Thalidomide thromboprophylaxis in multiple myeloma: A review of current evidence

Authors


Ms Marliese Renee Alexander B. Pharm Hons, Pharmacy Department, Peter MacCallum Cancer Centre, St Andrews Place, East Melbourne 3002, Melbourne, Victoria 3002, Australia. Email: marliese.alexander@petermac.org

Abstract

Currently multiple antithrombotic agents are used for thalidomide thromboprophylaxis in multiple myeloma patients. Agents used include low-dose aspirin, fixed low-dose and therapeutic warfarin and prophylactic low molecular weight heparin. To evaluate the evidence for the efficacy and safety of aspirin, warfarin and low molecular weight heparin thromboprophylaxis in multiple myeloma patients on thalidomide a literature search was conducted in May and June 2011. Databases searched included the Cochrane Database of Systemic Reviews and the Database of Abstracts of Reviews of Effects, Evidence Based Medicine Reviews and Ovid MEDLINE. The search was restricted to English language articles and limited to articles published from 2005 to 2011. Most studies consisted of small prospective cohort studies not originally designed to assess thromboprophylaxis as an outcome. A single comparative randomized trial, several retrospective review articles, two meta-analyses and two clinical practice guidelines were also identified. Current evidence fails to demonstrate a clear advantage of any particular thromboprophylaxis strategy. Results from the only prospective comparative randomized trial found no significant differences among aspirin, warfarin and low molecular weight heparin. More studies are required that consider not only efficacy and safety, but also costs, lifestyle burden and patient preference.

INTRODUCTION

Patients with multiple myeloma (MM) are at increased risk of thromboembolism due to both disease-specific and treatment-specific risk factors.1 Thalidomide, an immunomodulatory agent with potent anti-angiogenic and anti-inflammatory properties, has been shown to be an effective treatment for MM with high response rates and increased survival.2 However, thalidomide-based regimens have been shown to further increase the risk of venous thromboembolism (VTE) in MM patients with the incidence of VTE without thromboprophylaxis ranging from 7–36%.3 Thalidomide is used as single-agent or multi-agent therapy as part of induction, consolidation and maintenance regimens.2 Thalidomide as a single agent does not appear to significantly increase the risk of VTE; however, in combination with steroids, anthracyclines and other chemotherapeutic agents, is seen to significantly increase the risk of VTE.3 Bortezomib, a proteasome inhibitor used for its anti-proliferative and pro-apoptotic effects in MM is seen to act as a thrombo-protective agent and reduces the risk of VTE associated with thalidomide treatment.4

Although there is a clear indication for thromboprophylaxis in MM patients taking thalidomide as part of a multi-agent regimen, evidence for the use of a particular agent is lacking.3–12 The introduction of antithrombotic agents has been reported to reduce the incidence of VTE to less than 10%.2 This narrative review aims to gather available evidence to evaluate the safety and relative efficacy of aspirin, warfarin and low molecular weight heparin (LMWH) as agents for thromboprophylaxis in thalidomide-based regimens. In this review any report of deep vein thrombosis or pulmonary embolism is categorized under the single term of venous thromboembolism (VTE).

METHOD

Data search

In May and June 2011 a literature search was conducted using the Cochrane Database of Systemic Reviews (CDSR) and the Database of Abstracts of Reviews of Effects (DARE), Evidence Based Medicine (EBM) Reviews and Ovid MEDLINE. Searches were conducted using both MeSH terms and text words. Text words searched included multiple myeloma, plasma cell neoplasm$, cancer of plasma cell$, myeloma$, plasmacytoma$, Kahler$ disease, thalidomide, aspirin, acetylsalicylic acid, warfarin, LMWH and enoxaparin. Thrombo* was used to collectively search for thromboembolism, thrombosis, thromboembolic event, venous thrombosis, venous thromboembolism, pulmonary embolism and deep vein thrombosis and prophyl* for prophylaxis or prophylactic and prevent* for prevention/ preventative. MeSH terms searched included multiple myeloma and anticoagulation. In MEDLINE multiple myeloma as a text terms maps to multiple myeloma as the MeSH term as well. The search was restricted to English language articles only, excluding case reports and it was limited to articles published from 2005 to 2011. The reference lists of retrieved articles were also searched for additional articles.

Study selection

To be included in this review studies had to satisfy the following criteria: (i) patients with newly diagnosed or previously treated MM; (ii) patients on a thalidomide-based regimen (i.e., not lenalidomide); (iii) report incidence of VTE as an outcome measure.

