OL Oncology Letters 1792-1074 1792-1082 D.A. Spandidos 10.3892/ol.2023.14118 OL-26-6-14118 Articles Role of high‑sensitivity C‑reactive protein in patients with sarcoma Aggerholm-PedersenNinna 1 2 3 Baad-HansenThomas 3 4 MøllerHolger Jon 3 5 Sandfeld-PaulsenBirgitte 3 6 Department of Experimental Oncology, Aarhus University Hospital, 8200 Aarhus, Denmark Department of Oncology, Aarhus University Hospital, 8200 Aarhus, Denmark Department of Clinical Medicine, Aarhus University, 8200 Aarhus, Denmark Department of Orthopaedic Surgery, Aarhus University Hospital, 8200 Aarhus, Denmark Department of Clinical Biochemistry, Aarhus University Hospital, 8200 Aarhus, Denmark Department of Clinical Biochemistry, Viborg Regional Hospital, 8800 Viborg, Denmark Correspondence to: Dr Ninna Aggerholm-Pedersen, Department of Oncology, Aarhus University Hospital, 99 Palle Juul-Jensens Boulevard, 8200 Aarhus, Denmark, E-mail: ninnpede@rm.dk 12 2023 30 10 2023 26 6 531 22062023 08092023 Copyright: © Aggerholm-Pedersen et al. 2023 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

Immunotherapy has shown promising results in lung cancer and melanomas; however, the responses have been poor in patients with sarcoma. Understanding the relationship between the immune system and sarcoma is essential to develop improved immunotherapy approaches. High-sensitivity C-reactive protein (hs-CRP) has been proposed as a prognostic marker in other cancer types; however, to the best of our knowledge, the association between hs-CRP levels and mortality in patients with sarcoma has not been investigated. The present prospective, non-randomised, non-interventional explorative study investigated the prognostic value of hs-CRP in patients with sarcoma. Patients referred to the sarcoma centre of Aarhus University Hospital (Aarhus, Denmark) were included between April 2014 and December 2020. Clinical data were obtained from the national quality sarcoma database and biomarkers other than hs-CRP were obtained from the clinical laboratory information system. The study cohort consisted primarily of patients with localised sarcoma. hs-CRP was significantly higher in patients with bone sarcoma (P=0.022) and soft tissue sarcoma (STS; P<0.001) compared with control patients. For STS, grade III tumours but not metastatic disease were associated with a higher hs-CRP level (P=0.0001). Elevated hs-CRP levels were associated with increased overall mortality [hazard ratio (HR), 1.91; 95% CI, 1.33–2.75; P=0.001]. Furthermore, elevated hs-CRP levels were also associated with decreased progression-free survival (HR, 1.64; 95% CI, 1.17–2.29; P=0.004). Furthermore, for patients with hs-CRP <8 mg/l, higher hs-CRP was associated with an increased risk of recurrent disease and reduced overall survival compared with those of patients with low hs-CRP. In conclusion, the present study demonstrated that hs-CRP was a prognostic factor for overall mortality and progression-free survival in patients with localised sarcoma at the time of diagnosis. Further studies are required to investigate the mechanism behind the association between hs-CRP and sarcoma prognosis and its potential use in clinical practice.

 sarcoma prognosis high-sensitive C-reactive protein survival Danish Cancer Society R248-Ai4683 The present study was supported by a grant from the Danish Cancer Society (grant no. R248-Ai4683). Introduction

Immunotherapy has transformed cancer treatment in less than a decade. It is used primarily in lung cancer and melanomas, significantly improving overall survival for patients suffering from these cancers (1,2). However, for sarcoma patients, the responses to checkpoint inhibitors have been disappointing, with an objective response rate of only about 4% and progression-free survival of 2.4 months (3). It is proposed that sarcomas possess immune-evading mechanisms, which are not addressed by existing immunotherapeutic drugs. Therefore, understanding the connection between the immune system and sarcomas is crucial in developing improved immunotherapy approaches for sarcoma patients.

One of the earliest and most robust responses to inflammatory conditions is an increased C-reactive protein (CRP) level. CRP is an acute-phase reactant primarily synthesised in the liver and shed into the bloodstream as an early response to acute inflammation and, to a lesser extent, during chronic inflammation (4,5). However, it has been shown that small quantities of CRP are produced by other than liver cells, such as smooth muscle cells (6,7), epithelial cells (8), fat cells (9), and even by immune-modulating macrophages (10). Different cytokines stimulate the production of CRP, such as IL-6, IL-1 and tumour necrosis factor. CRP interacts with the vascular endothelium, contributes to monocyte-endothelium adhesion, increases reactive oxygen species, and triggers platelet aggregation, as shown in rodents (11). Furthermore, CRP binds to lysophosphatidylcholine on the surface of dead or dying eukaryotic cells and bacteria, activating the complement system and an essential player in the innate immune response (12). In vitro studies have shown that CRP inhibits the proliferation and activation of T-cells and is, therefore, believed to play a part in both the innate and adaptive immune response (13).

