Prognostic significance of pretreatment systemic immune-inflammation index in patients with prostate cancer: a meta-analysis

Background The SII (systemic immune-inflammation index) has been extensively reported to have a prognostic value in prostate cancer (PCa), despite the unconformable results. The purpose of this meta-analysis is to quantify the effect of pretreatment SII on survival outcomes in patients with PCa. Methods The following databases were searched: Web of Science, Cochrane Library, PubMed, Embase, and China National Knowledge Infrastructure (CNKI). For exploration of the SII’s correlations with the overall survival (OS) and the progression-free survival/biochemical recurrence-free survival (PFS/bRFS) in PCa, the pooled hazard ratios (HRs) were assessed within 95% confidence intervals (CIs). Results The present meta-analysis covered 10 studies with 8133 patients. Among the PCa population, a high SII was linked significantly to poor OS (HR = 2.63, 95% CI = 1.87–3.70, p < 0.001), and worse PFS/bRFS (HR = 2.49, 95% CI = 1.30–4.77, p = 0.006). However, a high SII was not linked significantly to T stage (OR = 1.69, 95% CI = 0.86–3.33, p = 0.128), the metastasis to lymph node (OR = 1.69, 95% CI = 0.69–4.16, p = 0.251), age (OR = 1.41, 95% CI = 0.88–2.23, p = 0.150), or the Gleason score (OR = 1.32, 95% CI = 0.88–1.96, p = 0.178). Conclusions For the PCa sufferers, the SII might be a promising prognostic biomarker, which is applicable to the high-risk subgroup identification, and provide personalized therapeutic strategies. Supplementary Information The online version contains supplementary material available at 10.1186/s12957-022-02878-7.

antigen (PSMA) expression in circulating tumor cells (CTC)+ had no significant impact on PSA drop and survival in mCRPC patients [7]. The PRIMERA trial validated the predictive importance of CTC detection in mCRPC patients as a result [7]. Hence, identification of novel biomarkers and treatment targets is imperative, in order to enhance the prognosis for patients with PCa. As indicated by growing evidence, the systemic inflammatory reactions are significant determinants of cancer development and survival outcomes in various cancer types [8]. Many serum inflammatory parameters, including the ratios of neutrophils/lymphocytes [9], platelets/ lymphocytes [10], C-reactive proteins/albumin [11], and the SII (systemic immune-inflammation index) [12], have been reported to be effective prognostic markers in different cancer types. The SII was calculated as follows: platelet quantity × neutrophil quantity/lymphocyte quantity. Its role as a prominent prognostic biomarker has been demonstrated in numerous types of carcinomas, such as the hepatocellular [13], pancreatic [14], breast [15], and non-small cell lung [16] carcinomas. Despite the prior explorations on SII's prognostic significance among the PCa population, unconformable results have been yielded [17][18][19][20][21][22][23][24][25][26]. High SII in PCa has been considered a valid prognostic biomarker for the poor outcome by several researchers [19,21,24], whereas others have denied this association [20]. Hence, the objective of the present meta-analysis is to evaluate SII's prognostic value in PCa based on current evidence.

Study guideline and ethics approval
The present meta-analysis was carried out as per the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [27]. The PRISMA checklist was shown in supplemental file 1. This meta-analysis has been registered with INPLASY (Registration No. INPLASY2022110155) and is available from https:// inpla sy. com/ inpla sy-2022-11-0155/. Being a literature-based study in nature, ethical approval was unnecessary; moreover, no data containing individual patient information were used.

Data sources and literature search
The following electronic databases were searched thoroughly: Web of Science, Cochrane Library, PubMed, Embase, and China National Knowledge Infrastructure (CNKI). The search duration was from the databases' inception to November 27, 2022. The search heading terms and keywords included "systemic immune-inflammation index, " "SII, " "prostate cancer, " "PCa, " "prostate carcinoma, " and "prostate neoplasm. " All searches used both subject headings of Medical Subject Headings (MeSH) terms and free test words. The detailed literature strategies for each database are shown in supplemental file 2. No language restrictions were applied. The reference lists were manually searched to find eligibility records.

Selection criteria
The inclusion criteria were identified according to the PICOS (population, intervention, comparator, outcomes, and study) criteria. The inclusion criteria were formulated as shown below: (i) P (population): Patients whose PCa was confirmed pathologically (ii) I (intervention): The SII level was examined pretreatment for PCa patients, and studies identified a cutoff value of SII for stratifying patients as low/ high SII. (iii) C (comparator): PCa patients with low SII level (iv) O (outcomes): Studies reported association between SII and PCa survival outcomes; presented any of such survival outcomes as bRFS (biochemical recurrencefree survival), RFS (recurrence-free survival), DFS (disease-free survival), PFS (progression-free survival), and OS (overall survival); and provided HRs (hazard ratios) and corresponding 95% CIs (confidence intervals) for survival outcomes or provided sufficient data to calculate them. (v) S (study design): Cohort studies, including prospective and retrospective cohorts published in English or Chinese.
Studies were excluded when any of the following criteria was satisfied: (i) case reports, meeting abstracts, reviews or letters, (ii) studies with overlapping patients, (iii) studies with inadequate data for making HR and 95% CI estimations, and (iv) nonhuman studies.

