Abstract
Background
Increasing epidemiological evidence supported that chronic inflammatory factors might be involved in the carcinogenesis and progression of various cancers. The present study tried to investigate the prognostic value of perioperative C-reactive protein (CRP) in prognosis of patients with epithelial ovarian carcinoma (EOC) from a tertiary university teaching hospital.
Methods
The cutoff value of CRP was calculated according to receiver operating characteristic (ROC) curve. Variables were compared using Chi-square test. Progress-free survival (PFS) and overall survival (OS) time were assessed by Kaplan–Meier (KM) survival analysis and Log rank test based on serum CRP level. Univariate and multivariate Cox regression analyses were applied for assessing the relationship between clinicopathological parameters and survival.
Results
Higher perioperative CRP levels (preoperative ≥5.15 mg/L and postoperative ≥72.45 mg/L) were significantly associated with serous tumor, high-grade, advanced stage, elevated preoperative CA125, suboptimal surgery, chemotherapy resistance, recurrence and death in EOC (P < 0.01). KM analysis suggested patients with elevated preoperative, postoperative and perioperative CRP had shorter survival (P < 0.01). Elevated perioperative CRP was an independent risk factor for PFS (HR 1.510, 95% CI 1.124–2.028; P = 0.006) and OS (HR 1.580, 95% CI 1.109–2.251; P = 0.011). Similar results were obtained for elevated preoperative CRP. Subgroup analysis further suggested that elevated perioperative CRP was also an independent risk factor for prognosis in advanced stage and serous EOC.
Conclusion
Elevated perioperative CRP was an independent risk factor for poorer prognosis of EOC, particularly in advanced stage and serous patients.
Keywords: epithelial ovarian carcinoma, perioperative C-reactive protein, prognosis
Introduction
Epithelial ovarian carcinoma (EOC) remains the leading cause of death from gynecologic tumors.1–3 Despite the improvements in surgical techniques and chemotherapeutic regimens, the 5-year survival rate for EOC is still poor.4,5 Although the exact cause of EOC has not been fully elucidated, increasing epidemiological evidence supported that chronic inflammation might be one mechanism of carcinogenesis and progression in various cancers.6–10 Thus, the evaluation of the relationships between inflammatory markers and disease progression of EOC might help guide clinical management and predict the prognosis of EOC.
C-reactive protein (CRP), released predominantly by hepatocytes upon tissue injury and inflammation, is an important and non-specific inflammatory factor.11,12 Accumulating evidences have revealed the association between CRP and the risk of various cancers.6,13–19 Peres et al6 found women with CRP concentrations >10mg/L showed a 67% increased risk of ovarian cancer compared to <1mg/L (OR=1.67, 95% CI 1.12–2.48; P=0.01). And CRP concentration >10mg/L was also positively associated with risk of mucinous (OR=9.67, 95% CI 1.10–84.80; P=0.04) and endometrioid carcinoma (OR=3.41, 95% CI 1.07–10.92; P=0.03).
In addition to the role in carcinogenesis, promising data on the prognostic role of preoperative CRP in various malignancies including EOC have been reported.11,20–24, However, the literature about EOC was limited and the sample size of most studies was small,25–29 which would prevent well-powered analyses of potential heterogeneity of the association between clinical parameters and patient prognosis. Even in Hefler’s report which recruited 623 patients with EOC, they did not analyze the relationship of perioperative (combined preoperative with postoperative) CRP and survival or conduct subgroup analysis,11 which might weaken the evidences for CRP as an independent prognostic factor of EOC. Furthermore, the cutoff value of preoperative CRP in different studies was varied.11,24 Lu et al24 found that the average preoperative CRP level in 107 Chinese patients with EOC was lower than previous reports for Caucasian cases, and deduced it might reflect ethnic variations. However, no study was reported about the role of both preoperative and postoperative (perioperative) CRP in the prognosis of EOC up to date. Therefore, the aim of the present retrospective cohort study was to comprehensively investigate the clinical relationship between perioperative serum CRP and prognosis of EOC in a relatively large sample size of Chinese population, which permitted us reduce disease heterogeneity by subgroup analysis. Hope to provide better understanding on how CRP influences prognosis and provide insights on the potential strategy of clinical management of EOC.
Materials and Methods
Patients and Data Collection
We retrospectively reviewed the records of ovarian cancer at Women’s hospital, Zhejiang University School of Medicine between 2002.01.01 and 2016.12.31. The study was approved by the Ethical Committee of women’s hospital, Zhejiang University School of Medicine (IRB-20200230-R). Owing to the retrospective character and the difficulty of recalling all enrolled patients, informed consent was specifically waived by the ethics committee. All the researcher declared to protect patient data confidentiality and compliance with the Declaration of Helsinki. The enrolled patient should meet all the following inclusion criteria: (1) initial treatment was surgery including comprehensive surgical staging or cytoreductive surgery, followed by platinum-based chemotherapy in patients with stage Ic-IV, (2) histological diagnosis of EOC confirmed by Paraffin Section, (3) preoperative blood routine showed normal white blood cell and neutrophil count, (4) available serum CRP within 3 days before operation (preoperative) or within 7 days after operation (postoperative), and (5) available follow-up data of recurrence and death. Exclusion criteria included (1) primary other cancer; (2) the increase of CRP was caused by infection, connective tissue diseases or other inflammatory conditions, judged body temperature, clinical manifestation and auxiliary examination; (3) postoperative complications developed (including postoperative infection and massive bleeding); (4) the first dose of chemotherapy was delayed more than weeks after surgery; (5) only postoperative CRP results were available. Due to the retrospective character of the present study, patients with preoperative CRP results were included for evaluating the prognostic value of preoperative CRP, while patients with both preoperative and postoperative CRP results were included for perioperative CRP.
