Introduction
Venous thromboembolism (VTE) following breast cancer chemotherapy is not uncommon. In early breast cancer, VTE occurs in 5-10% of patients receiving chemotherapy
Previous work has demonstrated a hypercoagulable state in breast cancer patients, with elevated markers of coagulation, including thrombin-antithrombin (TAT)
Several small studies have reported alterations in markers of coagulation in response to breast cancer chemotherapy, which support the development of a chemotherapy-induced hypercoagulable state
The requirement for platelets in cancer metastasis was recognised about 30 years ago and it is now recognised that platelets are an integral part of the microthrombus that is thought to promote the arrest and lodgement of circulating tumour cells
Platelets are numerous in the circulation and contain large stores of factors known to be essential for angiogenesis. Vascular endothelial growth factor (VEGF) is one such protein that is stored in large quantities in platelet α-granules
We hypothesized that platelets are functionally altered in women with breast cancer and this may represent a possible pathophysiological mechanism for the observed VTE during chemotherapy. In the current study we prospectively followed early and advanced breast cancer patients commencing chemotherapy to establish early alterations in the circulating plasma and serum levels of VEGF together with platelet release of VEGF.
Materials and methods
Patients
A total of 123 female patients [median age 52 (range, 31-78) years] commencing chemotherapy for breast cancer were recruited. Of these, 87 were receiving adjuvant chemotherapy following curative surgery, and 36 receiving chemotherapy for radiographically-proven metastatic breast disease (Table
Chemotherapy regimens used in breast cancer patients
Chemotherapy Regimen
Number of patients
5-fluorouracil, epirubicin, cyclophosphamide
65
Cyclophosphamide, methotrexate, 5-fluorouracil
15
Epirubicin, cyclophosphamide
4
Epirubicin
3
Docetaxol
15
Cyclophosphamide, methotrexate, 5-fluorouracil
8
Epirubicin, docetaxol
6
Vinorelbine, mitomycin
3
Epirubicin
2
5-fluorouracil, epirubicin, cyclophosphamide
1
Vinorelbine, 5-fluorouracil
1
Control subjects
Sixty eight age matched female controls [median age 48 (range, 31-78) years], with no history of cancer acted as control subjects.
Protocol
A prospective cohort study was undertaken. Serum vascular endothelial growth factor [sVEGF] and plasma vascular endothelial growth factor [pVEGF]) were measured prior to chemotherapy and at 24 hours, four-, eight days and three months following commencement of chemotherapy in all patients. A clinical assessment for VTE was performed at the same time points. Duplex ultrasound imaging was performed one month following commencement of chemotherapy or if symptoms developed.
Blood sampling and analytical methods
Atraumatic venous blood sampling was performed at the antecubital fossa and all specimens separated and stored within two hours after being collected into tubes containing citrate as anticoagulant. Citrate samples were immediately taken onto ice whilst serum samples were allowed to clot at room temperature. All samples were centrifuged at 2500 g for 20 minutes at 4°C and the plasma or serum decanted before being stored at -80°C in 300 μl aliquots. Platelet depleted plasma was prepared for the analysis of VEGF as detailed: Citrated tubes were immediately plunged into ice and taken to the laboratory were the sample was centrifuged at 4°C for twenty minutes at 3500 g. The supernatant was removed and re-centrifuged for 20 mins at 3500 g at 4°C following which the platelet depleted plasma (PDP) was aliquoted and the last 0.5 ml discarded. All samples were stored at -80°C until analysis.
Platelet count was measured using the Advia 120 Haematology System (Bayer Diagnostic, UK)
VEGF Assay
Serum and platelet depleted plasma VEGF were analysed using an enzyme-linked immunosorbent assay (ELISA) by R&D Systems®, Oxon, UK, with a sensitivity of 9 μg/ml.
In duplicate, sample (100 μl) diluted with 100 μl of assay diluent was incubated with the capture antibody for 2 hours at room temperature. Following washing steps (×3), 200 μl of detection antibody conjugate was added and the assay incubated for 2 hours at room temperature. The plate was again washed (×3) and the detection substrate added and incubated for 25 minutes at room temperature. Stop solution (25 μl) was added and the optical density read at 450 nm.
Platelet VEGF
The amount of VEGF released per platelet was calculated using the following previously validated formula
Ethical approval
The study was approved by the South Manchester Local Research Ethics Committee and all patients gave written informed consent.
