Abstract
Background
We compared survival outcomes and risk of venous thromboembolism (VTE) among patients with advanced and early-stage ovarian clear cell carcinoma (OCCC) and serous ovarian carcinoma (SOC), as well as potential links with interleukin-6 (IL-6) levels.
Methods
A multicenter case-control study was conducted in 370 patients with OCCC and 938 with SOC. In a subset of 200 cases, pretreatment plasma IL-6 levels were examined.
Findings
Patients with advanced OCCC had the highest 2-year cumulative VTE rates (advanced OCCC 43.1%, advanced SOC 16.2%, early-stage OCCC 11.9% and early-stage SOC 6.4%, P < 0.0001) and the highest median levels of IL-6 (advanced OCCC 17.8 pg/mL, advanced SOC 9.0 pg/mL, early-stage OCCC 4.2 pg/mL and early-stage SOC 5.0 pg/mL, 0.006). Advanced OCCC (hazard ratio [HR] 3.38, 0.0001), thrombocytosis (HR 1.42, P = 0.032) and elevated IL-6 (HR 8.90, P = 0.046) were independent predictors of VTE. In multivariate analysis, patients with advanced OCCC had significantly poorer 5-year progression-free and overall survival rates than those with advanced SOC (P < 0.01), and thrombocytosis was an independent predictor of decreased survival outcomes (P < 0.01). Elevated IL-6 levels led to poorer al rates in patients with OCCC (50% versus 87.5%, HR 4.89, P = 0.016) than in those with SOC (24.9% versus 40.8%), HR 1.40, P = 0.07).
Interpretation
Advanced OCCC is associated with an increased incidence of VTE and decreased survival outcomes, which has major implications for clinical management of OCCC.
Keywords: Ovarian cancer, Clear cell carcinoma, Venous thromboembolism, IL-6, Survival outcome
1. Introduction
Epithelial ovarian cancer comprises various histologic subtypes, including ovarian clear cell carcinoma (OCCC), which represents the second most common histologic subtype [1,2]. Accumulating evidence suggests that OCCC has distinct clinical and molecular characteristics compared with other histologic subtypes of epithelial ovarian cancer [3,4]. Clinically, patients with advanced-stage OCCC have poorer survival outcomes than those with advanced-stage serous ovarian carcinoma (SOC), whereas patients with early-stage OCCC and SOC have similar survival outcomes [5,6]. However, the reasons for this discrepancy and the mechanisms leading to poor survival outcomes in patients with advanced-stage OCCC have yet to be completely elucidated.
Venous thromboembolism (VTE) is a relatively common complication in ovarian cancer and is associated with decreased survival outcomes [7,8]. Epidemiologic studies have shown that patients with OCCC have an increased risk of VTE compared with patients with other histologic subtypes of epithelial ovarian cancer, and VTE adversely affects survival outcomes in patients with OCCC [9–11]. One possible biomarker linking OCCC with the increased risk of VTE is the proinflammatory cytokine interleukin 6 (IL-6). OCCC is known to be associated with higher IL-6 expression than in other subtypes of epithelial ovarian cancer [12], and IL-6 is a pivotal marker of paraneoplastic thrombocytosis, which is a prognostic factor for decreased survival of ovarian cancer [13, 14]. IL-6 could directly increase the risk of VTE by inducing procoagulant factors or indirectly increase the risk by inducing thrombocytosis [15, 16]. However, it is unclear whether VTE events and IL-6 levels differ by stage of disease. The aim of our study was to compare the survival outcomes and risk of VTE among patients with early- and advanced-stage OCCC and SOC, as well as to examine the relationship of these outcomes to IL-6 levels.
2. Patients and methods
2.1. Clinical information
A large-scale multicenter international case-control study was conducted in 10 academic institutions, including five from the United States, four from Japan and one from England. Institutional Review Board approval was obtained at each participating institution. Consecutive patients diagnosed with OCCC and SOC between January 1, 2000 and December 31, 2012 were identified from institutional databases. The STROBE guidelines for case-control studies were followed [17]. All patients had primary OCCC or SOC that was histologically confirmed from surgical specimens obtained in cytoreductive surgery. Those with a mixed histologic type were excluded from our analysis. A fraction of this study population was used in our previous study [13]. Medical records were retrospectively reviewed to obtain the following data: age, tumour markers, cytoreductive status, VTE characteristics, tumour characteristics and survival outcomes.
Tumour markers included CA-125 levels and the presence of thrombocytosis (based on platelet counts) at the time of diagnosis. The cutoff for thrombocytosis (platelet count ≥ 400 × 109 /L) was determined on the basis of prior work [18]. Among those with recurrent or progressive disease, CA-125 levels and platelet counts were also collected at the time of first recurrence or progression. Residual tumour size at the end of cytoreductive surgery was grouped as > 1 cm versus ≤ 1 cm. VTE characteristics included type of VTE (deep vein thrombosis [DVT], pulmonary embolism [PE] and others) and date of VTE diagnosis. Information for VTE was searched in both medical records and radiology reports for Doppler study, computed tomography scan and lung scan. Tumour characteristics included histologic subtype (OCCC or SOC) and cancer stage (early or advanced) based on International Federation of Gynecology and Obstetrics (FIGO) criteria. Early-stage disease was defined as FIGO stage I–II disease and advanced-stage disease was defined as FIGO stage III–IV disease [19].
For survival outcomes, we determined progression-free survival (PFS) and overall survival (OS). PFS was defined as the time interval between the date of ovarian cancer diagnosis and the date of the first recurrence or progression of disease or last follow-up if there was no recurrence or progression. OS was defined as the time interval between the date of ovarian cancer diagnosis and the date of death or last follow-up if the patient was still alive. OS after first recurrence or progression was also examined in the subset of patients who experienced recurrence or progression.
2.2. IL-6 measurement
Pretreatment plasma samples were available for consecutive patients in two institutions. Blood samples were obtained prior to surgery and centrifuged at 3000 rotations per minute for 10 min. Plasma was collected and stored in I-mL aliquots in a ‒80 °c freezer until it was processed. Plasma levels of IL-6 were examined using a Human IL-6 Quantikine enzyme-linked immunosorbent assay (ELISA) Kit (R&D Systems, Minneapolis, MN), according to the manufacturer’s instructions. The ELISA plates were read using Multiskan JX (ThermoFisher Scientific, Waltham, MA). All tests were done in triplicate. Written informed consent was obtained from each subject prior to the blood sampling.
