Venous Thrombosis and Long Term Use of Central

Venous Catheter in Cancer Patients treated with

Chemotherapy:

Epidemiology and Risk Factors

Giuseppe Curigliano

Dipartimento di Medicina, Divisione di Oncologia Medica,

Unità di Farmacologia Clinica e Nuovi Farmaci

Istituto Europeo di Oncologia,

Milano

M. Verso and G. Agnelli. Venous Thromboembolism Associated With Long-Term Use of Central Venous Cathetersin Cancer Patients. J Clin Oncol 21:3665-3675, 2003.

Epidemiology of catheter related Venous

Thromboembolism

(CVC-related VTE) in cancer patients

The reported incidence of symptomatic CVC-related VTE varies from 0.3 to 28.3%, with a rate of 12% in pediatric patients

The incidence of CVC-related VTE assessed by venography has been reported to vary from 27 to 66%. Most of the thrombi in these studies were asymptomatic.

There is no conclusive evidence that a particular type of CVC is more or less thrombogenic than others

Time course analysis of CVC-related VTE indicates the first 6 weeks after CVC insertion at higher risk of thromboembolic complication

Authors Study No. Patients CVC-related VTE (%)

Smith, 1983 RetrospectiveAdults

2800 0.3

Soto-Velasco, 1984

RetrospectiveAdults

1611 0.7

Cassidy, 1987 ProspectiveAdults

416 2.6

Gould, 1993 ProspectiveAdults

255 14.5

Eastridge, 1995

ProspectiveAdults

322 10

Kock, 1996 RetrospectiveAdults

1500 2.5

Schwarz, 2000 ProspectiveAdults

923 3.1

Biffi, 2001

Kurtakose, 2002

ProspectiveAdults

ProspectiveAdults

304

422

6.6

7.1

Incidence of clinically overt CVC-related VTE in cancer patients

Authors Study No. Patients CVC-related VTE (%)

Stoney, 1976 ProspectiveAdults

203 31

Ladefoged, 1978

RetrospectiveAdults

48 27.1

Brismar, 1982 ProspectiveAdults

53 35.8

Bern, 1990 RetrospectiveAdults

42 37.5

Monreal, 1996 RetrospectiveAdults

26 62

De Cicco, 1997 ProspectiveAdults

127 66

Martin, 1999 ProspectiveAdults

60 58.3

Glaser, 2001 ProspeciveChildren

24 50

Incidence of venographic CVC-related VTE in cancer patients

Risk factors for developement of CVC-related DVT

in cancer patients

CVC features Patients features

Chemical structure High platelet counts

Catheter diameter Cancer related activation of coagulation

Number of lumens CVC-related activation of coagulation

Side of insertion Chemotherapy-related activation of coagulation

CVC-related infection Thrombofilic molecular abnormalities

Insertion techniques

Venous thromboembolism (VTE) during

chemotherapy

Chemotherapy itself can increase the risk of thromboembolic disease:

Acute damage on vessel walls (bleomycin, carmustine, vinca alkaloids, adriamycin)

Decrease of natural coagulation inhibitors (reduced levels of protein C and S with cyclophosphamide, methotrexate and fluorouracil and reduced levels of antithrombin III with L-aspariginase)

Neil Goldenberg, Susan R. Kahn, and Susan Solymoss. Markers of Coagulation and Angiogenesis in Cancer-AssociatedVenous Thromboembolism. J Clin Oncol 21:4194-4199, 2003

Markers of coagulation and angiogenesis in cancer-

associated venous thromboembolism

Peripheral blood was collected before anticoagulant therapy from

cancer patients with acute deep venous thrombosis (DVP), those

without DVP and patients with acute DVT and no cancer

Median levels of thrombin-antithrombin complex (p=0.0001), prothrombin fragments 1 2 (p=0.003), and von Willebrand factor antigen were significantly greater (p<0.001) in the DVT cancer group than in the cancer control and DVT control groups

Median levels of tissue-type plasminogen activator were also significantly higher (p=0.0005), while protein C activity was lower in the DVT cancer group than in the DVT control group (p=0.0008).

