Prognostic and clinicopathological significance of Systemic Immune-Inflammation Index in cancer patients receiving immune checkpoint inhibitors: a meta-analysis

Abstract Background Among malignant neoplasm patients taking immune checkpoint inhibitors (ICIs), it remains unknown how the systemic immune-inflammation index (SII) affects their clinical prognosis. We therefore performed the present meta-analysis by collecting the most recent data, so that SII’s prognostic value among ICI-receiving carcinoma patients could be fully clarified. Methods For the prognostic significance evaluation of SII in ICI-receiving carcinoma patients, the combined hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated. Results The number of studies enrolled in the present meta-analysis totaled 17, where 1,990 patients were involved. Among the ICI-treated carcinoma patients, a high SII was linked significantly to inferior overall survival (OS) (HR = 2.62, 95% CI = 1.76–3.90), as well as progression-free survival (PFS) (HR = 2.09, 95% CI = 1.48–2.95) (p both <.001). Contrastively, SII was linked insignificantly to the age (OR = 1.08, 95% CI = 0.39–2.98, p = .881), gender (OR = 1.01, 95% CI = 0.59–1.73, p = .959), lymph node (LN) metastasis (OR = 1.41, 95% CI = 0.92–2.17, p = .117), or metastatic site quantity (OR = 1.49, 95% CI = 0.90–2.46, p = .119). Conclusion There are prominent associations of elevated SII with the poor survival outcomes (both short- and long-terms) among the ICIreceiving carcinoma patients. SII has potential as a reliable and cheap prognostic biomarker in the clinic for carcinoma patients receiving ICIs.


Introduction
Being the leading cause of death, cancer crucially impedes prolongation of the anticipated life span on a global scale [1].According to the GLOBOCAN 2020 estimates in 185 countries around the world, there were 19,292,789 new cancer cases, as well as 9,958,133 cases died due to cancer in 2020 [1].Regarding treatment for patients with cancer, surgery, chemotherapy, and radiotherapy are mainstay in clinical practice.In recent years, immune checkpoint inhibitors (ICIs) have been more and more commonly applied and have dramatically improved survival in various malignancies [2].ICIs contain therapeutic antibodies that disrupt negative immune regulatory checkpoints and therefore enhance antitumour immune responses [3].ICIs such antibodies targeting programmed cell death 1 (PD-1), cytotoxic T lymphocyte-associated protein 4 (CTLA-4) and PD1 ligand 1 (PD-L1) have showed promising therapeutic effect and have been applied for many cancer types in the clinic [4,5].Novel and reliable prognostic biomarker identification is crucial to attaining individualized therapy among carcinoma patients receiving ICIs in clinical settings [6].

Study guideline
The present meta-analysis was carried out as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [31].

Ethics statement
Since all the data in the present meta-analysis were retrieved from public databases, neither the institutional review board nor the ethics committee was required.

Search strategy
Literature was retrieved in a systematic and holistic manner by utilizing online databases like Web of Science, PubMed, Cochrane Library and Embase between the dates of inception and 1 January 2023.We registered this meta-analysis protocol in INPLASY.The registration number is INPLASY202310018.The search terms shown below were used: (SII or Systemic Immune-Inflammation Index) and (immune check point inhibitor or PD-L1 or PD-1 or avelumab or CTLA-4 or durvalumab or nivolumab or ipilimumab or atezolizumab or pembrolizumab or camrelizumab or tislelizumab or toripalimab or penpulimab or immune checkpoint blockade or immunotherapy).Only the articles published in English were retrieved.Possible inclusions were recognized also by prudently checking the references in qualified researches, as well as relevant reviews.

Inclusion and exclusion criteria
The inclusion criteria were as follows: (1) patients were pathologically diagnosed with cancer and treated with ICIs; (2) available data of pretreatment SII; (3) studies investigating the relationship between SII and prognosis of patients with cancer undergoing ICIs; (4) a cutoff value was identified to stratify patients as high/low SII groups; (5) any survival outcomes were reported including overall survival (OS), progression-free survival (PFS), recurrence-free survival (RFS) etc., (6) studies published in English language.The exclusion criteria were: (1) reviews, letters, meeting abstracts, and case reports; (2) studies did not provide survival data; (3) duplicate studies and (4) non-human studies.