2.3. Statistical analysis
The primary outcome of interest was the impact of stage-specific histologic subtype (advanced-OCCC, advanced-SOC, early-OCCC or early-SOC) on cumulative risk of VTE and survival outcomes. The secondary outcomes of interest were the correlation between IL-6 levels and VTE incidence and between IL-6 levels and survival outcomes, by stage-specific histologic subtype. Continuous variables were assessed for normality distribution using Kolmogorov–Smirnov test, and results were expressed using mean (±standard deviation) or median (range) as appropriate. Statistical significance of continuous variables was determined using Student t-test or Mann–Whitney U-test, depending on normality. Median values across the four groups were examined using Kruskal–Wallis test. For categorical variables, chi-square test or Fisher exact test was used as appropriate.
VTE is a time-dependent event after ovarian cancer diagnosis. Therefore, the statistical significance of cumulative risk of VTE was determined using log-rank test in univariate analysis. Cox proportional hazards regression modelling in multivariate analysis was also performed to identify independent predictors of VTE. For survival analysis, log-rank test in univariate analysis and Cox proportional hazards regression modelling in multivariate analysis were used, and results were expressed as hazard ratios (HR) and 95% confidence intervals (CI). In multivariate analysis, variables used in the final model were based on clinical relevance and impact in ovarian cancer; these included age ( ≥ 60 or <60 years), CA-125 levels (>35 or ≤ 35 IU/L), thrombocytosis (yes or no), residual tumour size (≤ 1 cm or > 1 cm), VTE (yes or no) and stage-specific histologic subtype (advanced-OCCC, advanced-SOC, early-OCCC, or early-SOC). Kaplan–Meier method was used to construct survival curves. P < 0.05 was considered statistical significant (two-tailed test). Statistical Package for the Social Sciences (SPSS, Inc., version 12.0, Chicago, IL) was used for all statistical analyses.
3. Results
3.1. Patient characteristics
We evaluated the records of 1,308 patients with ovarian cancer in the current study. Although most of the 938 patients with SOC had advanced disease (n = 836, 89.1%), most of the 370 patients with OCCC had early-stage disease (n = 264, 71.4%; Table Sl).
Stage-specific characteristics are shown in Table 1. The mean age of the entire cohort was 58.5 years. Both the advanced-OCCC and advanced-SOC groups had a high proportion of patients with elevated CA-125 levels at the initial cancer diagnosis (advanced-OCCC 96.2%, and advanced-soc 87.0%). The advanced-SOC group had the highest proportion of patients with thrombocytosis at initial diagnosis (41.6%), followed by the advanced-OCCC group (32.1%). The advanced-SOC group had the highest proportion of patients with residual tumour size > 1 cm at cytoreductive surgery (37.6%), followed by advanced-OCCC (26.4%). The median follow-up time of the entire cohort was 31.3 months, and there were 787 cases of recurrent or progressive disease (60.2%) and 487 deaths (37.2%) reported in the records.
Table 1.
Patient characteristics by stage-specific histologic subtype in our cohort (n = 1308).
| Characteristic | No. (%) |
P-valuea | |||
|---|---|---|---|---|---|
| OCCC (n = 370) |
SOC (n = 938) |
||||
| Advanced | Early | Advanced | Early | ||
| No. of patients | 106 (28.6) | 264 (71.4) | 836 (89.1) | 102 (10.9) | |
| Mean age (±standard deviation) | 57.6 years (±10.2 years) | 53.1 years (±10.2 years) | 60.3 years (±11.2 years) | 58.5 years (±10.6 years) | <0.0001 |
| <60 years | 62 (58.5) | 194 (73.5) | 394 (47.1) | 55 (53.9) | |
| ⩾60 years | 44 (41.5) | 70 (26.5) | 442 (52.9) | 47 (46.1) | |
| Median CA-125 levels at diagnosis (range) | 358 IU/L (17–21120 IU/L) | 59 IU/L (6–20887 IU/L) | 841 IU/L (8–86591 IU/L) | 223 IU/L (4–21634 IU/L) | <0.0001 |
| ⩽35 IU/L | 3 (2.8) | 87 (33) | 16 (1.9) | 26 (25.5) | |
| >35 IU/L | 102 (96.2) | 167 (63.3) | 727 (87.0) | 65 (63.7) | |
| Unknown | 1 (0.9) | 10 (3.8) | 93 (11.1) | 11 (10.8) | |
| Median CA-125 levels at first recurrence or progression (range)b | 108 IU/L (7–1019 IU/L) | 180 IU/L (6–1320 IU/L) | 351 IU/L (4–12957 IU/L) | 170 IU/L (5–681 IU/L) | 0.38 |
| ⩽35 IU/L | 15 (18.8) | 11 (33.3) | 80 (12.5) | 8 (25) | |
| >35 IU/L | 47 (58.8) | 19 (57.6) | 262 (40.8) | 18 (56.3) | |
| Unknown | 18 (22.5) | 3 (9.1) | 300 (46.7) | 6 (18.7) | |
| Median platelet counts at diagnosis (range) | 336 × 109/L (145–696 × 109/L) | 287 × 109/L (115–834 × 109/L) | 368 × 109/L (82–979 × 109/L) | 269 × 109/L (97–594 × 109/L) | <0.0001 |
| <400 × 109/L | 71 (67) | 217 (82.2) | 480 (57.4) | 92 (90.2) | |
| ⩾400 × 109/L | 34 (32.1) | 43 (16.3) | 348 (41.6) | 9 (8.8) | |
| Unknown | 1 (0.9) | 4 (1.5) | 8 (1.0) | 1 (1.0) | |
| Median platelet counts at first recurrence or progression (range)b | 271 × 109/L (54–1329 × 109/L) | 234 × 109/L (96–739 × 109/L) | 243 × 109/L (34–1178 × 109/L) | 239 × 109/L (127–779 × 109/L) | 0.028 |
| <400 × 109/L | 46 (57.5) | 26 (78.8) | 287 (44.7) | 23 (71.9) | |
| ⩾400 × 109/L | 17 (21.3) | 3 (9.1) | 45 (7.0) | 2 (6.3) | |
| Unknown | 17 (21.3) | 4 (12.1) | 310 (48.3) | 7 (21.9) | |
| Residual tumour size after surgery | <0.0001 | ||||
| >1 cm | 28 (26.4) | 3 (1.1) | 314 (37.6) | 8 (7.8) | |
| ⩽1 cm | 72 (67.9) | 261 (98.9) | 470 (56.2) | 94 (92.2) | |
| Unknown | 6 (5.7) | 0 (0) | 52 (6.2) | 0 (0) | |
| Recurrent or progressive disease | <0.0001 | ||||
| No | 26 (24.5) | 231 (87.5) | 194 (23.2) | 70 (68.6) | |
| Yes | 80 (75.5) | 33 (12.5) | 642 (76.8) | 32 (31.4) | |
| Death | <0.0001 | ||||
| No | 44 (41.5) | 244 (92.4) | 444 (53.1) | 89 (87.3) | |
| Yes | 62 (58.5) | 20 (7.6) | 392 (46.9) | 13 (12.7) | |
| Type of venous thromboembolism | <0.0001 | ||||
| None | 67 (63.2) | 230 (87.1) | 704 (84.2) | 92 (90.2) | |
| DVT alone | 17 (16.0) | 13 (4.9) | 69 (8.3) | 3 (2.9) | |
| PE alone | 9 (8.5) | 4 (1.5) | 41 (4.9) | 2 (2.0) | |
| DVT + PE | 12 (11.3) | 14 (5.3) | 21 (2.5) | 4 (3.9) | |
| Other | 1 (0.9) | 3 (1.1) | 1 (0.1) | 1 (1.0) | |
| Any | 39 (36.8) | 34 (12.9) | 132 (15.8) | 10 (9.8) | |
Abbreviations: OCCC, ovarian clear cell carcinoma; SOC, serous ovarian carcinoma; CA-125, cancer antigen 125; DVT, deep vein thrombosis; PE, pulmonary embolism.
