Fax: (011) + 81-59-231-5211
Article first published online: 14 JUL 2011
Copyright © 2011 American Cancer Society
Volume 118, Issue 4, pages 1055–1061, 15 February 2012
How to Cite
Nakamura, T., Matsumine, A., Matsubara, T., Asanuma, K., Uchida, A. and Sudo, A. (2012), Clinical significance of pretreatment serum C-reactive protein level in soft tissue sarcoma. Cancer, 118: 1055–1061. doi: 10.1002/cncr.26353
Each author certifies that his institution has approved the human protocol for this investigation and that all investigations were conducted in conformity with the ethical principles of research.
We thank Yamada Tomomi (Department of Translational Medical Science at Mie University Graduate School of Medicine) for her valuable help with the statistical analyses.
- Issue published online: 3 FEB 2012
- Article first published online: 14 JUL 2011
- Manuscript Accepted: 20 MAY 2011
- Manuscript Revised: 18 APR 2011
- Manuscript Received: 20 FEB 2011
- C-reactive protein;
- soft tissue sarcoma;
The aim of this study was to determine whether circulating C-reactive protein (CRP) levels before treatment predict the overall survival and disease-free survival in soft tissue sarcoma patients.
A total of 102 primary soft tissue sarcoma patients from 2003 to 2009 were retrospectively reviewed. The CRP levels were obtained before treatment for all patients. The patients who presented with metastases at diagnosis were excluded from this study.
Elevated CRP levels were seen in 18 patients. The tumor histological grade and American Joint Committee on Cancer stage in the patients with elevated CRP levels were significantly higher than those in patients with normal CRP levels. Patients with elevated CRP levels before initial treatment had a poorer overall survival than patients with normal CRP levels (P = .01). The overall survival estimates at 3 and 5 years were 75.3% and 53.8%, respectively, versus 90.3% and 81.3%, respectively. Patients with elevated CRP levels before initial treatment had poorer event-free survival after initial treatment than patients with normal CRP levels (P < .001). The event-free survival estimates at 2 and 5 years were 53.2% and 33.2%, respectively, versus 83.2% and 81.3%, respectively. A multivariate analysis also showed the preoperative CRP level to be an independent predictor of events.
The pretreatment serum CRP level may be a marker of aggressive tumor characteristics. Pretreatment elevated CRP levels were found to be a poor prognostic factor for overall survival in a univariate analysis, and for disease-free survival in a multivariate analysis, for soft tissue sarcoma patients. Cancer 2012;. © 2011 American Cancer Society.
C-reactive protein (CRP) is known to be induced in the acute phase of an inflammatory response. The production of CRP in hepatocytes is principally induced at the transcriptional level by interleukin (IL) 6, and can be enhanced by IL-1 beta.1 Increased CRP levels are reported in patients with many diseases, including cardiovascular disease,2, 3 type 2 diabetes,4 arthritis, and many types of cancer.5-13 Elevated CRP levels are associated with poor survival in patients with many types of cancer, such as renal cell carcinoma, gastric cancer, breast cancer, colorectal cancer, inoperative nonsmall cell lung cancer, prostate cancer, gastrointestinal cancer, pancreatic cancer, and esophageal cancer.5-13 The inflammatory status may also be a prognostic factor for soft tissue sarcoma. However, few studies have so far assessed the relation between biomarkers of inflammation and survival in soft tissue sarcoma patients.14 The aim of this study was to determine whether circulating CRP levels before treatment predict overall survival and disease-free survival in soft tissue sarcoma patients.