Statistical methods
Data on sVEGF and pVEGF was parametric after log conversion and so reported as geometric mean (confidence interval). Platelet releasate VEGF concentration was expressed as median (range). Comparative group analysis (early, advanced breast cancer and controls) of pre-chemotherapy patient values was by ANOVA, with further analysis of subgroups using Scheffe. Comparative group analysis (VTE within three months, VTE free) of patient values was by independent T-test. Changes in patient serum or plasma values with chemotherapy as compared with pre-treatment values were performed by paired T test, however to minimise errors induced by multiple tests a repeated measures analysis (Greenhouse Geiser correction) to compare trends over time in patients with and without VTE was used. Comparative group analysis (VTE within three months, VTE free) of changes in coagulation parameters with chemotherapy as compared with pre-treatment values were performed by analysis of covariance. A significance of p < 0.05 was used. Binary logistic regression to identify predictors of VTE was also performed. Analysis was performed on baseline data, and change from baseline. Appropriate corrections were made for cancer stage and age.
Results
Of 123 breast cancer patients receiving chemotherapy, 12 (9.8%) developed VTE within 3 months of chemotherapy, of which eight (66.7%) were symptomatic. Six of 36 (17%) metastatic breast cancer patients and six of 87 (6.9%) early breast cancer paients receiving adjuvant chemotherapy developed VTE.
Baseline data: prior to chemotherapy
Prior to chemotherapy, serum and plasma VEGF were all increased in advanced breast cancer patients compared to controls. Further serum and plasma VEGF were increased in advanced breast cancer compared to early breast cancer. Platelet count was increased in advanced breast cancer compared to controls. Calculated VEGF release per platelet from advanced cancer patients was elevated compared to early breast cancer patients and the control group (Table
Baseline biomarker results prior to chemotherapy
sVEGF μg/ml, geometric mean (CI)
344.0 (270.7-437.2)
181.0 (155.9-210.2)*
28.8 (25.3-32.8)
pVEGF μg/ml, geometric mean (CI)
22.8 (17.6-29.4)
14.2 (12.1-16.6)
15.1 (12.7-17.9)
Platelet count × 109/l, mean (CI)
326.7* (286.9-366.5)
309.6 (293.0-326.2)
278.6 (257.7-299.6)*
Platelet VEGF VEGF μg/ml per platelet × 109
median (range)
1.02*†‡ (0.17-3.53)
0.53* (0.05-2.89)
0.53 ‡(0.007-2.48)
Analysis of the difference between groups used Analysis of Variance (ANOVA). Where differences were found, further analysis (between pairs of groups, with respective pairs for each molecule marked with *†
Baseline data: Prior to chemotherapy - Development of VTE
Prior to chemotherapy plasma VEGF was significantly higher in those who developed VTE within 3 months of commencing chemotherapy compared to those who did not (27.8 [14.3-54.1] vs 15.4 [13.5-17.7] μg/ml, p = 0.01 [independent T-test]). Interestingly serum VEGF was not significantly different between both groups (282 [136.5-585.3] vs 212 [185.6-243.1] μg/ml, p = 0.2 [independent T-test]) In addition, platelet release of VEGF did not demonstrate a significant difference in those who developed VTE within 3 months compared to those who did not. (0.9 [0.4-1.9] vs 0.6 [0.5-0.7] VEGF μg/ml per platelet × 109, median [range], p = 0.6 [independent T-test]). Prior to chemotherapy however a 100 μg/ml increase in sVEGF was associated with a 40% increased risk of VTE (P = 0.003). A 10 μg/ml increase in pVEGF was associated with a 20% increased risk of VTE (P = 0.007)
Response to chemotherapy
The mean or geometric mean (CI) of plasma and serum VEGF together with platelet releasate VEGF (median (range)) at baseline, 24 hours, four and eight days and three months following chemotherapy are given in table
Alterations in biomarker parameters induced by chemotherapy in breast cancer patients
sVEGF μg/ml,
geometric mean (CI)
218.0 (190.3-249.6)
203.4 (176.5-234.4)