The clinical information of each selected patient was collected from the hospital database, and survival status was followed up by phone. The variables included age at diagnosis, histological type, FIGO stage, tumor grade, preoperative and postoperative serum CRP, preoperative serum CA125, postoperative residual tumor after primary surgery, chemotherapy sensitivity and the time of recurrence, death or last follow-up. Serum CRP was detected by immunoturbidimetry as part of the clinical routine management. Chemotherapy resistance was defined as having a time with recurrence of disease ≤6 months after completion of primary chemotherapy. Overall survival (OS) time was calculated as the interval between the date of primary surgery and the date of last follow-up or death. PFS was calculated as the interval from the date of primary surgery to the time of detected recurrence or progression.
Statistical Analysis
SPSS 20.0 statistical software was used for statistical analyses. The cutoff value of preoperative and postoperative CRP was 5.15 and 72.45mg/L, respectively, which was determined by Youden Index of the ROC curve. Variables were compared by Chi-square test. Spearman correlation analysis was used to analyze the correlation between preoperative CRP and postoperative CRP. PFS and OS were assessed by Kaplan–Meier survival (KM) analysis and Log rank test based on serum CRP level. Univariate and multivariate Cox regression analyses were applied for assessing the relationship between clinicopathological parameters and survival. For all analyses, an alpha level <0.05 was considered statistically significant.
Results
The Clinical-Pathological Characteristics of EOC Patients and Their Relationship with Perioperative CRP Level
A total of 654 EOC patients who met the included and excluded criteria were included for evaluating the prognostic value of preoperative CRP. Due to the retrospective character of the present study, 172 out of 654 EOC patients did not receive postoperative CRP measurement. Thus, only 482 EOC patients were included for evaluating the prognostic value of perioperative CRP (both preoperative and postoperative CRP). The median follow-up period of 654 EOC was 49 months, ranged from 3 to 190 months. There was a positive correlation between preoperative CRP and postoperative CRP (p=0.000). However, the correlation coefficient was only 0.315, which may be affected by pathological type, grade, FIGO stage, postoperative residual lesions and other clinical parameters.
As shown in Table 1 and Supplementary Table S1, Chi-square test suggested that older age, serous carcinoma, high grade, advanced stage, higher preoperative CRP, higher preoperative CA125, higher postoperative CRP, higher perioperative CRP (both preoperative CRP ≥5.15mg/L and postoperative CRP ≥72.45mg/L), chemotherapy resistance and larger postoperative residual tumor (≥1cm) significantly correlated with poorer prognosis of EOC patients. However, subgroup analysis according to tumor stage found that histological type and tumor grade were no longer associated with the prognosis of EOC, except for the relationship between histological type and prognosis in advanced-stage subgroup from 482 EOC patients (Table 1). Most of the non-serous tumors were in early stage with good prognosis. But once they progressed into advanced stage, the mortality of non-serous EOC would be higher than that of serous tumors (Table 1).
Table 1.
The Clinicopathological Characteristics of EOC Patients with Perioperative CRP
| Criterion | Total Patients | No Recurrence | Recurrence | P | Survival | Death | P | |
|---|---|---|---|---|---|---|---|---|
| Number | 482 | 249 (51.7%) | 233 (48.3%) | 326 (67.6%) | 156 (32.4%) | |||
| Age; Median (range) | 51.5 (16–79) | 51 (16–75) | 52 (23–79) | 0.002 | 50 (16–77) | 53 (23–79) | 0.001 | |
| Histological type | 0.000 | 0.009 | ||||||
| Serous | 343 (71.2%) | 145 (42.3%) | 198 (57.7%) | 217 (63.3%) | 126 (36.7%) | |||
| Clear cell | 61 (12.7%) | 44 (72.1%) | 17 (27.9%) | 47 (77.0%) | 14 (23.0%) | |||
| Mucinous | 36 (7.5%) | 32 (88.9%) | 4 (11.1%) | 32 (88.9%) | 4 (11.1%) | |||
| Endometrioid | 18 (3.7%) | 10 (55.6%) | 8 (44.4%) | 12 (66.7%) | 6 (33.3%) | |||
| Other | 24 (5.0%) | 18 (75.0%) | 6 (25.0%) | 18 (75.0%) | 6 (25.0%) | |||
| Tumor grade | 0.000 | 0.000 | ||||||
| Low-grade | 111 (23.0%) | 90 (81.1%) | 21 (18.9%) | 94 (84.7%) | 17 (15.3%) | |||
| High-grade | 371 (77.0%) | 159 (42.9%) | 212 (57.1%) | 232 (62.5%) | 139 (37.5%) | |||
| FIGO stage | 0.000 | 0.000 | ||||||
