Abstract

Background: Acquired and inherited risk factors for venous thromboembolism (VTE) and the incidence of symptomatic VTE were investigated in patients on adjuvant chemotherapy for breast or gastrointestinal cancer (GI).

Patients and methods: In a prospective observational study (January 2003 and February 2006), 199 GI (82 women/117 men; age range, 26–84 years) and 182 breast (180 women/2 men; age range, 29–85 years) cancer patients were enrolled and followed-up for symptomatic VTE during adjuvant chemotherapy. The effect of acquired (i.e. age, chemotherapy, tumour histotype, history of thrombosis, body mass index and smoking) and inherited risk factors [i.e. antithrombin, protein C (PC), protein S, homocysteine, activated PC resistance, factor V Leiden (FVL) and prothrombin (PT) mutations) was prospectively evaluated.

Results: Overall, 30 VTE events (7.87%) were recorded: 28 (7.35%) during treatment and 2 (0.52%) during the subsequent follow-up. Among all the 381 cancer patients, FVL was detected in 14 cases (3.67%) and PT mutation in 10 cases (2.62%). Multivariate analysis showed a significant association between the development of VTE and both thrombocytosis [hazard ratio (HR) 1.65; 95% confidence interval (CI), 1.04–2.637, P <0.0341] and a prior episode of thrombosis (HR 7.6; 95% CI, 1.77–33.1, P <0.006). FVL and PT mutations were not associated with the risk for VTE.

Conclusion: The present data indicate thrombocytosis and history of thrombosis as risk factors for development of a thrombotic event during adjuvant chemotherapy in patients with malignant diseases.

introduction

In 1865, Trousseau described the high incidence of venous thromboembolism (VTE) in a cohort of patients with gastrointestinal (GI) carcinoma. Although the association between cancer and hypercoagulability is well established, in these patients, pathogenesis of VTE is not entirely elucidated [1]. Chemotherapy increases the risk for VTE by damaging the vascular endothelium and decreasing plasma levels of naturally occurring coagulation inhibitors [2].

For patients with solid tumours, following a radical surgical resection, medical oncologists consider adjuvant chemotherapy to reduce the risk for tumour recurrence. The decision of whether or not to administer chemotherapy depends on several factors including the pathological stage and biological characteristics of the tumour.

The relationship between cancer, chemotherapy and thrombosis has been investigated most extensively in breast cancer patients. Levine et al. [3] clearly showed that chemotherapy causes thrombosis in stage II breast cancer patients receiving adjuvant chemotherapy . Among 205 patients randomly assigned to treatment, 14 episodes of VTE (6.8%) occurred during 979 patient-months of chemotherapy; in comparison, there were no such events during 2413 patient-months without therapy.

After this report, few studies have been conducted on cancer patients receiving adjuvant chemotherapy and the few that were carried out all concentrated on breast cancer patients. No similar data are available for GI cancer patients receiving adjuvant chemotherapy. Therefore, the information on concerning risk factors that specifically predispose one to VTE is limited in this patients group.

Current guidelines do not recommend VTE prophylaxis for cancer outpatients [4]. The ability to stratify risk would permit appropriate use of VTE prophylaxis only in patients at the highest risk.

Therefore, the decision was made to determine the incidence of symptomatic VTE and to investigate the acquired and inherited risk factors involved in developing VTE in cancer patients receiving adjuvant chemotherapy. The study was carried out through a prospective study that evaluated a cohort of consecutive cancer patients starting a new adjuvant chemotherapy. Since all patients were free of any evidence of disease, we were able to study the risk factors associated primarily with cancer-related adjuvant chemotherapy.

materials and methods

patients and study design

The study population comprised 381 consecutive patients who received adjuvant chemotherapy in the Division of Medical Oncology. Patients were followed prospectively. Patients who had enrolled between January 2003 and February 2006 and who had completed at least one cycle of chemotherapy were included in this analysis. Prerequisites for inclusion were as follows: (i) a histologically confirmed diagnosis of cancer, with targeted enrolment of specific tumour types (i.e. breast, GI cancer); (ii) an adjuvant chemotherapy regimen at the outset; and (iii) age of 18 years or older, and capable of providing informed consent. Patients were ruled out of the study if they (i) were receiving concurrent cytotoxic, biological or immunological therapy for other conditions, or continuous single-agent chemotherapy; (ii) had a diagnosis of acute leukaemia or myeloma; (iii) were pregnant or lactating; (iv) had an active infection requiring treatment; (v) were currently participating in a double-blind study; or (vi) had received a stem cell transplant. Cancer patients with active cancer, advanced disease, acute medical illnesses, hospitalisation or terminal conditions who are already at high risk for development of VTE were not included.

data gathering

Data were collected prospectively during chemotherapy and follow-up. This included baseline information regarding cancer stage and histopathology, age, gender, current medications, recent surgery, comorbidities, and the planned chemotherapy regimen. Data on body mass index (BMI), history of smoking, previous thromboembolic events and central venous catheter were all gathered. Before starting chemotherapy, each patient was screened for protein C (PC), protein S, antithrombin (AT), homocysteine, G1691A factor V and G20210A prothrombin (PT).

