Prognostic significance of C-reactive protein in patients with cervical cancer: a meta-analysis

Background Numerous studies have investigated the significance of pretreatment C-reactive protein (CRP) levels for determining the prognosis of cervical cancer (CC). The results of these studies, however, have been inconsistent. The present meta-analysis, therefore, focused on identifying the exact relationship of CRP levels with CC prognoses. Methods We searched the following databases from their inception until April 18, 2023: PubMed; Web of Science; Embase; and Cochrane Library. From the search results, we estimated the significance of CRP levels in determining the prognosis of CC, based on combined hazard ratios (HRs) and relevant 95% confidence intervals (CIs). Results The present meta-analysis included 12 studies, encompassing 2,204 patients. Based on combined data, an increased CRP level was significantly related to an unfavorable overall survival (OS) of patients with CC (HR = 1.63; 95% CI = 1.36–1.95; P < 0.001). Moreover, an increased CRP level was significantly associated with shortened progression-free survival (PFS) in patients with CC (HR = 1.68; 95% CI = 1.39–2.03; P < 0.001). According to the subgroup and sensitivity analyses, CRP level was a reliable factor in determining CC prognoses. Conclusion Based on the results of our present analyses, increased CRP levels were significant predictors of poor OS and PFS in patients with CC. CRP level, therefore, could be an independent and inexpensive factor for determining the prognosis of patients with CC in clinical settings. Systematic review registration INPLASY, identifier INPLASY202360074.


Introduction
Cervical cancer (CC) ranks the fourth most common cancer in women; besides, it is also the fourth frequent cause of cancer death among women globally (1).There were 604,127 new cases of CC and 6 341,831 deaths due to CC in 2020 worldwide (1,2).The primary etiological factor for CC is chronic human papillomavirus (HPV) infection (3,4), and surgical resection is the preferred treatment for patients diagnosed with International Federation of Gynecology and Obstetrics (FIGO) stages ≤ IIA, while chemoradiation is recommended for patients diagnosed with higher stages (3,5).It is estimated, however, that 20-25% of patients with CC will experience a recurrence after completing their primary treatment, and that locally advanced CC has a poor 5-year survival rate (6).To some extent, the lack of efficient prognostic markers may have a correlation to poor CC survival.Therefore, establishing reliable and easily available biomarkers for determining the prognosis of patients with CC would aid the development of treatment strategies.
A growing body of evidence suggests that inflammation is involved in the pathogenesis and development of solid tumors (7,8).When infection, tissue injury, trauma, neoplastic growth, or surgery interferes with the homeostasis of an organism, an acutephase response (APR) is immediately triggered (9).C-reactive protein (CRP), a sensitive and well-recognized systemic inflammatory marker produced by the liver in response to factors such as interleukin (IL)-6, IL-1, along with tumor necrosis factor-a (TNF-a) (10).CRP is an acute phase reactant that is widely considered to be a marker of both acute and chronic inflammation (11).CRP has several well-defined functions, including acting as a pattern recognition receptor with calciumsensitive binding pockets for ligands expressing phosphocholine (PC) moieties (12).CRP can also bind to activated cell membranes in which the PC groups on phospholipids become accessible when diacylphospholipids are hydrolyzed into monoacylphospholipids (12).In addition to being found in serum, CRP has recently been found to exist in an array of cyclic pentameric discs.The effects of CRP on the behavior of cells and the microenvironment are, therefore, dependent on its structural configuration.Pentameric CRP (pCRP) binds to PC on the surface of the cell, causing it to dissociate into its distinct monomeric isoform (mCRP), which subsequently reduces its aqueous solubility (13).It has been suggested that CRP levels are an effective factor in determining the prognosis of various cancers, including esophageal cancer (14), renal cell carcinoma (RCC) (15), prostate cancer (16), and gastric cancer (17).The relationship between CRP levels and survival outcomes in patients with CC has been extensively investigated; however, no consistent results have been reported thus far (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29).Elevated CRP levels have been reported to be a significant prognostic factor for CC in some studies (18,20,22), while others have demonstrated a non-significant association between CRP levels and CC prognosis (19,24).Given these inconsistent results, we aimed to comprehensively search relevant articles in the present meta-analysis to quantitatively identify the true relationship of CRP levels and CC prognoses.

Study guidelines
The present meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (30), and was registered with the International Platform of Registered Systematic Review and Metaanalysis Protocols (INPLASY) platform.The registration number of this meta-analysis on INPLASY is INPLASY202360074.

