PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
N Engl J Med. Author manuscript; available in PMC Aug 16, 2012.
Published in final edited form as:
PMCID: PMC3296780
NIHMSID: NIHMS360029
Paraneoplastic Thrombocytosis in Ovarian Cancer
Rebecca L. Stone, M.D., Alpa M. Nick, M.D., Iain A. McNeish, Ph.D., Frances Balkwill, Ph.D., Hee Dong Han, Ph.D., Justin Bottsford-Miller, M.D., Rajesha Rupaimoole, M.S., Guillermo N. Armaiz-Pena, Ph.D., Chad V. Pecot, M.D., Jermaine Coward, Ph.D., Michael T. Deavers, M.D., Hernan G. Vasquez, Ph.D., Diana Urbauer, M.S., Charles N. Landen, M.D., Wei Hu, M.D., Ph.D., Hannah Gershenson, Koji Matsuo, M.D., Mian M.K. Shahzad, M.D., Erin R. King, M.D., Ibrahim Tekedereli, M.D., Bulent Ozpolat, M.D., Ph.D., Edward H. Ahn, M.D., Virginia K. Bond, M.D., Rui Wang, B.S., Angela F. Drew, Ph.D., Francisca Gushiken, M.D., Katherine Collins, B.S., Koen DeGeest, M.D., Susan K. Lutgendorf, Ph.D., Wah Chiu, Ph.D., Gabriel Lopez-Berestein, M.D., Vahid Afshar-Kharghan, M.D., and Anil K. Sood, M.D.
Departments of Gynecologic Oncology and Reproductive Medicine (R.L.S., A.M.N., H.D.H., J.B.-M., W.H., H.G., K.M., M.M.K.S., E.R.K., A.K.S.), Cancer Biology (R.R., G.L.-B., A.K.S.), Experimental Therapeutics (G.N.A.-P., I.T., B.O., G.L.-B.), Hematology and Oncology (C.V.P.), Pathology (M.T.D.), Benign Hematology (H.G.V., V.A.-K.), Biostatistics (D.U.), and Leukemia (F.G.), and the Center for RNA Interference and Non-Coding RNA (H.D.H., G.L.-B., A.K.S.), University of Texas M.D. Anderson Cancer Center, the Department of Biochemistry and Molecular Biology, Baylor College of Medicine (R.W., W.C.), and the Department of Nanomedicine and Bioengineering, UT Health (G.L.-B.) — all in Houston; the Barts Cancer Institute, Queen Mary, University of London, London (I.A.M., F.B., J.C.); the Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Alabama, Birmingham (C.N.L.); the Department of Obstetrics and Gynecology, University of Maryland, Baltimore (E.H.A., V.K.B.); the Department of Cancer and Cell Biology, University of Cincinnati, Cincinnati (A.F.D.); and the Departments of Psychology (K.C., S.K.L.), Urology (S.K.L.), and Obstetrics and Gynecology (K.D., S.K.L.), and the Holden Comprehensive Cancer Center (K.D., S.K.L.), University of Iowa, Iowa City
Address reprint requests to Dr. Sood at the Department of Gynecologic Oncology and Reproductive Medicine, M.D. Anderson Cancer Center, Unit 1362, P.O. Box 301439, Houston, TX 77230-1439, or at ; asood/at/mdanderson.org
BACKGROUND
The mechanisms of paraneoplastic thrombocytosis in ovarian cancer and the role that platelets play in abetting cancer growth are unclear.
METHODS
We analyzed clinical data on 619 patients with epithelial ovarian cancer to test associations between platelet counts and disease outcome. Human samples and mouse models of epithelial ovarian cancer were used to explore the underlying mechanisms of paraneoplastic thrombocytosis. The effects of platelets on tumor growth and angiogenesis were ascertained.