It has been shown that conventional measurement of CRP, where a clinical cut point of 8 mg/l has been used, is a prognostic factor for patients with metastatic soft tissue sarcoma (14,15) and localised soft tissue sarcoma (16,17) and bone sarcoma (18,19). However, conventional CRP measurements are limited by the sensitivity to detect low values of CRP.

The high-sensitivity C-reactive protein (hs-CRP) quantifies CRP using an assay with a very low detection level. This allows for detecting and graduating low-grade inflammatory states such as obesity, diabetes, and cardiovascular disease (20,21). Hs-CRP is elevated in patients with soft tissue sarcoma (STS) compared to patients without sarcoma, and it has been proposed that it can be a diagnostic marker (22).

Macrophage-related biomarkers (sCD163 and SIRPα) were shown to be prognostic for overall survival and add prognostic values to a model containing known prognostic factors such as grade and age (23). However, the prognostic value and the association of hs-CRP levels and other inflammatory parameters have not been investigated in sarcoma patients.

This study investigates the association between serum hs-CRP and mortality among patients with primarily localised sarcoma.

 Materials and methods <sec> <title>Study cohort

This prospective, non-randomised, non-interventional explorative study investigates the prognostic value of hs-CRP in sarcoma patients. Patients referred to the sarcoma centre of Aarhus University Hospital, Denmark, between April 2014 and December 2020 were included in the present study. All patients signed an informed consent form before inclusion. The inclusion criteria were age (18 years or older) and willingness to donate a blood sample. Several patients were diagnosed with conditions other than sarcoma and served as a control group in this study. This control cohort was chosen because they were referred to the sarcoma centre with the suspicion of having a sarcoma. The control groups also included patients with desmoid tumour which is a local aggressive benign tumour and therefore not regarded as a sarcoma. However, these patients include patients with benign tumours such as lipomas or unspecific tissue reactions, to mention a few. After inclusion and following the national guidelines, patients were diagnosed and treated by an experienced multidisciplinary sarcoma team. Most patients were diagnosed with localised grade II and III STS and treated with surgery with or without pre- or postoperative radiation therapy. We have previously published results on immune suppressive macrophages using the same patient cohort (23).

 Data sources

Clinical data were obtained from the national quality sarcoma database, which contains comprehensive clinical information on individual sarcoma patients since 2009 in Denmark. The patients' records were reviewed to fill in missing values.

Biomarkers other than hs-sensitive CRP were obtained from the clinical laboratory information system (LABKA) database, which reports all blood tests taken according to the international nomenclature, properties, and Units (NPU) coding system. The values selected for analysis were the biomarkers closest to the day of the sarcoma diagnosis. LABKA results were included from 60 days before sarcoma diagnosis until 60 days after. Monocyte count, C-reactive protein (CRP), albumin, leucocyte count, and neutrophil count were selected for inclusion in this study. Each biomarker was categorised into normal or high/low according to the reference value at Aarhus University Hospital. High monocyte count was defined as ≥0.7×109 cells/l; low albumin levels were defined as <36 g/l. Elevated CRP was defined as values ≥8 mg/l; elevated leucocyte count was defined as values ≥10×109 cells/l, and high neutrophil count was defined as values ≥7×109 cells/l.

 Hs-sensitive CRP measurements

Before initiating the treatment, 30 ml of peripheral blood was collected in sodium citrate tubes and centrifuged at 2,000 or 2,500 g for ten minutes. The plasma was isolated and stored at −80°C until measurement according to the instructions by Danish Cancer Biobank, Bio- and GenomeBank, Denmark. Hs-sensitive CRP was measured with a chemiluminescent immunometric assay using the Atellica CH (Siemens, Germany). The upper limit of normal for hs-CRP is in our institution 3.0 mg/l, which is based on a Danish population. The hs-CRP values were used to allocate patients into high and low hs-CRP groups based on the medium value of hs-CRP.

 Prognostic profile

A predictive profile was created as described in our previous publication (23). In short, sCD163 and sSIRPα were divided into low or high groups based on their median serum concentration. CRP was separated into two categories, low or high, based on a threshold of 8 mg/l. Each categorical variable was assigned a score of 1 or 2, while the grade was assigned a score of 1, 2, or 3, depending on grade. The sum of all the assigned scores made up the final profile. The profile was then divided into three risk stratification groups: low-risk (score 4–5), intermediate risk (score 6–7), and high-risk (score 8–9). This profile is named profile 1 (23). In this paper, hs-CRP replaced the normal CRP measurement named profile 2.

 Statistical analysis

Clinical data and information on biomarkers were linked by the unique 10-digit civil personal registration (CPR) number, allowing for individual linkage between different reporting systems.