VTE in cancer patients: the role of chemotherapy

Clahsen et al suggested a contributing role of perioperative chemotherapy

to VTE especially in postmenopausal women and women who underwent mastectomy

Incidence of deep venous thrombosis:

2.1% assigned to perioperative chemotherapy 0.8% assigned to no treatment

Clahsen PC, et Al. Thromboembolic complication after perioperative chemotherapy in women with early breast cancer: an EORTC study. J Clin Oncol 1994: 12: 1266-71

VTE in cancer patients: the role of chemotherapy

Prospective/Randomized trial:

12 weeks of chemo-hormone therapy (CMF-VP-A-T)36 weeks of chemotherapy CMF-VP

Incidence of VTE: 6.8%

4.9% chemo-hormone therapy arm (CMF-VP-A-T)8.8% chemotherapy arm CMF-VP

Levine et al. The thrombogenic effect of anticancer drug therapy in women with stage II breast cancer. New Engl J Med 1988, 318: 404-407

VTE in cancer patients: the role of chemotherapy

179 consecutive germ cell cancer patients

Cisplatin containing regimens

15 patients (8.4%) developed 18 major VTE between the start of chemotherapy and 6 weeks after administration of the last cycle in first line treatment

Of these 18 events, 3 (16.7%) were arterial events, including 2 cerebral ischemic strokes and 15 (83.3%) were VTE including 11 pulmonary embolism. One (5.6%) of the 18 events was fatal.

Weijl et al.Thromboembolic events during chemotherapy for germ cell cancer: a cohort study and review of the literature. J Clin Oncol 2000, 18: 2169-78

VTE in cancer patients: the role of hormone

therapy and chemo-endocrine therapy

Authors observed one or more thromboembolic events in 48 of 353 women (13.6%) randomized to receive tamoxifen plus CMF compared to 5 of 352 women (2.6%) randomized to receive tamoxifen alone (p=0.001).

Significantly more women developed severe VTE in the T plus CMF arm than in the T arm (34 vs 5: p=0.0001).

Most thromboembolic events occurred while women were actually receiving chemotherapy ( 39 of 54, p<0.0001).

Pritchard K. Et al. Increased thromboembolic complications with concurrent tamoxifen and chemotherapy in a randomized trial of adjuvant therapy for women with breast cancer. J Clin Oncol, 1996: 14: 2731-2737

Authors Study Regimens VTE incidence

(%)

Clahsen et al ProspectiveRandomized

FAC perioperativeNo treatment

2.10.8

Levine et al ProspectiveRandomized

CMFVP + A + TCMFVP 36 weeks

4.98.8

IBCSG et al ProspectiveRandomized

CMF + AF periopNo treatment

0.50

Weiss et al ProspectiveRandomized

CMF (2 years)CMF (2 years)

CMFbCG (2 yrs)

3.56.35.4

Pritchard et al ProspectiveRandomized

CMF + TT

13.62.6

Fisher et al ProspectiveRandomized

TamoxifenMTF

CMF + T

1.24.24.5

Tormey et al ProspectiveRandomized

CMFCMF PT

03.8

Wils et al ProspectiveRandomized

E + TT

0.030.015

Kirwan et Al. Prophylaxis for venous thromboembolism during treatment for cancer: questionnaire survey. BMJ, 2003; 327: 597-598

Venous thrombosis in cancer patients:

A questionnaire Survey

29% of medical oncologists respondents to questionnaire consideredtheir patients not at risk of VTE

More than a quarter of medical oncologists do not recognize the thrombogenic effects of cancer treatments

A.K. Kakkar, M. Levine et Al.. Venous thrombosis in cancer patients: Insights from the FRONTLINE survey. The Oncologist 2003; 8: 381-388

Venous thrombosis in cancer patients:

Insights form the FRONTLINE (Fundamental

Research in Oncology and Thrombosis) Survey

35% of respondents to questionnaire were medical oncologists

62% or respondents practice in Western Europe or North America

29 questions were reported for non surgical patients

10 questions were related to vascular access devices

A.K. Kakkar, M. Levine et Al.. Venous thrombosis in cancer patients: Insights from the FRONTLINE survey. The Oncologist 2003; 8: 381-388

Venous thrombosis in cancer patients:

Insights form the FRONTLINE (Fundamental

Research in Oncology and Thrombosis) Survey

80% of respondents to questionnaire considered the use of central venous lines

to be associated with a high risk of VTE

Western Europeans typically considered CVC to be associated with a 0-20% risk of VTE (74% perceived the risk to be in this range), while North American respondents considered the risk, without

prophylaxis, between 11-30%.