RESULTS
Thrombocytosis was significantly associated with advanced disease and shortened survival. Plasma levels of thrombopoietin and interleukin-6 were significantly elevated in patients who had thrombocytosis as compared with those who did not. In mouse models, increased hepatic thrombopoietin synthesis in response to tumor-derived interleukin-6 was an underlying mechanism of paraneoplastic thrombocytosis. Tumor-derived interleukin-6 and hepatic thrombopoietin were also linked to thrombocytosis in patients. Silencing thrombopoietin and interleukin-6 abrogated thrombocytosis in tumor-bearing mice. Anti–interleukin-6 antibody treatment significantly reduced platelet counts in tumor-bearing mice and in patients with epithelial ovarian cancer. In addition, neutralizing interleukin-6 significantly enhanced the therapeutic efficacy of paclitaxel in mouse models of epithelial ovarian cancer. The use of an antiplatelet antibody to halve platelet counts in tumor-bearing mice significantly reduced tumor growth and angiogenesis.
CONCLUSIONS
These findings support the existence of a paracrine circuit wherein increased production of thrombopoietic cytokines in tumor and host tissue leads to paraneoplastic thrombocytosis, which fuels tumor growth. We speculate that countering paraneoplastic thrombocytosis either directly or indirectly by targeting these cytokines may have therapeutic potential. (Funded by the National Cancer Institute and others.)
Platelets are highly reactive cellular effectors of hemostasis, immunity, and inflammation.1 The concept that platelets play key roles in cancer growth and metastasis is long-standing. In fact, the clinical observation that thrombocytosis (defined as a platelet count of >450,000 per cubic millimeter) occurs in patients with solid tumors was made more than 100 years ago.2,3 Nearly 40% of persons incidentally found to have platelet counts exceeding 400,000 per cubic millimeter in the absence of iron deficiency and benign inflammatory conditions have an occult cancer, most commonly a primary gastrointestinal, lung, breast, or ovarian cancer.4 Beyond clinical observations, experimental evidence suggests that platelets actively promote cancer progression through diverse mechanisms, including protection of cancer cells from immune surveillance, negotiation of cancer-cell arrest in the micro-vasculature, and stimulation of angiogenesis.5
Given the short life span of a circulating platelet, approximately 1011 platelets must be produced daily to maintain a normal platelet count in adults.6,7 This high baseline level of production has the potential to markedly increase in the context of cancer. The mechanisms underlying this surge in platelets as well as its biologic significance are not well understood and are the focus of the current study. We investigated the potential clinical and biologic implications of a paracrine circuit that promotes megakaryopoiesis and thrombocytosis in the context of ovarian cancer.
CLINICAL DATA
After approval by the local institutional review boards, clinical data, including history of coexisting inflammatory conditions, were collected on 619 consecutive patients with primary epithelial ovarian cancer from four U.S. academic medical centers. The effect of treatment with an antihuman interleukin-6 antibody (siltuximab [CNTO 328], at a dose of 5.4 mg per kilogram of ideal body weight infused once every 2 weeks) on platelet counts in 18 patients with ovarian cancer was evaluated as part of a phase 1 and 2 study (EudraCT number, 2006-005704-13).8 Patients provided written informed consent for participation in this trial and for use of their banked specimens in the laboratory studies pertaining to this investigation.
MOUSE MODELS
All in vivo experiments in animals were approved by the institutional animal care and use committee. The development and characterization of the orthotopic mouse models of epithelial ovarian cancer have been described previously.913 Additional details regarding in vivo experiments with small interfering RNA (siRNA) delivery (Fig. 1A and 1B in the Supplementary Appendix, available with the full text of this article at NEJM.org), siltuximab, antiplatelet antibody (Emfret Analytics) (Fig. 2A through 2E in the Supplementary Appendix), inflammatory agents, and platelets from transgenic mice that express yellow fluorescent protein (YFP) are described in the Supplementary Appendix.
