Prognostic and clinicopathological value of fibrinogen-to-albumin ratio in colorectal cancer: a meta-analysis.

Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and the second leading cause of cancer-related mortality [1]. Notably, cancers of unknown primary site (CUP) account for approximately 1–3% of all malignant tumors [2]. Among these, CUP with a colon cancer profile is increasingly recognized as a distinct favorable subgroup [2]. According to GLOBOCAN data, 1,880,725 new cases of CRC and 615,880 deaths attributed to the disease were recorded globally in 2020 [3]. Advanced or intermediate-stage CRC is often associated with hematogenous metastasis to the liver, which significantly worsens prognosis and increases the risk of mortality [4]. Despite advancements in multimodal treatment approaches and tailored chemoradiotherapy, achieving long-term survival and effective disease management remains a significant challenge for patients with metastatic CRC [5]. Notably, the risk of CRC increases with age, with over two-thirds of cases occurring in individuals aged 65 years and older [6]. Although older patients are at higher risk for severe postoperative complications, no consensus exists regarding the impact of age on survival outcomes [7]. The prognosis for older patients might be affected by factors such as the stage of disease at diagnosis, tumor location, preexisting health conditions, and treatment modalities[8]. Although immunotherapy has recently emerged as a treatment option for patients with CRC, the prognosis for advanced and metastatic CRC cases remains poor [9,10]. Programmed cell death-ligand 1 (PD-L1) expression in immune cells is significantly elevated in mismatch repair (MMR)-deficient (MSI-H) CRC compared to MMR-proficient (MSI-L) tumors, with consistent expression across various MSI-H molecular subtypes [9]. Screening for defective DNA MMR using immunohistochemistry (IHC) and/or microsatellite instability testing is recommended, though converting the biological and technical variability of microsatellite instability tests into practical clinical data remains a challenge [11]. Research suggests that IHC testing of the MMR may produce inconsistent outcomes for specific germline mutations, potentially due to accompanying somatic mutations [12,13]. The 5-year survival rate for stage IIIC CRC is 53%, whereas for metastatic CRC, it drops to only 12% [10]. Biomarkers play a crucial role in guiding CRC therapy and improving survival rates [14]. Consequently, biomarker testing is recommended as a part of the standard diagnostic process for CRC [1]. The RAS mutation, a key genetic alteration in CRC, is associated with increased tumor aggressiveness and resistance to chemotherapy [14]. Moreover, microRNAs (miRNAs) function as both tumor suppressors and oncogenes, and their diagnostic, prognostic, and predictive potential is currently under investigation [13,14]. For instance, in metastatic CRC, resistance to anti-vascular endothelial growth factor or anti-epidermal growth factor receptor inhibitors has been linked to specific miRNA expression patterns. An increase in miR-126 is associated with resistance to bevacizumab, while resistance to cetuximab is observed with overexpression of miR-31, miR-100, and miR-125b, as well as downregulation of miR-7 [1,14]. Therefore, an urgent need exists to identify novel prognostic markers to enhance the clinical management of CRC.
Emerging evidence highlights the critical roles of nutrition and inflammation in CRC metastasis and progression [12,15]. Various inflammatory parameters, such as platelet-to-lymphocyte ratio [16], prognostic nutritional index [17], lymphocyte-to-monocyte ratio [18], controlling nutritional status score [19], and geriatric nutritional risk index[20], have been identified as prognostic markers for CRC. The fibrinogen–albumin ratio (FAR) is a novel inflammatory and nutritional biomarker that has demonstrated prognostic efficiency across diverse cancers, including esophageal cancer [21], hepatocellular carcinoma [22], osteosarcoma [23], non-small-cell lung cancer [24], and ovarian cancer [25]. Although recent studies have explored FAR’s potential in predicting CRC prognosis, no consistent findings have been obtained [26–35]. Some studies report that a higher FAR is significantly associated with poor CRC prognosis [27,29,34], whereas others found no significant correlation between FAR and CRC survival [30,33]. Therefore, this meta-analysis aimed to identify the precise role of FAR in predicting CRC prognosis and explore its relationship with clinicopathological factors in CRC.