| I | 148 (30.7%) | 135 (91.2%) | 13 (8.8%) | 142 (95.9%) | 6 (4.1%) | |||
| II | 40 (8.3%) | 29 (72.5%) | 11 (27.5%) | 34 (85.0%) | 6 (15.0%) | |||
| III | 279 (57.9%) | 81 (29.0%) | 198 (71.0%) | 143 (51.3%) | 136 (48.7%) | |||
| IV | 15 (3.1%) | 4 (26.7%) | 11 (73.3%) | 7 (46.7%) | 8 (53.3%) | |||
| Preoperative CRP; Median (range) | 3.65 (0–168.00) | 2.60 (0–136.80) | 5.50 (0–168.00) | 0.003 | 2.70 (0–136.80) | 6.66 (0–168.00) | 0.002 | |
| Postoperative CRP; Median(range) | 55.50 (0–334.00) | 49.10 (0–282.30) | 63.20 (1.80–334.00) | 0.001 | 50.90 (0–282.30) | 73.30 (4.30–334.00) | 0.010 | |
| Perioperative CRP | 0.000 | 0.000 | ||||||
| Other | 382 (79.3%) | 217 (56.8%) | 165 (43.2%) | 276 (72.3%) | 106 (27.7%) | |||
| Both increased* | 100 (20.7%) | 32 (32.0%) | 68 (68.0%) | 50 (50.0%) | 50 (50.0%) | |||
| Preoperative CA125 Median (range) | 238.65 (6.60–22,289.0) | 102.90(6.60–9744.0) | 562.2(10.20–22,289.00) | 0.000 | 152.90(6.60–9868.00) | 571.50(12.40–22,289.0) | 0.000 | |
| Postoperative residual tumor | 0.000 | 0.000 | ||||||
| <1cm | 395 (82.0%) | 237 (60.0%) | 158 (40.0%) | 302(76.5%) | 93(23.5%) | |||
| ≥1cm | 87 (18.0%) | 12 (13.8%) | 75 (86.2%) | 24 (27.6%) | 63 (72.4%) | |||
| Chemotherapy resistance | 0.000 | 0.000 | ||||||
| No | 403 (83.6%) | 249 (61.8%) | 154 (38.2%) | 316 (78.4%) | 87 (21.6%) | |||
| Yes resistance | 79 (16.4%) | 0 (0.0%) | 79 (100.0%) | 10 (12.7%) | 69 (87.3%) | |||
| Tumor grade | 0.000 | 0.000 | ||||||
| Early stage | Low-grade | 95 (50.5%) | 86 (90.5%) | 9 (9.5%) | 0.172 | 89 (93.7%) | 6 (6.3%) | 0.970 |
| High-grade | 93 (49.5%) | 78 (83.9%) | 15 (16.1%) | 87 (93.5%) | 6 (6.5%) | |||
| Advanced stage | Low-grade | 16 (5.4%) | 4 (25.0%) | 12 (75.0%) | 0.943 | 5 (31.2%) | 11 (68.8%) | 0.104 |
| High-grade | 278 (94.6%) | 81 (29.1%) | 197 (70.9%) | 145 (52.2%) | 133 (47.8%) | |||
| Histological type | 0.000 | 0.000 | ||||||
| Early stage | Non-serous | 111 (59.0%) | 97 (87.4%) | 14 (12.6%) | 0.940 | 101 (91.0%) | 10 (9.0%) | 0.143 |
| Serous | 77 (41.0%) | 67 (87.0%) | 10 (13.0%) | 75 (97.4%) | 2 (2.6%) | |||
| Advanced stage | Non-serous | 28 (9.5%) | 7 (25.0%) | 21 (75.0%) | 0.631 | 8 (28.6%) | 20 (71.4%) | 0.012 |
| Serous | 266 (90.5%) | 78 (29.3%) | 188 (70.7%) | 142 (53.4%) | 124 (46.6%) | |||
Notes: *preoperative CRP ≥5.15mg/L and postoperative CRP ≥72.45mg/L.
Since CRP levels were significantly associated with the prognosis of EOC, the relationship between CRP level (preoperative, postoperative and perioperative CRP, respectively) and clinical-pathological characteristics of EOC were further analyzed. The results suggested higher preoperative, postoperative and perioperative CRP levels were all significantly associated with advanced stage, postoperative residual tumor (≥1cm), chemotherapy resistance, recurrence, and death in EOC patients (all P<0.01). Higher preoperative and perioperative CRP were also associated with high-grade tumor and increased CA125 level, while postoperative and perioperative CRP levels were both significantly associated with serous tumor (P<0.05) (Table 2).
Table 2.
The Relationship Between Perioperative CRP Level and Clinicopathological Characteristics of EOC Patients
| Criterion | n | Preoperative CRP | P | n | Postoperative CRP | P | Perioperative CRP | P | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No Increased | Increased | No Increased | Increased | Other | Both Increased* | |||||||
| Age | <51y | 307 | 190(61.9%) | 117(38.1%) | 0.238 | 224 | 146(65.2%) | 78(34.8%) | 0.134 | 186(83.0%) | 38(17.0%) | 0.056 |
| ≥51y | 347 | 199(57.3%) | 148(42.7%) | 258 | 151(58.5%) | 107(41.5%) | 196(76.0%) | 62(24.0%) | ||||
| Histological type | Non-serous | 190 | 121(63.7%) | 69(36.3%) | 0.161 | 139 | 102(73.4%) | 37(26.6%) | 0.001 | 125(89.9%) | 14(10.1%) | 0.000 |
| Serous | 464 | 268(57.8%) | 196(42.2%) | 343 | 195(56.9%) | 148(43.1%) | 257(74.9%) | 86(25.1%) | ||||
| FIGO stage | I–II | 241 | 174(72.2%) | 67(27.8%) | 0.000 | 188 | 132(70.2%) | 56(29.8%) | 0.002 | 172(91.5%) | 16(8.5%) | 0.000 |
| III–IV | 413 | 215(52.1%) | 198(47.9%) | 294 | 165(56.1%) | 129(43.9%) | 210(71.4%) | 84(28.6%) | ||||
| Tumor grade | Low-grade | 146 | 100(68.5%) | 46(31.5%) | 0.012 | 111 | 75(67.6%) | 36(32.4%) | 0.142 | 99(89.2%) | 12(10.8%) | 0.003 |
| High-grade | 508 | 289(56.9%) | 219(43.1%) | 371 | 222(59.8%) | 149(40.2%) | 283(76.3%) | 88(23.7%) | ||||
| Preoperative CA125 | <35U/mL | 102 | 80(78.4%) | 22(21.6%) | 0.000 | 76 | 52(68.4%) | 24(31.6%) | 0.184 | 71(93.4%) | 5(6.6%) | 0.001 |