MATERIALS AND METHODS
A total of 102 primary soft tissue sarcoma patients treated with surgical resection (100 patients) or radiation therapy (2 patients) for primary tumor from April 2003 to December 2009 were retrospectively reviewed. Written informed consent was obtained from each patient. The patients who presented with metastases at diagnosis were excluded from this study. Pretreatment workup included brain, lung, abdomen, and pelvic computed tomography scans with and without contrast medium. The histopathological diagnosis and tumor grade determined using the French Federation of Cancer Centers Sarcoma Group system for all patients was reviewed and confirmed by independent pathologists. The serum CRP levels were measured using a Denka Seiken (Tokyo, Japan) X-2 autoanalyzer as part of a routine clinical examination. The CRP levels were obtained before treatment for all patients. If the CRP levels were elevated pretreatment, the CRP levels were measured once again just before the treatment. The latest CRP levels were used for these patients in the present study. The normal serum CRP level is ≤0.3 mg/dL at our institution. A clinicopathological analysis was performed comparing the CRP levels to the various factors, including age, sex, body mass index (BMI), history of hypertension and/or diabetes mellitus, tumor size, depth, tumor site, and tumor histological grade.
The statistical association of the clinicopathological factors was evaluated using the Mann-Whitney U test for quantitative data, and the chi-square test or Fisher exact test for qualitative data. The duration of overall and disease-free survival was defined as the interval between the date of initial treatment for the primary tumor and that of death and local recurrence or metastasis, respectively. Survival curves were constructed using the Kaplan-Meier method. The log-rank test was used to compare the survival of the patients with high and normal CRP levels. A multivariate analysis was performed using a Cox proportional hazard model. The variables included in the multivariate analysis were the significant factors identified in the univariate analysis. A value of P < .05 was considered to be significant in all statistical analyses.
Patient, Tumor, and Treatment Characteristics
The mean age at diagnosis was 55 years (range, 10-89 years). There were 56 male and 46 female patients. The mean follow-up period after the date of the initial treatment was 35 months (range, 7-87 months). The mean tumor size at diagnosis was 8.4 cm (range, 2-30 cm). The mean BMI was 22.8 kg/m2 (range, 15.4-30.7 kg/m2). Nineteen patients had cardiovascular disease, and 4 patients had diabetes mellitus. Forty-four patients had grade 3 sarcomas according to the French Federation of Cancer Centers Sarcoma Group system, 40 patients were histologically classified as having grade 1, and 18 patients had grade 2. The 102 patients were histologically classified as follows: 22 well-differentiated liposarcomas, 13 myxofibrosarcomas, 12 leiomyosarcomas, 11 malignant fibrous histiocytomas (MFHs), 9 myxoid liposarcomas, 6 malignant peripheral nerve sheath tumors (MPNSTs), 5 extraskeletal chondrosarcomas, 5 synovial sarcomas, 5 dermatofibrosarcoma protuberances, and 14 other tumors. The primary tumor sites were the thighs (n = 32), legs (n = 15), buttocks (n = 9), back (n = 8), upper arms (n = 6), forearms (n = 6), shoulders (n = 4), inguinal lesions (n = 4), and other sites (n = 16). The tumor depth was superficial in 34 patients and deep in 68 patients. According to the criteria established by the American Joint Committee on Cancer (AJCC) classification of soft tissue sarcomas, 40 patients were classified as stage 1, 37 were stage 2, and 25 were stage 3. One hundred of the 102 soft tissue sarcoma patients underwent surgery, and 2 patients received radiation therapy for the primary site tumor. Fifteen of the 102 sarcoma patients underwent adjuvant chemotherapy.
CRP as a Tumor Marker
Elevated CRP levels (>0.3 mg/dL) were seen in 18 patients (range, 0.4-20.4 mg/dL, average; 2.3 mg/dL). These included 7 with leiomyosarcomas (58%), 4 with MFHs (22%), 2 with well-differentiated liposarcomas (11%), 2 with MPNSTs (11%), and 3 with other tumors (17%) (Table 1). The CRP levels became normal after resection of the primary tumor in 16 patients. The CRP levels were obtained before postoperative adjuvant chemotherapy for the patients who received chemotherapy. The CRP level remained elevated in 1 patient who received radiation therapy for the primary lesion. The CRP level also remained elevated in another patient who developed a wound infection after surgery. Eleven of the 18 patients with elevated CRP levels (61%) developed a local recurrence and/or metastasis during the postoperative follow-up period. Five of these 11 showed re-elevation of the serum CRP level. Another patient maintained elevated CRP levels after radiation therapy before the diagnosis of lung metastasis. In addition, 1 of the 11 patients with metastases continued to have elevated CRP levels because of continuous local infections after surgery at the primary tumor site.
|Histology (No. of Patients)||Serum CRP Levels, mg/dL|
|Other sarcoma, n=37||37|
Thirteen of the 84 patients in the group with normal CRP levels (15%) developed local recurrence and/or metastasis. Two of the 13 patients who developed a local recurrence showed elevation of their CRP levels at the time of detection of the recurrence.