We did not screen patients for asymptomatic thrombosis; only symptomatic forms were checked. The medical charts and radiological history of all patients were checked for ultrasonography of the limbs, chest and abdominal computed tomography (CT) scan, and perfusion/ventilation lung scan. In all patients with proven VTE, the records and radiological reports were searched for indicators of tumour load at the time of the event. VTE was diagnosed by the treating clinician on the basis of clinical suspicion using the usual diagnostic procedures. The criterion for a diagnosis of VTE by compression ultrasonography was non-compressibility of a proximal vein. When symptoms indicative of pulmonary embolism (PE) developed, radionuclide lung scanning was carried out. In cases not conclusive for PE diagnosis, a spiral CT scan was required to confirm the PE. After entering the study, occurrence of a symptomatic VTE event was reported on the new-cycle examination, in the mid-cycle examination, or on the follow-up examination forms, depending upon the timing of the event.

The study protocol was approved by the local ethics committee, and a written informed consent was obtained from all patients.

laboratory studies

Plasma was separated by centrifugation of venous blood at 3000 g for 20 min and stored in aliquots at −80°C until the assay (<3 months). The PC (by chromogenic assay) study was designed to obtain a global evaluation of the PC pathway function. This is based on the ability of endogenous activated PC (APC)—generated after activation of PC by a snake venom extract (Protac)—to reduce the thrombin generation induced by a reagent containing tissue factor as described by Toulon et al. [5]. APC resistance factor V exclusively detected inherited APC resistance due to mutations in the factor V gene [6]. The free protein S was determined by measuring the increase in turbidity produced by the agglutination of two latex reagents (by immunoturbidimetric assay). AT activity was measured using a chromogenic assay (Coamatic Antithrombin; Chromogenics Instrumentation Laboratory, Monza, Italy) as described by Ellis et al. [7]. Finally, total homocysteine was determined by enzymatic assay [8].

genetic studies

Blood was collected from the antecubital vein without venous stasis using a Vacutainer PrecisionGlide needle 0.8 × 38 mm into four 3.15-ml Vacutainer® tubes containing 0.129 mol/l sodium citrate. The first 5 ml was discarded. Citrated platelet-poor plasma was prepared using two centrifugation steps: 5 min at 2150 g at room temperature and 10 min at 11 000 g at 4°C. The blood samples were stored at 40°C. Analyses for G1691A factor V and G20210A PT were carried out in all patients using the FV-PTH Stripassay (Vienna-Lab Labordiagnostika GmbH, Vienna, Austria) according to the manufacturer’s instructions. After genomic DNA isolation (according to the assay procedure), multiplex PCR was carried out using factor V- and factor II-specific primers. Genotype detection was carried out by Lipa technology, including hybridisation, stringent wash and colour development. Interpretation of results was determined by collector sheet. Analysis was carried out on blind samples. The physician was not aware of the mutation status. In addition, the technicians carrying out the assays were unaware of the patient’s VTE status.

statistical analysis

The association between thromboembolism and some clinical variables was studied using univariate analysis and reported as odds ratios (ORs) with 95% confidence intervals (CIs). The association results were re-evaluated using a multivariate logistic regression model, with VTE as the response variable. The variables that were associated with an increased risk for VTE (P < 0.05) in the univariate analysis were included in a multivariate logistic regression model. First-order interaction terms for the primary outcome of VTE were explored and reported.