Data extraction and quality assessment
All of the retrieved studies were assessed by 2 independent investigators (B. Z. and T. X.), who were also responsible for extracting information based on a designated form. Disputes were all resolved through negotiation until a consensus was reached. Information extracted included the name of first author, country, age, year of publication, sample size, research design, study duration, metastatic status of disease, therapeutic management, follow-up, SII cutoff, method for cutoff selection, quantity of patients having low/high SII, models of survival analysis, survival endpoints, study center, and HRs with 95% CIs. The methodological quality of enrolled studies was assessed by two reviewers (B. Z. and T. X.) independently on the NOS (Newcastle-Ottawa scale) [28], which achieves quality evaluation from 3 dimensions: selection, comparability, and outcome of interest. The NOS scores varied between 1 and 9 points, and the quality of studies was considered high when the NOS scores ≥ 6.

Statistical analysis
SII's prognostic significance for OS and PFS was assessed by estimating the pooled HRs with 95% CIs. For the evaluation of inter-study statistical heterogeneity, the χ 2 -based Q-test combined with Higgins' I 2 test was employed. Inter-study heterogeneity was considered significant when the p-value of Q-test (Ph) < 0.10 and I 2 > 50%; accordingly, we adopted the random-effects model. In other cases, a fixed-effects model was utilized. Further exploration was made via the subgroup analysis. The association of SII with the clinicopathological traits of PCa was examined through computation of ORs (odds ratios) and 95% CIs.

Prognostic of SII for OS in PCa
Eight studies with 1803 patients [17][18][19][20][21][22][23]26] provided HR and 95% CI statistics concerning SII for OS. Since the heterogeneity was significant (I 2 = 73.1%, p <0.001), we adopted the random-effects model. As is clear from the pooled results in Fig. 2 and Table 2, high SII was linked significantly to the poor OS (HR = 2.63, 95% CI = 1.87-3.70, p < 0.001). According to the subgroup analysis results, a SII elevation remained a prominent prognostic biomarker for OS, irrespective of research design, and investigated center, region, sample size, cutoff, metastatic state, therapeutic management, or type of survival analysis ( Table 2).
As demonstrated by the results in Fig. 4 and Table 4

Sensitivity analysis
To test the stability of this meta-analysis, sensitivity analysis was conducted through removing each study in turn to recalculate the combined data. The pooled HR estimates were not significantly changed, indicating that the results were stable (Fig. 5).