| ≥35U/mL | 552 | 309(56.0%) | 243(44.0%) | 406 | 245(60.3%) | 161(39.7%) | 311(76.6%) | 95(23.4%) | ||||
| Postoperative residual tumor | <1cm | 519 | 346(66.7%) | 173(33.3%) | 0.000 | 395 | 259(65.6%) | 136(34.4%) | 0.000 | 333(84.3%) | 62(15.7%) | 0.000 |
| ≥1cm | 135 | 43(31.9%) | 92(68.1%) | 87 | 38(43.7%) | 49(56.3%) | 49(56.3%) | 38(43.7%) | ||||
| Chemotherapy resistance | No | 538 | 348(64.7%) | 190(35.3%) | 0.000 | 403 | 260(64.5%) | 143(35.5%) | 0.003 | 334(82.9%) | 69(17.1%) | 0.000 |
| Yes | 116 | 41(35.3%) | 75(64.7%) | 79 | 37(46.8%) | 42(53.2%) | 48(60.8%) | 31(39.2%) | ||||
| Recurrence | No | 311 | 219(70.4%) | 92(29.6%) | 0.000 | 249 | 172(69.1%) | 77(30.9%) | 0.001 | 217(87.1%) | 32(12.9%) | 0.000 |
| Yes | 343 | 170(49.6%) | 173(50.4%) | 233 | 125(53.6%) | 108(46.4%) | 165(70.8%) | 68(29.2%) | ||||
| Death | No | 4055 | 273(67.4%) | 132(32.6%) | 0.000 | 326 | 221(67.8%) | 105(32.2%) | 0.000 | 276(84.7%) | 50(15.3%) | 0.000 |
| Yes | 249 | 116(46.6%) | 133(53.4%) | 156 | 76(48.7%) | 80(51.3%) | 106(67.9%) | 50(32.1%) | ||||
Notes: *preoperative CRP ≥5.15mg/L and postoperative CRP ≥72.45mg/L.
Independent Risk Factors Related to Prognosis in EOC Patients
As shown in Table 3, univariate Cox regression analysis identified histological type, tumor grade, FIGO stage, preoperative CA125 level, postoperative residual tumor size and preoperative CRP level are significant prognostic factors related with PFS and OS (all p<0.001). In addition, age (p=0.013) was significantly associated with OS, but not associated with PFS. Further multivariate analysis showed that elevated perioperative CRP (both increased) was an independent risk factor for PFS (HR 1.510, 95% CI 1.124–2.028; p = 0.006) and OS (HR 1.580, 95% CI 1.109–2.251; p = 0.011), in addition to serous tumor, advanced stage and suboptimal surgery (residual tumor ≥1cm). Similar results were also validated in 654 patients with preoperative CRP results (Supplementary Table S2), except for serous tumor (elevated preoperative CRP, advanced stage and suboptimal surgery for PFS:HR=1.506, 95% CI 1.206–1.881; 6.192, 95% CI 4.094–9.366 and 1.561, 95% CI 1.228–1.985; for OS:HR=1.646, 95% CI 1.270–2.134; 9.729, 95% CI 5.537–17.093 and 1.929, 95% CI 1.465–2.539 seperately; all p = 0.000).
Table 3.
Cox Regression Analysis of Factors Related to Survival in 482 Patients with Perioperative CRP
| Criterion | PFS | OS | ||||||
|---|---|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariate Analysis | Univariate Analysis | Multivariate Analysis | |||||
| HR(95% CI) | P | HR(95% CI) | P | HR(95% CI) | P | HR(95% CI) | P | |
| Age(year) (≥ 51 vs <51) | 1.260(0.972–1.633) | 0.081 | - | - | 1.508(1.092–2.082) | 0.013 | 1.076(0.772–1.500) | 0.666 |
| Histological type (Serous vs non-sereous) | 2.916(2.032–4.184) | 0 | 0.629(0.397–0.994) | 0.047 | 2.039(1.367–3.042) | 0 | 0.376(0.227–0.622) | 0 |
| Tumor grade (High-grade vs Low-grade) | 3.785(2.416–5.932) | 0 | 1.115(0.642–1.934) | 0.7 | 2.763(1.668–4.576) | 0 | 0.657(0.347–1.246) | 0.199 |
| FIGO stage(III–IV vs I–II) | 9.132(5.972–13.966) | 0 | 7.820(4.608–13.270) | 0 | 10.465(5.800–18.883) | 0 | 13.543(6.496–28.237) | 0 |
| Preoperative CA125 (U/mL) (≥35 vs <35) | 5.009(2.732–9.184) | 0 | 1.705(0.865–3.361) | 0.124 | 4.922(2.305–10.508) | 0 | 2.067(0.882–4.842) | 0.095 |
| Postoperative residual tumor (cm) (≥1 vs <1) | 3.655(2.761–2.837) | 0 | 1.754(1.311–2.345) | 0 | 4.610(3.340–6.362) | 0 | 2.425(1.716–3.427) | 0 |
| Perioperative CRP (Both increased vs Other) | 2.218(1.670–2.946) | 0 | 1.510(1.124–2.028) | 0.006 | 2.371(1.691–3.325) | 0 | 1.580(1.109–2.251) | 0.011 |
Subgroup analysis according to tumor stage suggested that no prognostic factor was associated with PFS and OS in EOC of early stage due to the good prognosis (Table 4). While for advanced stage subgroup, univariate and multivariate Cox regression analyses validated that elevated perioperative CRP and suboptimal surgery were independent risk factors for poorer prognosis (Table 4). As we know, non-serous tumor was an independent risk factor for OS in 482 patients with perioperative results and advanced-stage subgroup (Table 3–4), which was contrary to the results from 654 patients with preoperative CRP result (Table 2). Thus, subgroup analyses according to histological type were further conducted. Our results suggested perioperative CRP has no significant relationship with the prognosis in non-serous EOC. While in serous EOC, advanced stage, suboptimal surgery and elevated perioperative CRP were independent risk factors for poorer prognosis (Table 5).