Univariate and multivariate analyses (analysis of variance) were conducted for clinical factors [age, BMI, previous deep vein thrombosis (DVT), contraceptive, hormone therapy, smoking], haematological parameters (platelet count, leukocyte count, haemoglobin, haematocrit), acquired and inherited risk factors [homocysteine, AT, PC, protein S, APC resistance, factor V Leiden (FVL), PT mutation], and type of chemotherapy [anthracycline; fluorouracil; platinum; combination chemotherapy with cyclophosphamide, methotrexate and fluorouracil (CMF); other].

results

A total of 381 patients with cancer (262 women, 119 men; age range, 26–85 years) entered the study between January 2003 and February 2006. GI cancer (122 colon, 51 gastric, 16 rectal, and 10 pancreas) was the most common diagnosis, followed by breast (n = 182) cancer. All patients received adjuvant chemotherapy: fluorouracil (n = 174), CMF (n = 111), anthracycline (n = 71) and other (n = 21). Because of the high enrolment of breast cancer patients, women dominated the population set. The median duration of chemotherapy was 6 months (4–6 months) and the median follow-up 35 months (12–50 months). Table 1 describes the characteristics of the study population.

Table 1.

Patient characteristics according to VTE

 VTE
 
Total 
 No (n = 353), n (%) Yes (n = 28), n (%) 
Age in years, median (min–max) 62 (26–85) 64 (39–80) 62 (26–85) 
BMI [Weight Kg/Height2 (m2)], median (min–max) 23.6 (14.7–43.6) 23.9 (14.8–33.6) 23.7 (14.7–43.6) 
Sex 
    Male 114 (95.80) 5 (4.20) 119 
    Female 239 (91.22) 23 (8.78) 262 
T stage  
    1 112 (94.12) 7 (5.88) 119 
    2 80 (88.89) 10 (11.11) 90 
    3 129 (94.16) 8 (5.84) 137 
    4 32 (91.43) 3 (8.57) 35 
N stage 
    0 131 (92.91) 10 (7.09) 141 
    1 134 (94.37) 8 (5.63) 142 
    2 64 (90.14) 7 (9.86) 71 
    3 17 (85.00) 3 (15.00) 20 
    X 7 (87) 1 (13) 
Tumour site 
    Colon 113 (92.62) 9 (7.38) 122 
    Breast 166 (91.21) 16 (8.79) 182 
    Gastric 48 (94.12) 3 (5.88) 51 
    Rectal 16 (100.00) 0 (0.00) 16 
    Pancreas 10 (100.00) 0 (0.00) 10 
Previous VTE 
    No 347 (93.28) 25 (6.72) 372 
    Yes 6 (66.67) 3 (33.33) 
Hormone replacement therapy 
    No 340 (92.64) 27 (7.36) 367 
    Yes 12 (92.31) 1 (7.69) 13 
    Missing – 
Smoking 
    No 289 (93.83) 19 (6.17) 308 
    Yes 64 (87.67) 9 (12.33) 73 
Contraceptives 
    No 328 (93.18) 24 (6.82) 352 
    Yes 25 (86.21) 4 (13.79) 29 
Coagany 
    No 169 (91.85) 15 (8.15) 184 
    Yes 124 (96.88) 4 (3.13) 128 
    Missing 60 69 
CVC 
    No 322 (92.80) 25 (7.20) 347 
    Yes 31 (91.18) 3 (8.82) 34 
 VTE
 
Total 
 No (n = 353), n (%) Yes (n = 28), n (%) 
Age in years, median (min–max) 62 (26–85) 64 (39–80) 62 (26–85) 
BMI [Weight Kg/Height2 (m2)], median (min–max) 23.6 (14.7–43.6) 23.9 (14.8–33.6) 23.7 (14.7–43.6) 
Sex 
    Male 114 (95.80) 5 (4.20) 119 
    Female 239 (91.22) 23 (8.78) 262 
T stage  
    1 112 (94.12) 7 (5.88) 119 
    2 80 (88.89) 10 (11.11) 90 
    3 129 (94.16) 8 (5.84) 137 
    4 32 (91.43) 3 (8.57) 35 
N stage 
    0 131 (92.91) 10 (7.09) 141 
    1 134 (94.37) 8 (5.63) 142 
    2 64 (90.14) 7 (9.86) 71 
    3 17 (85.00) 3 (15.00) 20 
    X 7 (87) 1 (13) 
Tumour site 
    Colon 113 (92.62) 9 (7.38) 122 
    Breast 166 (91.21) 16 (8.79) 182 
    Gastric 48 (94.12) 3 (5.88) 51 
    Rectal 16 (100.00) 0 (0.00) 16 
    Pancreas 10 (100.00) 0 (0.00) 10 
Previous VTE 
    No 347 (93.28) 25 (6.72) 372 
    Yes 6 (66.67) 3 (33.33) 
Hormone replacement therapy 
    No 340 (92.64) 27 (7.36) 367 
    Yes 12 (92.31) 1 (7.69) 13 
    Missing – 
Smoking 
    No 289 (93.83) 19 (6.17) 308 
    Yes 64 (87.67) 9 (12.33) 73 
Contraceptives 
    No 328 (93.18) 24 (6.82) 352 
    Yes 25 (86.21) 4 (13.79) 29 
Coagany 
    No 169 (91.85) 15 (8.15) 184 
    Yes 124 (96.88) 4 (3.13) 128 
    Missing 60 69 
CVC 
    No 322 (92.80) 25 (7.20) 347 
    Yes 31 (91.18) 3 (8.82) 34 