Patient, tumour, and treatment-related variables were reported as frequencies, percentages or continuous variables expressed as medians with interquartile range (IQR). The variables were compared using the chi-squared, Fisher's exact test or Wilcoxon rank-sum test, depending on the variable in question, and stratified by the median value of the hs-CRP in the STS patients. When the median value of multiple groups was compared, the Kruskal Wallis test was used with the Dunn's test for multiple comparisons when significant results were obtained. The median values of hs-CRP were used to categorise patients into a high and low hs-CRP group.

The primary endpoints were recurrent disease and overall survival (OS). Time to recurrent disease was defined as the interval between the primary diagnosis and the first recurrent, local or metastatic. OS was defined as the time from the date of diagnosis until the death of any causes. Kaplan-Meier curves were used to visualise survival and the log rank test was used for comparison of groups. The study period ended on October 15th 2022, and patients alive at this date were censored. Crude (univariate) and adjusted (multivariate) analyses were performed by using the Cox proportional hazard model. A test for proportional hazard was used before including an additional variable in the analysis. Based on the literature, the following variables were included in the adjusted analysis: age, size of the primary tumour, and grade. Tumour size and age were included as continuous variables; all other variables were analysed as categorical. ROC curve for cut point determination using the common criteria with the point on the ROC curve where the sensitivity and specificity of the test are equal (24). Akaike's information criteria (AIC) and Harrell's concordance index determined the model with the minimum AIC values, regarded as the best model. Likelihood-ratio tests were used to evaluate whether the addition of a potential prognostic profile contributed significantly to the models' prognostic value. A Two-sided P<0.05 was considered statistically significant. All analyses were performed using Stata (version 17.1) software.

 Results <sec> <title>Demographic data

A total of 292 patients were included in this study. The majority of patients were diagnosed with sarcoma (n=188: STS=152, bone sarcoma=36), compared to patients with desmoid fibromatosis (n=26) and patients diagnosed with conditions other than sarcoma or desmoid fibromatosis and therefore defined as control patients (n=78) (Fig. 1).

A significantly higher median hs-CRP was observed in patients with STS or bone sarcoma compared to patients with desmoid tumours and control patients (Table I). For STS, grade III tumours were associated with a higher hs-CRP level (P=0.0001), whereas patients with metastatic disease did not have a higher hs-CRP level than those with localised disease. The median hs-CRP values with a 95% confidence interval (CI) are depicted for each subtype of STS, bone sarcoma, desmoid tumours, and the control group in Fig. 2. The median hs-CRP level varies between the different histologic subtypes, with undifferentiated pleomorphic sarcoma having the highest levels. The difference between subtypes is significant compared to the control group and between UPS and the other histological groups.

The diagnostic value of hs-CRP was evaluated by comparing patients with localised sarcoma (bone and soft tissue sarcoma) with patients without sarcoma disease (desmoid fibromatosis and control patients). Fig. 3 shows the ROC curve with the sensitivity and specificity for hs-CRP. The area under the curve is 0.66 (95% CI: 0.59–0.73), and according to the ROC curve, the optimal cut point is 1.41 mg/l for local recurrence and 1.72 mg/l for overall survival.

 Prognostic value of hs-CRP

Only patients with localised soft tissue sarcoma (n=133) were included in analysing the effect of hs-CRP.

Clinical and tumour characteristics for patients with localised STS are shown in Table II. Patients were divided into a high and a low hs-CRP group based on the median value (2.19 mg/l). For patients with a high level of hs-CRP, the tumours were more extensive, were of a higher grade, and more patients experienced relapse after curative treatment. Six patients with a high level of hs-CRP also had increased leucocyte and neutrophil cell count; only two patients had increased liver enzymes. For patients with a low hs-CRP, 76% (95% CI: 64–84%) had not experienced a relapse of disease after five years of follow-up compared to 45% (95% CI: 32–57%) of the patient with a high hs-CRP. The 5-year overall survival for patients with a low hs-CRP was 83% (95% CI: 72–90%) compared to 59% (95% CI: 46–70%) for patients with a high hs-CRP.

The univariate analyses are shown in Table III. Patient age, tumour size, tumour grade, CRP level, and neutrophile count are prognostic for recurrent disease, and patient's age, tumour size, tumour grade, CRP level, albumin level, and neutrophile count were prognostic for overall survival. The multivariate analysis showed that a high hs-CRP was an independent prognostic factor for recurrent disease with a hazard ratio of 1.90 (95% CI: 1.03–3.52) and overall survival with a hazard ratio of 2.20 (95% CI: 1.13–4.29).

All patients with localised soft tissue sarcoma treated with curative intent were included in the model selection analysis (n=133). The c statistics showed that adding hs-CRP to known prognostic factors such as grade, tumour size, and age significantly improved the prognostic model. Table IV shows the c-statistics for comparing different models. Replacing normal CRP with hs-CRP lowered the AIC from 414 to 411 when evaluating the risk of disease or recurrent disease. However, when evaluating the ability to predict overall survival, AIC was unchanged (AIC=322 when CRP was included and AIC=323 when hs-CRP was included). Moreover, a model containing the macrophage markers sCD163 and sSIRPα significantly improved the model, including hs-CRP adjusted for age, tumour size, and grade (P=0.0021).