LABORATORY MEASURES
Plasma levels of key thrombopoietic factors were measured with the use of human cytokine immunobead panels (Milliplex, Millipore) coupled with a multiplex assay (involving xMAP technology, Luminex) or with enzyme-linked immunosorbent assays (ELISA) (Quantikine, R&D Systems). Real-time reverse-transcriptase–polymerase-chain-reaction (RT-PCR) assays were performed with whole RNA extracted from tumor specimens (obtained from 46 patients) or liver specimens (from 10 patients and 10 mice) with the use of primer sets specific for human interleukin-6 and murine or human thrombopoietin, respectively, as previously described.9,14,15 Proliferation index and microvessel density were evaluated by staining for Ki67 and CD31 antigens, respectively.16,17 Dual immunofluorescence staining for CD31 and desmin and for CD31 and terminal deoxynucleotidyl transferase dUTP biotin nick end labeling (TUNEL) was used to assess pericyte coverage and apoptotic tumor and endothelial cells, respectively.18 The in vitro effects of gel-filtered platelets (see the Supplementary Appendix) on cell migration and proliferation in epithelial ovarian cancer were evaluated with the use of two-chamber chemotaxis and bromodeoxyuridine flow-cytometric assays (Click-iT EdU, Invitrogen), respectively.19
STATISTICAL ANALYSIS
Associations between platelet count and clinico-pathological variables were tested in 619 patients with newly diagnosed epithelial ovarian cancer with the use of Fisher’s exact test. Kaplan–Meier survival curves were generated and compared with the use of a log-rank statistic. Multivariate analysis of these data was performed with the Cox proportional-hazards model. To determine whether plasma thrombopoietic cytokine levels were associated with platelet counts in 150 patients with newly diagnosed epithelial ovarian cancer, we calculated Spearman’s rank-correlation coefficient for each pairwise association and tested for a significant difference from 0. We used Wilcoxon rank-sum tests to determine differences in thrombopoietic cytokines between patients with normal platelet counts (150,000 to 450,000 per cubic millimeter) and those with high counts (>450,000 per cubic millimeter).
For interleukin-6 analysis, we used a level close to the median to classify patients with newly diagnosed epithelial ovarian cancer as having low interleukin-6 levels (=10 pg per milliliter, 143 patients) or high levels (>10 pg per milliliter, 167 patients). Pearson’s correlation was used to calculate correlations between continuous variables. For preclinical data, Student’s t-test was used to test differences in sample means, with a sample of 10 mice per group (80% power to detect a 50% reduction in tumor weight at a 5% level of statistical significance). All testing was two-sided, and a P value of less than 0.05 was considered to indicate statistical significance. All analyses were completed with SAS software, version 9.1 for PC (SAS Institute).
CLINICAL IMPLICATIONS OF PARANEOPLASTIC THROMBOCYTOSIS
The mean (±SD) platelet count in patients with epithelial ovarian cancer and thrombocytosis was 558,410±90,010 per cubic millimeter (median, 542,000 per cubic millimeter) and without thrombocytosis was 316,570±70,140 per cubic millimeter (median, 318,000 per cubic millimeter) (Table 1 in the Supplementary Appendix). Thirty-one percent of the patients (192 of 619) had thrombocytosis at the time of initial diagnosis of epithelial ovarian cancer. Less than 2% of the patients had a coexisting nonmalignant inflammatory condition, and none of these patients had thrombocytosis at initial presentation. Although women with thrombocytosis had slightly lower hemoglobin levels than those with normal platelet counts (mean hemoglobin level, 11.9 vs. 12.7 g per deciliter; P<0.001), mean corpuscular volume and serum ferritin levels did not differ significantly between the two groups (Table 2 in the Supplementary Appendix). These data suggest that other inflammatory conditions and iron deficiency are unlikely to be responsible for the thrombocytosis observed in women with epithelial ovarian cancer.
Patients with thrombocytosis were significantly more likely to have advanced-stage disease, vascular thromboembolic complications, and higher preoperative levels of cancer antigen 125 than those with normal platelet counts (Fig. 3 in the Supplementary Appendix). Women with platelet counts of more than 450,000 per cubic millimeter had a significantly shorter median time to disease progression than those with normal platelet counts (Fig. 1A). The median overall survival among patients with thrombocytosis was 2.62 years, as compared with 4.65 years among those with normal platelet counts (P<0.001) (Fig. 1B). In the multivariate model that included age, disease stage, tumor grade, histologic type, and extent of surgical cytoreduction, thrombocytosis remained an independent predictor of compromised survival (P<0.001) (Table 3 in the Supplementary Appendix).