VTE, venous thromboembolism; BMI, body mass index; CVC, central venous catheter.

The laboratory risk factors for thromboembolism are shown in Table 2 according to tumour site.

With regard to the acquired risk factors, FVL was detected in the heterozygous form in 14 of the 381 cancer patients (3.6%), whereas the PT G20210A mutation gene was found in the heterozygous form in 10 of the 381 cancer patients (2.5%). Only in one patient with gastric cancer was a combination of both FVL and the G20210A heterozygous variant of the PT gene found in the same subject.

No discrepancies were observed between APC resistance and FVL confirmation.

Among the 381 patients, 30 (7.8%) developed VTE. Twenty-eight episodes (7.34%) occurred during chemotherapy; on the contrary, only 2 episodes (0.55) were observed during follow-up. The incidence of VTE is reported in Table 3 according to tumour site. VTE developed in (i) eight breast cancer patients receiving CMF [7.4% (8/108)], (ii) six breast cancer patients [9.2% (8/65)] receiving an anthracycline-based chemotherapy and (iii) two patients receiving a taxane-based chemotherapy. VTE developed in 5% of the colorectal cancer patients (9/174). The VTE sites were as follows: 22 femoral–iliac DVT, 5 calf DVT and 2 patients developed PE. Finally, one patient developed a portal vein thrombosis, which was described during an abdominal CT scan. The two PE cases were first diagnosed with a lung scan, and thereafter, both events were confirmed with a chest CT scan. For the two patients who developed PE, chemotherapy was interrupted, and the patients were hospitalised and treated with low-molecular weight heparin. Patients with femoral–iliac, calf and portal deep vein thromboses received low-molecular weight heparin in an outpatient setting but continued the planned treatment.

Table 2.

‘Laboratory risk factors’ for thrombosis during chemotherapy according to VTE

 VTE
 
Total 
 No (n = 353) Yes (n = 28) 
Homocysteine (%), median (min–max) 12.2 (4.7–75.0) 11.2 (5.8–20.9) 12.2 (4.7–75.0) 
Protein S (% activity), median (min–max) 88 (24–164) 96 (39–118) 88 (24–164) 
Protein C (% activity), median (min–max) 104 (25–176) 105.5 (16–158) 104 (14–176) 
WBC (×103/μl), median (min–max) 6 (3–16) 6 (4–10) 6 (3–16) 
HB (g/dl), median (min–max) 13 (8–16) 13 (11–16) 13 (8–16) 
PLT (×103/μl), median (min–max) 218 (76–489) 259.5 (131–412) 221 (76–489) 
Factor V Leiden, n (%) 
    Negative 336 (92.31) 28 (7.69) 364 
    Positive 14 (100.00) 0 (0.00) 14 
Factor II, n (%) 
    Negative 342 (92.93) 26 (7.07) 368 
    Positive 9 (90.00) 1 (10.00) 10 
 VTE
 
Total 
 No (n = 353) Yes (n = 28) 
Homocysteine (%), median (min–max) 12.2 (4.7–75.0) 11.2 (5.8–20.9) 12.2 (4.7–75.0) 
Protein S (% activity), median (min–max) 88 (24–164) 96 (39–118) 88 (24–164) 
Protein C (% activity), median (min–max) 104 (25–176) 105.5 (16–158) 104 (14–176) 
WBC (×103/μl), median (min–max) 6 (3–16) 6 (4–10) 6 (3–16) 
HB (g/dl), median (min–max) 13 (8–16) 13 (11–16) 13 (8–16) 
PLT (×103/μl), median (min–max) 218 (76–489) 259.5 (131–412) 221 (76–489) 
Factor V Leiden, n (%) 
    Negative 336 (92.31) 28 (7.69) 364 
    Positive 14 (100.00) 0 (0.00) 14 
Factor II, n (%) 
    Negative 342 (92.93) 26 (7.07) 368 
    Positive 9 (90.00) 1 (10.00) 10 

VTE, venous thromboembolism; WBC, white blood cell; HB, haemoglobin; PLT, platelet.