 Analyses of patients with a hs hs-CRP <8 mg/l

One hundred patients had a hs hs-CRP <8 mg/l, the reference level defining elevated CRP levels in Denmark. In a subgroup analysis including only patients with hs-CRP <8 mg/l, patients were divided into a high and low group based on the median hs-CRP value (1.32 mg/l). Age and tumour size were associated with high levels of hs-CRP. Again, high hs-CRP was associated with an increased risk of recurrent disease as the 5-year recurrence-free survival for patients with a hs-CRP below the median value was 74% (95% CI: 57–85%) compared to patients with a hs-CRP above the median value 62% (95% CI: 48–73%) see Fig. 4. Furthermore, a higher hs-CRP was associated with a reduced overall survival demonstrated by the 5-year overall survival for patients with a hs-CRP below the median value: of 85% (95% CI: 69–93%) compared to patients with a hs-CRP above the median value: 72% (95% CI: 59–82%) see Fig. 5.

 Discussion

This study investigated the association between serum high-sensitivity C-reactive protein (hs-CRP) and progression-free survival in patients with localised sarcoma treated with curative intent. The results showed that hs-CRP is an adverse prognostic factor for progression-free survival in these patients, even after adjusting for known prognostic factors.

Several studies have investigated the relationship between hs-CRP and the presence of cancer. In this article, the optimal cut-off for the prognostic values of hs-CRP was determined as the median values and not the cut-off based on the ROC curve. The ROC curve cut-off is based on the endpoint investigated and, therefore, this can only be applied when using a test and validation cohort. Other studies used the median value.

A study by Lee et al (25) found that hs-CRP levels are significantly higher in different cancer patients than in healthy controls, with a median hs-CRP of 0.77 and 0.59 mg/l for men and women, respectively. Nakamura et al (22) found a threshold level of 0.95 mg/l for discriminating between sarcoma and healthy control patients. These results are in alliance with our findings. However, the median level in our study was higher than that reported by Lee et al (25) and Nakamura et al (22). This difference could be due to the body mass index (BMI) being higher in Caucasians than the Asian participants (25). We did not have information about BMI in our study. Nevertheless, this is supported by Cho et al (26), who found higher hs-CRP among breast cancer survivors with a larger body mass index than patients with a lower BMI.

Lee et al (25) found that a one mg/l increase in hs-CRP was associated with increased mortality after a 17-year follow-up, but the association was found only in women. Our study did not show any difference in the CRP levels between sexes; however, we observed that an increase in CRP by one mg/l led to a 2.69 increased risk of dying (median follow-up time was 6.9 years). Similarly, in the study by Nakamura et al (22), hs-CRP levels were associated with poor prognosis and decreased survival in patients with soft tissue sarcoma. All these results suggest a relationship between hs-CRP and sarcoma.

Desmoid tumours included in this study represent locally aggressive neoplasms that do not metastasise but are challenging to treat. As for other cancers, the tumour immune microenvironment (TME) is essential when considering immunotherapy. One study investigated the immune expression markers of 33 tissue samples from patients suffering from desmoid tumours. This study concluded that desmoid tumours have a solid immune infiltration in the tumour margins; however, PD-L1 was not present in the tumour cells. PD-L1 is a target for immunotherapy (27). In our study, the level of hs-CRP was not elevated compared to the control group, indicating that the inflammation in these patients may not play a pivotal role.

Several mechanisms have been proposed to explain the link between CRP and cancer. One mechanism involves the role of inflammation in promoting tumour initiation and progression (28,29). CRP is a marker of systemic inflammation, and chronic inflammation has been linked to several types of cancer, including sarcoma (15,17,18). In addition, an in vivo experiment has shown that CRP influences the tumour microenvironment by promoting suppressive tumour-associated macrophages (30). These macrophages promote angiogenesis, an essential process for tumour growth and metastases (31). Another proposed mechanism involves the role of CRP as a marker of immune system dysfunction (13).

The strengths of this study include its prospective design, large sample size and the use of a highly sensitive assay to measure hs-CRP, which allowed for the detection of low levels of inflammation. In addition, the comprehensive clinical data and biomarker measurements allowed for the adjustment of potential confounders in the analysis. Patients suspected of having sarcoma but who turned out to have a benign condition comprise the control group in this study. It is a strength in this study as the destining of sarcoma and non-sarcoma patients is essential. However, there are also some limitations to this study. First, the study evaluated soft tissue sarcoma patients as one collective group of patients, which could blur variations between different histological subtypes. Second, the study was not designed to investigate the underlying mechanisms linking hs-CRP and mortality in sarcoma patients, and further studies are needed to elucidate these mechanisms.