Figure 1
Figure 1
Clinical Significance of and Thrombopoietic Cytokines Associated with Paraneoplastic Thrombocytosis in Patients with Epithelial Ovarian Cancer
PARANEOPLASTIC THROMBOCYTOSIS IN MOUSE MODELS OF EPITHELIAL OVARIAN CANCER
On the basis of the clinical findings described above, we evaluated the effect of cancer on platelet counts in orthotopic mouse models of epithelial ovarian cancer. As compared with platelet counts in control mice, those in mice with epithelial ovarian cancer were increased by 30 to 130% (P = 0.009 for HeyA8, P = 0.003 for A2780ip2, and P = 0.001 for 2774 mouse models) (Fig. 4A in the Supplementary Appendix). Platelet counts correlated significantly with tumor burden (r = 0.78, P<0.001) and number of intraperitoneal metastases (r = 0.65, P = 0.004) (Fig. 4B and 4C in the Supplementary Appendix). Platelet counts were also increased by 59 to 64% in two syngeneic mouse models of ovarian cancer (P=0.01) (Fig. 4D in the Supplementary Appendix). To assess generalizability, we also quantified platelets in mouse models of breast cancer (GILM2 cells), uterine cancer (Ishikawa cells), and pancreatic cancer (heterotopic human pancreatic xenografts). Platelet counts were significantly increased in the models of uterine and pancreatic cancer but not in the breast-cancer model (Fig. 4E in the Supplementary Appendix).
To better understand the temporal relationship between tumor growth and thrombocytosis, we carried out a longitudinal assessment of platelet counts in HeyA8–luciferase-positive tumor-bearing mice. Platelet counts began to rise when a tumor was detected by bioluminescence imaging and increased in parallel with grossly measurable disease (Fig. 5 in the Supplementary Appendix).
AN UNDERLYING MECHANISM OF PARANEOPLASTIC THROMBOCYTOSIS
To test the hypothesis that factors produced by cancer cells or host tissues (or both) stimulate megakaryopoiesis and paraneoplastic thrombocytosis, we first enumerated megakaryocytes in the bone marrow and spleens from tumor-bearing mice. Mean platelet counts strongly correlated with mean medullary and splenic megakaryocyte counts (r = 0.95, P = 0.01). As compared with medullary and splenic megakaryocyte counts in controls, those in tumor-bearing mice were increased by a factor of 2 to 3 and 7 to 13, respectively (P<0.001) (Fig. 6A in the Supplementary Appendix). To determine whether the findings would be similar in patients with epithelial ovarian cancer, we surveyed our clinical data set for women who underwent a bone marrow biopsy. Among the eight patients in whom a bone marrow biopsy was performed, platelet counts correlated with megakaryocyte density (two patients with thrombocytosis had megakaryocyte hyperplasia, four patients had normal platelet and megakaryocyte counts, and two patients had thrombocytopenia and megakaryocyte hypoplasia) (Fig. 6B in the Supplementary Appendix).
Next, we considered the possible involvement of humoral factors known to physiologically regulate megakaryopoiesis and platelet production. We quantified plasma levels of 10 key thrombopoietic factors in an independent cohort of 150 patients with newly diagnosed epithelial ovarian cancer (31% of whom had thrombocytosis, a frequency consistent with that in the initial cohort). Platelet counts correlated significantly with plasma interleukin-6 levels (Spearman’s rank-correlation coefficient, 0.37; P<0.001) and thrombopoietin levels (Spearman’s rank-correlation coefficient, 0.29; P<0.001) (Table 4 in the Supplementary Appendix). Plasma levels of interleukin-6 and thrombopoietin were significantly elevated in patients with thrombocytosis (Fig. 1C and 1D, and Table 5 in the Supplementary Appendix).
Plasma levels of human interleukin-6, but not mouse interleukin-6, were significantly elevated in all mouse models of epithelial ovarian cancer that were tested (Fig. 7A in the Supplementary Appendix). Thrombopoietin is mainly synthesized by the liver at a fixed rate, but little is known about the influence of disease states such as cancer on hepatic thrombopoietin production.20 In liver specimens resected from both control and tumor-bearing mice, levels of hepatic thrombopoietin messenger RNA were 38 to 64% higher in tumor-bearing mice (P = 0.43 for the 2774 mouse model and P<0.001 for the A2780ip2 and HeyA8 mouse models of epithelial ovarian cancer) (Fig. 7B in the Supplementary Appendix).