RESULTS

No results were found in the CDSR or DARE. A search of EBM Reviews identified three relevant results, all of which were also found searching MEDLINE. Using the previously defined search strategy the literature search yielded 30 results. Overall 15 results were excluded: six evaluated only lenalidomide, six evaluated the efficacy of thalidomide-based regimens, one the pathophysiology of immunomodulatory-induced thrombosis, one evaluated treatment of thromboembolism and one duplicated previously reported data. Five additional studies were retrieved from the reference lists of previously identified literature. Of the relevant publications, two were meta-analyses,5,6 eight review articles,3,4,7–12 seven prospective trials13–19 and two clinical practice guidelines.20,21 Of the prospective trials only one was a randomized head-to-head comparative trial of all three agents.19 Results from the prospective trials are summarised in Table 1. In all the studies, the authors recognize the lack of prospective comparative randomized trials and identify a need for further studies in this area before any antithrombotic regimen can be definitively recommended over another for this patient group. Sub-analyses within the group are also required with current studies identifying the effectiveness of antithrombotic regimens in certain regimens and subgroups only; for example, newly diagnosed versus previously treated, and thalidomide alone or with high-dose dexamethasone or multi-agent chemotherapy.

Table 1.  Published data thalidomide thromboprophylaxis
PublicationStudy designDisease statusAgentsPatients (N)ThromboprophylaxisThrombotic incidence rate
  1. CAD, cyclophosphamide, doxorubicin and dexamethasone; CTD, cyclophosphamide, thalidomide and dexamethasone; DCEP, dexamethasone, cyclophosphamide, etoposide and cisplatin; DvD-T, doxorubicin, vincristine, dexamethasone and thalidomide; MP-T, melphalan, prednisolone and thalidomide; TAD, thalidomide, doxorubicin and dexamethasone; TD, thalidomide and dexamethasone; VAD, vincristine, doxorubicin and dexamethasone; VMPT-VT, bortezomib, melphalan, prednisone and thalidomide followed by bortezomib and thalidomide maintenance; VTD, bortezomib, thalidomide and dexamethasone.

Webber et al.13Prospective studyNewly diagnosedTD24Low-dose warfarin25%
16Therapeutic warfarin0%
Zangari et al.14Prospective studyNewly diagnosedMulti-agent chemotherapy (VAD, DCEP, CAD) + thalidomide87None34%
35Low-dose warfarin31%
68LMWH15%
Palumbo et al.15Prospective studyNewly diagnosedMP-T129None20%
LMWH3%
Minnema et al.16Prospective studyNewly diagnosedTAD211LMWH9%
Baz et al.17Prospective studyNewly diagnosed & relapsed refractoryDVd-T19None58%
58Aspirin from beginning19%
26Aspirin after ≥1 cycle15%
Niesvizky et al.18Retrospective analysisNewly diagnosed & relapsed refractoryCTD60None33%
Aspirin0%
Prospective studyNewly diagnosedTD15Aspirin6.6%
Palumbo et al.19Randomized controlled trialNewly diagnosedTD, VTD, VMPT-VT667Aspirin6.4%
Warfarin8.2%
LMWH5.0%

META-ANALYSES

Carrier et al.5 searched MEDLINE, EMBASE and OVID Health Star (1950–2010) as well as a handsearch of potentially relevant journals and the American Society of Hematology conference proceedings from 2005 to 2009.5 The aim of their review was to determine the absolute rates of VTE with and without different thromboprophylactic agents in patients with newly diagnosed or previously treated MM receiving thalidomide-based or lenalidomide-based regimens. The review included 61 articles (11 randomized controlled trials and 50 prospective cohort studies) that assessed thalidomide-based regimens including 24 with newly diagnosed MM and 37 with previously treated MM. Listed as the rate of VTE per 100 patient cycles, the authors report the following figures based on a variety of thalidomide-based regimens; (i) for newly diagnosed MM patients: aspirin 2.3 (95% CI, 0.0–2.79), fixed low-dose warfarin 2.8 (95% CI, 2.0–3.9), prophylactic LMWH 2.1 (95% CI, 1.1–3.6) and therapeutic anticoagulation 1.6 (95% CI, 0.2–4.1); (ii) for previously treated MM patients: aspirin 0.4 (95% CI, 0.01-0.9), fixed low-dose warfarin 2.4 (95% CI, 1.3–4), prophylactic LMWH 0 (95% CI, 0–1.72) and therapeutic anticoagulation 0 (95% CI, 0–2.2). The authors concluded that therapeutic anticoagulation (therapeutic LMWH or warfarin to a target International Normalised Ratio (INR) 2–3) produces the largest absolute risk reduction. However, given the available data, formal comparisons between thromboprophylactic agents (aspirin, fixed low-dose warfarin and prophylactic LMWH) could not be performed.