However, it is known that tumour-associated macrophages (TAM) are believed to be one of the major immunosuppressive components in cancers. The association between high levels of serum CRP and the number of tumour-associated macrophages of the subtypes CD204 and CD163 has been shown for patients with hepatocellular cancer (32). The activation of monocytes to immunosuppressive macrophages is believed to be facilitated by CRP along with the activation of endothelial cells, platelets and the complement system, which are important in the innate immune response leading to macrophage recruitment and activation (33). Fig. 6 is a proposed association between CRP and the innate immune system.

Additionally, it is essential to note that sarcoma is a rare and complex disease, and more research is needed to understand the underlying mechanisms fully and develop new treatments and therapies for sarcoma patients. Using biomarkers such as hs-CRP and inflammation can help improve the diagnosis, prognosis, and management of sarcoma, but it is crucial to use them in conjunction with other diagnostic tests and imaging studies to make a definitive diagnosis. Furthermore, integrating these biomarkers with other omics technologies can provide a comprehensive understanding of the disease and help identify new diagnostic and therapeutic targets.

In conclusion, High-Sensitive C-reactive protein and inflammation are important biomarkers where high levels of the individual biomarker are linked to more advanced stages of sarcoma and increased risk of mortality. These biomarkers can also influence a patient's response to therapy; therefore, treatment plans for sarcoma must be tailored to account for an individual inflammatory profile. More research is needed to understand the underlying mechanisms and develop new treatments and therapies for sarcoma patients.

 Acknowledgements

The Danish CancerBiobank (Aarhus, Denmark) is acknowledged for handling and storing biological material.

 Availability of data and materials

The data that support the findings of this study are available from the Danish Regions but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of Danish Ethical commission.

 Authors' contributions

NAP, TBH, HJM and BSP conceived the study. NAP and BSP wrote the original draft. NAP, TBH, HJM and BSP wrote, reviewed and edited the manuscript. BSP and NAP confirm the authenticity of all the raw data. All authors have read and approved the final manuscript.

 Ethics approval and consent to participate

The regional Ethics Committee of Denmark (approval no. 1-10-72-58-14; Region Midtjylland, Viborg, Denmark) and the Danish Agency of Data Protection (approval no. 1-16-02-112-14; Valby, Denmark) have approved the study. Before obtaining the blood samples, written informed consent was obtained from the patients.

 Patient consent for publication

Not applicable.

 Competing interests

The authors declare that they have no competing interests.

