C‐Reactive Protein and Management of Hepatocellular Carcinoma: Bridging Systemic Inflammation and Clinical Prognosis

Hepatocellular carcinoma (HCC) remains a global health burden and a leading cause of cancer-related mortality [1]. While curative modalities exist for early-stage HCC, many patients are diagnosed at advanced stages where systemic therapies, particularly immune checkpoint inhibitors (ICIs), represent the cornerstone of management [2, 3]. Currently, no serum biomarker can reliably predict response to such therapies [1, 4]. C-reactive protein (CRP), an acute-phase reactant produced by hepatocytes in response to interleukin-6 (IL-6) signalling [5], is a key component of the innate immune response and systemic inflammation. In their multicenter retrospective study, Hatanaka et al. [6] investigated the prognostic role of serum CRP levels in 167 patients with advanced HCC treated with durvalumab and tremelimumab (Dur/Tre) across Japanese institutions. The low-CRP group showed significantly better median progression-free survival (PFS; 3.6 vs. 2.4 months, p = 0.02) and overall survival (OS; not reached vs. 7.9 months, p < 0.001) compared to the high-CRP group. Multivariate analyses confirmed low CRP as an independent favourable predictor of PFS (Hazard ratio [HR]: 1.50, 95% confidence interval [CI]: 1.02–2.21, p = 0.04) and OS (HR: 3.62, 95% CI: 2.00–6.54, p < 0.001). No significant differences were observed in ORR, disease control rate or immune-related adverse events between the groups. Limitations of this study include its retrospective design, limited number of patients and relatively short follow-up, use of Dur/Tre in later-line settings and baseline imbalances in liver function and tumour burden [6]. However, these findings provide evidence supporting CRP's prognostic value across all HCC stages. CRP has been recognised as a biomarker in early-stage HCC. Lin et al. developed the PACE score in a large multicenter cohort of 2770 patients undergoing hepatectomy [7]. By integrating CRP (≥ 10 mg/L) with alpha-fetoprotein (AFP ≥ 400 ng/mL), the score stratified patients into three prognostic groups with distinct 5-year OS (59.6%, 43.9% and 20.6%) and recurrence-free survival. The PACE score model retained its prognostic value even within BCLC stages A and B. Yasuda et al. [8] evaluated patients undergoing resection beyond the Milan criteria and introduced the CALLY index—combining CRP, albumin and lymphocyte count—as a significant predictor of very early (within 6 months) recurrence. Patients with a CALLY index < 2.8 experienced poorer survival. Similarly, Iida et al. confirmed these findings in 384 patients who underwent surgical resection for HCC [9]. Those with CALLY values < 5 had markedly lower 5-year survival (46%) compared to those with higher scores (71%), a finding reproduced in an independent cohort of 267 patients. Song et al. [10] assessed two additional CRP-based metrics—the inflammation-immunity-nutrition score (IINS) and the inflammatory burden index (IBI)—in 701 patients. Both indices independently correlated with long-term outcomes, and models incorporating IINS outperformed the traditional staging systems. In the liver transplantation setting, an analysis by Carr et al. [11] involving 330 patients revealed that CRP showed an independent association with OS in multivariate analysis. Elevated CRP levels were linked to larger tumour burden, increased portal vein thrombosis and worse survival, irrespective of AFP levels or tumour size. Together, these studies underscore the growing role of CRP—alone or integrated within composite indices—in stratifying risk and refining prognosis in the spectrum of early-stage HCC management. Emerging evidence highlights the prognostic relevance of serum CRP in patients with intermediate-to-advanced HCC. In a cohort of 152 patients treated with ICI-based hepatic artery infusion chemotherapy (HAIC), Zeng et al. [12] demonstrated that elevated baseline CRP (≥ 1 mg/dL) was associated with shorter median OS (20 months vs. not reached, p = 0.010). Multivariate analysis confirmed CRP as an independent predictor of poor prognosis. Importantly, a combined cholesterol–CRP prognostic score (CCPS) stratified patients into low-, intermediate- and high-risk groups, which showed distinct survival outcomes and significantly different ORR (70.6%, 46.6% and 44.1%, respectively; p = 0.038). In a large retrospective study of 313 patients with early- and intermediate-stage HCC treated with transarterial chemoembolization (TACE), Hayashi et al. introduced a CRP–AFP classification system. Patients in the lowest-risk class (CRP < 0.2 mg/dL and AFP < 100 ng/mL) had a