Table 4.
Cox Regression Analysis of Factors Related to Survival in Early and Advanced Stage Subgroup of 482 Patients
| Criterion | PFS | OS | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariate Analysis | Univariate Analysis | Multivariate Analysis | ||||||
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | ||
| Early stage | Age (years)(≥ 51 vs <51) | 1.787(0.794–4.025) | 0.161 | – | – | 2.470(0.744–8.206) | 0.140 | – | – |
| Histological type (Serous vs non-serous) | 1.080(0.479–2.435) | 0.853 | - | - | 0.334(0.073–1.534) | 0.159 | - | - | |
| Tumor grade (High-grade vs Low-grade) | 1.676(0.733–3.830) | 0.221 | - | - | 1.055(0.339–3.283) | 0.926 | - | - | |
| Preoperative CA125 (U/mL)(≥35 vs <35) | 1.407(0.583–3.395) | 0.447 | - | - | 1.203(0.362–3.995) | 0.763 | - | - | |
| Postoperative residual tumor (cm)(≥1 vs <1) | - | - | - | - | - | - | - | - | |
| Perioperative CRP (Both increased vs Other) | 0.891(0.210–3.791) | 0.876 | - | - | 0.043(0.000–297.045) | 0.486 | - | - | |
| Advanced stage | Age (years)(≥ 51 vs <51) | 0.929(0.706–1.222) | 0.598 | - | - | 1.155(0.826–1.616) | 0.399 | - | - |
| Histological type (Serous vs non-serous) | 0.721(0.458–1.133) | 0.156 | - | - | 0.515(0.321–0.828) | 0.006 | 0.462(0.276–0.773) | 0.003 | |
| Tumor grade (High-grade vs Low-grade) | 0.666(0.371–1.193) | 0.172 | - | - | 0.428(0.231–0.793) | 0.007 | 0.548(0.281–1.069) | 0.078 | |
| Preoperative CA125 (U/mL)(≥35 vs <35) | 1.701(0.632–4.576) | 0.293 | - | - | 1.469(0.468–4.615) | 0.510 | - | - | |
| Postoperative residual tumor (cm)(≥1 vs <1) | 1.877(1.411–2.498) | 0.000 | 1.766(1.321–2.360) | 0.000 | 2.482(1.784–3.453) | 0.000 | 2.539(1.803–3.575) | 0.000 | |
| Perioperative CRP (Both increased vs Other) | 1.638(1.221–2.197) | 0.001 | 1.495(1.109–2.016) | 0.008 | 1.824(1.291–2.577) | 0.001 | 1.587(1.108–2.271) | 0.012 | |
Table 5.
Cox Regression Analysis of Factors Related to Survival in Serous and Non-Serous Stage Subgroup of 482 Patients
| Criterion | PFS | OS | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariate Analysis | Univariate Analysis | Multivariate Analysis | ||||||
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | ||
| Non-serous | Age (years)(≥ 51 vs <51) | 1.219(0.628–2.366) | 0.558 | - | - | 1.250(0.610–2.564) | 0.542 | - | - |
| FIGO stage (III–IV vs I–II) | 11.171(5.616–22.219) | 0.000 | 10.820(4.797–24.409) | 0.000 | 13.267(6.149–28.622) | 0.000 | 12.455(4.967–31.228) | 0.000 | |
| Tumor grade (High-grade vs Low-grade) | 2.418(1.245–4.694) | 0.009 | 0.776(0.358–1.679) | 0.519 | 2.792(1.356–5.752) | 0.005 | 0.670(0.280–1.602) | 0.368 | |
| Preoperative CA125 (U/mL) (≥35 vs <35) | 4.201(1.629–10.836) | 0.003 | 3.319(1.264–8.715) | 0.015 | 4.576(1.596–13.119) | 0.005 | 3.165(1.065–9.402) | 0.038 | |
| Postoperative residual tumor (cm)(≥1 vs <1) | 8.124(3.101–21.278) | 0.000 | 1.279(0.456–3.592) | 0.640 | 10.489(3.948–27.868) | 0.000 | 1.782(0.623–5.096) | 0.281 | |
| Perioperative CRP (Both increased vs Other) | 2.208(0.916–5.324) | 0.078 | - | - | 2.327(0.889–6.094) | 0.085 | - | - | |
| Serous | Age (years)(≥ 51 vs <51) | 1.205(0.909–1.599) | 0.195 | - | - | 1.548(1.077–2.226) | 0.018 | 1.078(0.741–1.568) | 0.693 |
| FIGO stage (III–IV vs I–II) | 8.627(4.561–16.318) | 0.000 | 7.966(3.877–16.370) | 0.000 | 22.459(5.553–90.836) | 0.000 | 23.939(5.121–11.902) | 0.000 | |
| Tumor grade (High-grade vs Low-grade) | 2.957(1.098–7.962) | 0.032 | 1.960(0.713–5.386) | 0.192 | 1.497(0.553–4.055) | 0.427 | - | - | |
| Preoperative CA125 (U/mL) (≥35 vs <35) | 2.987(1.324–6.737) | 0.008 | 0.519(0.208–1.297) | 0.160 | 3.227(1.026–10.149) | 0.045 | 0.283(0.079–1.015) | 0.053 | |
| Postoperative residual tumor (cm)(≥1 vs <1) | 2.855(2.123–3.839) | 0.000 | 1.827(1.347–2.478) | 0.000 | 3.864(2.718–5.493) | 0.000 | 2.492(1.717–3.615) | 0.000 | |
| Perioperative CRP (Both increased vs Other) | 1.926(1.424–2.605) | 0.000 | 1.599(1.166–2.192) | 0.004 | 2.144(1.487–3.092) | 0.000 | 1.613(1.107–2.350) | 0.013 | |
Elevated Perioperative CRP Level Was Associated with Shorter PFS and OS
As shown in Figure 1, EOC patients with elevated preoperative CRP had shorter PFS (22.0 vs 119.0 months) and OS (67.0 vs not reached) compared to preoperative CRP <5.15mg/L. EOC patients with elevated postoperative CRP also had shorter PFS (29.0 vs 119.0 months) and OS (76.0 vs not reached) compared to postoperative CRP <72.45 mg/L. Consistently, EOC patients with elevated perioperative CRP had shorter PFS (17.0 vs 119.0 months) and OS (50.0 vs not reached) compared to other patients. These results suggested that elevated perioperative CRP level (both preoperative CRP ≥5.15 mg/L and postoperative CRP ≥72.45 mg/L) is a consolidated predictive factor for poorer prognosis in EOC patients. The predicting capacity of perioperative CRP (combined with preoperative and postoperative CRP) was significantly higher than that predicted by preoperative and postoperative CRP alone.