 References SchwartzbergLKorytowskyBPenrodJRZhangYLeTKBatenchukCKrugLReal-world clinical impact of immune checkpoint inhibitors in patients with advanced/metastatic non-small cell lung cancer after platinum chemotherapyClin Lung Cancer20287296.e4201910.1016/j.cllc.2019.04.00431130450 LiJGuJEfficacy and safety of PD-1 inhibitors for treating advanced melanoma: A systematic review and meta-analysisImmunotherapy1012931302201810.2217/imt-2018-011630474476 SaerensMBrusselaersNRotteySDecruyenaereACreytensDLapeireLImmune checkpoint inhibitors in treatment of soft-tissue sarcoma: A systematic review and meta-analysisEur J Cancer152165182202110.1016/j.ejca.2021.04.03434107450 Kolb-BachofenVA review on the biological properties of C-reactive proteinImmunobiology183133145199110.1016/S0171-2985(11)80193-21937562 ClyneBOlshakerJSThe C-reactive proteinJ Emerg Med1710191025199910.1016/S0736-4679(99)00135-310595891 CalabróPWillersonJTYehETHInflammatory cytokines stimulated C-reactive protein production by human coronary artery smooth muscle cellsCirculation10819301932200310.1161/01.CIR.0000096055.62724.C514530191 GuoFLiuJTWangCJPangXMPravastatin inhibits C-reactive protein generation induced by fibrinogen, fibrin and FDP in isolated rat vascular smooth muscle cellsInflamm Res61127134201210.1007/s00011-011-0396-422089528 XingXQDuanSWuXWGanYZhaoSPChenPWuSJAtorvastatin reduces lipopolysaccharide-induced expression of C-reactive protein in human lung epithelial cellsMol Med Rep4753757201121584495 Peyrin-BirouletLGonzalezFDubuquoyLRousseauxCDubuquoyCDecourcelleCSaudemontATachonMBéclinEOdouMFMesenteric fat as a source of C reactive protein and as a target for bacterial translocation in Crohn's diseaseGut617885201210.1136/gutjnl-2011-30037021940721 LiMLiuJTPangXMHanCJMaoJJEpigallocatechin-3-gallate inhibits angiotensin II and interleukin-6-induced C-reactive protein production in macrophagesPharmacol Rep64912918201210.1016/S1734-1140(12)70886-123087143 BanaitTWanjariADanadeVBanaitSJainJRole of high-sensitivity C-reactive protein (Hs-CRP) in non-communicable diseases: A reviewCureus14e30225202236381804 Du ClosTWFunction of C-reactive proteinAnn Med32274278200010.3109/0785389000901177210852144 YoshidaTIchikawaJGiuroiuILainoASHaoYKrogsgaardMVassalloMWoodsDMStephen HodiFWeberJC reactive protein impairs adaptive immunity in immune cells of patients with melanomaJ Immunother Cancer8e000234202010.1136/jitc-2019-00023432303612 NakamuraTKatagiriHShidoYHamadaSYamadaKNaganoAYamadaSTsukushiSIshimuraDMatsumineAAnalysis of factors for predicting survival in soft-tissue sarcoma with metastatic disease at initial presentationAnticancer Res3731371341201728551655 Aggerholm-PedersenNMaretty-KongstadKKellerJSafwatASerum biomarkers as prognostic factors for metastatic sarcomaClin Oncol (R Coll Radiol)31242249201910.1016/j.clon.2019.01.01130744933 NakamuraTMatsumineAMatsubaraTAsanumaKUchidaASudoAClinical significance of pretreatment serum C-reactive protein level in soft tissue sarcomaCancer11810551061201210.1002/cncr.2635321761398 Maretty-KongstadKAggerholm-PedersenNKellerJSafwatAA validated prognostic biomarker score for adult patients with nonmetastatic soft tissue sarcomas of the trunk and extremitiesTransl Oncol10942948201710.1016/j.tranon.2017.09.00229031130 Aggerholm-PedersenNMaretty-KongstadKKellerJBaerentzenSSafwatAThe prognostic value of serum biomarkers in localized bone sarcomaTransl Oncol9322328201610.1016/j.tranon.2016.05.00627567955 FunovicsPTEdelhauserGFunovicsMALauxCBerzaczyDKubistaBKotzRIDominkusMPre-operative serum C-reactive protein as independent prognostic factor for survival but not infection in patients with high-grade osteosarcomaInt Orthop3515291536201110.1007/s00264-011-1208-821249357 KandeloueiTAbbasifardMImaniDAslaniSRaziBFasihiMShafiekhaniSMohammadiKJamialahmadiTReinerŽSahebkarAEffect of statins on serum level of hs-CRP and CRP in patients with cardiovascular diseases: A systematic review and meta-analysis of randomized controlled trialsMediators Inflamm20228732360202210.1155/2022/873236035125965 GhuleAKambleTKTalwarDKumarSAcharyaSWanjariAGaidhaneSAAgrawalSAssociation of serum high sensitivity C-reactive protein with pre-diabetes in rural population: A two-year cross-sectional studyCureus13e19088202134868746 NakamuraTMatsumineAIinoTMatsubaraTAsanumaKUchidaASudoARole of high-sensitivity C-reactive protein in the differentiation of benign and malignant soft tissue tumorsAnticancer Res34933936201424511035 Aggerholm-PedersenNFriisHNBaad-HansenTMøllerHJSandfeld-PaulsenBMacrophage biomarkers sCD163 and sSIRPα in serum predict mortality in sarcoma patientsCancers (Basel)151544202310.3390/cancers1505154436900335 UnalIDefining an optimal cut-point value in ROC analysis: An alternative approachComput Math Methods Med20173762651201710.1155/2017/376265128642804 LeeSAKwonSOSongMChoiJYShinAShuXOZhengWLeeJKKangDThe association of serum high-sensitivity C-reactive protein level with the risk of site-specific cancer mortality: The health examinees (HEXA) study cohortAm J Epidemiol19120022013202210.1093/aje/kwac14135916370 ChoHJSongSKimZYounHJChoJMinJWKimYSChoiSWLeeJEAssociations of body mass index and weight change with circulating levels of high-sensitivity C-reactive protein, proinflammatory cytokines, and adiponectin among breast cancer survivorsAsia Pac J Clin Oncol19113125202310.1111/ajco.1377935590398 SiozopoulouVMarcqEJacobsJZwaenepoelKHermansCBraunsJPauwelsSHuysentruytCLammensMSomvilleJDesmoid tumors display a strong immune infiltration at the tumor margins and no PD-L1-driven immune suppressionCancer Immunol Immunother6815731583201910.1007/s00262-019-02390-031511925 AggarwalBBShishodiaSSandurSKPandeyMKSethiGInflammation and cancer: how hot is the link?Biochem Pharmacol7216051621200610.1016/j.bcp.2006.06.02916889756 CoussensLMWerbZInflammation and cancerNature420860868200210.1038/nature0132212490959 MarjonKDMarnellLLMoldCDu ClosTWMacrophages activated by C-reactive protein through Fc gamma RI transfer suppression of immune thrombocytopeniaJ Immunol18213971403200910.4049/jimmunol.182.3.139719155486 FujiwaraTHealeyJOguraKYoshidaAKondoHHataTKureMTazawaHNakataEKunisadaTRole of tumor-associated macrophages in sarcomasCancers (Basel)131086202110.3390/cancers1305108633802565 WangYLiZHuangZYuXZhengLXuJC-reactive protein is an indicator of the immunosuppressive microenvironment fostered by myeloid cells in hepatocellular carcinomaFront Oncol11774823202210.3389/fonc.2021.77482335070979 McFadyenJDKieferJBraigDLoseff-SilverJPotempaLAEisenhardtSUPeterKDissociation of C-reactive protein localizes and amplifies inflammation: Evidence for a direct biological role of C-reactive protein and its conformational changesFront Immunol91351201810.3389/fimmu.2018.0135129946323