Discussion
In former researches, the potential of SII as a prognostic biomarker has been explored for the PCa population [17][18][19][20][21][22][23][24][25][26], despite the unconformable results. In our current meta-analysis, the exact prognostic role of SII in PCa was clarified by pooling the data from 10 studies involving 8133 patients. According to our findings, a high SII represented an independent prognostic biomarker for PFS/bRFS and OS among the PCa patients. Besides, SII exhibited a reliable prognostic power across varying subgroups. However, elevated SII also not correlated with LN metastasis, T stage, Gleason score, or age in PCa. Based on the evidence obtained from this meta-analysis, we recommend the application of SII as a cost-efficient new biomarker for guiding the management and followup for the PCa population. Increasing evidence has suggested the tight linkage of immunoreactions to the cancer occurrence, development, and metastasis [29,30]. Tumor-derived proinflammatory cytokines like IL (interleukin)-6, IL-8, VEGF, tumor necrosis factor-α, and interferon-γ can be secreted into the tumor microenvironment, leading to chronic inflammation, thus facilitating tumor progression [31]. The SII is a combination of neutrophil, lymphocyte, and platelet counts; thus, it can be elevated in the event of high quantities of neutrophils, platelets, and/or a low quantity of lymphocytes. Current evidence indicates that a microenvironment is offered by the tumor-infiltrating neutrophils, which facilitates the tumor cell growth; they also promote angiogenesis and cell mobility [32]. By constraining the cytolytic potential of various immunocytes, high neutrophil counts inhibit the immunity system [29]. Platelets can release various matrix metalloproteinases to facilitate the degradation of the extracellular matrix, thereby promoting metastasis of cancer cells [33]. In addition, through the proangiogenic cytokine discharge inside the microvasculature of cancer cells, the platelet aggregation is capable of facilitating the tumor growth [34]. In contrast, for the tumor development suppression, lymphocytes are involved critically in the immunosurveillance for cancer [35]. Lymphocytes, including subsets such as CD8+ and CD3+ T cells, correlated with good prognosis in various cancers [36]. Therefore, a high SII is a promising indicator of a combination of neutrophils, lymphocytes, and platelets. The exact prognostic significance of PSA (prostatespecific antigen), the most extensively applied PCa biomarker [37], in PCa has been reported in several important latest studies [38][39][40]. A study on 148 patients showed the validity of PSA as a biomarker for forecasting the PCa prognosis when its levels were from 20 to 70 ng/mL [38]. According to another study focusing on PCa sufferers having low levels of PSA, the tumor stage was more advanced when the PSA level at diagnosis was < 3.5 ng/ml compared to that between 3.5 and 10 ng/ml [39]. As suggested by a study enrolling 90 PCa patients whose PSA levels were > 100 ng/mL, the differences in OS or CSS were insignificant among the 3 PSA groups, namely the slightly high (100-200 ng/mL), moderately high (200-1000 ng/mL), and considerably high (> 1000 ng/mL) groups [40]. According to the results of our current meta-analysis, the prognosis of patients was poor when their SII was high. SII is an independent risk factor for PCa and could be a better screening method for PCa before biopsy [41]. Therefore, the SII could be used as a biomarker for improving the diagnostic and prognostic accuracy of PSA. Furthermore, a recent study suggested that the level of CCL2 (serum CC motif (chemokine) ligand 2) is a valid prognostic indicator for poor survival in PCa [42]. Patients with serum CCL2 levels ≥ 320 pg/ mL had relatively worse OS, CSS, and CRPC-free survival than those with CCL2 concentrations < 320 pg/mL [42]. The relationship between CCL2 levels and the SII needs to be investigated in future studies. Notably, this meta-analysis included four studies with the endpoint of PFS [19,20] and bRFS [24,25]. Two studies enrolled patients with mCRPC [19,20], whereas two studies recruited the localized PCa patients [24,25]. The definition of these endpoints in four studies is not all the same. In Fan's study [19] with mCRPC patients, PFS referred to an interval from the commencement of the first mCRPC therapy (i.e., docetaxel-prednisone or abiraterone) until the time of radiographic progression. In Kobayashi, H. 's work [20] with mCRPC patients, an interval from the disease diagnosis to progression was regarded as PFS. Besides, disease was considered progressed when the serum level of PSA was elevated by > 2 ng/mL, the rise over nadir was 50%, and/or a new lesion emerged, or the known lesions classified as per the RECIST (ver. 1.1)  increased by one or more [43]. In Rajwa' s study with localized PCa patients, bRFS was defined as the interval from radical prostatectomy to the first PSA rise of two consecutive PSA values > 0.2 ng/ml [24]. Although the definitions are not all the same, our results demonstrated that in PCa, a high SII was a prominent prognostic biomarker for poor PFS/bRFS (Table 3).
To probe deeper into SII's prognostic function among patients with different metastatic status, subgroup analysis was conducted. The results showed that a high SII was not associated with poor PFS/bRFS in localized PCa patients (p = 0.160), as well as not with worse PFS/ bRFS in mCRPC patients (p = 0.183) ( Table 3). Because of the relatively small size of samples, our results still need be verified through large-scale researches. SII's prominent prognostic value has been pinpointed for varying solid tumors by extensive recent meta-analyses [44][45][46][47]. As suggested by a meta-analysis involving 7 studies, an elevated pretreatment SII was linked to inferior carcinoma-specific survival/DFS/PFS and poor OS in pancreatic carcinoma patients [44]. According to another meta-analysis covering 4236 patients, a high  pretreatment SII forecasted poor OS in gastric carcinoma [48]. A high SII was also reported to be linked to the poor OS among the renal cell carcinoma sufferers [47]. Through a meta-analysis involving 2,796 patients, Wang et al. reported that elevated pretreatment SII was related to lower OS and earlier time-to-recurrence in hepatocellular carcinoma [49]. As indicated by a latest meta-analysis enrolling 12 studies, high levels of SII were correlated pronouncedly with worse PFS and OS among the colorectal cancer population [50]. Our present results on SII's prognostic role agree with those in other types of carcinomas.
Regarding several shortcomings of our meta-analysis, first of all, the optimal SII cut-off was not determined. The included studies used different cutoff thresholds, which might have contributed to the heterogeneity among studies. Second, majority of the enrolled studies were retrospective, while there was merely 1 enrolled prospective study. Thus, differences in unadjusted factors could lead to selection bias. Third, our meta-analysis included qualified published studies in English or Chinese only, while failing to enroll relevant articles in other languages, which is also likely to result in inherent heterogeneity.