Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner
Report on the Relative Efficacy of Oral Cancer Therapy for Medicare Beneficiaries Versus Currently Covered Therapy

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Part 4. Thalidomide for Multiple Myeloma

Executive Summary

Multiple myeloma is a progressive, debilitating malignancy characterized by the proliferation and accumulation of cancerous plasma cells and the overabundance of monoclonal paraprotein. It is part of a spectrum of diseases ranging from monoclonal gammopathy of unknown significance (MGUS) to plasma cell leukemia. Extensive skeletal destruction with osteolytic lesions, osteopenia, and/or pathologic fractures is common, as well as anemia, hypercalcemia, and kidney dysfunction.  Although treatable, multiple myeloma is considered incurable and accounts for approximately 2 percent of all cancer deaths.1  Historically, intermittent oral melphalan and prednisone (MP) was standard therapy for untreated symptomatic multiple myeloma.  In more recent years, newer combination chemotherapy regimens have been used both as initial first-line chemotherapy and as salvage chemotherapy, with better response rates but little effect on overall survival. 

Example combination chemotherapy programs include VBCMP (vincristine, carmustine, cyclophosphamide, melphalan, and prednisone) and VAD (vincristine, doxorubicin, and dexamethasone).  There is a survival benefit when patients responding to chemotherapy such as VAD are treated with high dose chemotherapy plus single or double autologous stem cell transplantation.  Nonetheless, over 80 percent of patients still relapse within 7 years.  Treatment programs that include transplantation have limited applicability due to toxicity and associated age, performance status, and organ function requirements.  Nearly all patients with multiple myeloma will eventually relapse and become resistant to further treatment.  Median survival remains approximately 4 years.

Thalidomide, a glutamic acid derivative, was used as sedative in the late 1950s until it was withdrawn from the market because it caused severe birth defects. Thalidomide's anti-angiogenic properties were appreciated in the 1990s and because bone marrow angiogenesis plays a substantial role in the development of multiple myeloma, thalidomide has been tried in multiple myeloma.  Since the first publication documenting objective responses with thalidomide in patients with refractory myeloma was published in 1999, there has been a rapid proliferation of published and abstract reports on the use of thalidomide in multiple myeloma.  In 1998, the Food and Drug Administration (FDA) approved thalidomide for use in treating leprosy (Hansen's disease); it is not currently FDA-approved for multiple myeloma.  Thalidomide can only be prescribed under the System for Thalidomide Education and Prescribing Safety (S.T.E.P.S.) program, patented by Celgene Corporation.

Scope and Key Questions

The key questions for this review were developed with experts in the field of oncology, health economics, and health policy.  The key questions are as follows:

  1. For patients with relapsed or refractory multiple myeloma, what is the effect of thalidomide compared to standard chemotherapy regimens (e.g., VBMCP (vincristine, carmustine, melphalan, cyclophosphamide, and prednisone) and VAD (vincristine, doxorubicin, and dexamethasone)) on 2-year survival, disease-free survival, CR, PR (m-protein), and quality of life?
  2. For patients with relapsed or refractory multiple myeloma, what is the effect of thalidomide compared to standard chemotherapy regimens (e.g., VBMCP (vincristine, carmustine, melphalan, cyclophosphamide, and prednisone) and VAD (vincristine, doxorubicin, dexamethasone)) on adverse effects, tolerability, and compliance?
  3. What patient or tumor characteristics distinguish treatment responders from non-responders and have potential to be used to target therapy?

As there was emerging information regarding the role of thalidomide for newly diagnosed and smoldering multiple myeloma, these groups were also considered as part of this review.


Search Strategy

Primary studies were sought in a computerized bibliographic search of MEDLINE® (1966 through September 2004, updated August 2005) and limited to articles published in the English language. Additional strategies included searching ancillary bibliographic databases, searching abstracts presented at the American Society of Clinical Oncology and American Society of Hematology professional meetings since 2004, querying experts, and checking references of included studies and review articles.

Selection Criteria

Each citation identified from the search strategies was evaluated according to the following selection criteria.  Evaluations were performed by the authors.