The chi-square test was used for P values.
Advanced OCCC: n = 80; early-stage OCCC: n = 33; advanced SOC: n = 642; early-stage SOC: n = 32.
3.2. Characteristics of VTE
There were 215 VTEs reported in the entire cohort (16.4%). DVT alone (n = 102, 47.4%) was the most common type of VTE, followed by PE alone (n = 56, 26.0%) and DVT + PE (n = 51, 23.7%). The advanced-OCCC group had the highest incidence of any VTE (36.8%) among the 4 groups, as well as the highest incidence of DVT + PE (11.3%, P < 0.0001, Table 1).
The cumulative prevalence of VTE was examined by stage-specific histologic subtype (Table 2). In univariate analysis, advanced-OCCC was associated with the highest cumulative risk of VTE among the four groups (2-year cumulative VTE rates, advanced-OCCC 43.1%, advanced-soc 16.2%, early-OCCC 11.9% and early-soc 6.4%, P < 0.0001, Fig. 1A). In addition, age ≥ 60 years (19.3% versus 14.1%, P = 0.009), elevated CA-125 levels (17.0% versus 5.8%, P = 0.003) and thrombocytosis (22.3% versus 13.8%, P = 0.0001) were associated with increased risk of VTE. Residual tumour size ≤ 1 cm at cytoreductive surgery showed a protective effect against VTE (13.6% versus 20.2%, P = 0.028).
Table 2.
Risk factors for venous thromboembolism in our cohort (n = 1308).
| Variable | No. | 2-Year cumulative rate, % | Univariate |
Multivariate |
||
|---|---|---|---|---|---|---|
| HR (95%CI) | P-valuea | HR (95%CI) | P-valuea | |||
| Age | 0.009 | 0.016 | ||||
| <60 years | 705 | 14.1 | 1 | 1 | ||
| ⩾60 years | 603 | 19.3 | 1.42 (1.09–1.86) | 1.46 (1.07–1.99) | ||
| CA-125 levels | 0.003 | 0.025 | ||||
| ⩽35 IU/L | 132 | 5.8 | 1 | 1 | ||
| >35 IU/L | 1061 | 17.0 | 2.52 (1.33–4.76) | 2.38 (1.11–5.07) | ||
| Thrombocytosisb | 0.0001 | 0.032 | ||||
| No | 860 | 13.8 | 1 | 1 | ||
| Yes | 434 | 22.3 | 1.66 (1.26–2.18) | 1.42 (1.03–1.96) | ||
| Residual tumour size after surgery | 0.028 | 0.52 | ||||
| >1 cm | 353 | 20.2 | 1 | 1 | ||
| ⩽4 cm | 897 | 13.6 | 0.72 (0.53–0.97) | 0.89 (0.63–1.27) | ||
| Stage-specific histologic type | <0.0001 | |||||
| Advanced SOC | 836 | 16.2 | 1 | 1 | ||
| Advanced OCCC | 106 | 43.1 | 2.74 (1.92–3.92) | 3.38 (2.28–5.01) | <0.0001 | |
| Early-stage SOC | 102 | 6.4 | 0.54 (0.28–1.03) | 0.80 (0.38–1.68) | 0.55 | |
| Early-stage OCCC | 264 | 11.9 | 0.73 (0.50–1.06) | 1.23 (0.77–1.95) | 0.39 | |
Abbreviations: HR, hazard ratio; CI, confidence interval; CA-125, cancer antigen 125; SOC, serous ovarian carcinoma; OCCC, ovarian clear cell carcinoma.
The log-rank test was used for univariate analysis and Cox proportional hazards regression modelling was used for multivariate analysis.
Platelet count ⩾ 400 × 109/L.
Fig. 1.
Venous thromboembolism (VTE), interleukin-6 (IL-6) and survival of ovarian cancer. The log-rank test or chi-square test were used to generate the P values. (A) Cumulative risk of thromboembolism after diagnosis of ovarian cancer. (B) Overall survival by stage-specific histologic subtype. (C) Overall survival after recurrence or progression by stage-specific histologic subtype. (D) Proportion of patients with IL-6 levels ≥ 10 pg/mL (prior to treatment), by stage-specific histologic subtype. (E) Progression-free survival by IL-6 level in patients with ovarian clear cell carcinoma (OCCC). (F) Progression-free survival by IL-6 level in patients with serous ovarian carcinoma (SOC). Abbreviation: HR, hazard ratio.
To identify independent risk factors for VTE, we performed multivariate analysis (Table 2). After controlling for age, CA-125 levels, thrombocytosis, cytoreductive status and stage-specific histologic subtype, we found that advanced-OCCC remained an independent risk factor for VTE compared with advanced-SOC (HR 3.38, 95%CI 2.28–5.01, P < 0.0001). Thrombocytosis also remained an independent risk factor for VTE (HR 1.42, 95%CI 1.03–1.96, P = 0.032).