El Aaccaoui et al.6 reviewed 50 articles identified from a search of PubMed in 2006. The aim of the study was to assess the risk of VTE in patients taking thalidomide and whether the risk is affected by concomitant administration of other medications, and to study the effect of anticoagulation and anti-platelet medications. In an analysis of the 3322 MM patients the relative risk for prophylactic LMWH or therapeutic warfarin (INR 2–3) was 0.6 (95% CI, 0.4–0.8), low-dose aspirin 0.7 (95% CI, 0.5–1.1) and fixed low-dose warfarin 1.0 (95% CI, 0.6–1.7). The authors concluded that prophylactic LMWH and therapeutic warfarin (INR 2–3) significantly reduced the risk of VTE (P < 0.01). However, neither low-dose aspirin nor fixed low-dose warfarin was found to significantly reduce the risk of VTE events (P > 0.1). Overall, the number of patients analysed who were taking aspirin (237) and low-dose warfarin (154) was significantly less than those on therapeutic warfarin or prophylactic LMWH (576).

Limitations of meta-analyses

As most of the studies evaluating thromboprophylaxis were small in numbers it is difficult to evaluate the true treatment effect. Thus although meta-analyses provide the highest level of evidence, they have several limitations. Even using the best available evidence, both analyses relied heavily on retrospective data and data from studies not designed to assess thromboprophylaxis as a primary outcome. Furthermore, the meta-analyses were conducted prior to the publication of the only (at time of writing this review) randomized comparative trial comparing the three commonly proposed regimens of low-dose aspirin, fixed-low-dose warfarin and prophylactic LMWH.19

No language restriction was applied to the search results of either review; however El Aaccaoui et al.6 restricted articles to those freely available from the American University of Beirut medical library;6 which introduces bias into the type of research articles reviewed – 16 articles were omitted based on this process.

In both meta-analyses data were analysed according to thalidomide regimen (alone, with dexamethasone, with multi-agent chemotherapy); however El Aaccaoui et al.6 did not differentiate between patients with newly diagnosed or with relapsed/refractory disease. Newly diagnosed MM patients are at increased risk of VTE compared with patients with relapsed/refractory disease.3 The distribution of patients was not reported and, if not evenly spread, may have introduced a confounding factor into the data.

Clinical trials – warfarin

Both Webber et al.13 and Zangari et al.14 evaluated prophylactic fixed low-dose warfarin (1 mg/day) in newly diagnosed MM patients on a variety of thalidomide-based regimens. Following high incidence rates of VTE (25 and 34%, respectively) both trial protocols were amended to use full-dose warfarin and prophylactic LMWH. Zangari et al.14 reported a reduced incidence of VTE with 40 mg/day LMWH (15%) but Webber et al.13 failed to give details on the effectiveness of full-dose warfarin.

Clinical trials –LMWH

In the previously mentioned study by Zangari et al.14 40 mg/day LMWH thromboprophylaxis eliminated the difference in VTE incidence between the two study arms, with an incidence rate of 15% for both the thalidomide and non-thalidomide treatment groups.

In a study by the Italian Myeloma Network15 including elderly MM patients (> 65 years old) randomized to receive melphalan, prednisone and thalidomide, the introduction of enoxaparin prophylaxis (40 mg/day) reduced the rate of thromboembolism from 20 to 3% (P = 0.005).

A prospective study by Minnema et al.16 evaluated vincristine, dexamethasone and doxorubicin and thalidomide-dexamethasone-doxorubicin (TAD) chemotherapy with thromboembolism as a secondary outcome measure. The incidence of VTE was reported at 9% in the TAD arm with the use of prophylactic LMWH (nadroparine) – a comparable rate to that typically seen with intensive chemotherapy.6

Clinical trials – aspirin

Baz et al.17 evaluated the efficacy of low-dose aspirin thromboprophylaxis in both newly diagnosed and previously treated MM patients (N = 103) on a chemotherapy regimen of pegylated doxorubicin, vincristine, dexamethasone and thalidomide. Low-dose aspirin (81 mg/day) was introduced following high rates of VTE without thromboprophylaxis. Aspirin decreased the overall rate of VTE to 17.8% (P < 0.001) compared to 58% with no thromboprophylaxis. Moreover, it was well tolerated by patients and associated with no bleeding complications.

Niesvizky et al.18 describes the use of low-dose aspirin (81 mg/day) thromboprophylaxis in a subset of MM patients randomized to receive thalidomide and dexamethasone. The study included small numbers of participants (15), making it difficult to assess the true meaning of the data. Thromboembolism was reported in 6.6% of patients with aspirin prophylaxis on thalidomide and dexamethasone compared with 21.4% of patients in the control arm (dexamethasone alone).