Ethnics statement
The present meta-analysis did not require the involvement of an institutional review board or ethics committee, as all of the data acquired were derived from previously published studies.

Literature search
In the present study, we thoroughly searched the PubMed, Web of Science, Embase, and Cochrane Library databases from their inception until April 18, 2023, for the following keywords/search terms: (C-reactive protein OR CRP OR c-reactive protein) AND (cervical cancer OR cervical carcinoma OR uterine cervix cancer OR cervical neoplasm OR cervix cancer).Only English studies were considered.After the initial search, the references from each of the included articles were manually searched to identify additional relevant articles.

Eligibility criteria
The inclusion criteria were as follows: (i) CC diagnosis confirmed based on pathology; (ii) studies explored the between serum CRP levels and any survival outcomes of patients with CC; (iii) available hazard ratios (HRs) and associated 95% confidence intervals (CIs) for prognoses from studies or calculable data based on the information provided in the articles; (iv) a cut-off value was defined with which to determine low/high CRP levels; and (v) studies were published in English.The exclusion criteria were as follows: (i) articles formatted as reviews, meeting abstracts, case reports, comments, or letters; (ii) patients with new infections, or chronic infection diseases, autoimmune diseases, organ dysfunctions, hematologic diseases, and patients with another type of tumor; (iii) articles involved animal studies; and (iv) duplicate articles.

Extraction of data and evaluation of study quality
Two researchers (SY and ZZ) independently screened all eligible studies and extracted the information required from each, which included the following: first author; publication year; country; sample size; study duration; age; FIGO stage; treatment; CRP level cut-off; cut-off determination method; survival endpoints; follow-up; survival analysis type; and HRs with 95% CIs.All disagreements were resolved by reaching a consensus after discussion with a third reviewer (LS).The overall survival (OS) was selected as the primary outcome, and progression-free survival (PFS) as the secondary outcome.Additionally, we utilized the Newcastle-Ottawa Scale (NOS) score to evaluate the quality of the included articles (31).The NOS evaluates studies based on 3 factors cohort selection; comparability; and results.A NOS score ≥ 6, based on a 0-9 point scale, suggests highquality research.

Statistical analysis
Pooled HRs and 95% CIs were calculated to estimate the significance of CRP levels in determining the prognosis for patients with CC.In general, a combined HR > 1 with a 95% CI not overlapping 1 indicated a significant association with poor prognosis, while a combined HR < 1 with a 95% CI not overlapping 1 indicated a better prognosis.Inter-study heterogeneity was evaluated using Cochran's Q-test (32) and I 2 statistics (33).I 2 was used to quantify the degree of heterogeneity among the studies, as follows: I 2 < 25%, low degree; 25-75%, moderate degree; and > 75%, high degree of heterogeneity (33,34).To analyze the pooled data, we used two different computational models, based on the traits of the included studies, and the cut-off for significant heterogeneity was set at I 2 > 50% (35-38).When high heterogeneity was determined based on I 2 > 50% and Q-test P < 0.10, and a random-effects model (REM; DerSimonian-Laird method) was used (39); otherwise, a fixed-effects model (FEM; Mantel-Haenszel method) was used (40).Subgroup analyses of OS and PFS were conducted to identify possible sources of heterogeneity, and we conducted a sensitivity analysis by removing one article at a time, in order, to evaluate the robustness of the combined results.Funnel plots and Begg's test (41) were utilized to evaluate publication bias.All statistical analyses were conducted using Stata software version 12.0 (Stata Corporation, College Station, TX, USA), with P < 0.05 indicating a significant difference.

Significance of CRP in predicting OS
A total of 11 articles, encompassing 2,089 patients (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28), presented data on the relationship between the CRP value and OS in patients with CC.REM was utilized in these articles because significant heterogeneity was detected (I 2 = 71.6%;P < 0.001).In Table 2 and Figure 2, as seen based on the pooled data, increased CRP expression was significantly tied to extremely poor OS in patients with CC (HR = 1.63; 95% CI = 1.36-1.95;P < 0.001).Subgroup analyses were conducted based on various factors.As seen in Table 2, subgroup analysis results indicated that elevated levels of serum CRP were still a significant marker for predicting OS in patients with CC regardless of sample size, country, FIGO stage, treatment, CRP cut-off, cut-off determination, or survival analysis type (all, P < 0.05).