Figure 1.
The relationship between perioperative CRP level and prognosis in EOC patients (A) Kaplan–Meier curves for PFS depending on the preoperative CRP level; (B) Kaplan–Meier curves for OS depending on the preoperative CRP level; (C) Kaplan–Meier curves for PFS depending on the postoperative CRP level; (D) Kaplan–Meier curves for OS depending on the postoperative CRP level; € Kaplan–Meier curves for PFS depending on the perioperative CRP level; (F) Kaplan–Meier curves for OS depending on the perioperative CRP level.
Discussion
Although the prognosis of EOC depends on a variety of factors, clinical decision-making is still based on established histopathologic prognosticators.11 Previous studies have recognized that inflammatory-related cytokines play important regulatory roles in tumorigenesis, cancer progression and metastasis.18,19 The tumor microenvironment of EOC is rich in a variety of proinflammatory cytokine and chemokines, such as interleukin (IL)-1, IL-6, transforming growth factor-β and interferon-r, which can affect cellular communication, stimulate CRP production and are critical for tumor growth, invasion, and migration.30,31 Increasing evidences supported that inflammatory factors including CRP, were not only secreted by hepatocytes as an inflammatory response to infection, trauma and malignant tumors but also derived from tumor cells themselves.32–34 Compared with other inflammatory factors, serum CRP is a marker detected in daily clinical practice, which would be easy to perform.
The association between elevated pretreatment CRP levels and poor prognosis has been studied in different cancers including EOC.6,11,17,20–24,35–37 Knittelfelder et al validated that pre-treatment CRP level represented an independent prognostic factor for survival in patients with oral and oropharyngeal cancer, particularly in those treated with definitive chemo-radiotherapy.20 Hefler et al also reported that preoperative serum CRP could serve as clinically useful marker in 623 patients with EOC and found that the patients with CRP ≤1 mg/dl had better 5-year OS than those >1 mg/dl (82% vs 58.5%).11 While Lu et al found that CRP > 8 mg/l was related with poorer 5-year survival in 107 EOC patients.24 Furthermore, the ratio of CRP and albumin has recently been suggested as a novel independent marker of poor prognosis among EOC.28 Consistent with previous study, we found that preoperative CRP (≥5.15 mg/L) was an independent risk factor for survival in patients with EOC, in addition to the previous established prognosticators including FIGO stage and postoperative residual lesion.11 KM analysis also revealed better prognosis in lower preoperative CRP. The potential cutoff values in different studies were varied, which might be due to the different cancer types and study population.
In accordance with the predictive value of preoperative CRP level, the present study firstly revealed significant relationships between the elevated perioperative CRP and advanced tumor stage, low grade, serous carcinoma, elevated preoperative serum CA125, chemotherapy resistance and surgical residue lesions. The relationship between CRP level and tumor stage in the present study supported the hypothesis that CRP production could be from malignant cells.20 As we know, tumor stage, grade and postoperative residual tumor are the most reliable predictors for clinical prognosis of EOC.11,38,39 Thus, we speculated that perioperative CRP level could be adopted as a union factor to predict the prognosis of EOC.
As we anticipated, the levels of preoperative and postoperative serum CRP were both significantly higher in patients suffering chemoresistance, relapse and death than those in other patients. The relationship between CRP and chemoresistance suggested that increased perioperative CRP level could be used for selecting patients who would benefit from platinum-based chemotherapy. Moreover, Cox regression analysis revealed higher perioperative CRP (both preoperative ≥5.15 mg/L and postoperative ≥72.45 mg/L) was an independent risk factor for recurrence and death of EOC in all enrolled patients. Further subgroup analysis according to tumor stage and histology confirmed similar results in advanced stage and serous EOC patients. KM analysis also revealed EOC patients with elevated perioperative CRP (both increased) suffered shorter PFS and OS. As an inflammatory factor, the elevated CRP in patients with poor prognosis proposed that anti-inflammatory therapy could be a potentially effective strategy for EOC treatment.40
Conclusions
In conclusion, as a study with a relatively larger sample size, we firstly validated that elevated perioperative CRP might serve as an independent prognostic predictor for EOC with shorter PFS and OS, especially in patients with advanced stage and serous EOC. The level of perioperative CRP could be an identifier to screen the potential effective strategy for clinical management of EOC. Nevertheless, due to the retrospective character of the present study, further prospective and experimental studies are warranted to verify the prognostic value of CRP and clarify the intrinsic mechanism of CRP in tumor progression.
Funding Statement
This research was supported by Hangzhou Municipal Health and Family Planning Commission (0020190447).
Abbreviations
CRP, C-reactive protein; EOC, epithelial ovarian carcinoma; KM, Kaplan–Meier; OS, overall survival; PFS, progression-free survival; MST, median survival time.