158.3 (136.6-183.4)
154.3 (132.5-179.8)
236.7 (205.0-273.4)
pVEGF μg/ml,
geometric mean (CI)
16.4 (14.2-18.8)
14.9 (13.2-16.9)
16.3 (14.0-18.0)
20.5 (18.3-13.0)
21.6 (18.8-24.9)
Platelet count × 109/l
Mean (CI)
314 (298.4-330.9) [123]
306 (287.2-325) [99]
263.3 (249.7-276.8) [108]
242.3 (227.3-257.2) [117]
310 (285.0-334) [112]
Platelet VEGF VEGF μg/ml per platelet × 109
median (range) [n]
0.62 (0.54-0.72)
[120]
0.61 (0.52-0.71)
[99]
0.51 (0.43-0.61)
[108]
0.53 (0.45-0.64)
[114]
0.72 (0.62-0.83)
[106]
Alterations in biomarker parameters induced by chemotherapy in breast cancer patients developing VTE within three months of chemotherapy
sVEGF μg/ml,
geometric mean (CI)
282.6 (136.5-585.3)
267.8 (136.5-525.4)
220.1 (110.7-437.3)
226.0 (116.3-439.2)
358.8 (116.3-774.2)
pVEGF μg/ml,
geometric mean (CI)
27.8 (14.3-54.1)
19.5 (10.4-36.6)
19.0 (9.4-38.3)
23.4 (13.2-41.5)
18.7 (13.7-25.5)
Platelet count × 109/l mean (CI)
342.5 (265.8-419.2) [12]
338.9 (246.4-431.4) [10]
269.6 (225.5-313.8) [11]
274 (216.8-331.2) [11]
386.2 (238.3-534.1) [10]
Platelet VEGF VEGF μg/ml per platelet × 109
median (range) [n]
0.89 (0.42-1.87)
[11]
0.89 (0.47-1.68)
[10]
0.86 (0.39-1.86)
[11]
0.83 (0.39-1.75)
[11]
1.02 (0.53-1.97)
[9]
Alterations in biomarkers induced by chemotherapy in breast cancer patients remaining free of VTE within three months of chemotherapy
sVEGF μg/ml,
geometric mean (CI)
212.4 (185.6-243.1)
197.5 (171.1-228.0)
152.8 (131.7-177.4)
148.3 (126.8-173.3)
228.0 (197.4-263.4)
pVEGF μg/ml,
geometric mean (CI)
15.4 (13.5-17.7)
14.5 (12.9-16.4)
16.0 (13.7-18.7)
20.2 (18.0-22.7)
21.9 (18.8-25.5)
Platelet count × 109/l, mean (CI)
311.6 (295.2-328) [111]
302.4 (283.4-321.4) [89]
262.5 (248.1-277) [97]
239 (223.4-254.5) [106]
302.5 (279.6-325.5) [102]
Platelet VEGF VEGF μg/ml per platelet × 109
median (range) [n]
0.61 (0.52-0.71)
[108]
0.58 (049-0.69)
[89]
0.49 (0.41-0.58)
[97]
0.52 (0.43-0.62)
[103]
0.70 (0.61-0.81) [97]
Analysing all patients together, irrespective of subsequent development of VTE, serum and plasma VEGF, together with platelet release of VEGF showed a significant trend over time (repeated measures analysis). Serum VEGF decreased during the first 8 days, however plasma VEGF increased (P < 0.001). By 3 months both showed trend for increased levels (P < 0.001). In patients with and without VTE, by four days following chemotherapy, platelet count was reduced, however it remained within normal limits. Levels returned to baseline by three months (Table
A differing pattern of change was observed in serum VEGF between advanced and early breast cancer patients which was not reflected in plasma VEGF. The difference was apparent in the first eight days following chemotherapy, where sVEGF levels in advanced cancer showed a greater decrease, from baseline, at 24 hours (p = 0.004) and four days (p = 0.002) compared to early breast cancer. The difference between the trends seen in response to chemotherapy in advanced and early breast cancer patients was no longer apparent by three months (p = 0.2). The rapid decrease in sVEGF levels, in advanced breast cancer, as an early response to chemotherapy was not mirrored by pVEGF levels. (Table
Table showing the chemotherapy induced changes in serum VEGF and platelet release of VEGF in advanced and early breast cancer
Procoagulant
Group
Baseline
Day 1
Day 4
Day 8
3 Months
Significant Difference in Trend
i) pre-chemo to day 8
ii)Pre-chemo to 3 months
sVEGF
μg/ml
Geometric Mean (CI) [n]
ABC
344 (271-437) [35]
287 (219-376) [33]
240 (182-317) [33]
253 (196-327) [33]
344 (248-477) [26]
i) 0.01*
ii) 0.2*
EBC
181 (156-210) [86]
177 (151-207) [81]
134 (114-157) [81]
127 (107-151) [83]
211 (181-246) [83]
i) 0.01**
ii) 0.2**
VEGF released
per platelet
μg/ml/platelet
Geometric mean (CI) [n]
ABC
1.02 (0.8-1.30) [35]
0.92 (0.68-1.23) [28]
0.85 (0.61-1.17) [30]
0.92 (0.72-1.19) [31]
1.03 (0.77-1.39) [25]
i) 0.007*
II) 0.03*
EBC
0.53 (0.45-0.62) [82]
0.51 (0.42-0.63) [67]
0.43 (0.35-0.52) [73]
0.43 (0.34-0.