Patient selection according to histological subtype, tumour grade and missing value of hs-CRP. The number indicated in the parentheses is the number of patients with a hs-CRP measurement. hs-CRP, high-sensitivity C-reactive protein.

Median hs-CRP value in mg/l according to histological subtype and for control patients. The median value is shown with a 95% confidence interval. The comparisons among different groups were performed using the Kruskal-Wallis test with Dunn's test as post hoc analysis. For patients with liposarcoma (P<0.0001), UPS (P<0.0001), myxofibromsarcoma (P=0.02), other (P=0.01) and bone sarcoma (P=0.02) a significant difference was observed compared with control patients. For patients with UPS compared with the other histological groups, a significant difference was observed for liposarcoma (P=0.003), myxofibrosarcoma (P=0.04), angiosarcoma (P=0.01), others (P=0.007), chondrosarcoma (P=0.005) and desmoid tumours (P<0.001). Hs-CRP, high-sensitivity C-reactive protein; UPS, undifferentiated pleomorphic sarcoma.

Diagnostic value hs-CRP in all patients with localised sarcoma (n=167) combined with all patients with benign conditions (n=104), including patients without sarcoma and desmoid tumours. All patients with metastatic disease were excluded from the analysis. The risk of relapse includes all patients (n=133) with localised STS at the time of diagnosis. The risk of death included all patients (n=133) with localized STS at the time of diagnosis. hs-CRP, high-sensitivity C-reactive protein; STS, soft tissue sarcoma.

Time to relapse for patients with localised soft tissue sarcoma at the time of diagnosis. (A) All patients with a hs-CRP measurement. (B) Patients with a hs-CRP level <8 mg/l. The log-rank test was used to compare groups. hs-CRP, high-sensitivity C-reactive protein.

Overall survival for patients with localised soft tissue sarcoma at diagnosis. (A) All patients with hs-CRP measurements. (B) Patients with a hs-CRP level <8 mg/l. The log-rank test was used to compare groups. hs-CRP, high-sensitivity C-reactive protein.

Illustration of the proposed role of CRP and the tumour microenvironment. The liver produces pCRP as a result of inflammatory cytokines, such as IL-6 and IL1β, excreted from cancer cells. pCRP is dissociated into mCRP on the surface of different cells, such as monocytes (1) and endothelial cells (2), through the binding to a ligand called phosphocholine. 1, mCRP together with other inflammatory markers might promote the production of immune-inhibitory TAMs called M2-directed macrophages. 2, The recruitment and activation of M2-directed TAMs could also be due to the endothelial activation by CRP. The endothelial activation might facilitate leukocyte recruitment, including recruitment of monocytes to the tumour. The recruitment of monocytes to the tumour increases and promotes the development into TAMs (M2 directed). Furthermore, platelet and complement activation are facilitated, which are essential in the innate immune response. CRP, C-reactive protein; mCRP, monomeric CRP; pCRP, pentamer of CRP; TAM, tumour-associated macrophage.

Median concentration of hs-CRP for different subgroups at diagnosis.

CRP, mg/l
Clinical variables No. Median Range Kruskal Wallis test P-value Dunn's test P-value
Histology 0.0004
  Control 78 1.09 0.20–8.51
  Desmoid tumours 26 0.68 0.20–23.11 0.340a
  Bone sarcoma 34 2.16 0.23–20.36 0.018a
  Soft tissue sarcoma (localized only) 133 2.19 0.28–71.69 <0.001a
Soft tissue sarcoma 0.0001
  Grade I 22 1.27 0.23–5.42
  Grade II 38 1.34 0.26–19.75 0.220b
  Grade III 73 3.43 0.46–109.17 <0.001b; <0.001c
Soft tissue sarcoma 0.9244
  Localised 133 2.23 0.28–71.69
  Metastatic 17 1.53 0.21–38.78

Compared with the control group;

Compared with grade I;

Compared with grade II. The range is reported as the 5-95th percentile. The Kruskal-Wallis test was used with Dunn's test for multiple comparisons when significant results were obtained. hs-CRP, high-sensitivity C-reactive protein.