Materials and methods

Study guideline
This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guideline [36].

Ethics statement
As this study is a meta-analysis of previously published research, it did not require ethical approval.

Literature search
A comprehensive search of PubMed, Web of Science, Cochrane Library, and Embase databases was conducted up to January 14, 2025. The search strategy included the terms (albumin-to-fibrinogen OR albumin/fibrinogen OR fibrinogen-to-albumin OR fibrinogen/albumin OR Alb to Fib) AND (colonic neoplasms OR colon cancer OR colorectal neoplasms OR rectal cancer OR rectal tumor OR rectum cancers OR colorectal cancer OR CRC OR colorectal carcinoma OR colorectal tumor). Both Medical Subject Headings terms and free-text keywords were utilized. Only studies published in English were included in this meta-analysis. Additionally, the reference lists of identified articles were reviewed to identify potentially relevant studies.

Inclusion and exclusion criteria
The inclusion criteria were as follows: (1) studies involving patients with a pathological diagnosis of CRC; (2) studies that examined the relationship between FAR and clinical outcomes in CRC cases; (3) studies with extractable or computable hazard ratios (HRs) and 95% confidence intervals (CIs); (4) available FAR thresholds; and (5) studies published in English. The exclusion criteria were: (1) letters, reviews, comments, and conference abstracts; (2) studies with insufficient data for meta-analysis; and (3) animal studies.

Data extraction and quality assessment
Two investigators (Y.S. and J.C.) independently screened eligible studies and extracted data. Disagreements were resolved through discussion with a third reviewer (X.L.) until a consensus was reached. Extracted data included first author, publication year, country, sample size, sex, age, study period, study design, histology, TNM stage, treatment details, FAR threshold, threshold determination method, survival endpoints, survival analysis types, follow-up duration, HRs, and 95% CIs. Overall survival (OS) and progression-free survival (PFS) were the primary and secondary endpoints, respectively. The quality of the included studies was assessed using the Newcastle–Ottawa Scale (NOS), which covers three domains: selection, comparability, and outcome assessment [37]. NOS scores range from 0 to 9 points, and studies scoring ≥6 points were considered high-quality.

Statistical analysis
In this study, pooled HRs and 95% CIs were computed to evaluate the prognostic value of FAR for predicting OS and PFS in CRC. Inter-study heterogeneity was assessed using Cochran’s Q test and I2 statistics. A threshold of p < 0.10 and I2 > 50% indicated significant heterogeneity, in which case a random-effects model was used; otherwise, a fixed-effects model was applied. Subgroup analyses were also conducted. The association between FAR and clinicopathological factors in CRC was explored using pooled odds ratios (ORs) and 95% CIs. Additionally, sensitivity analysis was performed by sequentially omitting one article at a time to assess the robustness of our results. Publication bias was evaluated using Begg’s and Egger’s tests. Statistical analysis was performed using Stata version 12.0 software (Stata Corporation, College Station, TX, USA), with statistical significance set at p < 0.05.

Results

Literature retrieval process
The initial literature search identified 106 studies, with 73 remaining after duplicates were removed (Figure 1). After title and abstract screening, 53 studies were excluded due to irrelevance or being animal studies. Twenty articles were further assessed through full-text evaluation, with 10 excluded due to unavailable survival data (n = 7), irrelevance of FAR (n = 2), and being a review article (n = 1). Finally, 10 articles comprising 4,704 cases were included in the meta-analysis [26–35] (Figure 1; Table 1).

