Prognostic and clinicopathological significance of C-reactive protein in patients with ovarian cancer: a meta-analysis

Background Many studies have explored the relationship between C-reactive protein (CRP) levels and survival outcomes in patients with ovarian cancer (OC); however, consistent results have not been reported. As such, this meta-analysis was performed to accurately assess the prognostic and clinicopathological roles of CRP in OC. Methods The PubMed, Web of Science, Embase, and Cochrane Library databases were systematically searched for relevant studies published from inception to April 7, 2023. The effect of CRP level(s) and OC prognostic outcomes was analyzed by computing the combined hazard ratio (HR) and corresponding 95% confidence interval (CI). Thereafter, the association between CRP level(s) and clinicopathological factors was evaluated using a combined odds ratio (OR) and corresponding 95% CI. Results The present meta-analysis included 15 studies comprising 3202 subjects. According to the combined data, higher CRP levels were markedly associated with unfavorable overall survival (OS) (HR 1.23 [95% CI 1.11–1.37]; p < 0.001) and progression-free survival (PFS) (HR 1.55 [95% CI 1.30–1.84]; p < 0.001) in patients with OC. Furthermore, the results indicated that high CRP levels were significantly correlated with International Federation of Gynecology and Obstetrics (FIGO) stages III–IV (p < 0.001), residual tumor size ≥ 1 cm (p < 0.001), histological grade 3 (p = 0.040), and ascites volume ≥ 500 mL (p < 0.001). Conclusion The results of this meta-analysis demonstrated that higher serum CRP levels were strongly associated with dismal OS and PFS in subjects with OC. High CRP levels were also significantly associated with clinical factors implicated in tumor aggressiveness and the development of OC. Supplementary Information The online version contains supplementary material available at 10.1186/s12957-023-03290-5.


Background
In recent decades, ovarian cancer (OC), a frequently observed malignancy among females, has been characterized by high mortality and morbidity rates worldwide [1].OC accounts for 1.6% of newly diagnosed cancer cases and 2.1% of cancer-associated mortality worldwide annually [2].Approximately 313,959 new cases of OC and 207,252 cases of OC-related death were reported globally in 2020 [3].Despite the progress in diagnosis, surgery, chemotherapy, radiotherapy, and immunotherapy of OC over the past decade [4,5], 5-year survival and recurrence rates remain only at 39% and 70%, respectively [6,7].Poor prognosis and a high incidence of OC recurrence may, in part, be associated with insufficient efficient markers for prognosis prediction.Consequently, the identification of new and reliable prognostic biomarkers for OC is necessary to inform and support clinical management.
Current evidence has revealed that inflammation and immunity contribute to the initiation, progression, development, and metastasis of cancer [8].The relationship between chronic inflammation and cancer has attracted increasing attention, and inflammation is now considered to be a facilitating feature [9].Inflammation can promote tumor progression and metastasis [10].C-reactive protein (CRP) is an acute-phase protein mostly generated in the liver after infection, inflammation, and tissue injury [11].As reported by many studies, serum CRP levels are elevated in multiple cancers [12,13].Previous investigations have reported that high serum CRP levels predict dismal prognosis in different cancer types, such as breast cancer [14], diffuse large B-cell lymphoma (DLBCL) [15], nasopharyngeal carcinoma [16], renal cell carcinoma [17], and colorectal cancer [18].Furthermore, current evidence indicates that high CRP levels are associated with an increased risk for OC [19].According to a multicenter study, CRP is implicated in ovarian carcinogenesis and inflammation and is particularly linked to endometrioid and mucinous carcinomas [19].Moreover, a previous study suggested that high CRP levels were correlated with OC stage and tumor size [20].The utility of CRP levels in predicting the prognosis of OC has been widely explored [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35]; however, consistent results have not been reported.For example, a higher CRP level has been reported to be markedly associated with poor survival of patients with OC in some studies [26,32,35].However, other researchers failed to identify any relationship between CRP and survival in those with OC [24,31].As such, we performed a comprehensive literature search to investigate the utility of CRP in accurately predicting the prognosis of patients with OC.Additionally, the relationship between CRP level(s) and the clinicopathological characteristics of patients with OC was also explored.

Study guideline
The present meta-analysis was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [36].The PRISMA checklist is provided as Additional file 1.The protocol of this meta-analysis was registered in INPLASY (ID: INPLASY202380097).The link of this protocol is https:// inpla sy.com/ inpla sy-2023-8-0097/.

Ethics statement
This meta-analysis did not require ethics approval because the data did not contain personal information, which precluded any privacy concerns.