Inclusion criteria were as follows:

Patients: Patients with multiple myeloma.
Interventions: Thalidomide
Comparators: Any

Study designs:

  • For efficacy questions:  Prospective clinical trials; may be phase II uncontrolled, or phase III randomized controlled trials.
  • For studies of adverse effects: May be retrospective or prospective case series, cohort studies, or clinical trials provided the number of patients treated (at risk for adverse effects) as well as the number with adverse effects can be ascertained.
  • For studies of predictors of response: May be retrospective or prospective case series, cohort studies, case-control studies, or clinical trials provided the response can be ascertained for patients with and without the predictor.


  • For efficacy questions:  Survival, quality of life (QOL), and the following intermediate outcomes:
    • Complete response:
      • Lack of detectable M-protein in serum or urine by immunoelectrophoresis & immunofixation, maintained for a minimum of 6 weeks.
      • Bone marrow biopsy with <5 percent plasma cells.
      • No increase in size or number of bone lesions.
      • Disappearance of plasmacytomas. 
    • Partial response:
      • Reduction in serum M-protein by at least 50 percent, maintained for at least 6 weeks.
      • Reduction in urine Bence Jones protein by at least 90 percent or <200 mg, maintained for at least 6 weeks.
      • If non-secretory, reduction in bone marrow plasma cells by at least 50 percent, maintained for at least 6 weeks.
      • No increase in size or number of bone lesions.
  • For studies of adverse effects:Adverse effects, tolerability, and compliance with treatment.
  • For studies of predictors of response: Predictive value of patient or tumor characteristics that are associated with clinically important differences in treatment response that are:
    • Related to the mechanism of action of the drug (i.e., molecular target).
    • Candidates for diagnostic testing (even if not commercially or clinically available currently [e.g., Polymerase Chain Reaction]).

The Evidence

1. For patients with relapsed or refractory multiple myeloma, what is the effect of thalidomide compared to standard chemotherapy regimens (e.g., VBMCP (vincristine, carmustine, melphalan, cyclophosphamide, and prednisone) and VAD (vincristine, doxorubicin, and dexamethasone)) on 2-year survival, disease-free survival, CR, PR (m-protein), and quality of life (QOL)?

While the original question was about relapsed or refractory multiple myeloma, we expanded our review of the topic to include untreated myeloma because many of the newer studies of thalidomide focused on this setting.  Also, we included some studies of asymptomatic myeloma although the current standard is not to treat this group but rather adopt an approach of "watchful waiting." The breadth of studies, myeloma treatment settings (first-line, relapsed, asymptomatic, peri-transplantation), and drug combinations highlights the many ways that thalidomide is quickly becoming incorporated into myeloma treatment regimens.  Key clinical issues include the mechanism of this prototype drug, managing toxicity, and finding the most effective dose, schedule, and medication combinations. Nonetheless, thalidomide's most critical contribution to the array of anti-myeloma treatments is as an oral medication with a tolerable side effect profile that has efficacy in the relapsed or refractory setting and can be administered to the elderly and/or debilitated patients typical of the multiple myeloma population.

VBCMP and VAD are the comparators.  No studies have randomized patients to thalidomide versus these interventions.  As such, historical rates and survival estimates from previous trials including these agents must be used as the comparison group.  Two-year survival rates were rarely reported except in the Samson, et al., study of VAD for untreated patients where 83 percent of responders were alive at 2 years.  In the Mineur, et al., trial of bolus VAD vs. VDD for untreated myeloma, median time to progression was 24 months.  Median overall survival had not been reached and was expected to exceed 40 months with both arms.

It is difficult to directly compare numbers between categories as response criteria for the various studies vary widely and very few of the thalidomide data presented are from randomized studies (only thalidomide-dexamethasone vs. dexamethasone or MP in untreated myeloma).  Our use of PPR 25 percent as the summary response criteria for thalidomide is supported in another recent literature review for multiple myeloma. This is notably different than the PPR 50 percent criteria described for most of the older trials.  It can be misleading to compare the PPR 50 percent, as some studies report PPR 50 percent to mean all responses that were greater than 50 percent (i.e., 50-100 percent) and others indicate just those reflected in that response level (e.g., 50-74 percent with next response level at 75 percent).  Response ranges for thalidomide are broad, reflecting heterogeneity among studies and study populations, including the volume and intensity of previous myeloma treatments, study quality, and study size.  Also, participant populations may be represented multiple times in the different published analyses of these studies; it is difficult to determine.