Table 3.

Incidence of VTE during and after chemotherapy according to the tumour site

 Incidence of VTE during chemotherapy, n (%) Incidence of VTE after chemotherapy, n (%) 
All patients 28 (7.35) 2 (0.52) 
Colon 9 (7.38) – 
Gastric 3 (5.88) 1 (1.96) 
Rectal – – 
Pancreas – – 
Breast 16 (8.79) 1 (0.55) 
 Incidence of VTE during chemotherapy, n (%) Incidence of VTE after chemotherapy, n (%) 
All patients 28 (7.35) 2 (0.52) 
Colon 9 (7.38) – 
Gastric 3 (5.88) 1 (1.96) 
Rectal – – 
Pancreas – – 
Breast 16 (8.79) 1 (0.55) 

VTE, venous thromboembolism.

With regard to VTE occurring during chemotherapy, the median time to thrombosis was 4 months (2–6 months). No fatal event was observed.

In the univariate analysis (Table 4), the following covariates showed a statistically significant association with the development of symptomatic VTE during adjuvant chemotherapy: prechemotherapy platelet count of ≥300 × 109/l and a previous episode of VTE thrombosis. In the multivariate analysis (Table 5), the following variables were independently associated with the risk for VTE during adjuvant chemotherapy: prechemotherapy platelet count of ≥300 × 109/l [hazard ration (HR) 2.8; 95% CI, 1.13–7.30, P <0.026] and a previous episode of VTE thrombosis (HR 12.4; 95% CI, 2.48–62.6, P <0.0026).

Table 4.

Risk factors for VTE during chemotherapy: univariate analysis

Variable Odds ratio 95% confidence interval P value 
Breast versus other 1.502 0.691–3.267 0.3049 
Age 1.017 0.983–1.053 0.3220 
Previous VTE 6.941 1.638–29.414 0.0086 
BMI [Weight Kg/Height2 (m2)] 1.015 0.927–1.111 0.7529 
FVL Not estimated 0.9761 
Factor II 1.462 0.178–11.984 0.7237 
Hormone replacement therapy 1.050 0.132–8.378 0.9630 
Homocysteine (%) 0.941 0.853–1.038 0.2228 
WBC (×103/μl) 0.868 0.665–1.134 0.2996 
HB (g/dl) 1.091 0.792–1.503 0.5925 
PLT (×103/μl) 1.598 1.013–2.520 0.0439 
Sex 2.194 0.813–5.920 0.1207 
T stage 1.021 0.693–1.504 0.9160 
N stage 1.282 0.844–1.948 0.2436 
M stage Not estimated 0.9890 
Protein S 1.004 0.984–1.025 0.7005 
Protein C 1.000 0.983–1.018 0.9915 
Smoke 2.139 0.925–4.945 0.0754 
Contraceptives 2.187 0.704–6.796 0.1762 
Coagany 0.363 0.118–1.122 0.0784 
CVC 1.247 0.356–4.364 0.7302 
Variable Odds ratio 95% confidence interval P value 
Breast versus other 1.502 0.691–3.267 0.3049 
Age 1.017 0.983–1.053 0.3220 
Previous VTE 6.941 1.638–29.414 0.0086 
BMI [Weight Kg/Height2 (m2)] 1.015 0.927–1.111 0.7529 
FVL Not estimated 0.9761 
Factor II 1.462 0.178–11.984 0.7237 
Hormone replacement therapy 1.050 0.132–8.378 0.9630 
Homocysteine (%) 0.941 0.853–1.038 0.2228 
WBC (×103/μl) 0.868 0.665–1.134 0.2996 
HB (g/dl) 1.091 0.792–1.503 0.5925 
PLT (×103/μl) 1.598 1.013–2.520 0.0439 
Sex 2.194 0.813–5.920 0.1207 
T stage 1.021 0.693–1.504 0.9160 
N stage 1.282 0.844–1.948 0.2436 
M stage Not estimated 0.9890 
Protein S 1.004 0.984–1.025 0.7005 
Protein C 1.000 0.983–1.018 0.9915 
Smoke 2.139 0.925–4.945 0.0754 
Contraceptives 2.187 0.704–6.796 0.1762 
Coagany 0.363 0.118–1.122 0.0784 
CVC 1.247 0.356–4.364 0.7302 

VTE, venous thromboembolism; BMI, body mass index; FVL, factor V Leiden; WBC, white blood cell; HB, haemoglobin; PLT, platelet; CVC, central venous catheter.