3.3. Survival analysis
In the univariate analysis for 5-year PFS rates, age years (24.6% versus 38.4%), CA-125 > 35 IU/L (27.4% versus 81.3%), thrombocytosis (15.3% versus 40.7%) and VTE (15.7% versus 35.3%) were associated with decreased PFS (P < 0.0001; Table 3). Five-year PFS rates were 13.3% for advanced-OCCC, 19.7% for advanced-SOC, 84.7% for early-OCCC and 66.9% for early-soc (P < 0.0001).
Table 3.
Factors influencing 5-year progression-free survival (PFS) rates in our cohort (n = 1308).
| Variable | No. | 5-Year PFS, % | Univariate |
Multivariate |
||
|---|---|---|---|---|---|---|
| HR (95%CI) | P-valuea | HR (95%CI) | P-valuea | |||
| Age | <0.0001 | 0.35 | ||||
| <60 years | 705 | 38.4 | 1 | 1 | ||
| ⩾60 years | 603 | 24.6 | 1.45 (1.26–1.67) | 1.08 (0.92–1.27) | ||
| CA-125 levels | <0.0001 | 0.002 | ||||
| ⩽35 IU/L | 132 | 81.3 | 1 | 1 | ||
| >35 IU/L | 1061 | 27.4 | 6.95 (4.40–11.0) | 2.29 (1.35–3.88) | ||
| Thrombocytosisb | <0.0001 | <0.0001 | ||||
| No | 860 | 40.7 | 1 | 1 | ||
| Yes | 434 | 15.3 | 2.09 (1.80–2.41) | 1.39 (1.18–1.64) | ||
| Residual tumour size after surgery | <0.0001 | <0.0001 | ||||
| >1 cm | 353 | 8.5 | 1 | 1 | ||
| ⩽4 cm | 897 | 43.3 | 0.35 (0.30–0.41) | 0.64 (0.54–0.76) | ||
| VTE | <0.0001 | 0.02 | ||||
| No | 1093 | 35.3 | 1 | 1 | ||
| Yes | 215 | 15.7 | 1.78 (1.49–2.12) | 1.28 (1.04–1.58) | ||
| Stage-specific histologic type | <0.0001 | |||||
| Advanced SOC | 836 | 19.7 | 1 | 1 | ||
| Advanced OCCC | 106 | 13.3 | 1.31 (1.04–1.66) | 1.45 (1.12–1.86) | 0.004 | |
| Early-stage SOC | 102 | 66.9 | 0.23 (0.16–0.32) | 0.36 (0.24–0.55) | <0.0001 | |
| Early-stage OCCC | 264 | 84.7 | 0.09 (0.06–0.13) | 0.14 (0.10–0.21) | <0.0001 | |
Abbreviations: HR, hazard ratio; CI, confidence interval; CA-125, cancer antigen 125; VTE, venous thromboembolism; SOC, serous ovarian carcinoma; OCCC, ovarian clear cell carcinoma.
The log-rank test was used for univariate analysis and Cox proportional hazards regression modelling was used for multivariate analysis.
Platelet count ⩾ 400 × 109/L.
In the multivariate analysis controlling for age, CA-125, thrombocytosis, cytoreductive status, VTE and stage-specific histologic subtype, patients with advanced-OCCC had poorer 5-year PFS rates than those with advanced-soc (HR 1.45, 95%CI 1.12–1.86, P = 0.004; Table 3). Thrombocytosis (HR 1.39, 95%CI 1.18–1.64, P < 0.0001) and VTE (HR 1.28, 95%CI 1.04–1.58, P = 0.02) also remained independent prognostic factors for decreased 5-year PFS rates.
For 5-year OS rates, in the univariate analysis, age years (44.7% versus 58.0%), CA-125 > 35 IU/L (49.1% versus 90.7%), thrombocytosis (35.8% versus 60.7%) and VTE (33.0% versus 55.8%) were associated with decreased 5-year OS rates (P < 0.0001; Table S2). Five-year OS rates were 28.2% for advanced-OCCC, 39.8% for advanced-SOC, 89.5% for early-OCCC and 82.1% for early-soc (Fig 1B, P < 0.0001).
In the multivariate analysis controlling for age, CA-125, thrombocytosis, cytoreductive status, VTE and stage-specific histologic subtype, those with advanced-OCCC had poorer 5-year OS rates than those with advanced-soc (HR 2.15, 95%CI 1.60–2.88, P < 0.0001; Table S2). In addition, thrombocytosis (HR 1.45, 95%CI 1.18–1.79, P = 0.001) remained an independent prognostic factor for decreased 5-year OS rates. In a post hoc multivariate analysis, 5-year OS rates did not differ between those with early-OCCC and those with early-soc (HR 0.62, 95%CI 0.27–1.39, P = 0.24).
3.4. Recurrent or progressive tumours
In our cohort, 787 patients experienced recurrence or progression. Among these patients, prognostic factors for OS after the first recurrence or progression of the tumour were examined (Table 4). Patterns of CA-125 and thrombocytosis at recurrence or progression were distinctively different from those at the initial ovarian cancer diagnosis. Specifically, CA-125 levels were similar across the 4 stage-specific histologic subtype groups (P = 0.38; Table 1). Although thrombocytosis was more common in those with advanced-SOC (41.6%) than in those with advanced-OCCC (32.1%) at the time of the initial cancer diagnosis, at the time of the first recurrence or progression, thrombocytosis was more common in those with advanced-OCCC (27.0%) than in those with advanced-SOC (13.6%, P = 0.028). Splenectomy at the time of cytoreductive surgery (n = 15, 1.1%) was not associated with thrombocytosis at the time of the first recurrence or progression of disease (9.1% versus 15.1%, P = 1.0).
Table 4.