Prognostic value of CRP for PFS
Of the 12 included articles, 6 studies, encompassing 1,415 patients (18,20,23,24,28,29), provided data on the role of CRP values in predicting the PFS of patients with CC.The FEM was utilized with these articles, due to their low level of heterogeneity (I 2 = 22.9%; P = 0.262; Figure 3).In line with our aforementioned analysis, however, increased CRP levels were significantly correlated with shortened PFS in patients with CC (HR = 1.68; 95% CI = 1.39-2.03;P < 0.001; Figure 3, Table 3).Similar to what we observed with the subgroup analysis for OS, an increased CRP level was significant in predicting PFS regardless of sample size, country, FIGO stage, treatment, CRP cut-off, cut-off determination, or survival analysis type (P < 0.05; Table 3).

Sensitivity analysis
The impact of each article on the combined results was analyzed by performing a sensitivity analysis, which revealed stable outcomes for both OS (Figure 4A) and PFS (Figure 4B).

Publication bias
Begg's test and funnel plots were used to assess publication bias.As shown in Figure 5, symmetry was observed in the funnel plots.For Begg's test, the resulting P-value for OS was 0.082, while that for PFS was 0.452, indicating the absence of significant publication bias in the present study.

Discussion
Although previous studies have evaluated the significance of CRP levels in determining the prognoses of patients with CC, the results were inconsistent.The present meta-analysis, which included 12 articles encompassing 2,204 cases, aimed to definitively identify the relationship of CRP levels with CC prognoses.Based on the findings of the present meta-analysis, increased serum CRP levels are significantly predictive of OS and PFS in patients with CC, and as demonstrated by the subgroup and sensitivity analyses, this relationship was very reliable.Taken together, these results suggest that serum CRP level may be a useful prognostic biomarker for patients with CC.To the best of our knowledge, this is the first meta-analysis to explore the prognostic significance of serum CRP levels in patients with CC.
CRP is a well-established biomarker for inflammation and can be easily assayed in clinical practice (42).Additionally, as  an acute-phase protein generated by the liver, CRP is indicative of whether inflammation exists, as well as its level in the body, while maintaining certain advantages, such as a stable half-life, cost-effectiveness, simple measurement, and standardization (43).There are multiple mechanisms underlying the prognostic value of CRP in patients with CC.First, the presence of tumor tissue causes inflammation, which, in turn, increases serum CRP levels (44); therefore, increased CRP levels indicate tumor necrosis or local tissue injury.Second, IL-6 can induce hepatocytes to produce CRP, as can other factors, such as IL-1, TNF-a, and transforming growth factor b (TGFb) (45), while via vascular endothelial growth factor (VEGF), IL-6 may promote angiogenesis in the presence of inflammation (46).Third, it has been shown that CRP induces an inflammatory tumor microenvironment that activates certain signaling pathways important for tumor proliferation, angiogenesis, and metastasis (47).It is likely, therefore, that patients with CC and a systemic inflammatory response, as indicated by elevated serum CRP levels, will be given a poor prognosis.The present meta-analysis has several strengths.To the best of the authors' knowledge, this is the first meta-analysis to explore the relationship between CRP levels and survival outcomes in patients with CC.Second, all included studies were of high quality (NOS score ≥ 6), which contributes to the overall statistical power of this meta-analysis.Third, sensitivity analyses and publication bias tests confirmed the reliability of the results of the present meta-analysis.
Numerous recent meta-analyses have reported the use of CRP in determining the prognoses of a variety of solid tumors.Forest plots of the prognostic role of CRP for OS in patients with CC.
Forest plots of the prognostic role of CRP for PFS in patients with CC.The present meta-analysis does have some limitations.First, the included studies were all of a retrospective nature, which may have introduced selection bias.Second, the cut-off values of CRP levels ranged from 2.6-10, but we were unable to determine an optimal cut-off value for determining the prognosis of CC based on CRP levels.Third, our sample size was relatively small, as many of the included studies had relatively small sample sizes.Therefore, larger prospective studies are required to verify our findings.

Conclusions
In conclusion, the results of the present meta-analysis indicated that increased CRP levels are markedly associated with poor OS and PFS in patients with CC.Therefore, serum CRP level could be an inexpensive and independent factor for determining the prognoses of patients with CC in a clinical setting.

FIGURE 1 PRISMA
FIGURE 1 PRISMA flowchart of the established screening strategy.

TABLE 1
Basic characteristics of included studies in this meta-analysis.

TABLE 2
Subgroup analysis of prognostic role of CRP for OS in patients with CC.

TABLE 3
Subgroup analysis of prognostic role of CRP for PFS in patients with CC.