On the basis of data suggesting that interleukin-6 enhances hepatic thrombopoietin synthesis,21,22 we measured platelet counts and plasma thrombopoietin levels in tumor-bearing (ID8VEGF-164 syngeneic model) mice that lacked functional hepatocyte interleukin-6 receptors. Thrombocytosis was not observed in these mice (Fig. 2A). Furthermore, the mean plasma thrombopoietin level was 33% lower in the tumor-bearing mice that lacked functional hepatocyte interleukin-6 receptors than in tumor-bearing mice with functional receptors (flox/flox controls) (Fig. 7C in the Supplementary Appendix). These findings suggest that increased hepatic thrombopoietin synthesis in response to tumor-derived interleukin-6 is an underlying mechanism of paraneoplastic thrombocytosis (Fig. 2B).
Figure 2
Figure 2
Underlying Mechanism of Paraneoplastic Thrombocytosis in Epithelial Ovarian Cancer
To determine whether this paracrine signaling is the basis for paraneoplastic thrombocytosis in humans, we quantified tumor and plasma interleukin-6 levels in a subset of 46 patients with newly diagnosed epithelial ovarian cancer, using a quantitative RT-PCR assay and ELISA, respectively. A strong correlation between tumor and plasma interleukin-6 levels was observed (r = 0.8, P<0.001) (Fig. 8A in the Supplementary Appendix). Moreover, mean tumor interleukin-6 levels were significantly higher in patients with thrombocytosis than in those with normal platelet counts (Fig. 8B in the Supplementary Appendix). Interleukin-6 levels were also significantly associated with plasma thrombopoietin levels (Spearman’s rank-correlation coefficient, 0.6; P<0.001). We also used a quantitative RT-PCR assay to examine 10 liver samples removed from patients with epithelial ovarian cancer during cytoreductive surgery. Hepatic thrombopoietin expression closely paralleled platelet counts (r = 0.64, P = 0.04) (Fig. 8C in the Supplementary Appendix).
To better understand the clinical significance of this paracrine-mediated paraneoplastic thrombocytosis, we evaluated the effect of circulating interleukin-6 levels on progression-free and overall survival among 310 patients with epithelial ovarian cancer. In univariate analyses, patients with plasma interleukin-6 levels that were 10 pg per milliliter or lower had improved progression-free survival (Fig. 8D in the Supplementary Appendix) and overall survival (Fig. 2C), as compared with patients who had plasma interleukin-6 levels that were higher than 10 pg per milliliter. In multivariate models, plasma interleukin-6 levels of more than 10 pg per milliliter and platelet counts of more than 450,000 per cubic millimeter remained independent predictors of a poor prognosis even after adjustment for stage, grade, and volume of residual disease after cytoreductive surgery.
Next, to ascertain whether interleukin-6 and thrombopoietin play functional roles in paraneoplastic thrombocytosis, we tested the effect of silencing these factors on thrombocytosis in the A2780ip2 orthotopic mouse model. The combination of interleukin-6 and thrombopoietin siRNA was most effective, completely abrogating thrombocytosis (Fig. 3A). We also explored the therapeutic value of targeting interleukin-6 with siltuximab (humanized anti–interleukin-6 antibody), paclitaxel (a chemotherapeutic agent commonly used in epithelial ovarian cancer), or the two agents combined. Siltuximab or paclitaxel monotherapy significantly reduced tumor growth in mouse models of epithelial ovarian cancer (2774 and A2780ip2 models). Combination treatment was most effective, reducing tumor growth by more than 90% in both models. Assessment of platelet counts at the conclusion of treatment revealed that siltuximab blocked the thrombocytosis characteristic of untreated tumor-bearing mice (Fig. 9A and 9B in the Supplementary Appendix). Moreover, platelet counts rebounded by 34% within 10 days after the end of siltuximab therapy, pointing to the direct role of tumor-derived interleukin-6 in thrombocytosis.
Figure 3
Figure 3
Functional Role of Interleukin-6 and Thrombo-poietin in Paraneoplastic Thrombocytosis
To support these in vivo data with human clinical data, we tracked platelet counts in a cohort of patients with ovarian cancer who received treatment with single-agent siltuximab every 2 weeks for 12 weeks. Anti–interleukin-6 treatment resulted in a significant and sustained reduction in platelet counts, from a mean count of 356×109±147×109 per liter before treatment to 243×109±85×109 per liter after 3 weeks of treatment (P = 0.009) (Fig. 3B).