Factors influencing overall survival (OS) after recurrence or progression in a portion of our cohort (n = 787).
| Variable | No. | 2-Year OS rate after recurrence, % | Univariate |
Multivariate |
||
|---|---|---|---|---|---|---|
| HR (95%CI) | P-valuea | HR (95%CI) | P-valuea | |||
| Age | 0.046 | 0.15 | ||||
| <60 years | 384 | 46.8 | 1 | 1 | ||
| ⩾60 years | 403 | 39.4 | 1.20 (1.00–1.44) | 1.21 (0.94–1.56) | ||
| CA-125 levelsb | <0.0001 | <0.0001 | ||||
| ⩽35 IU/L | 114 | 64.0 | 1 | 1 | ||
| >35 IU/L | 346 | 41.8 | 1.97 (1.43–2.72) | 2.02 (1.44–2.84) | ||
| Thrombocytosisb | <0.0001 | 0.002 | ||||
| No | 382 | 49.3 | 1 | 1 | ||
| 67 | 28.7 | 2.00 (1.47–2.71) | 1.67 (1.21–2.30) | |||
| VTE | 0.05 | 0.97 | ||||
| No | 625 | 45.3 | 1 | 1 | ||
| 162 | 35.2 | 1.24 (1.00–1.54) | 1.01 (0.74–1.37) | |||
| Stage-specific histologic subtype | <0.0001 | |||||
| Advanced SOC | 642 | 46.3 | 1 | 1 | ||
| Advanced OCCC | 80 | 17.0 | 2.24 (1.70–.94) | 2.68 (1.90–3.79) | <0.0001 | |
| Early-stage SOC | 32 | 61.3 | 0.59 (0.33–1.05) | 0.84 (0.44–1.60) | 0.60 | |
| Early-stage OCCC | 33 | 29.5 | 1.20 (0.77–1.89) | 2.12 (1.30–3.44) | 0.002 | |
Abbreviations: HR, hazard ratio; CI, confidence interval; CA-125, cancer antigen 125; VTE, venous thromboembolism; SOC, serous ovarian carcinoma; OCCC, ovarian clear cell carcinoma.
The log-rank test was used for univariate analysis and Cox proportional hazards regression modelling was used for multivariate analysis.
CA-125 and platelet counts were measured at the time of first recurrence or progression of disease.
The median OS duration after the first recurrence or progression was 13.0 months. In the multivariate analysis controlling for age, CA-125, thrombocytosis, VTE and stage-specific histologic subtype, thrombocytosis at the first recurrence or progression remained an independent prognostic factor for decreased 2-year OS rates after the first recurrence or progression (28.7% versus 46.3%; HR 1.67, 95%CI 1.21–2.30, 0.002). In addition, advanced-OCCC (17.0% versus 46.3%; HR 2.68, 95%CI 1.90–3.79, P < 0.0001) and early-OCCC (29.5% versus 46.3%; HR 2.12, 95%CI 1.30–3.44, P = 0.002) remained independent prognostic factors for decreased 2-year OS rates after the first recurrence or progression compared to advanced-SOC (Table 4 and Fig. 1C).
3.5. IL-6 levels
Plasma samples were available for 200 patients in the cohort (OCCC n = 38, and SOC n = 162). The median plasma IL-6 level among these 200 patients was 7.6 pg/mL, and 85 of the patients (42.5%) had IL-6 levels ≥ 10 pg/mL. Thrombocytosis was associated with IL-6 levels ≥ 10 pg/mL (66.7% versus 27.0%; HR 5.39, 95%CI 2.91–10.0, 0.0001). Patients with VTE had significantly higher IL-6 levels than those without VTE (median IL-6 levels: 14.4 versus 7.1 pg/mL, P = 0.003).
In a multivariate analysis controlling for age, CA-125 levels, thrombocytosis, cytoreductive status and stage-specific histologic subtype, high pretreatment IL-6 levels remained an independent predictor of VTE compared with lower IL-6 levels (2-year cumulative VTE rates, 5–19.9 versus <5 pg/mL, 14.0% versus 1.4%; HR 7.98, 95%CI 0.99–64.0, P = 0.051; and ≥ 20 versus pg/mL, 17.1% versus 1.4%; HR 8.90, 95%CI 1.04–76.0, P = 0.046; Table 5).
Table 5.
Interleukin-6 (IL-6) and risk of venous thromboembolism in a portion of our cohort (n = 200).
| Variable | No. | 2-Year cumulative rate, % | Univariate |
Multivariate |
||
|---|---|---|---|---|---|---|
| HR (95%CI) | P-value | HR (95%CI) | P-value | |||
| Age | 0.57 | 0.78 | ||||
| <60 years | 99 | 12.4 | 1 | 1 | ||
| ⩾60 years | 101 | 9.1 | 0.78 (0.33–1.86) | 0.87 (0.34–2.22) | ||
| CA-125 levels | 0.19 | 0.95 | ||||
| ⩽35 IU/L | 15 | 0.0 | 1 | 1 | ||
| >35 IU/L | 185 | 11.6 | 22.7 (0.02–28,568) | na | ||
| Thrombocytosisa | 0.028 | 0.31 | ||||
| No | 122 | 7.0 | 1 | 1 | ||
| Yes | 78 | 16.7 | 2.58 (1.07–6.22) | 1.65 (0.63–4.33) | ||
| Residual tumour size after surgery | 0.96 | 0.84 | ||||
| >1 cm | 63 | 11.5 | 1 | 1 | ||
| ⩽1 cm | 137 | 10.0 | 1.03 (0.39–2.67) | 1.11 (0.41–3.04) | ||
| Stage-specific histologic subtype | 0.039 | |||||
| Advanced SOC | 151 | 13.4 | 1 | 1 | ||
| Advanced OCCC | 6 | 40.0 | 5.07 (1.13–22.7) | 3.43 (0.76–15.6) | 0.11 | |
| Early-stage SOC | 11 | 0.0 | 0 | na | 0.99 | |
| Early-stage OCCC | 32 | 17.5 | 2.17 (0.77–6.13) | 4.00 (1.25–12.8) | 0.019 | |
| IL-6 levels prior to treatmentb | 0.012 | |||||
| <5 pg/mL | 68 | 1.4 | 1 | 1 | ||
| 5–19.9 pg/mL | 79 | 14.0 | 9.66 (1.25–74.9) | 7.98 (0.99–64.0) | 0.051 | |
| ⩾20 pg/mL | 53 | 17.1 | 12.1 (1.53–95.5) | 8.90 (1.04–76.0) | 0.046 | |
The log-rank test was used for univariate analysis and Cox proportional hazards regression modelling was used for multivariate analysis.
Abbreviations: HR, hazard ratio; CI, confidence interval; CA-125, cancer antigen 125; VTE, venous thromboembolism; SOC, serous ovarian carcinoma; OCCC, ovarian clear cell carcinoma; na, not available.
Platelet count ⩾ 400 × 109/L.