The most notable findings are the following:

  • Thalidomide has activity in both the untreated and resistant/refractory settings.
  • Generally, survival and responses are better when dexamethasone has been added.
  • Response rates and survival estimates do not appear to be substantially different from that seen with VBCMP or VAD.

Thalidomide's place in the multiple myeloma therapeutic armamentarium is clarified as these similar response rates are considered in terms of the comparative adverse events, ease of administration, and ability to be combined with other treatments.

  • First, thalidomide (or thalidomide plus dexamethasone) has a different toxicity profile than the combination chemotherapy regimens.  Until head to head studies are done, it will be difficult to be certain; however thalidomide appears to have less intense toxicity with fewer treatment-related deaths.  Deaths such as those related to neutropenic fever from VBCMP and VAD and cardiotoxicity with VAD are not reported for thalidomide.  The unexpected thromboembolic risk of thalidomide can be mitigated by adding enoxaparin.  Thalidomide's peripheral neuropathy is cumulative and will need further consideration.  Sedation can be minimized by slowly escalating the dose.
  • Second, thalidomide is oral and can be managed in the outpatient setting.  It does not require venous access or central venous catheters.  This is balanced by the increased burden of the S.T.E.P.S. program, an important reminder and safeguard for the known teratogenicity of thalidomide.
  • Third, thalidomide can be administered in elderly, immunocompromised patients and those with renal or cardiac dysfunction.  It is unlikely that the true magnitude of this advantage is represented across the efficacy studies, as such ill patients are often excluded from the study populations.
  • Fourth, it has activity even when patients have been heavily pretreated with VAD, VBCMP, or high dose chemotherapy plus autologous stem cell transplant.  Hence, thalidomide can be added to the list of appropriate options for treatment of multiple myeloma and the timing of its use is considered based upon the needs of the individual.
  • Fifth, evidence of maximal response is seen early so thalidomide does not need to be continued for long periods if it is not effective.  In the 2001 Barlogie, et al., study of thalidomide only in refractory/relapsed myeloma, 70 percent of patients achieving a PPR >25 percent did so within 2 months and 90 percent within 4.5 months. 

  • Sixth, it can be combined with other agents with additive effect.  In particular, lack of severe myelosuppression with thalidomide makes this possible.  Thalidomide plus MP appears to be superior to MP alone and there are many promising combinations.
  • Seventh, thalidomide can be used in the pre- and post-transplantation settings although some recent data suggest that it may be better not to use thalidomide for post-transplant maintenance but rather save the intervention for future relapse states.

Should thalidomide always be combined with dexamethasone?  Pre-clinical data suggests synergistic effects when thalidomide is combined with dexamethasone. Dexamethasone is the main active agent in VAD.  Weber, et al., reported that thalidomide restored the sensitivity of myeloma cells to dexamethasone-induced apoptosis.  Generally, survival and responses are better when dexamethasone has been added, with fewer side effects.  Thalidomide doses are generally lower when dexamethasone is added.  Dexamethasone dosing is variable across studies.  Unless a patient has a contraindication to high dose dexamethasone (e.g., severe labile diabetes, history of steroid psychosis), the addition of dexamethasone is quickly becoming standard when thalidomide is used.

The ideal dose of thalidomide is unclear.  The 2001 Barlogie, et al., study demonstrated that patients who received >42 g of thalidomide in the first 3 months had significantly better response rates and survival.  Similar findings were noted in both of the predictors study on the topic.  Recent studies have looked to decreasing the thalidomide dose though, predominantly in an effort to decrease adverse effects.  This is most noticeable across the range of thalidomide plus dexamethasone studies, some of which start at 50 mg and many of which fix the thalidomide dose at 200 mg.

The role of thalidomide in soft tissue plasmacytomas is also unclear.  Some authors report poorer responses in this setting.  More data are needed.