Table 5.

Risk factors for VTE during chemotherapy: multivariate analysis

Variable Odds ratio 95% confidence interval P value 
Previous VTE 7.666 1.775–33.112 0.0064 
PLT 1.655 1.0.39–2.637 0.0341 
Variable Odds ratio 95% confidence interval P value 
Previous VTE 7.666 1.775–33.112 0.0064 
PLT 1.655 1.0.39–2.637 0.0341 

VTE, venous thromboembolism; PLT, platelet.

discussion

This is the first study evaluating the role of acquired and inherited risk factors for the development of VTE in cancer patients receiving adjuvant chemotherapy.

The association between cancer and VTE has been well described, but most of the published studies analysed a heterogeneous cancer population, including newly diagnosed patients, those receiving active therapy, hospitalised patients and those receiving terminal care. Here, instead, we carried out a prospective study of the incidence of symptomatic VTE and its associated risk factors in cancer outpatients who were beginning their first adjuvant chemotherapy regimen. Hence, by enrolling patients without the evidence of disease, after radical surgery, this study allowed us to analyse the role chemotherapy plays without affecting the active cancer.

The present data are consistent with those of a population-based study of patients with a new diagnosis of VTE [9]. In this study, patients receiving chemotherapy showed a sixfold increase in VTE when compared with patients not receiving such treatment. In the prospective study by Levine et al. [3], the incidence of VTE was similar to that reported in the present study. Furthermore, in the study by Pritchard et al. [10], the authors found a remarkable incidence of VTE (12.6%) in postmenopausal breast cancer patients receiving concomitant chemohormonotherapy.

According to two recent guidelines, patients receiving adjuvant chemotherapy are generally considered at low risk for the development of VTE [4, 11]. Since most of these patients do not develop VTE, it is important to identify those who are at high risk for VTE.

Our study indicates that, when prospectively evaluated, the incidence of VTE is relevant. The aforementioned studies confirm our data [3, 9, 10]. There are several factors that may explain why VTE is underestimated in this subgroup of patients. Indeed, many clinical trials published in recent years have reported few VTE-related toxic effects in the adjuvant setting. Furthermore, in its initial version, the USA National Cancer Institute criteria for common toxicity did not include VTE, indicating that this type of complication had been underestimated.

Many studies have demonstrated the role of inherited APC resistance—via the FVL mutation—in the aetiology of VTEs. Similarly, the PT G20210A mutation is also associated with an increased risk of developing VTE. In our study, FVL was detected in the heterozygous form in 14 of 381 cancer patients (3.6%) and the PT G20210A mutation gene in the heterozygous form in 10 of 381 cancer patients (2.5%). It was surprising that, in the present study, only one patient was heterozygous for PT G20210A and none of the patients with FVL developed VTE. Furthermore, no role was found for homocysteine, PC, protein S and AT III. One large recently reported study calculated that among 10 000 cancer patients, only 8–34 VTE events may be directly related to FVL or G20210A gene mutation [12]. If we assume that a prophylactic anticoagulant therapy would be effective in 80% of these patients, we need to screen 10 000 cancer patients to prevent 7–27 events. Overall, this indicates that screening would not be cost effective.

We found a strong association between elevated basal platelet counts and the risk for VTE during chemotherapy. A recent study of cancer outpatients identified an admission platelet count of >350 000/mm3 as predictive for VTE [13]. In that article, the authors basically described patients with active disease. Our article, instead, is the first to report the association between elevated basal platelet count and the risk of developing VTE during adjuvant chemotherapy. In the present study, 13 of 79 (16.5%) patients with a platelet count of >300 000/mm3 developed VTE as compared with 17 of 302 (5.6%) for those with <300 000/mm3. These results indicate that patients with thrombocytosis should be considered as high risk for developing VTE during chemotherapy. The association of platelet count with thrombosis may simply reflect an inflammatory state induced by a more advanced tumour, a previous surgery or infection. Most of our patients received chemotherapy 30–40 days after surgical tumour excision. We cannot rule out the possibility that thrombocytosis may be an acute phase effect. It is known that platelets play a complex role in haemostasis and thrombosis. Platelet activation leads to exocytosis of granular constituents, release of newly synthesised mediators and discharge of membrane-bound transcellular signalling molecules. Platelets may be an important link in physiological or pathological processes including inflammation, malignancy and the immune response.