Data extraction and quality assessment
Relevant researches were collected by two independent researchers (Y.W. and Q.N.) through retrieval and scanning of identified titles and abstracts.All disagreements were addressed through negotiation until reaching consensus.Information extracted from qualified researches included: first author's name, country, sample size, publication year, study design, age, gender, cancer type, study period, ICIs type, tumour-node-metastasis (TNM) stage, study centre, survival outcomes, cut-off value of SII, follow-up, hazard ratio (HR), survival assessment, as well as 95% confidence interval (CI).Pooling of HRs and 95% CIs was accomplished from multivariate assessments if provided.Otherwise, univariate assessments were the source of HRs and 95% CIs.The enrolled studies were subjected to quality evaluation via the Newcastle-Ottawa Scale (NOS) [32].This scale assesses the research quality regarding selection, comparability and prognosis.Studies scoring !6 on this nine-point NOS are considered to have high quality.

Statistical analysis
SII's prognostic significance among the ICI-receiving carcinoma patients was evaluated by estimating the combined HRs and 95% CIs.Cochran's Q test was employed to assess the inter-study heterogeneity by utilizing I 2 statistics.In the case of insignificant heterogeneity (p > .05for v 2 test or I 2 < 50%), a fixed-effect model was adopted.Otherwise, a random-effects model was selected.Heterogeneity source was identified by conducting subgroup analysis stratified by diverse variables.How SII was associated with the clinicopathological traits of cancer patients using ICIs was assessed by pooling odds ratios (ORs) and 95%CIs.Possible publication bias was evaluated through Begg's test.Stata 12.0 (College Station, TX) was utilized to analyse the entire data.p < .05 was regarded as statistically significant.

Association between SII and clinicopathological features
Four studies including 636 patients [17,19,26,28] reported the associations of SII with the clinicopathological traits for carcinoma patients receiving ICIs.As shown in

Publication bias
Possible publication bias was assessed by employing Begg's test combined with funnel plot test.As shown in Figure 4, the funnel plots for OS and PFS were not evidently asymmetric in shape.With the Begg's test, the p values for OS and PFS were .583and .956,respectively.Thus, insignificant bias was present in our meta-analysis.

