The Rise, Fall and Subsequent Triumph of Thalidomide: Lessons Learned in Drug Development

Introduction

Historic perspective

Thalidomide was developed by the Swiss pharmaceutical company CIBA in 1953 and then was introduced by the German pharmaceutical company Chemi Grunenthal in 1956 [Rajkumar, 2004]. Initially marketed as Contergan, thalidomide was prescribed as a nonbarbiturate hypnotic sedative able to produce deep sleep without hangover or risk of dependency. Testing in rodent models failed to establish a median lethal dose and the drug was generally believed to be nontoxic to humans [Lenz, 1988]. In contrast to the extensive testing performed today, in that era formal testing for harmful teratogenic effects was not undertaken. Soon available worldwide, the drug became popular for its anti-emetic effect in pregnant woman suffering with morning sickness. Much of the drug's popularity was due to its wide availability as it was accessible without prescription and also relatively inexpensive. In Germany it quickly became a top-selling sedative and an estimated 14.6 tons were sold in 1960 [Eriksson et al. 2001].

In 1961 Dr William McBride, an Australian obstetrician, and Dr Widukind Lenz, a German pediatrician and geneticist, made independent observations linking thalidomide use in pregnancy to congenital malformations [Lenz, 1962; McBride, 1961]. These findings were confirmed by multiple cases worldwide and thalidomide ultimately was withdrawn from the marketplace. Initial reports identified limb and bone abnormalities including amelia, phocomelia, syndactyly and underdeveloped long bones among other deformities [Lenz, 1962; Mellin and Katzenstein, 1962a, 1962b; McBride, 1961] (Figure 1). Additional observations included atresia of the esophagus, duodenum and anus as well as cardiac abnormalities and aplasia of the gallbladder and appendix [Mellin and Katzenstein, 1962b; McBride, 1961]. The majority of malformations occurred when thalidomide was ingested between 34 and 49 days after the last menstrual period, with even a single dose being associated with increased risk [Lenz, 1988]. Up to 40% of affected infants died within 1 year.

Figure 1.

(a) Single views of upper extremities in a patient exposed to thalidomide in utero. White arrow: fusion at the elbow joint and absence of fingers; Yellow arrow: absence of radius and shortening of ulna. (Reproduced with permission from LearningRadiology.com.) (b) Phocomelia in a baby.

In the United States thalidomide was briefly available as an investigational agent. The drug was endorsed as an anxiolytic but never was approved for marketing. Dr Frances Kelsey, a physician and pharmacologist, was the FDA officer assigned to review the drug application; she denied approval based on a lack of safety data. Principal in Kelsey's decision were emerging data linking thalidomide to neurologic toxicities, including peripheral neuritis [Kelsey, 1988]. For her efforts in preventing thalidomide from being marketed and thus averting a major tragedy in the United States, Dr Kelsey was honored with the President's Award for Distinguished Federal Civilian Service from President John F. Kennedy in 1962 (Figure 2).

Figure 2.

Dr Frances Kelsey is awarded the President's Award for Distinguished Federal Civilian Service from President John F. Kennedy in 1962.

An estimated 10,000 infants were affected worldwide with more uncounted stillborn or miscarried pregnancies [Franks et al. 2004]. A lasting impact of these tragic events has been in the positive change in the drug regulation process. Problems with animal models and inefficiencies in the pharmaceutical agent approval process were rectified by new legislation which revamped the FDA regulatory process, expanded patient informed consent procedures and called for more transparency from drug manufacturers. As means of restitution, committees were organized in Germany to assign payment to those affected most severely. Similar organizations were formed in Britain, Canada, and Sweden. Thalidomide was withdrawn from most commercial markets by 1961 and banned worldwide by the end of the decade.

Pharmacology

Thalidomide, α-(N-phthalimido) glutarimide, is a racemic derivative of glutamic acid consisting of equal amounts of R-(+) and S-(-) enantiomers [Figg et al. 1999] (Figure 3). The enantiomers undergo rapid chiral interconversion under physiological conditions[Eriksson et al. 2001]. The (S)-isomer is an inhibitor of tumor necrosis factor (TNF)-α release from mononuclear blood cells [Wnendt et al. 1996] whereas the (R)-form correlates with sedative effects [Hoglund et al. 1998]. These differences, however, are not clinically relevant because of the rapid interconversion between enantiomers. The thalidomide molecule is subject to spontaneous hydrolysis [Schumacher et al. 1965a]. Metabolism to active metabolites appears to be responsible for the drug's activity. Of interest, these metabolites are species specific, explaining, in part, why the actions of thalidomide are species dependant [Bauer et al. 1998; Schumacher et al. 1965b].

Figure 3.

Thalidomide interconverts between (R)- and (S)-enantiomers with protein binding of 55% and 65%, respectively. The (R)-form is responsible for sedative effects and the (S)-form is responsible for immunomodulatory effects.