Study characteristics
Table 1 summarizes the basic characteristics of the included articles [26–35]. All eligible studies were conducted in China and published in English [26–35] between 2018 and 2024. Sample sizes ranged from 71 to 1,533 (median: 335). Nine studies were retrospective [26,28–35], and one was a prospective trial [27]. Nine studies included patients with CRC [26–28,30–35], while one focused on patients with rectal cancer [29]. Regarding cancer staging, three studies included patients with stages I–IV CRC [31–33], three focused on stages II–III cases [29,34,35] and two articles each enrolled patients with stages I–III [26,30] and IV [27,28] CRC. The FAR threshold ranged from 8.43 to 12, with a median of 10.75. Five articles utilized the receiver operating characteristic curve [28,29,32–34] to determine the FAR threshold, while five others applied the X-tile software [26,27,30,31,35]. Seven studies highlighted the significance of FAR in predicting OS [26,27,29,31–34] and eight provided data on the association between FAR and PFS [27–31,33–35]. Multivariate regression was used to derive HRs and 95% CIs in eight studies [26,29–35], while two adopted univariate regression [27,28]. NOS scores ranged from 7 to 9, indicating that all included articles were of high quality (Table 1).

FAR and OS
Seven studies involving 2,943 cases [26,27,29,31–34] evaluated the role of FAR in predicting OS. A fixed-effects model was used due to insignificant heterogeneity (I2 = 39.7%, p = 0.127). Elevated FAR was significantly associated with poor OS among patients with CRC (HR = 1.59, 95% CI = 1.38–1.83, p < 0.001; Figure 2; Table 2). Subgroup analysis revealed that high FAR remained a significant predictor of poor OS, regardless of study design, histology, treatment, FAR threshold, threshold determination method, and type of survival analysis (all p < 0.05, Table 2). Additionally, FAR was strongly associated with poor OS in subgroups with a sample size >300 (p < 0.001) and among patients with TNM stages II–III (p < 0.001) and IV (p < 0.001) (Table 2).

FAR and PFS
Eight studies involving 3,802 patients [27–31,33–35] explored the relationship between FAR and PFS in CRC. A fixed-effects model was used due to insignificant heterogeneity (I2 = 0, p = 0.825). The pooled analysis showed that elevated FAR was significantly associated with worse PFS (HR = 1.65, 95% CI = 1.44–1.90, p < 0.001; Figure 3; Table 3). Subgroup analysis confirmed that elevated FAR was consistently correlated with worse PFS, regardless of study design, sample size, histology, treatment, FAR threshold, threshold determination method, or survival analysis type (all p < 0.05, Table 3). Moreover, subgroup analysis indicated that FAR was significantly correlated with poor PFS among patients with stages II–III, I–IV, and IV CRC (p < 0.05, Table 3).

FAR and CRC clinicopathological factors
Seven studies involving 2,507 cases provided data on the relationship between FAR and clinicopathological factors in CRC [26,28,29,31–34]. The pooled analysis revealed that elevated FAR was significantly associated with age ≥60 years (OR = 1.56, 95% CI = 1.31–1.85, p < 0.001), male sex (OR = 1.20, 95% CI = 1.01–1.43, p = 0.042), and poor tumor differentiation (OR = 1.63, 95% CI = 1.26–2.10, p < 0.001) (Figure 4; Table 4). However, no significant correlation was observed between FAR and TNM stage (OR = 1.45, 95% CI = 0.99–2.11, p = 0.055), tumor size (OR = 1.51, 95% CI = 0.75–3.05, p = 0.247), N stage (OR = 1.29, 95% CI = 0.87–1.89, p = 0.205), perineural invasion (OR = 0.92, 95% CI = 0.68–1.24, p = 0.595), or vascular invasion (OR = 1.18, 95% CI = 0.77–1.80, p = 0.449) (Figures 4 and 5; Table 4).

Sensitivity analysis
A sensitivity analysis was conducted to assess the stability of our meta-analysis outcomes. The combined analysis showed that the association of FAR with OS and PFS remained consistent, regardless of the exclusion of any individual study (Figure 6). These findings confirm the reliability and robustness of our results.

Publication bias
Funnel plots, along with Begg’s and Egger’s tests, were used to examine potential publication bias. As shown in Figure 7, no significant publication bias was observed for OS (Begg’s test: p = 0.548; Egger’s test: p = 0.589) or PFS (Begg’s test: p = 0.536; Egger’s test: p = 0.812).