IL-6 levels were grouped into lower third (1–33‰, <5 pg/mL), mid third (34–66‰, 5–19.9 pg/mL) and upper third (⩾67‰, ⩾20 pg/mL).
Across the four stage-specific histologic subtype groups, the advanced-OCCC group had the highest proportion of IL-6 levels ≥ 10 pg/mL (advanced-OCCC 83.3%, advanced-soc 47.7%, early-OCCC 15.6% and early-soc 27.3%, P = 0.001, Fig. 1D). Similarly, the advanced-OCCC group had the highest median IL-6 levels among the four groups (advanced-OCCC 17.8 pg/mL, advanced-SOC 9.0 pg/mL, early-OCCC 4.2 pg/mL and early-SOC 5.0 pg/mL, P = 0.006). Similar to the results from the entire cohort, advanced-OCCC was associated with the highest risk of VTE across the 4 groups (2-year cumulative VTE risk: advanced-OCCC 40%, advanced-SOC 13.4%, early-OCCC 17.5% and early-soc 0%, P = 0.039).
Among the 200 patients in whom IL-6 levels were examined, IL-6 levels ≥ 10 pg/mL were associated with decreased 2-year PFS rates (27.4% versus 47.6%; HR 1.58, 95%CI 1.11–2.26, P = 0.013). However, when the patients were stratified by histologic subtype, the magnitude of the difference was larger in those with OCCC (2-year PFS rates, IL-6 ≥ 10 versus<10 pg/mL, 50% versus 87.5%, HR 4.89, 95%CI 1.17–20.5, P = 0.016, Fig. 1E) than in those with SOC (24.9% versus 40.8%, HR 1.40, 95%CI 0.97–2.03, P = 0.07, Fig. 1F).
4. Discussion
In the current study, advanced-OCCC, but not early-OCCC, was associated with a substantially increased risk of VTE compared with advanced-SOC. Patients with advanced-OCCC also had the highest frequency of elevated IL-6 levels, and those with advanced-OCCC had poorer survival outcomes than those with advanced-SOC. These findings suggest that the IL-6 pathway plays a pivotal role in the progression of OCCC. A proposed schema of the role of IL-6 in OCCC, based on our results, is shown in Fig. SI.
A recent study showed that paraneoplastic thrombo cytosis contributed to poor outcomes in ovarian cancer. The authors showed that IL-6 stimulates hepatocytes to induce thrombopoietin, which further induces megakaryocytes in bone marrow to produce platelets [13]. These IL-6–induced platelets promote tumour progression by providing platelet-related cancer cell protection from the host immune system, providing growth factors and promoting tumour angiogenesis. In our study, survival analyses showed that both advanced-OCCC and thrombocytosis were the common prognostic factors for decreased survival outcomes (Tables 3 and 4). In addition, elevated IL-6 levels were associated with thrombocytosis and advanced disease. These findings provide evidence for the existence of IL-6–mediated paraneoplastic thrombocytosis in advanced-OCCC.
Although IL-6–mediated paraneoplastic thrombocytosis is a proposed mechanism for tumour progression in advanced-OCCC, our study showed that thrombocytosis was less common in patients with advanced-OCCC than in those with advanced-SOC (32.1% versus 41.6%, Table 1). This implies that the IL-6–mediated paraneoplastic thrombocytosis is more prominent in SOC, and there may be an alternative IL-6–related pathway that contributes to tumour progression in OCCC, such as a direct autocrine pathway. Recently, the tumour microenvironment was reported to be a source of IL-6 in certain types of malignancy [20,21]. Some researchers have even speculated that the tumour microenvironment may supply IL-6 in SOC [22]. Therefore, hypothetically, there are 2 possible pathways for IL-6 interaction in ovarian cancer: (i) direct pathway via autocrine IL-6 signalling and (ii) IL-6–mediated indirect pathway via paraneoplastic thrombocytosis [13,221. In the IL-6 autocrine pathway in OCCC, downstream signalling of the IL-6 receptor is activated via the HIFlA-STAT3 cascade and ultimately induces VEGF, a key mediator for tumour angiogenesis [12,23,24]. Indeed, VEGF expression in OCCC is significantly higher than in other histologic types of epithelial ovarian cancer [25]. Collectively, our findings suggest that both direct and indirect IL-6 pathways may lead to tumour progression in OCCC, and the direct pathway may be more active in OCCC.
VTE is a common issue in OCCC [10]. Quality of life in cancer patients can be compromised as a result of the VTE itself and from drug injections, treatment costs and decreased survival outcomes. In our study, those with VTE had significantly higher levels of IL-6 than those who did not have VTE, and this association has not been well studied in cancer patients [26]. In addition, VTE was associated with thrombocytosis (Table 2), which has recently been recognised in the oncology field [27]. Our findings also showed that elevated IL-6 is associated with thrombocytosis, and those with advanced-OCCC had the highest frequency of multiple-site VTE (DVE + PE; 11.3%) compared with the other groups. Taken together, our findings indicate that VTE in OCCC is a clinical manifestation and surrogate marker of the aggressiveness of IL-6–driven tumour progression, and the decreased survival outcomes in those who develop VTE among OCCC patients is more likely from the aggressive tumour behaviour than from cardiovascular collapse.
Although overexpression of IL-6 in OCCC has been reported by various investigators, the exact mechanism driving IL-6 overexpression in OCCC has yet to be determined [28]. Although a fraction of IL-6 may come from the tumour microenvironment, a substantial fraction of IL-6 is speculated to come from tumour cells in OCCC given its high IL-6 expression. To date, ARIDIA mutation (loss of function) and PIK3CA mutation (gain of function) have been identified as common occurrences in OCCC (40–60%) [2,29]. Available evidence suggests a possible link between the PIK3CA mutation and IL-6 overexpression. Specifically, the PIK3CA mutation upregulates NF-KB, which leads to IL-6-dependent STAT3 activation [30]. Moreover, a recent pre-clinical study has shown that additional loss of ARIDIA function in the setting of PIK3CA overexpression is a key step in pathogenesis of OCCC [31]. Interestingly, IL-6 transcription in this OCCC model of ARIDIA and PIK3CA mutation is found to be elevated. Therefore, it is paramount to see if this correlation seen in pre-clinical study also exists in human samples of OCCC by sequencing ARIDIA and PIK3CA correlating with plasma IL-6 levels.