Clinical trials – randomized comparative trials

Palumbo et al.19 recently (2011) published a head-to-head randomized trial comparing low-dose aspirin (100 mg/day), fixed low-dose warfarin (1.25 mg/day) and prophylactic LMWH (40 mg/day) for thalidomide thromboprophylaxis in MM patients. At the time of writing this review, this is the only known published randomized comparative study of thromboprophylactic agents in this setting. The trial recruited 667 newly diagnosed MM patients who were randomized in a 1:1:1 ratio to low-dose aspirin, fixed low-dose warfarin and prophylactic LMWH. The trial was well constructed with no significant prognostic differences between the three intervention arms and the median follow up was 24 months. Overall, the trial demonstrated no statistically significant differences in the incidence of primary outcome (thromboembolism, acute cardiovascular event or sudden death within the first 6 months of treatment) between the three agents. When the participants were analysed in subgroups according to specific chemotherapy regimens differences in thromboprophylaxis efficacy were seen. Specifically, elderly patients on bortezomib, melphalan, prednisone and thalidomide followed by continuous bortezomib and thalidomide, were at increased risk of VTE (+11%, P = 0.006) when treated with fixed low-dose warfarin compared to prophylactic LMWH. There were no significant differences in this setting between the aspirin and LMWH groups and the incidence of adverse events was similar across all arms.

Review articles

Many review articles have been published within the past 5 years.3,4,7–12 A range of studies are evaluated, including small cohort studies and retrospective analysis as well as some of the prospective studies evaluated in the current narrative review. All reviews acknowledge limitations in regard to the quality of data available and conclude that randomized controlled trials are required to compare different thromboprophylactic regimens.

DISCUSSION

Since the publication of recent meta-analyses5,6 that concluded that head-to-head comparative trials are required to compare thromboprophylactic regimens, a robust multicentre phase 3 randomized clinical trial comparing aspirin, fixed low-dose warfarin and prophylactic LMWH has been published.19 The data suggest no overall significant differences between the three regimens, indicating a need for further studies to validate these results and analyze their cost and benefits.

Most studies evaluate the safety profiles of thromboprophylactic regimens (reported as incidence of bleeding). However, few studies consider patient preference for oral or injectable administration and their ability to comply with INR monitoring. Compliance indicators are rarely reported, introducing bias into studies not analysed using an intention-to-treat principle. The cost-benefit ratio should also be considered, as LMWH is significantly more expensive than either warfarin or aspirin. The cost per 30 days of treatment, based on the dispensing prices listed under the general schedule of the Australian Pharmaceutical Benefits Scheme (PBS), are: $163.32 for 40 mg daily enoxaparin, $12.42 for 1–2 mg daily warfarin and $2.15 for 100 mg daily aspirin.22 Under the PBS (current at 2011) the patient contribution is capped at $33.30 (general patients) or $5.60 (concession patients) with the Australian government absorbing the remaining costs.22 With new evidence demonstrating the efficacy of both aspirin and fixed low-dose warfarin, there is scope for significant pharmaceutical cost savings.

The International Myeloma Working Group8 recommend a stratified approach to thromboprophylaxis based on risk factors for VTE. Aspirin is recommended for patients with up to one individual-related or myeloma-related risk factor, LMWH or full-dose warfarin in the presence of at least two risk factors and LMWH or full-dose warfarin for all patients on high-dose dexamethasone, doxorubicin or multi-agent chemotherapy, independent of the presence of additional risk factors. The authors state that the goal should be to use the safest and least cumbersome form of thromboprophylaxis that reduces the risk of VTE to at least below 10%.8

The American Society of Clinical Oncology (ASCO) 2007 Guidelines for venous thromboembolism20 state that LMWH or adjusted-dose warfarin (INR 1.5) thromboprophylaxis should be used in all MM patients being treated with thalidomide in combination with dexamethasone or any other chemotherapy agents.

The National Comprehensive Cancer Network (NCCN) Guidelines on venous thromboembolic disease (version 2.2011)21 recommends the use of aspirin (81–325 mg/day) for low-risk MM patients only, and therapeutic warfarin (INR 2–3) or prophylactic LMWH (40 mg enoxaparin/day) for high-risk MM patents. High-risk patients include those on regimens of thalidomide and high-dose dexamethasone, doxorubicin or multi-agent chemotherapy, or for patients with two or more individual or myeloma risk factors.

CONCLUSION

Studies consistently show evidence for the safety and efficacy of both low-dose aspirin17–19 and prophylactic LMWH,14–16,19 with greater evidence for LMWH over a wider range of thalidomide-based regimens. The NCCN guidelines21 reflect current evidence for the safety of low-dose aspirin; however limit its use to low-risk patients. ASCO is yet to include aspirin in its guidelines for thalidomide thromboprophylaxis.20 Results for low-dose warfarin are inconsistent13,14,19 and further studies are required to evaluate its safety and efficacy.

Additional randomized comparative studies are required with sub-analyses of different MM groups based on the current risk stratification models for VTE: thalidomide alone, and with dexamethasone, anthracyclines or multi-agent chemotherapy. A cost-benefit analysis should be conducted taking into account patients' preference and their compliance with oral or injectable administration.

Ancillary