Literature retrieval
The PubMed, Web of Science, Embase, and Cochrane Library databases were searched for relevant studies, published from inception until April 7, 2023, using the following search strategies and terms: (C-reactive protein or C-reactive protein or CRP) and (ovarian cancer or ovarian neoplasm or ovarian carcinoma or ovarian tumor).The detailed search strategies for each database are shown in Additional file 2. The literature search was restricted to studies published in English.In addition, the reference lists of eligible studies were manually searched to identify other potentially relevant works.

Inclusion and exclusion criteria
Studies were included based on the following criteria: OC diagnosed by pathology, reporting an association between pretreatment CRP levels and any survival outcome in OC, available hazard ratios (HRs) and 95% confidence intervals (CIs) for prognosis or calculability based on available data, a threshold identified to stratify low and high CRP levels, and published in English.Review articles, meeting abstracts, letters, case reports, comments, studies with no survival data, and animal studies were excluded.

Data acquisition and quality evaluation
Two researchers (WZ and ZZ) reviewed the potentially eligible studies and collected the data.Disagreements were discussed with a third researcher (LQ) until a consensus was reached.The following information was extracted from each of the included studies: first author, publication year, country, sample size, age, study period, International Federation of Gynecology and Obstetrics (FIGO) stage, study center, treatment, threshold CRP level (mg/L), threshold determination approach, survival endpoint, survival analysis, followup, and HRs with corresponding 95% CIs.Overall survival (OS) and progression-free survival (PFS) were the primary and secondary outcomes, respectively.The Newcastle-Ottawa Quality Assessment Scale (NOS) was used to evaluate the methodological quality of the included studies [37].More specifically, study quality was divided into three categories: participant selection (0-4 points), study comparability (0-2 points), and outcome ascertainment (0-3 points), with a total score of 0-9.Studies with NOS scores ≥ 6 were considered to be of high quality.

Statistical analysis
Combined HR and 95% CI were determined to evaluate whether CRP could be used to predict the prognosis of patients with OC.Heterogeneity across the included studies was explored using Cochran's Q test and the I 2 statistic.Studies with I 2 > 50% and/or p < 0.10 indicated obvious heterogeneity; accordingly, combined HR was calculated using a random-effects model; otherwise, a fixed-effects model was used.Subgroup analyses according to different factors were performed to identify potential sources of heterogeneity.In addition, the relationship between CRP level(s) and clinicopathological factors in patients with OC was assessed using a combined odds ratio (OR) and corresponding 95% CI.Funnel plot symmetry was visually inspected to assess publication bias using Begg's and Egger's tests.Statistical analysis was performed using Stata version 12.0 (StataCorp LLC, College Station, TX, USA).Differences with p < 0.05 were considered to be statistically significant.

CRP level and PFS in patients with OC
Seven studies enrolling 1790 patients [24,25,27,31,32,34,35] analyzed the effect of CRP level on the prognosis of OC.Owing to non-obvious heterogeneity, a fixed-effects model was adopted (I 2 = 9.3%, p = 0.358).Combined data demonstrated that high CRP levels exhibited an obvious relationship with poor PFS in those with OC (HR 1.55 [95% CI 1.30-1.84];p < 0.001) (Table 3, Fig. 3).As revealed by subgroup analysis, the role of CRP in predicting PFS was not influenced by the threshold determination approach or type of survival analysis (p < 0.05) (Table 3).Additionally, elevated CRP levels remained the obvious factor predicting dismal PFS for the following subgroups: Asian region (HR  3).

Publication bias
Funnel plots, together with Begg's and Egger's tests, were used to investigate publication bias.Visual inspection of the funnel plots revealed no significant asymmetry in OS or PFS (Fig. 5).Moreover, the findings also demonstrated no obvious publication bias with regard to OS (p = 0.913 and p = 0.761 according to Begg's and Egger's tests, respectively) and PFS (p = 0.881 and p = 0.666 according to Begg's and Egger's tests, respectively).Based on these findings, publication bias was not detected in the included studies.