Venous thrombosis in cancer patients

In conclusion, the pathogenesis of UL-DVT in patients with CVC is probably multifactorial.

Early thromboembolic events are essentially related to the loss of vessel integrity caused by CVC placement.

Late thromboembolic events are probably related to CVC features, insertion technique, catheter tip position, and occurrence of catheter infection.

The role of thrombophilic molecular abnormalities is less clear.

VTE in cancer patients

European Institute of Oncology Policy

Selection of patients at high risk

Influence of Factor V Leiden and the G20210A prothrombin mutation on the development of deep vein thrombosis in cancer patients treated with chemotherapy

Influence of Factor V Leiden and the G20210A prothrombin mutation on the development of deep vein thrombosis in cancer patients treated with chemotherapy

AIMS

The major objective of the study is to evaluate retrospectively the influence of Factor V Leiden and the G20210A prothrombin mutation in the development of deep vein thrombosis in cancer patients treated with infusional chemotherapy.

Influence of Factor V Leiden and the G20210A prothrombin mutation on the development of deep vein thrombosis in cancer patients treated with chemotherapy

EXCLUSION CRITERIA

· Metastatic cancer patients with one or more of the following deficiency: antithombin III, protein C, or protein S.

· Autoimmune desorders, including primary antiphospholipid-antibody syndrome.

· Deep vein thrombosis not diagnosed by objective methods such as phlebography, unltrasonographic examination, or impedance plethysmography.

Patients and methods

Between January 1999 and February 2001, we retrospectively analyzed the incidence of first deep vein thrombosis in 300 consecutive patients with locally advanced or metastatic breast cancer treated in a single institution with a combination of chemotherapy administered continuously through a totally implanted access port.

We identified 25 women (study group) with catheter-related deep vein thrombosis. For each of the 25 patients we selected two women eligible for identical chemotherapy, with similar age, stage of disease and prognostic features as control.

The prothrombotic factor V Leiden and prothrombin mutation G20210A genotype analyses were performed in all patients.

Analyses were performed on blinded samples and all patients signed a specific informed consent.

Results

•A total of 25 cases (thrombosis) and 50 frequency-matched controls were evaluated for Factor V Leiden.

•Five cases and 2 controls were found with the mutation in the factor V Leiden for a frequency of 20% (95% CI= 9%-39%) and 4% (95% CI=1%-14%) respectively.

•Only one subject (case) was found with the G20210A mutation in the prothrombin gene.

•Time from start of infusion chemotherapy to thrombosis was not significantly different between those with (median=31 days) and without the mutation (median 43 days, p=0.6).

•Thus, the frequency of the mutation was significantly higher in the cases than in controls (p=0.04) and a logistic regression analysis, adjusted by age, yielded an odds ratio of 6.1 (95% CI=1.1-34.3; p=0.04).

Controls (n. 50)

Cases (n. 25)

p values

Age 50 43-54 51 46-55 0.4

Menopausal status

Pre 24 49% 11 44% 0.8

Post 25 51% 14 56%

Tumor stage

Locally advanced 29 59% 15 60% 0.9

Metastatic 20 41% 10 40%

Number of cycles

6 4-6 3 2-5 0.9

Mutations

Prothrombin 0 0% 1 4% 0.3

Factor V Leiden 2 4% 5 20% 0.04

Frequency 1-14% 9-39&

Conclusions

The identification of a subgroup of patients at high risk for thrombosis could be a fair strategy to either avoid medical devices which increase the risk of thromboembolism, or to implement the compliance in prescribing anticoagulant prophylaxis in patients with cancer.

We are not able to recommend thrombosis prophylaxis and immediate catheter removal after treatment, advising testing for the factor V Leiden in all patients receiving continuous infusion chemotherapy.

It is however reasonable to offer testing for factor V Leiden in cases when there’s history of thromboembolism (even when not cancer related), to allow proper anticoagulation for those patients who were so identified.