Discussion
The prognostic value of FAR for predicting CRC prognosis has been explored in previous studies, yielding inconsistent results. This meta-analysis pooled data from 10 articles involving 4,704 cases [26–35], demonstrating that elevated FAR levels were significantly correlated with poor OS and PFS in CRC cases. The prognostic value of FAR remained significant across various subgroups. Moreover, a high FAR was significantly linked to age ≥60 years, male sex, and poor tumor differentiation in CRC. Publication bias tests and sensitivity analyses confirmed the robustness and reliability of these findings. Elevated FAR significantly predicted both short- and long-term prognostic outcomes in CRC. To our knowledge, this meta-analysis is the first to investigate the value of FAR for predicting CRC prognosis.
FAR is calculated using fibrinogen and albumin levels, with an elevated FAR resulting from increased fibrinogen and/or decreased albumin levels. The exact mechanisms underlying the prognostic value of FAR in CRC are not fully understood and are analyzed below. First, fibrinogen, a vital component of the coagulation system, plays a significant role in systemic inflammation. As an acute-phase glycoprotein, fibrinogen contributes to tumor invasion and progression by regulating coagulation, immune function, and inflammation [38]. When converted to fibrin with the help of thrombin, it forms a protective barrier surrounding cancer cells, promoting cancer cell survival and playing a crucial role in tumor development [39]. Second, serum albumin serves as a marker for both nutritional and immune status, with lower levels indicating poor nutrition and weakened immunity [40]. Hypoproteinemia has been linked to the release of inflammatory cytokines, such as interleukin (IL)-6 and tumor necrosis factor-alpha, which are associated with poor prognosis [41]. In addition, hypoalbuminemia is associated with impaired immune function due to macrophage activation [42]. IL-6 further exacerbates this by stimulating the liver to produce acute-phase proteins like C-reactive protein, increasing amino acid demand and depleting albumin levels. Therefore, FAR is a potential reliable prognostic marker for patients with CRC.
Notably, a high FAR value is associated with low albumin levels. Generally, in patients with cancer, albumin levels tend to drop more significantly during the middle and late stages of the disease, leading to hypoalbuminemia [43], which is indicative of malnutrition. Malnutrition in patients with cancer arises from various factors, including inflammation, disequilibrium between anabolic and catabolic processes, adverse effects of anti-cancer treatments, reduced food intake, and hormonal imbalances [44]. Decreased albumin levels can impair immune function, leading to a weaker response to cancer cells and promoting tumor growth [45]. Additionally, preoperative malnutrition is a common concern in patients with CRC and significantly impacts their postoperative recovery and outcomes [46]. Therefore, improving the nutritional status of patients with cancer during anti-cancer treatment is essential for enhancing their prognosis and overall well-being.
Recent meta-analyses have highlighted the prognostic value of FAR for predicting the prognosis of different cancers [47–50]. For instance, Zhang et al. performed a meta-analysis of 5,088 cases, concluding that elevated FAR was markedly associated with poor OS and worse disease-free survival (DFS) in malignant tumors [50]. Similarly, Sun et al. in a meta-analysis of 7,282 cases, demonstrated that high FAR predicted unfavorable outcomes, including OS, DFS, and PFS [49]. Additionally, Li et al. in a meta-analysis of 19 articles, found a strong association between higher FAR and poor cancer prognosis [47]. Our findings regarding the prognostic value of FAR in CRC are consistent with those observed in other cancers.
This study has certain limitations that should be acknowledged. First, all the included studies were conducted in China, necessitating validation of the prognostic value of FAR in CRC cases in other countries. Although no geographic restrictions were observed during article selection and only studies published in English were included, this geographic limitation may impact the generalizability of the findings. Second, the FAR threshold was inconsistent across the included articles, potentially introducing selection bias. Third, the majority of the included studies were retrospective in design, which may have contributed to heterogeneity in the results. To address these limitations, large-scale, multi-regional prospective studies are needed to validate our findings.

Conclusions
In summary, elevated FAR is significantly associated with poor OS and PFS in patients with CRC. Additionally, higher FAR is strongly linked to older age and poor tumor differentiation in CRC.

Supplementary Material

PRISMA_2020_checklist.docx