFS in patients with NSCLC treated with ICIs. Baseline LDH and dNLR values could be convenient and affordable predictive biomarkers, which can select and stratify patients who could benefit from ICIs, promoting the development of individualized therapy. More well-designed and large-scale studies are still needed to confirm these results.

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding authors.

YZ and LM had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. QZ, YZ, and LM: Concept and design. All authors: Acquisition, analysis, or interpretation of data. QZ, YZ, and LM: Drafting of the manuscript. QZ, XG, LS, and LM: Critical revision of the manuscript for important intellectual content. QZ: Statistical analysis. YZ and LM: Administrative, technical, or material support and study supervision. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.