Discussion
CRP, a frequently used inflammatory biomarker, is produced in the liver and atherosclerotic plaques.Its role in predicting prognosis in patients with OC has been widely analyzed; however, consistent results have not been reported [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35].This study combined data from 15 studies involving 3202 subjects to precisely determine the prognostic utility of CRP levels for predicting prognosis in OC.Our results indicated that elevated CRP levels were markedly associated with shortened OS and inferior PFS in patients with OC.Furthermore, higher CRP levels exhibited a significant relationship with advanced FIGO stage, larger residual tumor size, higher histological grade, and ascites volume ≥ 500 mL.
Collectively, these data suggest that elevated CRP level is a prognostic marker for poor short-and long-term survival in patients with OC.Increased CRP levels are also predictive of clinicopathological factors, indicating high disease aggressiveness.To our knowledge, this is the first meta-analysis to investigate whether CRP levels can be used to predict the prognosis of patients with OC.
Higher CRP levels are associated with tissue damage, infection, atherosclerosis, arterial hypertension, obesity, diabetes, and/or cancers [38].The mechanisms underlying the relationship between high CRP levels and poor OC survival are discussed below.First, chronic and persistent inflammation may lead to carcinogenesis or angiogenesis, which promotes tumor cell proliferation [39].In particular, certain inflammatory cells can generate cytokines and chemokines in the blood, such as interleukin (IL)-6, IL-8, and tumor necrosis factor-α, which promote the production of CRP in the liver [40].Second, inflammation can promote tumor development by generating growth factors to sustain cell growth and survival, limit cell death, and produce proangiogenic factors that accelerate neovascularization [41].Importantly, inflammation in the tumor microenvironment may be reflected by circulating CRP levels and proteins related to early inflammation and have important effects [42].Third, as supported by increasing evidence, inflammatory factors, such as CRP, are produced by hepatocytes after trauma, infection, and cancer; moreover, they can also be produced by cancer cells [43,44].Therefore, CRP level is an easy and credible marker for predicting the prognosis of patients with OC.
In the current meta-analysis, we included 15 studies and expected CRP to be a significant prognostic marker in patients with OC for the following reasons.Previous evidence suggests a biological function of CRP in ovarian carcinogenesis [19,20].Second, the included Fig. 2 Forest plot of the prognostic role of CRP for OS in patients with OC studies provided controversial results regarding the prognostic role of CRP in OC [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35].More than one-half of the studies yielded positive results.Third, the significant correlation among FIGO stage, tumor size, and histological grade also met our expectations because these results were in accordance with those of a previous study [20].
Recently, many meta-analyses have explored whether CRP can be used to predict the prognosis of different solid tumors [45][46][47].According to a meta-analysis including 16 studies by Zhou et al. [45], higher CRP levels were associated with worse OS, cancer-specific survival, and PFS in prostate cancer.In a meta-analysis of 1287 subjects, Chen et al. [46] reported that patients with metastatic colorectal cancer with higher CRP levels exhibited markedly reduced OS relative to those with lower CRP levels.Based on a meta-analysis including 4449 subjects, Chen et al. [48] reported that higher CRP levels predicted dismal OS, cancer-specific survival, and PFS for head and neck squamous cell carcinoma.A recent meta-analysis of 3000 subjects indicated that higher CRP levels before treatment were associated with poor OS and PFS in diffuse large B-cell lymphoma [49].Another meta-analysis of 5215 patients revealed that elevated serum CRP levels were associated with worse OS and distant metastasis-free survival in nasopharyngeal carcinoma [50].Our findings in OC confirmed the prognostic value of CRP for additional cancers.
The present investigation had some limitations.First, the included studies had a retrospective design, and some HRs were calculated based on univariate regression, possibly causing an overestimation of effect sizes.Second, there was inherent heterogeneity in OS-likely due  to the retrospective design of the included studieswhich persisted after applying the random-effects model.Third, the threshold CRP level and threshold determination approaches were not uniform among the included studies.Therefore, large prospective studies using a standard threshold CRP level should be conducted for further validation.
The current meta-analysis is the first to identify the prognostic and clinicopathological roles of CRP in OC by integrating data from 15 studies.Future studies should focus on the optimal CRP cutoff value for patients with OC.Furthermore, clinical assessment tools that incorporate CRP levels should be developed to predict survival outcomes in patients with OC.

Conclusions
In conclusion, the results of the present study demonstrated that elevated serum CRP levels predicted poor OS and inferior PFS in patients with OC.High CRP levels were also significantly associated with clinical factors implicated in tumor aggressiveness and development.Therefore, CRP level could be adopted as an easy and credible marker to predict prognosis in patients with OC.

Table 1
Baseline characteristics of included studiesNR not reported, FIGO International Federation of Gynecology and Obstetrics, ROC receiver operating characteristics, OS overall survival, PFS progression-free survival, NAC neoadjuvant chemotherapy, NOS Newcastle-

Table 2
Subgroup analysis of the prognostic value of CRP for OS in patients with ovarian cancer

Table 3
Subgroup analysis of the prognostic value of CRP for PFS in patients with ovarian cancer Fig. 3 Forest plot of the prognostic role of CRP for PFS in patients with OC

Table 4
The association between CRP and clinicopathological features in patients with ovarian cancer