The unique characteristics of recurrent or progressive OCCC deserve further discussion. Generally, recurrent OCCC is resistant to therapy, with a response rate of < 10% [32]. This was also true in our study, and those with advanced-OCCC had poorer 2-year OS rates after recurrence or progression than those with advanced-SOC (Table 5). The advanced-OCCC group had a higher proportion of patients with thrombocytosis at the first recurrence or progression than the other groups, and this may represent a role for paraneoplastic thrombocytosis in the setting of advanced-OCCC. Interestingly, 2-year OS rates after recurrence or progression were also shorter in those with early-OCCC than in those with advanced-SOC. This implies that recurrent or progressive OCCC is quite chemoresistant regardless of original disease status. Our study did not have information for chemotherapy treatment after the first recurrence or progression and therefore we were unable to address this question.
A potential limitation of our study is that this is a retrospective study that may have some confounding factors. For example, we did not use the standard case record form to capture VTE events; however, all the participating institutions are tertiary care cancer centres and patient follow-up is quite consistent. In addition, central pathology review to confirm OCCC was not performed for the study. A potential weakness of our study is that central histopathologic slide review was not available for grading serous tumours [33,34]. Nevertheless, in post hoc analysis, high-grade SOC cases (n = 768) were compared to CCC cases (Tables S3–6). The results were consistent in that advanced-OCCC was significantly associated with increased risk of VTE (HR 3.34, 95%CI 2.23–4.99, P < 0.0001), decreased PFS (HR 1.55, 95%CI 1.20–2.01, P = 0.001), decreased OS (HR 2.19, 95%CI 1.62–2.97, P < 0.0001) and decreased OS after the first recurrence or progression of disease (HR 2.73, 95%CI 1.89–3.96, P < 0.0001) compared to advanced high-grade SOC in multivariate analysis. Lastly, a relatively small sample size for IL-6 assessment may limit generalisability.
In summary, our results indicate that advanced-OCCC is thrombogenic and may be a surrogate marker of tumour that is biologically more aggressive disease. Treatment involving both anti-thrombotic agents and blocking of IL-6 signalling may be an attractive approach in advanced-OCCC. A phase Il study examining the efficacy of the combination of a monoclonal antibody against IL-6 with siltuximab for platinum-resistant ovarian cancer showed that the combination had some therapeutic activity; however, this study was not solely for patients with OCCC [35]. Statin therapy is also suggested to reduce IL-6 activity and VTE risk by inhibiting inflammatory cytokines, resulting in reduced cancer-related mortality [36,37]. Currently (as of April 8, 2015), no ongoing clinical trial of a treatment targeting IL-6 for OCCC has been registered at clinicaltrials.gov. Further preclinical and clinical studies are warranted.
Supplementary Material
Acknowledgments
Grant support
This work was supported by the National Institutes of Health (CA016672, CA109298, CA177909, UH2TR000943, PSO CA083639, PSO CA098258), Cancer Prevention and Research Institute of Texas (RPI 10595, RP120214), Ovarian Cancer Research Fund, Inc. (Program Project Development Grant), Department of Defense (OC073399, OC120547), the Judi A Rees ovarian cancer research fund, Mr. and Mrs. Daniel P. Gordon, H.A. and Mary K. Chapman charitable foundation, the Blanton-Davis Ovarian Cancer Research Program, the Betty Anne Asche Murray Distinguished Professorship (A.K.S.) and Ensign Endowment for Ovarian Cancer Research (K.M. and L.D.R.). R.A.P. and J.M.H. were supported by the NCI-DHHS-NIH T32 Training grant (T32 CA101642).
Footnotes
Conflict of interest statement
None declared.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejca.2015.07.012.
References
- [1].Sung PL, Chang YH, Chao KC, Chuang CM. Global distribution pattern of histological subtypes of epithelial ovarian cancer: a database analysis and systematic review. Gynecol Oncol 2014;133:147–54. [DOI] [PubMed] [Google Scholar]
- [2].Anglesio MS, Carey MS, Kobel M, Mackay H, Huntsman DG. Clear cell carcinoma of the ovary: a report from the first Ovarian Clear Cell Symposium, June 24th, 2010. Gynecol Oncol 2011;121:407–15. [DOI] [PubMed] [Google Scholar]
- [3].Vaughan S, Coward Jl, Bast RC Jr, Berchuck A, Berek JS, Brenton JD, et al. Rethinking ovarian cancer: recommendations for improving outcomes. Nat Rev Cancer 2011;11:719–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].del Carmen MG, Birrer M, Schorge JO. Clear cell carcinoma of the ovary: a review of the literature. Gynecol Oncol 2012;126:481–90. [DOI] [PubMed] [Google Scholar]
- [5].El Braicu, Sehouli J, Richter R, Pietzner K, Denkert C, Fotopoulou C. Role of histological type on surgical outcome and survival following radical primary tumour debulking of epithelial ovarian, fallopian tube and peritoneal cancers. Br J Cancer 2011;105:1818–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Farley JH, Brady WE, Birrer MJ, Gershenson DM, Fleming GF. An evaluation of survival of ovarian cancer patients with clear cell carcinoma versus serous carcinoma treated with platinum therapy: A Gynecologic Oncology Group experience. J Clin Oncol 2013;3(suppl), abstr 5534. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Gunderson CC, Thomas ED, Slaughter KN, Farrell R, Ding K, Farris RE, et al. The survival detriment of venous thromboembolism with epithelial ovarian cancer. Gynecol Oncol 2014;134: 73–7. [DOI] [PubMed] [Google Scholar]
- [8].Fotopoulou C, duBois A, Karavas AN, Trappe R, Aminossadati B, Schmalfeldt B, et al. Incidence of venous thromboembolism in patients with ovarian cancer undergoing platinum/paclitaxel-containing first-line chemotherapy: an exploratory analysis by the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer study Group. J Clin Oncol 2008;26:2683–9. [DOI] [PubMed] [Google Scholar]
- [9].Diaz ES, Waits AE, Karlan BY, Walsh CS. Venous thromboembolism during primary treatment of ovarian clear cell carcinoma is associated with decreased survival. Gynecol Oncol 2013;131:541–5. [DOI] [PubMed] [Google Scholar]
- [10].Matsuura Y, Robertson G, Marsden DE, Kim SN, Gebski V, Hacker NF. Thromboembolic complications in patients with clear cell carcinoma of the ovary. Gynecol Oncol 2007;104:406–10. [DOI] [PubMed] [Google Scholar]
- [11].Duska LR, Garrett L, Henretta M, Ferriss JS, Lee L, Horowitz N. When ‘never-events’ occur despite adherence to clinical guidelines: the case of venous thromboembolism in clear cell cancer of the ovary compared with other epithelial histologic subtypes. Gynecol Oncol 2009;116:374–7. [DOI] [PubMed] [Google Scholar]
- [12].Anglesio MS, George J, Kulbe H, Friedlander M, Rischin D, Lemech C, et al. IL6-STAT3-HIF signaling and therapeutic response to the angiogenesis inhibitor sunitinib in ovarian clear cell cancer. Clin Cancer Res 2011;17:2538–48. [DOI] [PubMed] [Google Scholar]
- [13].Stone RL, Nick AM, McNeish IA, Balkwill F, Han HD, Bottsford-Miller J, et al. Paraneoplastic thrombocytosis in ovarian cancer. N Engl J Med 2012;366:610–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Davis AN, Afshar-Kharghan V, Sood AK. Platelet effects on ovarian cancer. Semin Oncol 2014;41:378–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Kerr R, Stirling D, Ludlam CA. Interleukin 6 and haemostasis. Br J Haematol 2001;115:3–12. [DOI] [PubMed] [Google Scholar]
- [16].Fadare O Clear cell carcinomas of the gynecologic tract and thromboembolic events: what do we know so far? Womens Health (Lond Engl) 2014;10:479–81. [DOI] [PubMed] [Google Scholar]
- [17].STROBE guideline at http://www.strobe-statement.org/ [accessed April 2015].
