Analysis of combination anti-angiogenesis therapy on markers of coagulation, platelet activation and angiogenesis in patients with advanced cancer
Abstract
The effects of combination anti-angiogenesis therapy (marimastat, captopril and fragmin) on plasma levels of coagulation initiator tissue factor (TF), platelet marker soluble P-selectin and angiogenic vascular endothelial growth factor (VEGF) were tested in 25 patients with advanced cancer. They had higher soluble P-selectin (P!0.001) and TF (P!0.001), but not VEGF (PZ0.066) than 25 age and sex-matched controls. VEGF and TF correlated significantly (rZ0.8, P!0.001) in cancer patients. Soluble P-selectin, TF and VEGF did not change at 4- and 8-weeks whilst on treatment. We provide further evidence linking coagulation and angiogenesis but combination anti-angiogenesis therapy does not influence plasma soluble P-selectin, TF or VEGF.
Keywords: Angiogenesis; Platelets; Soluble P-selectin; Tissue factor; VEGF
1. Introduction
Cancer carries a significant risk of thrombosis, possibly related to abnormalities in platelet activation and increased levels of coagulation proteins such as fibrinogen [1]. Recent evidence suggests that some component(s) of this hypercoagulable state may be related to angiogenesis, a process driven by growth factors such as vascular endothelial growth factor (VEGF) [2–4]. Tissue Factor (TF), the activator of blood coagulation, has recently been shown to have additional biological functions, including blood vessel development, cell migration, and inflammation [5]. Moreover, TF is expressed in a wide range of cancer cells, and plays an important part in tumour angiogenesis, tumour progression, and metastasis [6–9]. Platelet activation results in the release of various pro- or anti-antigenic factors, including VEGF [10]. The latter is essential for tumour metastasis [11], and we have previously reported raised levels in breast and haematological malignancies [12]. Recently, the determination of plasma levels of soluble P-selectin (sP-sel) has been proposed as an in vivo marker of platelet activation, and recent reports indicate raised levels in cancer [13].
Nonetheless, the effects of anti-angiogenic therapy on coagulation and platelet activation, and their relation- ship to angiogenic markers are unknown, although we have previously shown that sP-selectin is reduced after conventional chemotherapy in ovarian cancer patients [14].
As inappropriate angiogenesis is a feature of advanced, metastatic cancer, anti-angiogenic thera- pies may be useful, and is being tested in trials such as COMBAT (COMBination Anti-angiogeneis Trial). The rationale for the components of drugs tested in this trial is as follows: Captopril, an angiotensin converting enzyme (ACE) inhibitor, has both anti- angiogenic and anti-tumour activity in rodent models
[15] and inhibits the growth of human renal carcinoma xenografts [16]. Matrix metalloprotease inhibitors (MMPI’s) have previously been shown to block tumour growth and metastasis in a mouse model in combination with captopril [17]. The board spectrum orally available MMPI, marimastat, has been used in combination with captopril [18,19]. Finally, low molecular weight heparins (LWMH, such as fragmin) are anti-angiogenic in model systems, and prolong survival in animal models [20].
The present communication reports the testing of two hypothesises: (a) that there is a positive relationship between increased levels of markers of coagulation, platelet activation, and angiogenesis in advanced human cancer, and (b) that anti-angiogenic therapy will reduce levels of these markers. We tested our hypotheses in (a) a cross-sectional study of patients with advanced cancer compared to healthy controls, and (b) in a longitudinal intervention study of patients enrolled in the COMBAT study. Age and sex matched controls were also recruited to provide a perspective of levels of the plasma markers in health.
2. Patients and methods
Patients aged over 18 years with WHO perform- ance status 0–2 and histologically proven advanced malignancy refractory to conventional chemo- and radio-therapy treatment who gave written informed consent were eligible for the COMBAT, which was approved by the Oxford Research Ethics Committee according to principles in the Helsinki Declaration. Diagnoses were renal carcinoma (nZ8), colorectal
carcinoma [2], gastric carcinoma [4], mesothelioma [3] and one each of melanoma, sigmoid carcinoma, sacral carcinoma, transitional cell carcinoma of the pelvis and carcinoma of the bronchus, cervix, lung and bladder. WHO performance statuses were 0, nZ11, 1, nZ10 and 2, nZ4.