median OS of 42.2 months, compared to 13.2 months in the highest-risk class (CRP ≥ 1.0 mg/dL and AFP ≥ 100 ng/mL) [13]. The CRP–AFP classification outperformed traditional scores like STATE and mHAP in time-dependent ROC analysis. Similarly, Rekik et al. confirmed that CRP > 1 mg/dL predicted inferior median OS in a Western cohort undergoing TACE (8 vs. 20 months, p = 0.00186) [14]. The STATE score, incorporating CRP, albumin and tumour burden, further stratified patients (median OS: 6.7 vs. 22.2 months for STATE < 18 vs. ≥ 18, respectively; p = 0.0002). In advanced HCC, elevated CRP levels have consistently correlated with poor outcomes across systemic therapies. In a large multicentre Indian study, Singh et al. found that CRP > 1 mg/dL independently predicted non-response to atezolizumab plus bevacizumab [15], along with elevated AFP, PIVKA-II and diabetes. Similarly, Kaneko et al. [16] analysed 213 patients treated with the same regimen and reported that baseline CRP ≥ 1 mg/dL was associated with higher AFP levels, worse ALBI scores, a lower disease control rate (62.2%), and shorter OS (HR 2.292; p < 0.001). Multivariate analysis confirmed CRP as an independent prognostic factor. Dynamic changes in CRP also hold prognostic value. Qin et al. evaluated 132 patients on immune checkpoint blockade and stratified them into three groups based on early CRP kinetics. CRP flare-responders (initial rise then decline) had the highest objective ORR (41.2%) and significantly longer median PFS (11.6 months), compared to CRP responders (5.2 months) and non-responders (2.3 months; p < 0.001). These data emphasise the value of early CRP fluctuations as a dynamic and non-invasive biomarker. The CRAFITY score (CRP and AFP in ImmunoTherapY) was developed by Scheiner et al. [17] to predict prognosis in patients receiving PD-(L)1-based therapies. It assigns 1 point each for CRP ≥ 1 mg/dL and AFP ≥ 100 ng/mL. In a European multicentre study, median OS ranged from 27.6 months in the low group (0) to 6.4 months in the high group (2) (p < 0.001). Ueno et al. [18] validated the CRAFITY score in 302 Japanese patients on atezolizumab–bevacizumab, linking higher CRAFITY scores with shorter OS and PFS, as well as treatment refractoriness (24.6%, 44.6% and 57.9% of patients in CRAFITY 0, 1 and 2 groups, p < 0.001). CRP also retains prognostic value in second-line tyrosine kinase inhibitor (TKI) therapy. In a multicentre study by Okumura et al. [19], patients with unresectable HCC treated with lenvatinib and baseline CRP ≥ 0.5 mg/dL had shorter OS (7.8 vs. 22.9 months, p < 0.001) and time-to-treatment failure (4.4 vs. 8.5 months, p = 0.007). Likewise, a high CRAFITY score in sorafenib-treated HCC patients predicted shorter OS (4.6 months) compared to a low score (18.3 months; p < 0.001), but it did not predict disease control rate (DCR) [17]. These are summarised in Table 1. CRP incorporated in scores (PACE, CALLY) predicts OS, recurrence. Inflammatory indices (IINS, IBI) correlate with outcomes. CRP independently predicts post-transplant survival. Enhances surgical/transplant risk stratification. Supplements BCLC staging. Elevated CRP predicts worse OS with HAIC or TACE. CRP–AFP and STATE scores outperform traditional tools. Aids prognostication and treatment stratification. High CRP linked to inferior outcomes with ICIs and TKIs. CRP kinetics and CRAFITY score refine prognostic prediction. Supports CRP use in systemic therapy guidance. Collectively, CRP has emerged as a clinical biomarker with prognostic significance across all stages of HCC. Although the pathophysiology has not been completely clarified, its predictive role may involve IL-6 signalling and immune modulation, including CD4+/CD8+ T-cell responses [20]. The multicentre study by Hatanaka et al. highlights the potential of CRP not only in disease prognosis but also as a predictive marker of response to Dur/Tre. Further prospective studies are needed to validate and better clarify the clinical utility of CRP in therapeutic practice in HCC patients. The authors have nothing to report. George V. Papatheodoridis has served as advisor for Abbvie, Astra Zeneca, Genesis, Gilead, GlaxoSmithKline, Janssen, Ipsen, Merck Sharp & Dohme, Novartis, Novo Nordisk, Roche, Takeda and Vir, as lecturer for Abbvie, Astra Zeneca, Gilead, GlaxoSmithKline, Janssen, Ipsen, Merck Sharp & Dohme, Novartis, Novo Nordisk and Roche, and has received research grants from Abbvie, Gilead, Takeda and Vianex. The other authors declare no conflicts of interest. This article is linked to Hatanaka et al. paper. To view this article, visit https://doi.org/10.1111/liv.70192. The authors have nothing to report.