- [18].Matsuo K, Yessaian AA, Lin YG, Pham HQ, Muderspach Ll, Liebman HA, et al. Predictive model of venous thromboembolism in endometrial cancer. Gynecol Oncol 2012;128:544–51. [DOI] [PubMed] [Google Scholar]
- [19].Prat J Staging classification for cancer of the ovary, fallopian tube, and peritoneum. Int J Gynaecol Obstet 2013;124:1–5. [DOI] [PubMed] [Google Scholar]
- [20].Rosean TR, Tompkins VS, Olivier AK, Sompallae R, Norian LA, Morse HC 3rd, et al. The tumor microenvironment is the main source of IL-6 for plasma cell tumor development in mice. Leukemia 2015;29:233–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Zhang L, Yang J, Qian J, Li H, Romaguera JE, Kwak LW, et al. Role of the microenvironment in mantle cell lymphoma: IL-6 is an important survival factor for the tumor cells. Blood 2012;120:3783–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Yanaihara N, Anglesio MS, Ochiai K, Hirata Y, Saito M, Nagata C, et al. Cytokine gene expression signature in ovarian clear cell carcinoma. Int J Oncol 2012;41:1094–100. [DOI] [PubMed] [Google Scholar]
- [23].Nilsson MB, Langley RR, Fidler IJ. Interleukin-6, secreted by human ovarian carcinoma cells, is a potent proangiogenic cytokine. Cancer Res 2005;65:10794–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Coward JI, Kulbe H. The role of interleukin-6 in gynaecological malignancies. Cytokine Growth Factor Rev 2011;23:333–42. [DOI] [PubMed] [Google Scholar]
- [25].Yamamoto S, Konishi 1, Mandai M, Kuroda H, Komatsu T, Nanbu K, et al. Expression of vascular endothelial growth factor (VEGF) in epithelial ovarian neoplasms: correlation with clinicopathology and patient survival, and analysis of serum VEGF levels. Br J Cancer 1997;76:1221–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Reitter EM, Ay C, Kaider A, Pirker R, Zielinski C, Zlabinger G, et al. Interleukin levels and their potential association with venous thromboembolism and survival in cancer patients. Clin Exp Immunol 2014;177:253–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Connolly GC, Phipps RI), Francis CW. Platelets and cancer-associated thrombosis. Semin Oncol 2014;41:302–10. [DOI] [PubMed] [Google Scholar]
- [28].Glasspool R M, McNeish IA. Clear cell carcinoma of ovary and uterus. Curr Oncol Rep 2013;15:566–72. [DOI] [PubMed] [Google Scholar]
- [29].Okamoto A, Glasspool R M, Mabuchi S, Matsumura N, Nomura H, Itamochi H, et al. Gynecologic Cancer InterGroup (GCIG) consensus review for clear cell carcinoma of the ovary. Int J Gynecol Cancer 2014;24:S20–5. [DOI] [PubMed] [Google Scholar]
- [30].Hutti JE, Pfefferle AD, Russell SC, Sircar M, Perou CM, Baldwin AS. Oncogenic P13K mutations lead to NF-kappaB-dependent cytokine expression following growth factor deprivation. Cancer Res 2012;72:3260–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Chandler RL, Damrauer JS, Raab JR, Schisler JC, Wilkerson MD, Didion JP, et al. Coexistent ARIDIA-PIK3CA mutations promote ovarian clear-cell tumorigenesis through pro-tumorigenic inflammatory cytokine signalling. Nat Commun 2015;6:6118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Crotzer DR, Sun CC, Coleman RL, Wolf JK, Levenback CF, Gershenson DM. Lack of effective systemic therapy for recurrent clear cell carcinoma of the ovary. Gynecol Oncol 2007;105:404–8. [DOI] [PubMed] [Google Scholar]
- [33].Bodurka DC, Deavers MT, Tian C, Sun CC, Malpica A, Coleman RL, et al. Reclassification of serous ovarian carcinoma by a 2-tier system: a Gynecologic Oncology Group Study. Cancer 2012;118:3087–94. [DOI] [PubMed] [Google Scholar]
- [34].Malpica A, Deavers MT, Lu K, Bodurka DC, Atkinson EN, Gershenson DM, et al. Grading ovarian serous carcinoma using a two-tier system. Am J Surg Pathol 2004;28:496–504. [DOI] [PubMed] [Google Scholar]
- [35].Coward J, Kulbe H, Chakravarty P, et al. Interleukin-6 as a therapeutic target in human ovarian cancer. Clin Cancer Res 2011;17:6083–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [36].Rodriguez AL, Wojcik BM, Wrobleski SK, Myers DD Jr, Wakefield TW, Diaz JA. Statins, inflammation and deep vein thrombosis: a systematic review. J Thromb Thrombolysis 2012;33:371–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related mortality. N Engl J Med 2012;367: 1792–802. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.