Patients were required to have adequate haemato- logical, renal and hepatic function and have evaluable or measurable disease. Exclusion criteria for the trial were a second malignancy, treatment with anti-cancer agents in the previous 4 weeks (6 weeks in the case of nitrosureas and mitomycin), serious systemic illness and patients with contraindications to the study drugs. Thus, patients with previous bleeding disorder, including heparin induced thrombocytopaenia, renal artery stenosis, cardiac outflow obstruction, severe arteriosclerosis, or recent treatment with procaina- mide or allopurinol were excluded. Patients were not allowed to receive aspirin, warfarin, lithium, allopur- inol or procainamide whilst on study.
Drugs were administered according to the follow- ing schedule (Fig. 1). A test dose of 12.5 mg of captopril was given orally. If patients tolerated the first dose then they continued on captopril taking 12.5 mg twice daily for 1 week, followed by 25 mg twice daily for 1 week after which they took 50 mg twice daily. Marimastat was administered orally at a dose of 10 mg twice daily. Fragmin was administered by a once daily subcutaneous injection at a dose of 200 units/kg for 4 weeks after which the dose was reduced to 5000 units/day. Patients were compared to an equal number of age and sex matched controls recruited from hospital and laboratory staff, spouses of patients, and patients attending for endoscopy, hernia repair and cataract repair. Raised levels of the plasma markers in cancer compared to health are established: the controls are provided merely for comparison and no original hypotheses apply.
Blood samples were obtained from the patients at baseline, 28 and 56 days, from the controls at one time point only. Blood was taken into citrated vacutainers and plasma obtained after centrifugation at 2500 rpm (1000g) for 20 min. Plasma levels of VEGF, sP-sel (R&D Systems, Abingdon, UK), and Tissue Factor (Axis-Shield, UK) were assayed by ELISA. Data normally distributed is presented as mean and standard deviation and analysed by t-test. Data distributed non-normally is presented as median and range and analysed by the Mann–Whitney U-test. Data from 25 cases and 25 controls provides the power (1P!0.05 and 1KbZ0.8) to detect a difference of 2/3rd of standard deviation (SD) in a continuously variable index with a normal.
3. Results
The demographic and laboratory data of assays of plasma proteins are presented in Table 1. Age, sex and platelet counts were comparable. As expected, patients with cancer had raised sP-sel and TF but the VEGF difference was not significant. Data from the 22 patients who completed the trial are presented in Table 2. There were no significant differences between plasma levels of TF, sP-selectin or VEGF post- treatment, when compared to baseline (pre-treatment) levels. However, there was a trend for VEGF to rise with time so that, after eight weeks, levels were higher than those of the healthy controls (PZ0.0109). A strong positive correlation (Spearman rZ0.588, P!0.001) was found between TF and VEGF in cancer patients (Fig. 2) but not in controls (Spearman rZ0.061, PZ0.716). There were no significant correlations between any of the other markers in cancer patients or controls (data not shown).
4. Discussion
We investigated possible inter-relationships between coagulation, platelet activation, and angio- genesis in cancer, as defined by levels of plasma TF, sP-selectin and VEGF, respectively, and whether or even combination anti-angiogenic therapy does not improve the clinical condition. It is also plausible that the drugs were not given in strong enough doses to show a response. Again, as the first trial of its kind, and drawing on data from animal models alone, it is unlikely that these drug doses are optimal. One previous study, using a single VEGF-inhibitor (SU5416) also found no gross changes in coagulation factors [31].
In conclusion, we have provided further evidence that coagulation and angiogenesis are linked in human cancer, and shown for the first time that plasma levels of TF and VEGF have a strong significant correlation. We have also shown that combination anti-angiogenic therapy with captopril, marimastat and fragmin has no effect on plasma levels of TF, VEGF or sP-selectin. Targeting angiogenesis and/or the processes that are inter-related to angiogenesis (e.g. coagulation, plate- lets) BB-2516 remains an important strategy for cancer clinicians and scientists alike.