Only one study specifically evaluated QOL outcomes.  In an abstract presented at the American Society of Clinical Oncology meeting in May 2005, Mileshkin and colleagues investigated the effect of thalidomide plus celecoxib in 66 patients with relapsed multiple myeloma.  The EORTC QLQ-C30 was used to measure QOL.  Overall response to thalidomide (PPR 25 percent) was 42 percent.  Global health on the QLQ-C30 decreased (lower is worse) for 80 percent of participants over the first month of thalidomide treatment.  Among responders, QOL on this sub-scale increased for 29 percent of individuals.  Responders were more likely to have improvement in QOL than non-responders (61 percent vs. 27 percent, p=0.024).  Health-related QOL was also reported in a study of 65 patients with refractory/relapsed myeloma treated with thalidomide only.  The QLQ-C30 was again used as the measurement instrument.  Pain improved and constipation worsened with thalidomide, but otherwise it was difficult to determine the impact of thalidomide on QOL from this report.

2. For patients with relapsed or refractory multiple myeloma, what is the effect of thalidomide compared to standard chemotherapy regimens (e.g., VBMCP (vincristine, carmustine, melphalan, cyclophosphamide, and prednisone) and VAD (vincristine, doxorubicin, dexamethasone)) on adverse effects, tolerability and compliance?

The two most notable adverse effects with thalidomide are peripheral neuropathy and thromboembolism.  Bradycardias, skin toxicity, constipation, and neutropenia are also well described.  Using data from studies of thalidomide only, thalidomide side effects include constipation (3-11 percent grade 3 and 4), neurotoxicity predominantly evident as peripheral neuropathy (1-7 percent grade 3 or 4) and sedation (3-13 percent grade 3 or 4), cardiac insufficiency due to bradycardia (2-6 percent grade 3 or 4), leukopenia (2-31 percent grade 3 and 4), and blood clots (2-10 percent grade 3 or 4).  Side effects are dose dependent as evidenced in studies by Singhal, et al., Hus, et al., and Rajkumar, et al., that escalated thalidomide up to 800 mg with exaggeration of side effects including somnolence, neuropathy, and constipation.

In the 1998 Mineur, et al., randomized trial of VAD vs. VBCMP, toxicities described included neutropenic infections that led to four deaths (VAD 2 and VMBCP 2), corticosteroid effects in two cases both in the VAD arm (pancreatitis and diabetes mellitus for one case, candidal esophagitis for the other), cardiotoxicity after three cycles of VAD, and hematological toxicity after VAD requiring treatment modification. 

In the 2003 Dimopoulos, et al., randomized trial of VAD administered as intravenous bolus injection vs. VDD for patients with previously untreated myeloma, toxicities in the bolus VAD and VDD arms respectively were Grade 2 neutropenia (20 percent vs. 15 percent, p=0.7), Grade 2 thrombocytopenia (10 percent vs. 5 percent, p=0.2), Grade 2 nausea/vomiting (4 percent vs. 5 percent, p=0.8), Grade 1 alopecia (55 percent vs. 37 percent, p<0.001), Grade 2 mucositis (7 percent vs. 15 percent, p=0.3), Grade 2 erythrodysesthesia (2 percent vs. 13 percent, p=0.03), and Grade 2 neurotoxicity (13 percent vs. 15 percent, p=0.9).  Steroid-related side-effects occurred with equal frequency in both arms; Cushingoid features were noted in approximately one-fifth of patients, hyperglycemia in 15 percent of patients treated with bolus VAD bolus and in 12 percent treated with VDD, mood changes in <10 percent of patients in either arm and peptic ulcer disease, hiccups and proximal muscle weakness each occurred in <5 percent of patients.  Infections, which required antibiotics, including neutropenic fever, were noted in 17 percent of patients treated with bolus VAD and 18 percent treated with VDD.  Eleven patients (9 percent) in the bolus VAD and 14 (11 percent) in the VDD arm died within the first 4 months of treatment.  Among the 11 patients treated with bolus VAD, three deaths were due to infections and two were due to heart failure and/or myocardial infarction.  Of the 14 early deaths in the VDD arm, four were due to infections and three were due to heart failure and/or myocardial infarction.