No formal dose escalation studies have been conducted; however, single-dose thalidomide studies indicate that bioavailability, absorption, distribution, and elimination are dependent on dose and concurrent illness. The mean time to peak plasma concentrations range from 2.9 to 5.7 hours and peak levels are approximately 1 μg/ml. The mean half-life of elimination is approximately 5–7 hours [Teo et al. 2004; von Moos et al. 2003]. The exact metabolic route of thalidomide is not fully understood. Nonenzymatic hydrolysis leads to the formation of hydrolysis products which are primarily excreted through the urine; in total less than 1% is excreted as unchanged drug [Teo et al. 2004]. There appears to be minimal hepatic metabolism and pharmacokinetics properties in hepatic dysfunction have not been determined. No dosage modifications are needed for patients with renal dysfunction. Single clinical doses administered have ranged from 50 mg to up to 1200 mg; most clinical utility, however, appears to occur in the 50-400 mg per day range.

Mechanism of action

The molecular mechanisms of action for the effects of thalidomide are not fully understood. Thalidomide possesses immunomodulatory, anti-inflammatory, and anti-angiogenic properties related to complex modulation of inflammatory cytokines (Figure 4). One of the primary effects of thalidomide is selective inhibition of TNF-α production in human monocytes, presumably via enhanced degradation of TNF-α mRNA [Moreira et al. 1993; Sampaio et al. 1991]. In addition, thalidomide has been shown to inhibit nuclear factor (NF)-KB activity through a process involving inhibition of I-κB kinase activity [Keifer et al. 2001]. NF-κB is a DNA binding transcription factor involved in regulating the expression of genes that contribute to the immune response, including TNF-α, interleukin (IL)-8, and IL-12 [Barnes and Karin, 1997]. Thalidomide also inhibits IL-6 and IL-12 production [Vacca et al. 2005]. Other anti-inflammatory properties include inactivation of Caspase-1, an activator of pro-inflammatory cytokines including IL-1 [Vacca et al. 2005].

Figure 4.

Proposed mechanisms of action of thalidomide upon tumorous plasma cell and surrounding microenvironment. In addition to direct inhibitory effects on the malignant plasma cell, thalidomide enhances the immune system effector cells, suppresses angiogenesis, inhibits various cytokine mediators and interferes with the bone marrow stromal cell-malignant plasma cell interaction (adhesion molecule interference, yellow circle). NK, natural killer; TNF, tumor necrosis factor; IL-6, interleukin-6; VEGF, vascular endothelial growth factor. The circular ‘plus sign’ indicates stimulation and the circular ‘minus sign’ indicates inhibition. Illustration courtesy of Alessandro Baliani © [2011].

Thalidomide potently stimulates T cells with preferential proliferation in the CD8 ^{T-cell subset [}Haslett et al. 1998]. In addition, thalidomide is effective in T helper cell (Th) immunoregulation by preferentially inducing Th2 cytokine production and inhibiting Th1 cytokine production in peripheral mononuclear cells. This activity shifts the nature of T lymphocyte-dependent immune responses whereby enhancing IL-4 production and inhibiting interferon (IFN)-γ production [McHugh et al. 1995].

In 1971, Dr Judah Folkman proposed that tumors required angiogenesis for survival; hence, strategies designed to interfere with growth of new blood vessels could be effective cancer treatment [Folkman, 1971]. After this discovery, the knowledge regarding the different features of tumor vasculature compared with normal vasculature has proliferated dramatically. Identification of various signaling pathways involved in tumor angiogenesis such as Hypoxia-inducible factor-1, vascular endothelial growth factor, platelet-derived growth factor, and basic fibroblastic growth factor (bFGF) have led to the development of novel therapeutic agents [Cook and Figg, 2010; Eichholz et al. 2010]. The potent anti-angiogenic properties of thalidomide were first described using a rabbit cornea model of FGF-induced neovascularization. D'Amato and colleagues postulated that thalidomide-induced birth defects were due to inhibited blood vessel growth in the developing fetal limb bud and perhaps a similar mechanism may inhibit vasculogenesis in the tumor microenvironment [D'Amato et al. 1994]. Kenyon and colleagues described thalidomide as an inhibitor of angiogenesis in a mouse cornea model, hypothesizing that thalidomide significantly inhibits bFGF and vascular endothelial growth factor (VEGF)-induced corneal neovascularization [Kenyon et al. 1997]. They further characterized the S(-) enantiomer as responsible for the anti-angiogenic properties and identified phthaloylglutamic acid as one metabolite which retains anti-angiogenic activity in the cornea model [Kenyon et al. 1997].

Ito and coworkers proposed that the mechanism of thalidomide anti-angiogenesis, and thus its teratogenic properties, is related to direct thalidomide binding and inhibition of cereblon which is a component of an E3 ubiquitin ligase complex [Ito et al. 2010]. The ligase activity is important in the expression of FGF and therefore implicated in angiogenesis regulation, cell signaling and limb outgrowth [Toffalini and Demoulin, 2010; Hideshima et al. 2000]. In addition to anti-angiogenic properties, thalidomide has several actions against plasma cell myeloma. Thalidomide inhibits tumor cell adhesion and cell-cell communication important for proliferation and survival. Cytokines important for plasma cell growth in the bone marrow microenvironment include IL-6, IL-10, and TNF-α which are inhibited by thalidomide. Lastly, thalidomide may directly prevent myeloma cell growth by inducing free radical-mediated DNA damage [Segarra et al. 2010; Mitsiades et al. 2002; Hideshima et al. 2000]. These varied properties have justified use of this agent for anti-myeloma therapy.