BIOLOGIC SIGNIFICANCE OF PARANEOPLASTIC THROMBOCYTOSIS
To evaluate the biologic role of platelets in tumor growth, we used an antiplatelet antibody directed against mouse glycoprotein 1b a (at a dose of 0.5 μg per gram of body weight) to induce a sustained 50% reduction in platelets in the A2780ip2 mouse model. Halving the circulating platelet count inhibited tumor growth by 50% (P = 0.03) (Fig. 4A). Microscopical examination of tumors from mice treated with antiplatelet antibody revealed substantial coagulative necrosis (Fig. 4B). As compared with the use of control IgG, the use of antiplatelet antibody to deplete platelets resulted in a 44% decrease in proliferation (P<0.001) (Fig. 4B) and a 51% reduction in microvessel density (P<0.001) (Fig. 10A in the Supplementary Appendix). Platelet depletion reduced pericyte coverage by 60% (P<0.001); tumor microvessels were predominantly devoid of pericytes, and existing pericytes appeared to be loosely attached (Fig. 10B in the Supplementary Appendix).18,23 Platelet depletion with the use of antiplatelet antibody, as compared with control IgG, resulted in an increase in tumor-cell and endothelial-cell apoptosis by a factor of four and a factor of eight, respectively (P = 0.008 and P<0.001, respectively) (Fig. 10C in the Supplementary Appendix).
Figure 4
Figure 4
Antitumor Effects of Blocking Paraneoplastic Thrombocytosis
We next explored whether platelets extravasate into the tumor bed, ascites, or both, sites where direct interactions with cancer cells are more opportune. Platelets from transgenic mice that express YFP under the control of the promoter of platelet factor 4 were transfused into tumor-bearing recipient mice (Fig. 5A). YFP-labeled platelets were detected in the tumor bed, primarily in a perivascular location, and were found to be free-floating and adherent to tumor cells in ascites (Fig. 5B).
Figure 5
Figure 5
Platelets in the Tumor Microenvironment
To determine the extent to which these observations may be tumor-specific, we compared the number of native, extravascular platelets in tumor samples, in the peritoneum after the induction of chemical peritonitis, and in subcutaneous tissues after the induction of a foreign-body reaction. Extravasated platelets were not observed in the context of a foreign-body inflammatory reaction and were rarely observed in the peritoneum after a pro-inflammatory stimulus. However, abundant platelets were detected in the tumor microenvironment apart from the vasculature (Fig. 5C). On the basis of finding platelets in contact with tumor cells outside the bloodstream, we assessed the functional effects of plasma-purified platelets on the proliferation and migration of epithelial ovarian cancer cells. Platelets significantly stimulated cell proliferation (Fig. 11A in the Supplementary Appendix) and migration (P<0.001 for the three cell lines) (Fig. 11B in the Supplementary Appendix).
Our study shows that increased thrombopoietic cytokine production by tumor and host tissues is a major driving force for paraneoplastic thrombocytosis. In particular, this process is mediated by hepatic thrombopoietin synthesis that is increased in response to excessive tumor-derived interleukin-6, thereby increasing platelet counts, which in turn promote tumor growth, creating a feed-forward loop.
Although paraneoplastic thrombocytosis is a well-recognized phenomenon in patients with cancer, the mechanisms underlying this observation are not fully defined. Findings from this study, derived from converging clinical information as well as pharmacologic and genetic manipulations, not only reveal the complex paracrine signaling responsible for paraneoplastic thrombocytosis in ovarian cancer but also point to the pivotal role of platelets in driving the biologic mechanisms of malignant tumors. For example, the secretory activities of platelets may stabilize the tumor vasculature and prevent intratumoral hemorrhage. The platelet-derived factors involved in maintaining angiogenic microvessel integrity and the cell type that platelets influence most are not fully known.24 Given the role of pericytes in supporting and stabilizing small blood vessels, it is possible that platelets regulate tumor-vasculature homeostasis, in part by nurturing pericytes. Prospective collection and in-depth analysis of platelets from patients with cancer will be required to fully understand the cancer-promoting role of platelets.