Clinicopathological Significance of CRP
The relation between the clinicopathological features and preoperative serum CRP level is shown in Table 2. The tumor histological grade and AJCC stage in the patients with elevated CRP levels were significantly higher than those in patients with normal CRP levels. Other factors, such as age, sex, BMI, history of hypertension and/or diabetes, and depth had no significant relation with the CRP level. The tumor size showed a marginally significant difference between the patients with elevated CRP levels (mean, 10.4 cm) and those with normal CRP levels (mean, 8 cm) in the quantitative analysis (P = .057).
|Characteristics||CRP ≤ 0.3||CRP > 0.3||P|
|HTN and/or diabetes|
|1 or 2||70||7||.0002|
Overall Survival and Predictors of Mortality
Patients with elevated CRP levels before initial treatment had poorer overall survival than patients with normal CRP levels (P = .01). The overall survival estimates at 3 and 5 years were 75.3% and 53.8%, respectively, versus 90.3% and 81.3%, respectively (Fig. 1). A univariate analysis also revealed significantly poorer predictive values for tumor histological grade (P < .001; Table 3). Elevated CRP levels lost their prognostic significance in the multivariate analysis.
|Patient Characteristics||No.||Overall Survival||P|
|3 Years||5 Years|
In the 62 patients with high-grade sarcomas, which include grade 2 and 3 sarcoma according to the French Federation of Cancer Centers Sarcoma Group system, there was prognostic significance of elevated CRP levels compared with normal CRP levels (P = .045). The overall survival estimates at 3 and 5 years were 69.6% and 41.8%, respectively, versus 82.3% and 72.0%, respectively (Table 4). A univariate analysis also revealed significantly poorer predictive values for positive surgical margin (P = .04). Elevated CRP levels and surgical margin lost their prognostic significance in the multivariate analysis.
|Patient Characteristics||No.||Overall Survival||P|
|3 Years||5 Years|
Event-Free Survival Rate and Predictors of Events
Patients with elevated CRP levels before initial treatment had poorer event-free survival after initial treatment than patients with normal CRP levels (P < .001). The event-free survival estimates at 2, 3, and 5 years were 53.2%, 33.2%, and 33.2%, respectively, versus 83.2%, 81.3%, and 81.3%, respectively (Fig. 2). A univariate analysis also revealed significantly poorer outcomes for patients with high tumor histological grade (P < .001) and older age (P = .04; Table 5). A multivariate analysis showed that the pretreatment CRP levels and tumor histological grade were independent predictors of events (Table 6).
|Patient Characteristics||No.||5-Year Disease-Free Survival, %||P|
|Variables||HR (95% CI)||P|
|Female sex||0.84 (0.37-1.9)||.68|
|Grade 3||4.34 (1.67-11.31)||.003|
|CRP ≤0.3||0.36 (0.16-0.84)||.017|
In the 62 patients with high-grade sarcoma, a univariate analysis revealed significantly poorer outcomes for patients with an elevated CRP level (P = .0014) (Table 7). The event-free survival estimates at 3 and 5 years were 17.1% and 17.1%, respectively, versus 68.9% and 68.9%, respectively. A univariate analysis also revealed significantly poorer outcomes for patients with positive surgical margin (P = .003) and adjuvant chemotherapy (P = .01). A multivariate analysis showed that the preoperative CRP level was an independent predictor of events (Table 8).