There are no prospective comparative studies between thalidomide and VAD/VBCMP to specifically answer this question.  However, Cavo, et al., recently presented a retrospective review that compared the experience of 200 patients receiving thalidomide plus dexamethasone or VAD as preparative regimens for SCT. Patients were matched on age, disease stage, and Β2 microglobulin.  Grade 3/4 toxicity was presented.  Among patients receiving thalidomide plus dexamethasone, 15 percent developed DVT, 0 percent granulocytopenia, 9 percent constipation, 4 percent infections, 4 percent neuropathy, and 6 percent deaths during treatment.  Among patients receiving VAD, 2 percent developed DVT, 12 percent granulocytopenia, 3 percent constipation, 5 percent infections, 7 percent neuropathy, and 6 percent deaths during treatment.

A more complete review of the differences in administration and tolerability is provided in the previous section.  Compliance data were not identified during this review.

3. What patient or tumor characteristics distinguish treatment responders from non-responders and have potential to be used to target therapy?

Thus far, despite myriad studies reporting predictors of response, little consistent data support the use of any specific tests related to the mechanism of the disease.  TNFα polymorphisms at position -238 of the gene promoter were correlated with response and survival in the one study of the topic, but, as was seen across this group of studies, often a single study was positive but subsequent confirmations were negative.  Two studies of TNFα as a predictor suggested that TNFα correlated with survival, but one did not.  The same studies reported similar findings for IL6.  Studies of Vascular Endothelial Growth Factor (VEGF), Vascular Endothelial Growth Factor (VEGF), and other substances had very few consistent positive findings.  Taken together, these studies suggest that we have a lot to learn about the mechanism of action of thalidomide, that predictors related to angiogenesis are likely to be less helpful, and that cytokine like TNFα and IL-6 play may be more predictive after future study.

Other clinical and demographic factors that predict response include age and beta-2 microglobulin.  These findings do not substantially add to current care, as the findings were fairly consistent with the previously known predictors for myeloma.

Once large randomized trials are available, predictor analyses should be repeated to see if any new patterns or predictors emerge.

Current State of Clinical Use

The National Cancer Institute (NCI) guidelines at list thalidomide as a treatment option within the array of current options, without specifying where in the treatment order it should fall.  The guidelines argue that the choice of first-line and subsequent therapies should be individualized based upon patient age, general health, and patient preference.  A dose of thalidomide is not recommended and the guideline argues that more data are needed until clear recommendations about the role of dexamethasone and enoxaparin can be provided.  The NCCN does not have a guideline for multiple myeloma.

Implications for Future Research

As has been highlighted throughout this review, there is much work to be done on both the clinical and basic science levels.  Clinically, randomized data are needed.  The final results of the ongoing phase III trials are anxiously awaited.  These will guide subsequent directions for therapy.  It is unclear whether a randomized study of VAD versus thalidomide (or thal-dex) will be possible, as the older patient profile ideal for thalidomide may be able to tolerate the standard chemotherapy arm.  If the study is limited to only those who can tolerate VAD then the results may be less applicable across all of the patients for whom thalidomide is the best choice.  A randomized trial using VDD and thalidomide may be more feasible.  Certainly, data produced from these studies will be invaluable to assist with better understanding adverse event profiles and predictors of response.

Much work is ongoing to further elucidate the mechanism of action of thalidomide.  A focus on the cytokine milieu is evolving.  Use of gene array technology to profile multiple myeloma and match this information to thalidomide response is also ongoing.  Thalidomide represents the prototype of an emerging class of drugs, and it is imperative that its efficacy and mechanism of generating tumor response is well understood.  Other immunomodulatory analogs of thalidomide like CC-5013 (Revimid) are also in clinical testing.

Symptoms and QOL is another important future direction for thalidomide research.  How does thalidomide impact pain control, functional status, ability to return to work, and other QOL outcomes?

An invaluable improvement for this body of research would be a strategy of quality reporting and use of similar response criteria such as the Blade criteria.  The quality of reporting was clearly limited among studies in this review.  Similarly, the inconsistency of response criteria and outcomes reported limited comparisons across studies (e.g., variability in reporting and meaning of PPR).  An international standard would greatly improve the accuracy and utility of future systematic reviews on myeloma treatments.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care