Data Sharing Statement
All data generated or analyzed during this study are included in this article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Ethics Approval and Informed Consent
Approved by the Ethical Committee of women’s hospital, Zhejiang University School of Medicine (IRB-20200230-R).
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
The authors declare that they have no competing interests in this work.
References
- 1.Xian W, George S. Meeting report from the 2018 12th biennial ovarian cancer research symposium detection and prevention of ovarian cancer. Int J Gynecol Cancer. 2019;29(Suppl 2):s2–s6. doi: 10.1136/ijgc-2019-000454 [DOI] [PubMed] [Google Scholar]
- 2.Chen Z, Guo X, Sun S, Lu C, Wang L. Serum miR-125b levels associated with epithelial ovarian cancer (EOC) development and treatment responses. Bioengineered. 2020;11(1):311–317. doi: 10.1080/21655979.2020.1736755 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70(1):7–30. doi: 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
- 4.Marchetti C, Pisano C, Facchini G, et al. First-line treatment of advanced ovarian cancer: current research and perspectives. Expert Rev Anticancer Ther. 2010;10(1):47–60. doi: 10.1586/era.09.167 [DOI] [PubMed] [Google Scholar]
- 5.Reid BM, Permuth JB, Sellers TA. Epidemiology of ovarian cancer: a review. Cancer Biol Med. 2017;14(1):9–32. doi: 10.20892/j.issn.2095-3941.2016.0084 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Peres LC, Mallen AR, Townsend MK, et al. High levels of C-reactive protein are associated with an increased risk of ovarian cancer: results from the ovarian cancer cohort consortium. Cancer Res. 2019;79(20):5442–5451. doi: 10.1158/0008-5472.CAN-19-1554 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pignon JP, le Maître A, Maillard E, Bourhis J; MACH-NC Collaborative Group. Meta-analysis of chemotherapy in head and neck cancer (MACH-NC): an update on 93 randomised trials and 17,346 patients. Radiother Oncol. 2009;92(1):4–14. doi: 10.1016/j.radonc.2009.04.014 [DOI] [PubMed] [Google Scholar]
- 8.Lu L, Wu Y, Feng M, Xue X, Fan Y. A novel seven-miRNA prognostic model to predict overall survival in head and neck squamous cell carcinoma patients. Mol Med Rep. 2019;20(5):4340–4348. doi: 10.3892/mmr.2019.10665 [DOI] [PubMed] [Google Scholar]
- 9.Groblewska M, Mroczko B, Wereszczyńska-Siemiatkowska U, et al. Serum interleukin 6 (IL-6) and C-reactive protein (CRP) levels in colorectal adenoma and cancer patients. Clin Chem Lab Med. 2008;46(10):1423–1428. doi: 10.1515/CCLM.2008.278 [DOI] [PubMed] [Google Scholar]
- 10.Nakamura M, Bax HJ, Scotto D, et al. Immune mediator expression signatures are associated with improved outcome in ovarian carcinoma. Oncoimmunology. 2019;8(6):e1593811. doi: 10.1080/2162402X.2019.1593811 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hefler LA, Concin N, Hofstetter G, et al. Serum C-reactive protein as independent prognostic variable in patients with ovarian cancer. Clin Cancer Res. 2008;14(3):710–714. doi: 10.1158/1078-0432.CCR-07-1044 [DOI] [PubMed] [Google Scholar]
- 12.Thurner EM, Krenn-Pilko S, Langsenlehner U, et al. The elevated C reactive protein level is associated with poor prognosis in prostate cancer patients treated with radiotherapy. Eur J Cancer. 2015;51(5):610–619. doi: 10.1016/j.ejca.2015.01.002 [DOI] [PubMed] [Google Scholar]
- 13.Crozier JE, McKee RF, McArdle CS, et al. Preoperative but not postoperative systemic inflammatory response correlates with survival in colorectal cancer. Br J Surg. 2007;94(8):1028–1032. doi: 10.1002/bjs.5706 [DOI] [PubMed] [Google Scholar]
- 14.Katano A, Takahashi W, Yamashita H, et al. The impact of elevated C reactive protein level on the prognosis for oro-hypopharynx cancer patients treated with radiotherapy. Sci Rep. 2017;7(1):17805. doi: 10.1038/s41598-017-18233-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Graupp M, Schaffer K, Wolf A, et al. C-reactive protein is an independent prognostic marker in patients with tongue carcinoma-A retrospective study. Clin Otolaryngol. 2018;10:13102. [DOI] [PubMed] [Google Scholar]
- 16.Tai SF, Chien HT, Young CK, et al. Roles of preoperative C-reactive protein are more relevant in buccal cancer than other subsites. World J Surg Oncol. 2017;15(1):47. doi: 10.1186/s12957-017-1116-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Khandavilli SD, Ceallaigh PO, Lloyd CJ, Whitaker R. Serum C-reactive protein as a prognostic indicator in patients with oral squamous cell carcinoma. Oral Oncol. 2009;45(10):912–914. doi: 10.1016/j.oraloncology.2009.03.015 [DOI] [PubMed] [Google Scholar]
- 18.Jaiswal M, LaRusso NF, Burgart LJ, Gores GJ. Inflammatory cytokines induce DNA damage and inhibit DNA repair in cholangiocarcinoma cells by a nitric oxide-dependent mechanism. Cancer Res. 2000;60(1):184–190. [PubMed] [Google Scholar]
- 19.Kuper H, Adami HO, Trichopoulos D. Infections as a major preventable cause of human cancer. J Intern Med. 2000;248(3):171–183. doi: 10.1046/j.1365-2796.2000.00742.x [DOI] [PubMed] [Google Scholar]