A recent elegant study confirms that platelets may play a major role in the development of VTE in cancer patients [14]. This work identified a previous episode of VTE as a risk factor for VTE development. All these VTE events occurred before the cancer diagnosis. Among these patients, six cases were idiopathic since no risk factor was identified in the patient's medical history; three others developed femoral vein thrombosis after major surgery.

Contrary to a recent study, we did not find any difference in the incidence of VTE between breast and GI cancer patients [13]. Nevertheless, the different patient populations may explain these apparently different results. Indeed, only patients who had undergone radical resection and were free of any evidence of disease were enrolled in our study. Since cancer cells can activate the coagulation cascade, it is reasonable to think that only GI patients with active cancer are at increased risk for VTE as compared with those with breast cancer. This may explain the differences in our results versus those obtained by Khorana et al. [13].

We reported an increased risk for VTE in breast cancer patients receiving an anthracycline- or a CMF-based chemotherapy. Furthermore, the VTE rate was increased in GI cancer patients receiving a 5-fluorouracil-based chemotherapy. These results are consistent with previous reports [15, 16].

Our study has several limitations: it only considered symptomatic VTE and thus a number of asymptomatic VTEs may have been missed, which, in turn, would lead to an underestimation of the VTE risk. Furthermore, we did not investigate the role of the D-dimer, available thrombin generation markers and methylenetetrahydrofolate reductase (MTHFR) mutations.

Falanga et al. [17] evaluated markers of in vivo clotting activation (thrombin–AT complex, PT fragment 1 ± 2 and D-dimer) and natural anticoagulants (PC and AT). The prospective randomised study was carried out in 32 metastatic breast cancer patients receiving chemotherapy. In this case, the D-dimer was elevated in cancer patients as compared with normal controls and declined in 16 patients receiving warfarin prophylaxis [17]. Nevertheless, none of the laboratory variables could predict thrombosis in individual patients.

Finally, in a further study of patients with both normal D-dimer and ultrasonography, only one episode of VTE occurred during a 3-month follow-up period [19].

Thrombin generation is a very promising approach for detecting an individual’s coagulation potential. Various parameters may be calculated such as time to thrombin burst, thrombin peak value and total thrombin generated. The most important study evaluating the role of thrombin generation to predict VTE recurrence excluded cancer patients [19]. No prospective validated studies have been carried out in cancer patients and future trials are needed to draw conclusions.

Finally, with regard to the MTHFR gene, it is well known that cytosine to thymine substitution at position 677 results in a variant of the enzyme with reduced catalytic activity and elevated plasma homocysteine. Hyperhomocysteinaemia is considered a risk factor for VTE, but the relationship between thrombophilia and a mutation in the MTHFR C677T—which is associated with a thermolabile phenotype, decreased enzyme activity and mild hyperhomocysteinaemia—is still controversial. Given the conflicting results reported thus far, no definitive conclusions can be drawn on the role this mutation plays in patients with or without cancer [20, 21].

In sum, in a prospective study, we found that the incidence of VTE is not negligible and is relevant from a clinical point of view. Furthermore, we confirm that chemotherapy is a strong risk factor for VTE since most of the events occur during adjuvant chemotherapy. Finally, we identified two risk factors to predict VTE during chemotherapy. Our data may be potentially important in selecting patients at risk for VTE during chemotherapy and for the design of future trials to investigate the role of primary prophylaxis.

This paper is dedicated to Nico Fenili.