This aberrant paracrine signaling may have important clinical ramifications, since paraneoplastic thrombocytosis arises in a substantial proportion of patients with cancer. For example, one in three women who have received a diagnosis of ovarian cancer have thrombocytosis and are thus at substantially increased risk for advanced disease, vascular thromboembolic complications, and compromised disease-specific survival. Given that paraneoplastic thrombocytosis portends adverse clinical outcomes, countering it and using direct antiplatelet strategies have the potential to serve as a valuable therapeutic approach in humans.
Although the present investigation does not test this strategy in patients with cancer, such a premise is directly supported by epidemiologic studies of drugs known to interfere with crosstalk between platelets and tumor cells. Several prospective clinical trials have shown that low-molecular-weight heparin improves survival among patients with cancer, independently of the prevention of vascular thromboembolic complications due to anticoagulation.25 In addition, a recent large prospective cohort study revealed that daily use of aspirin after the diagnosis of colorectal cancer decreases cancer-specific and overall mortality.26,27 These data warrant further consideration, particularly given that paraneoplastic thrombocytosis may be not simply an epiphenomenon of cancer progression but a contributor to the process.
Supplementary Material
Supplement1
Acknowledgments
Supported in part by a Gail MacNeil KOH Research Award from the Gynecologic Cancer Foundation; grants (CA109298, P50 CA083639, P50 CA098258, CA128797, RC2GM092599, CA140933, CA016672, U54 CA151668, and R01CA104825) from the National Institutes of Health; a Program Project Development Grant from the Ovarian Cancer Research Fund; grants (OC073399, W81X-WH-10-1-0158, and BC085265) from the Department of Defense; the Baylor College of Medicine and M.D. Anderson Cancer Center Multidisciplinary Research Program; the Zarrow Foundation; the Marcus Foundation; the Estate of C.G. Johnson, Jr.; the United Kingdom Medical Research Council; the Blanton–Davis Ovarian Cancer Research Program; the Laura and John Arnold Foundation; the RGK Foundation; the Bettyann Asche Murray Distinguished Professorship; and training grants (T32 CA101642 supported Drs. Stone, Nick, Bottsford-Miller, and King and T32 CA009666 supported Dr. Pecot) from the National Cancer Institute. The siltuximab trial was funded by the United Kingdom Medical Research Council and sponsored and monitored by Queen Mary, University of London.
We thank Donna Reynolds, Robert Langley, Ph.D. (Department of Cancer Biology, M.D. Anderson Cancer Center), De-Yu Shen, and Nicholas B. Jennings for their technical assistance and helpful discussions, as well as Bueso-Ramos, M.D., Ph.D. (Department of Hematopathology, M.D. Anderson Cancer Center), for his hematopathology expertise.
Footnotes
Dr. Stone reports being listed as one of the inventors on a U.S. patent entitled “Use of Selective Adenosine A1 Receptor Allosteric Enhancers to Manipulate Angiogenesis.” No other potential conflict of interest relevant to this article was reported.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
1. Leslie M. Cell biology: beyond clotting: the powers of platelets. Science. 2010;328:562–4. [PubMed]
2. Trousseau A, Bazire V, Cormack J. Lectures on clinical medicine, delivered at the Hôtel-Dieu, Paris. London: R. Hardwicke; 1867.