- 20.Knittelfelder O, Delago D, Jakse G, et al. The pre-treatment C-reactive protein represents a prognostic factor in patients with oral and oropharyngeal cancer treated with radiotherapy. Cancers. 2020;12(3):626. doi: 10.3390/cancers12030626 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gockel I, Dirksen K, Messow CM, Junginger T. Significance of preoperative C-reactive protein as a parameter of the perioperative course and long-term prognosis in squamous cell carcinoma and adenocarcinoma of the oesophagus. World J Gastroenterol. 2006;12(23):3746–3750. doi: 10.3748/wjg.v12.i23.3746 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Brown DJ, Milroy R, Preston T, McMillan DC. The relationship between an inflammation-based prognostic score (Glasgow Prognostic Score) and changes in serum biochemical variables in patients with advanced lung and gastrointestinal cancer. J Clin Pathol. 2007;60(6):705–708. doi: 10.1136/jcp.2005.033217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Polterauer S, Grimm C, Tempfer C, et al. C-reactive protein is a prognostic parameter in patients with cervical cancer. Gynecol Oncol. 2007;107(1):114–117. doi: 10.1016/j.ygyno.2007.06.001 [DOI] [PubMed] [Google Scholar]
- 24.Lu Y, Huang S, Li P, et al. Prognostic evaluation of preoperative serum C-reactive protein concentration in patients with epithelial ovarian cancer. Exp Ther Med. 2015;9(5):2003–2007. doi: 10.3892/etm.2015.2350 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zeng F, Wei H, Yeoh E, et al. Inflammatory markers of CRP, IL6, TNFα, and soluble TNFR2 and the risk of ovarian cancer: a meta-analysis of prospective studies. Cancer Epidemiol Biomarkers Prev. 2016;25(8):1231–1239. doi: 10.1158/1055-9965.EPI-16-0120 [DOI] [PubMed] [Google Scholar]
- 26.Liu Y, Chen S, Zheng C, et al. The prognostic value of the preoperative C-reactive protein/albumin ratio in ovarian cancer. BMC Cancer. 2017;17(1):285. doi: 10.1186/s12885-017-3220-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Li J, Jiao X, Yuan Z, Qiu H, Guo R. C-reactive protein and risk of ovarian cancer: a systematic review and meta-analysis. Medicine. 2017;96(34):e7822. doi: 10.1097/MD.0000000000007822 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ose J, Schock H, Tjønneland A, et al. Inflammatory markers and risk of epithelial ovarian cancer by tumor subtypes: the EPIC cohort. Cancer Epidemiol Biomarkers Prev. 2015;24(6):951–961. doi: 10.1158/1055-9965.EPI-14-1279-T [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Dobrzycka B, Mackowiak-Matejczyk B, Terlikowska KM, et al. Serum levels of IL-6, IL-8 and CRP as prognostic factors in epithelial ovarian cancer. Eur Cytokine Netw. 2013;24(3):106–113. doi: 10.1684/ecn.2013.0340 [DOI] [PubMed] [Google Scholar]
- 30.Diakos CI, Charles KA, McMillan DC, Clarke SJ. Cancer-related inflammation and treatment effectiveness. Lancet Oncol. 2014;15(11):e493–e503. doi: 10.1016/S1470-2045(14)70263-3 [DOI] [PubMed] [Google Scholar]
- 31.Mantovani A, Allavena P, Sica A, et al. Cancer-related inflammation. Nature. 2008;454(7203):436–444. doi: 10.1038/nature07205 [DOI] [PubMed] [Google Scholar]
- 32.Nimptsch K, Aleksandrova K, Boeing H, et al. Association of CRP genetic variants with blood concentrations of C-reactive protein and colorectal cancer risk. Int J Cancer. 2015;136(5):1181–1192. doi: 10.1002/ijc.29086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Nozoe T, Korenaga D, Futatsugi M, Saeki H, Maehara Y, Sugimachi K. Immunohistochemical expression of C-reactive protein in squamous cell carcinoma of the esophagus-significance as a tumor marker. Cancer Lett. 2003;192(1):89–95. doi: 10.1016/S0304-3835(02)00630-4 [DOI] [PubMed] [Google Scholar]
- 34.Marnell L, Mold C, Du Clos TW. C-reactive protein: ligands, receptors and role and role in inflammation. Clin Immunol. 2005;117(2):104–111. doi: 10.1016/j.clim.2005.08.004 [DOI] [PubMed] [Google Scholar]
- 35.Peter F, Wittekindt C, Finkensieper M, Kiehntopf M, Guntinas-Lichius O. Prognostic impact of pretherapeutic laboratory values in head and neck cancer patients. J Cancer Res Clin Oncol. 2013;139(1):171–178. doi: 10.1007/s00432-012-1320-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Oluleye OW, Folsom AR, Nambi V, et al. B-type natriuretic peptide, C-reactive protein, and cause-specific mortality. Ann Epidemiol. 2013;23(2):66–73. doi: 10.1016/j.annepidem.2012.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Zuo H, Ueland PM, Ulvik A, et al. Plasma biomarkers of inflammation, the kynurenine pathway, and risks of all-cause, cancer, and cardiovascular disease mortality: the Hordaland health study. Am J Epidemiol. 2016;183(4):249–258. doi: 10.1093/aje/kwv242 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Poole EM, Lee IM, Ridker PM, Buring JE, Hankinson SE, Tworoger SS. A prospective study of circulating C-reactive protein, interleukin-6, and tumor necrosis factor α receptor 2 levels and risk of ovarian cancer. Am J Epidemiol. 2013;178(8):1256–1264. doi: 10.1093/aje/kwt098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Friedlander ML. Prognostic factors in ovarian cancer. Semin Oncol. 1998;25(3):305–314. [PubMed] [Google Scholar]
- 40.Zhang Z, Chen F, Shang L. Advances in antitumor effects of NSAIDs. Cancer Manag Res. 2018;10:4631–4640. doi: 10.2147/CMAR.S175212 [DOI] [PMC free article] [PubMed] [Google Scholar]