|Patient Characteristics||No.||Overall Survival||P|
|3 Year||5 Year|
|Variables||HR (95% CI)||P|
|Positive surgical margin||2.64 (0.96-7.28)||.06|
|Adjuvant chemotherapy||0.68 (0.26-1.78)||.43|
|CRP ≤0.3||0.40 (0.16-0.98)||.046|
The overall 5-year survival rate in patients with soft tissue sarcomas of all stages remains only 50% to 60%.15 Several large series have defined several clinical prognostic factors that are associated with survival.15-18 The measurement of CRP is familiar to most physicians, and the measurement of the CRP level is easy and cost effective. Elevated preoperative serum CRP levels are found in a variety of cancers, and an elevated serum CRP level is an indicator of a poorer prognosis in many cancers, such as renal cell carcinoma, gastric cancer, breast cancer, colorectal cancer, inoperative nonsmall cell lung cancer, prostate cancer, gastrointestinal cancer, pancreatic cancer, and esophageal cancer.5-13 The current univariate analyses showed that elevated CRP levels were associated with decreased overall survival, and univariate and multivariate analyses showed an association of elevated CRP levels with decreased disease-free survival in 102 soft tissue sarcoma patients. Furthermore, we evaluated whether elevated CRP levels predict the overall survival and the disease-free survival in 62 patients with high-grade sarcomas. Two variables related to the treatment, surgical margin and adjuvant chemotherapy, were added in this analysis. Similarly, elevated CRP levels were associated with decreased overall survival, and univariate and multivariate analyses showed an association of elevated CRP levels with decreased disease-free survival in only high-grade sarcoma patients.
There are several possible mechanisms for the relation between CRP and cancer. First, tumor growth can cause tissue inflammation, and hence elevate the CRP levels.19 Second, the immune response of the host to tumor growth itself can cause an elevation in CRP levels.20, 21 Third, there is evidence that cancer cells can increase the production of inflammatory proteins, which could explain the high CRP levels in patients with malignancies.22-24 In the tumor microenvironment, inflammatory cells produce cytokines, particularly IL-6, in response to tumor cells, tissue necrosis, and the associated inflammation. Rutkowski et al23 showed that increased serum levels of IL-6 were found in >50% of soft tissue sarcoma patients, and that its expression correlated with tumor size and grade. They also showed that the pretreatment IL-6 serum levels correlated as an independent factor with prognosis; this was also observed in many other cancers, including renal cell cancer, colorectal cancer, lung cancer, and melanoma, and also appears to be true for our present cases.25-28 The production of CRP in hepatocytes appeared to be principally induced at the transcriptional level following the elevation of circulating IL-6.
Elevated CRP levels are detected in 13% to 72% of patients with cancers, which is significantly higher than those in healthy control subjects.8, 11-13, 29 Nakanishi et al14 found elevated pretreatment serum CRP levels in 65% of their 46 MFH patients. The tumor histological grade and AJCC stage in the patients with elevated CRP levels in the current series was significantly higher than those in patients with normal CRP levels. Although elevated CRP levels were detected in only 18% of the total 102 patients, 14 (32%) of the 44 grade 3 sarcoma patients had elevated CRP levels. Elevated CRP levels are therefore likely to be associated with progressive disease and worse survival for soft tissue sarcoma patients.
There are a few limitations to the present study. First, the presence of general diseases associated with possible higher levels of markers of inflammation, like chronic bronchitis and collagen disease, were not taken into consideration because of lack of information, although there was no correlation between chronic inflammatory conditions (cardiovascular disease and diabetes mellitus) and elevated CRP levels in the statistical analysis. In addition, it is not likely that the patients with elevated CRP levels in the present series had such chronic inflammatory diseases, because all but 2 of the patients had normal CRP levels after tumor resection. However, the retrospective design of the study was another limitation.
Serum CRP levels before initial treatment appear to represent a marker of aggressive characteristics of soft tissue sarcomas. Elevated CRP levels before initial treatment were found to be a poor prognostic factor for the overall survival in a univariate analysis, and the disease-free survival in a multivariate analysis for soft tissue sarcoma patients.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
Each author certifies that he has no commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements) that might pose a conflict of interest in connection with the submitted article.