References

1.
Rickels
FR
Levine
M
Edwards
RL
Hemostatic alterations in cancer patients
Cancer Metastasis Rev
 , 
1992
, vol. 
1
 (pg. 
237
-
248
)
2.
Levine
MN
Lee
AYY
Treatment of venous thrombosis in cancer patients
Acta Haematol
 , 
2001
, vol. 
106
 (pg. 
81
-
87
)
3.
Levine
MN
Gent
M
Hirsch
I
, et al.  . 
The thrombogenic effect of anticancer drug therapy in women with stage II breast cancer
N Engl J Med
 , 
1988
, vol. 
318
 (pg. 
404
-
407
)
4.
Mandalà
M
Falanga
A
Piccioli
A
, et al.  . 
Venous thromboembolism and cancer: guidelines of the Italian Association of Medical Oncology (AIOM)
Crit Rev Oncol Hematol
 , 
2006
, vol. 
59
 (pg. 
194
-
204
)
5.
Toulon
P
Smirnov
M
Triscott
M
, et al.  . 
A new chromogenic assay (HemosIL ThromboPath) is sensitive to major prothrombotic risk factors affecting the protein C pathway. Results of a multicenter study
Thromb Res
 , 
2009
, vol. 
124
 (pg. 
137
-
143
)
6.
Trossaërt
M
Conard
J
Horellou
MH
, et al.  . 
The modified APC resistance test in the presence of factor V deficient plasma can be used in patients without oral anticoagulant. Thromb
Haemost
 , 
1996
, vol. 
75
 (pg. 
520
-
526
)
7.
Ellis
MH
Manor
Y
Witz
M
Risk factors and management of patients with upper limb deep vein thrombosis
Chest
 , 
2000
, vol. 
117
 (pg. 
43
-
46
)
8.
Tan
Y
Tang
L
Sun
X
, et al.  . 
Total-homocysteine enzymatic assay
Clin Chem
 , 
2000
, vol. 
46
 (pg. 
1686
-
1688
)
9.
Chew
HK
Wun
T
Harvey
D
, et al.  . 
Incidence of venous thromboembolism and its effect on survival among patients with common cancers
Arch Intern Med
 , 
2006
, vol. 
166
 (pg. 
458
-
464
)
10.
Pritchard
KI
Paterson
AH
Paul
NA
, et al.  . 
Increased thromboembolic complications with concurrent tamoxifen and chemotherapy in a randomized trial of adjuvant therapy for women with breast cancer. National Cancer Institute of Canada Clinical Trials Group Breast Cancer Site Group
J Clin Oncol
 , 
1996
, vol. 
14
 (pg. 
2731
-
2737
)
11.
Lyman
Gh
Khorana
AA
Falanga
A
, et al.  . 
American Society of Clinical Oncology guideline: recommendations for venous thromboembolism prophylaxis and treatment in patients with cancer
J Clin Oncol
 , 
2007
, vol. 
25
 (pg. 
5490
-
5505
)
12.
Blom
JW
Dogen
CJ
Osanto
S
Rosendaal
FR
Malignancies, prothrombotic mutations, and the risk of venous thrombosis
JAMA
 , 
2005
, vol. 
293
 (pg. 
715
-
722
)
13.
Khorana
A
Francis
CW
Culakova
E
Lyman
GH
Risk factors for chemotherapy-associated venous thromboembolism in a prospective observational study
Cancer
 , 
2005
, vol. 
104
 (pg. 
2822
-
2829
)
14.
Ay
C
Simanek
R
Vormittag
R
, et al.  . 
High plasma levels of soluble P-selectin are predictive of venous thromboembolism in cancer patients: results from the Vienna Cancer and Thrombosis Study (CATS)
Blood
 , 
2008
, vol. 
112
 (pg. 
2703
-
2708
)
15.
Saphner
T
Tormey
DC
Gray
R
Venous and arterial thrombosis in patients who receive adjuvant therapy for breast cancer
J Clin Oncol
 , 
1991
, vol. 
9
 (pg. 
286
-
294
)
16.
Otten
HM
Mathijssen
J
ten Cate
H
, et al.  . 
Symptomatic venous thromboembolism in cancer patients treated with chemotherapy, an underestimated phenomenon
Arch Intern Med
 , 
2004
, vol. 
164
 (pg. 
190
-
194
)
17.
Falanga
A
Levine
MN
Consonni
R
, et al.  . 
The effect of very-low-dose warfarin on markers of hypercoagulation in metastatic breast cancer: results from a randomized trial
Thromb Haemost
 , 
1998
, vol. 
79
 (pg. 
23
-
27
)
18.
Ten Wolde
M
Kraaijenhagen
RA
Prins
MH
Büller
HR
The clinical usefulness of D-dimer testing in cancer patients with suspected deep venous thrombosis
Arch Intern Med
 , 
2002
, vol. 
162
 (pg. 
1880
-
1884
)
19.
Hron
G
Kollars
M
Binder
BR
, et al.  . 
Identification of patients at low risk for recurrent venous thromboembolism by measuring thrombin generation
JAMA
 , 
2006
, vol. 
296
 (pg. 
397
-
402
)
20.
Ely
SF
Potential risk of a common hereditary thrombophilia and adjuvant chemotherapy for low-stage breast cancer
J Clin Oncol
 , 
2005
, vol. 
23
 (pg. 
4242
-
4243
)
21.
Pihusch
R
Danzl
G
Scholz
M
, et al.  . 
Impact of thrombophilic gene mutations on thrombosis risk in patients with gastrointestinal carcinoma
Cancer
 , 
2002
, vol. 
94
 (pg. 
3120
-
3126
)