Clinical, tumour and treatment characteristics of patients with localised soft tissue sarcoma and CRP measurements (n=133).

Characteristics Total Low CRP High CRP P-value
Sex, n (%)
  Male 59 (44) 29 (43) 30 (45) 0.801
  Female 74 (56) 38 (57) 36 (55)
Median age, years (p5-p95) 67 (27–85) 65 (25–84) 69 (47–85) 0.06
Histological subtype, n (%)
  Liposarcoma 32 (24) 15 (22) 17 (26) 0.792
  UPS 23 (17) 8 (12) 15 (23)
  Leiomyosarcoma 17 (13) 9 (13) 8 (12)
  Myxofibrosarcoma 17 (13) 10 (15) 7 (11)
  Angiosarcoma 8 (6) 4 (6) 4 (6)
  Synovial sarcoma 7 (5) 4 (6) 3 (5)
  Others 29 (22) 17 (25) 12 (18)
Median tumour size, cm (p5-p95) 6 (1–23) 5 (1–16) 8 (1–27) <0.01
Tumour grade, n (%)
  Low 22 (17) 16 (24) 6 (9) 0.002
  Intermediate 38 (29) 23 (34) 15 (23)
  High 73 (55) 28 (42) 45 (68)
Treatment, n (%)
  Surgery 132 (99) 67 (98) 65 (100) 0.496
  Radiation therapy 50 (38) 21 (31) 29 (44) 0.154
Treatment intent, n (%)
  Curativea 132 (99) 67 (100) 65 (98) 0.312
Relapse, n (%)
  Yes 51 (39) 17 (25) 34 (52) 0.001
  No 82 (61) 50 (75) 32 (48)

1 patient had local disease but was not treated with curative intent. The χ2 test and Fisher's exact test were used to compare category variables. Fisher's exact test was used if the expected count in >20% of the cells of the analysed contingency table was 5 or fewer. The Wilcoxon rank sum test was used for continuous variables. CRP, C-reactive protein; p5–p95, 5-95th percentile; UPS, undifferentiated pleomorphic sarcoma.

Univariate analyses in patients with localised soft tissue sarcoma (n=133) using Cox regression analysis.

Risk of relapse Overall survival
Characteristics Hazard ratio 95% CI P-value Hazard ratio 95% CI P-value
Age, years 1.03 1.01–1.05 0.007 1.06 1.03–1.09 <0.001
Sex
  Female 1 1
  Male 1.21 0.68–2.16 0.513 1.22 0.67–2.21 0.509
Size, cm 1.04 1.01–1.08 0.018 1.05 1.01–1.08 0.011
Grade
  I 1 1
  II 1.17 0.35–3.90 0.793 2.96 0.64–13.75 0.165
  III 4.26 1.52–11.97 0.006 8.22 1.97–34–39 0.004
Serum biomarkers
  Monocytes count, ×109 cells/l 1.09 0.55–2.16 0.796 1.60 0.83–3.10 0.159
  hs-CRP, mg/l 2.74 1.52–4.92 0.001 2.62 1.40–4.90 0.002
  CRP, mg/l 1.95 1.08–3.51 0.027 2.22 1.19–4.12 0.012
  Albumin, g/l 1.60 0.82–3.15 0.165 2.06 1.03–4.09 0.040
  Leucocyte count, ×109 cells/l 1.39 0.65–2.98 0.395 2.02 0.99–1.12 0.054
  Neutrophil count, ×109 cells/l 2.29 1.10–4.78 0.028 3.06 1.53–6.13 0.002

Monocytes count was categorized into normal ≤0.7×109 cells/l and high >0.7×109 cells/l levels. CRP was categorized into normal ≤8 mg/l and high >8 mg/l levels. Albumin was categorized into normal ≥36 g/l and low <36 g/l levels, Leucocyte count was categorized into normal ≤10×109 cells/l and high >10×109 cells/l levels. Neutrophil count was categorized into normal ≤7×109 cells/l and high >7×109 cells/l levels. hs-CRP, high-sensitivity C-reactive protein.

Model fitting.

Prognostic models Relapse Survival
Model AIC C-index AIC C-index
Grade 456 0.66 397 0.67
Age 466 0.61 392 0.70
Tumour size 468 0.63 409 0.65
hs-CRP 461 0.62 404 0.64
Grade + hs-CRP 451 0.79 394 0.71
Grade + age + hs-CRP 448 0.73 373 0.77
Grade + age + tumour size + hs-CRP 446 0.74 369 0.79
Grade + age + tumour size 448 0.73 373 0.78
Grade + age + tumour size + hs-CRP + sCD163 and sSIRPα 450 0.74 362 0.79

AIC, Akaike's information criteria; hs-CRP, high-sensitivity C-reactive protein; C-index, concordance index; sCD163, soluble CD163; sSIRPα, soluble signal regulatory protein α.