53) [78]
0.65 (0.54-0.77) [78]
i) 0.007**
ii) 0.03**
sVEGF = serum vascular endothelial growth factor (VEGF)
ABC = advanced breast cancer
EBC = early breast cancer
*Significant difference in trend compared to early breast cancer
**Significant difference in trend compared to advanced breast cancer
Platelet release of VEGF showed a significant trend in reponse to chemotherapy. There was a decline in platelet release of VEGF by day 4 and 8 (P < 0.001) increasing to above pre-chemotherapy levels by 3 months (Table
The alteration in platelet release of VEGF was significantly different in advanced breast cancer (Table
When patients who subsequently developed VTE were compared to patients who remained free of VTE (Tables
Platelet release of VEGF during chemotherapy did not demonstrate a significantly different response in those who developed VTE within 3 months compared to those who did not (P = 0.6).
Discussion
Consistent with previously published literature, sVEGF and pVEGF levels in this study are significantly elevated in advanced breast cancer patients
The higher VEGF seen in serum compared to plasma samples, is consistent with previous studies reporting platelet release of VEGF as the main source of serum VEGF
In the first eight days following chemotherapy, the increase in pVEGF is matched by a decrease in sVEGF. This may reflect increased platelet release of VEGF as a response to chemotherapy, with a consequent reduction in stored levels of VEGF in the platelet α-granules
Previous research has shown that platelet lysates from breast cancer patients have increased VEGF concentrations compared to controls
After an initial decrease in platelet VEGF content, possibly due to VEGF release stimulated by chemotherapy, the possible increase in platelet VEGF content in advanced cancer only, suggests an important store of this angiogenic molecule in active cancer. The profound effect of chemotherapy on platelet VEGF content in early breast cancer suggests an initial early release of VEGF by platelets in response to chemotherapy, hence depleting stores. This appears to be followed by a reactive increase in VEGF storage, which is mirrored by the absolute platelet count. Verheul and co-workers, in a study of 27 breast cancer patients receiving chemotherapy, reported an initial thrombocytopenia followed by a strong platelet rebound coinciding closely with a sVEGF peak. However, by not correcting this for background (plasma) levels of VEGF, or calculating VEGF per platelet, they concluded that sVEGF simply reflects platelet count
Prior to chemotherapy, plasma VEGF was increased in patients subsequently developing VTE, with a trend for increased sVEGF levels in the VTE patient group. Other investigators have reported sVEGF and pVEGF correlate with the haemostatic markers TAT, fibrinogen and D-dimer
Of the procoagulants we have studied, only pVEGF demonstrates an early altered response to chemotherapy in patients developing VTE, compared to those without VTE. The marked reduction in pVEGF, but not sVEGF at 24 hours in the VTE group, suggests rapid uptake of background VEGF in the hypercoagulable patients. Whether this uptake is by cancer cells, or perhaps more likely, by altered function of platelets, remains to be established.
There was no difference in pre-chemotherapy platelet count, or platelet release of VEGF, in all breast cancer patients with and without subsequent VTE. This supports the findings of Leibovitch, who although reporting reactive thrombocytosis as a common finding following major urological surgery, found no association with thromboembolic complications
In conclusion, raised pVEGF, prior to chemotherapy in patients developing VTE, may point to the procoagulant trigger for cancer-induced VTE. The increased baseline pVEGF, but not sVEGF, in patients developing VTE implies tumour VEGF rather than platelet VEGF is the important procoagulant here. However, the rapid reduction in pVEGF following chemotherapy, in the VTE group, suggests an increased VEGF uptake, either by tumour, or perhaps more likely, by platelets. This study provides evidence that a clinically useful profile could be produced to identify patients at increased risk of VTE. With such a profile, a trial of targeted anticoagulant prophylaxis could be developed.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
CCK: Designed the study, undertook patient recruitment, laboratory analysis and data analysis. Also prepared and critically reviewed the draft manuscript. Read and approved the final manuscript.
GJB: Designed the study, reviewed the data and critically reviewed the manuscript. Read and approved the final manuscript.
SK: Critically reviewed the manuscript and made important changes to the scientific content. Read and approved the final manuscript.
GMcD: Supervised the laboratory analysis and analysis of the data. Drafted and critically reviewed the manuscript for scientific content. Read and approved the final manuscript.