3. Riess L. Zur pathologischen anatomie des blutes. Arch Anat Physiol Wissensch Med. 1872;39:237–49.
4. Levin J, Conley CL. Thrombocytosis associated with malignant disease. Arch Intern Med. 1964;114:497–500. [PubMed]
5. Borsig L. The role of platelet activation in tumor metastasis. Expert Rev Anticancer Ther. 2008;8:1247–55. [PubMed]
6. Kaushansky K. Lineage-specific hematopoietic growth factors. N Engl J Med. 2006;354:2034–45. [PubMed]
7. Branehög I, Ridell B, Swolin B, Weinfeld A. Megakaryocyte quantifications in relation to thrombokinetics in primary thrombocythaemia and allied diseases. Scand J Haematol. 1975;15:321–32. [PubMed]
8. 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. [PMC free article] [PubMed]
9. Thaker PH, Han LY, Kamat AA, et al. Chronic stress promotes tumor growth and angiogenesis in a mouse model of ovarian carcinoma. Nat Med. 2006;12:939–44. [PubMed]
10. Kluger HM, Chelouche Lev D, Kluger Y, et al. Using a xenograft model of human breast cancer metastasis to find genes associated with clinically aggressive disease. Cancer Res. 2005;65:5578–87. [PubMed]
11. Kamat AA, Merritt WM, Coffey D, et al. Clinical and biological significance of vascular endothelial growth factor in endometrial cancer. Clin Cancer Res. 2007;13:7487–95. [PubMed]
12. Kim MP, Evans DB, Wang H, Abbruzzese JL, Fleming JB, Gallick GE. Generation of orthotopic and heterotopic human pancreatic cancer xenografts in immunodeficient mice. Nat Protoc. 2009;4:1670–80. [PubMed]
13. McFarland-Mancini MM, Funk HM, Paluch AM, et al. Differences in wound healing in mice with deficiency of IL-6 versus IL-6 receptor. J Immunol. 2010;184:7219–28. [PubMed]
14. Merritt WM, Lin YG, Han LY, et al. Dicer, Drosha, and outcomes in patients with ovarian cancer. N Engl J Med. 2008;359:2641–50. [Erratum, N Engl J Med 2010;363:1877.] [PMC free article] [PubMed]
15. Deng L, Broaddus RR, McCampbell A, et al. Identification of a novel estrogen-regulated gene, EIG121, induced by hormone replacement therapy and differentially expressed in type I and type II endometrial cancer. Clin Cancer Res. 2005;11:8258–64. [PubMed]
16. Lu C, Han HD, Mangala LS, et al. Regulation of tumor angiogenesis by EZH2. Cancer Cell. 2010;18:185–97. [PMC free article] [PubMed]
17. Shahzad MM, Lu C, Lee JW, et al. Dual targeting of EphA2 and FAK in ovarian carcinoma. Cancer Biol Ther. 2009;8:1027–34. [PMC free article] [PubMed]
18. Lu C, Thaker PH, Lin YG, et al. Impact of vessel maturation on antiangiogenic therapy in ovarian cancer. Am J Obstet Gynecol. 2008;198(4):477.e1–477.e9. [PMC free article] [PubMed]
19. Sood AK, Bhatty R, Kamat AA, et al. Stress hormone-mediated invasion of ovarian cancer cells. Clin Cancer Res. 2006;12:369–75. [PMC free article] [PubMed]
20. Kaushansky KK. Thrombopoietin. N Engl J Med. 1998;339:746–54. [PubMed]
21. Kaser A, Brandacher G, Steurer W, et al. Interleukin-6 stimulates thrombopoiesis through thrombopoietin: role in inflammatory thrombocytosis. Blood. 2001;98:2720–5. [PubMed]
22. Burmester H, Wolber EM, Freitag P, Fandrey J, Jelkmann W. Thrombopoietin production in wild-type and interleukin-6 knockout mice with acute inflammation. J Interferon Cytokine Res. 2005;25:407–13. [PubMed]
23. Blair P, Flaumenhaft R. Platelet alpha-granules: basic biology and clinical correlates. Blood Rev. 2009;23:177–89. [PMC free article] [PubMed]
24. Ho-Tin-Noé B, Goerge T, Wagner DD. Platelets: guardians of tumor vasculature. Cancer Res. 2009;69:5623–6. [PubMed]
25. Akl EA, van Doormaal FF, Barba M, et al. Parenteral anticoagulation for prolonging survival in patients with cancer who have no other indication for anticoagulation. Cochrane Database Syst Rev. 2007;3:CD006652. [PubMed]
26. Chan AT, Ogino S, Fuchs CS. Aspirin use and survival after diagnosis of colorectal cancer. JAMA. 2009;302:649–58. [PMC free article] [PubMed]
27. Rothwell PM, Wilson M, Elwin CE, et al. Long-term effect of aspirin on colorectal cancer incidence and mortality: 20-year follow-up of five randomised trials. Lancet. 2010;376:1741–50. [PubMed]