Discussion
The previous works have shown that SII was prognostically significant among carcinoma patients receiving ICIs, whereas the results were inconsistent.In the present meta-analysis, data were extracted from 17 studies involving 1990 patients for quantitatively analysing the prognostic role of SII for OS and PFS.Our results demonstrated that elevated SII was a significant biomarker of outcome for both OS and PFS among the ICI-receiving carcinoma patients.Besides, SII exhibited a reliable prognostic efficiency, which was unaffected in diverse subgroups.However, SII was found associated significantly with age, gender, number of metastasis sites, or LN metastasis in our meta-analysis.Assessment of publication bias revealed the reliability of the results.In summary, our meta-analysis implied the significance and validity of SII as a biomarker of outcome for the ICI-treated carcinoma patients.Those cancer patients treated with ICIs who have high pretreatment SII may experience high risk of disease progression and should be cautiously managed.
The SII was derived as the formula: SII ¼ (platelet quantity Ã neutrophil quantity)/lymphocyte quantity.Accordingly, a high SII would be indicative of high platelet and neutrophil quantities, and/or low quantity of lymphocytes.For the ICI-treated carcinoma patients, it is possible to explain the mechanisms of high SII with inferior survival prognosis in the following aspects: first, neutrophils can promote cancer development through directly interacting with tumour cells or indirectly remodelling the tumour microenvironment [33].Growth and metastasis of tumours can also be promoted by neutrophils through the proinflammatory cytokine and chemokine discharge in tumour microenvironment [34].Second, platelet quantity is an extra indicator for a systemic inflammatory reaction and possible microvascular thrombosis.Platelets also enables the angiogenesis facilitation via the cytokine VEGF, thereby promoting the growth of tumours [35].Upon activation, the platelets also probably regulate the tumour site migration of immunocytes and haematopoietic cells, and thus facilitates the cancer-related inflammation [36].Thirdly, as a major cellular immunity component in humans, lymphocytes are implicated in anti-cancer immunoresponses [37].T lymphocytes, in particular, function pivotally in the tumour cell identification and killing, thereby suppressing the multiplication and motility of these cells [38].Therefore, the reduction of lymphocytes can result in impairment of anti-tumour activity of the host.
The subgroup analysis was performed to examine the impact of treatment, cancer type, cut-off value, and study design on the results of this meta-analysis in Tables 3 and 4. The subgroup analysis indicated that sample size, study centre, country, study centre did not significantly affect the prognostic role of SII for both OS and PFS for cancer patients receiving ICIs.Additionally, SII showed obvious prognostic effect on OS and PFS for patients treated with Anti-PD-1, Anti-PD-1/PD-L1 þ Anti-CTLA-4, and Anti-PD-L1 (Tables 3 and 4).Inherent heterogeneity may exist in included studies; however, subgroup analysis and publication bias test confirmed the reliability of our meta-analysis.
ICIs are applied in multiple cancer types and become standard treatment strategy for some cancers.The use of PD-1 inhibitors as adjuvant therapy in high-risk resected stage III or IV melanoma has become standard practice [39].In the treatment of advanced gastric cancers, ICIs including nivolumab and pembrolizumab have emerged as a promising treatment option [40].Different ICIs have been successfully introduced into clinical medicine for lung cancer treatment since 2014, including pembrolizumab, atezolizumab, and durvalumab [41].Immune checkpoint inhibitors (ICIs) have been approved as first-line therapy for patients ineligible for cisplatin or as second-line therapy for patients with metastatic bladder cancer [42].
Recently, a series of studies also investigated SII's prognostic value in diverse solid tumours through meta-analysis [14,[43][44][45].According to a meta-analysis involving 3515 patients by Zhang and colleagues, an elevated pre-therapy SII was linked significantly to worse RFS/PFS and OS among BTC patients [14].Li et al. found significant associations of high SII with inferior PFS, cancer-specific survival (CSS) and OS among oesophageal squamous cell carcinoma (ESCC) patients receiving operation in a meta-analysis including nine retrospective studies [44].In a meta-analysis on 10 studies enrolling 7087 patients by J. Li et al.SII displayed association with poor OS and RFS among the bladder cancer patients [45].M. Li et al. demonstrated the connection of elevated pre-therapy SII to inferior CSS/disease-free survival (DFS)/PFS and poor OS for pancreatic cancer in their meta-analysis comprising 2132 patients [46].In addition, a meta-analysis containing 12 studies identified the prognostic role of high SII for worse OS and PFS in patients with colorectal cancer [47].Another recent meta-analysis with 2365 subjects reported the association between SII and short-term and long-term survival outcomes in patients with pancreatic cancer [48].Our current meta-analysis demonstrated the significant prognostic value of SII in cancer patients treated with ICIs, which were in accordance with observations in other cancer.The present meta-analysis has a few shortcomings.First of all, only those articles published in English were considered.Since the enrolled patients were  predominantly Asians and partially Caucasians, subjects other than Caucasian and Asian ethnicities were assessed inadequately.Second shortcoming was the rather small sample size.Despite the inclusion of 17 studies, only 1990 patients were enrolled.The small sample size may also lead to the negative results regarding SII's correlations with the clinicopathological parameters.Thirdly, the uniformed SII thresholds across the enrolled studies is a probable contributor to the meta-analytical heterogeneity.Hence, verifying our results by large-scale trials is necessary, where a unified SII threshold should be utilized.In summary, elevated SII was linked significantly to the inferior survival prognosis (both short-and longterms) among the ICI-receiving carcinoma patients.For these patients, SII may serve as a reliable and cheap prognostic biomarker in the clinical settings.

Figure 1 .
Figure 1.Flowchart representation of the steps of the literature search and selection in this meta-analysis.

Figure 2 .
Figure 2. Forest plot of HR for the relationship between SII and overall survival in cancer patients receiving ICIs.

Figure 3 .
Figure 3. Forest plot of HR for the association between SII and progression-free survival in cancer patients receiving ICIs.

Table 1 .
Basic characteristics of included studies in the current meta-analysis.

Table 2 .
The details of NOS scale for included studies in this meta-analysis.

Table 3 .
The prognostic value of SII for OS in cancer patients treated with ICIs.

Table 4 .
The prognostic value of SII for PFS in cancer patients treated with ICIs.

Table 5 .
The correlation between SII and clinicopathological variables in cancer patients undergoing ICIs treatment.