Historic perspective

Thalidomide was developed by the Swiss pharmaceutical company CIBA in 1953 and then was introduced by the German pharmaceutical company Chemi Grunenthal in 1956 [Rajkumar, 2004]. Initially marketed as Contergan, thalidomide was prescribed as a nonbarbiturate hypnotic sedative able to produce deep sleep without hangover or risk of dependency. Testing in rodent models failed to establish a median lethal dose and the drug was generally believed to be nontoxic to humans [Lenz, 1988]. In contrast to the extensive testing performed today, in that era formal testing for harmful teratogenic effects was not undertaken. Soon available worldwide, the drug became popular for its anti-emetic effect in pregnant woman suffering with morning sickness. Much of the drug's popularity was due to its wide availability as it was accessible without prescription and also relatively inexpensive. In Germany it quickly became a top-selling sedative and an estimated 14.6 tons were sold in 1960 [Eriksson et al. 2001].

In 1961 Dr William McBride, an Australian obstetrician, and Dr Widukind Lenz, a German pediatrician and geneticist, made independent observations linking thalidomide use in pregnancy to congenital malformations [Lenz, 1962; McBride, 1961]. These findings were confirmed by multiple cases worldwide and thalidomide ultimately was withdrawn from the marketplace. Initial reports identified limb and bone abnormalities including amelia, phocomelia, syndactyly and underdeveloped long bones among other deformities [Lenz, 1962; Mellin and Katzenstein, 1962a, 1962b; McBride, 1961] (Figure 1). Additional observations included atresia of the esophagus, duodenum and anus as well as cardiac abnormalities and aplasia of the gallbladder and appendix [Mellin and Katzenstein, 1962b; McBride, 1961]. The majority of malformations occurred when thalidomide was ingested between 34 and 49 days after the last menstrual period, with even a single dose being associated with increased risk [Lenz, 1988]. Up to 40% of affected infants died within 1 year.

Figure 1.

(a) Single views of upper extremities in a patient exposed to thalidomide in utero. White arrow: fusion at the elbow joint and absence of fingers; Yellow arrow: absence of radius and shortening of ulna. (Reproduced with permission from LearningRadiology.com.) (b) Phocomelia in a baby.

In the United States thalidomide was briefly available as an investigational agent. The drug was endorsed as an anxiolytic but never was approved for marketing. Dr Frances Kelsey, a physician and pharmacologist, was the FDA officer assigned to review the drug application; she denied approval based on a lack of safety data. Principal in Kelsey's decision were emerging data linking thalidomide to neurologic toxicities, including peripheral neuritis [Kelsey, 1988]. For her efforts in preventing thalidomide from being marketed and thus averting a major tragedy in the United States, Dr Kelsey was honored with the President's Award for Distinguished Federal Civilian Service from President John F. Kennedy in 1962 (Figure 2).

Figure 2.

Dr Frances Kelsey is awarded the President's Award for Distinguished Federal Civilian Service from President John F. Kennedy in 1962.

An estimated 10,000 infants were affected worldwide with more uncounted stillborn or miscarried pregnancies [Franks et al. 2004]. A lasting impact of these tragic events has been in the positive change in the drug regulation process. Problems with animal models and inefficiencies in the pharmaceutical agent approval process were rectified by new legislation which revamped the FDA regulatory process, expanded patient informed consent procedures and called for more transparency from drug manufacturers. As means of restitution, committees were organized in Germany to assign payment to those affected most severely. Similar organizations were formed in Britain, Canada, and Sweden. Thalidomide was withdrawn from most commercial markets by 1961 and banned worldwide by the end of the decade.

Pharmacology

Thalidomide, α-(N-phthalimido) glutarimide, is a racemic derivative of glutamic acid consisting of equal amounts of R-(+) and S-(-) enantiomers [Figg et al. 1999] (Figure 3). The enantiomers undergo rapid chiral interconversion under physiological conditions[Eriksson et al. 2001]. The (S)-isomer is an inhibitor of tumor necrosis factor (TNF)-α release from mononuclear blood cells [Wnendt et al. 1996] whereas the (R)-form correlates with sedative effects [Hoglund et al. 1998]. These differences, however, are not clinically relevant because of the rapid interconversion between enantiomers. The thalidomide molecule is subject to spontaneous hydrolysis [Schumacher et al. 1965a]. Metabolism to active metabolites appears to be responsible for the drug's activity. Of interest, these metabolites are species specific, explaining, in part, why the actions of thalidomide are species dependant [Bauer et al. 1998; Schumacher et al. 1965b].

Figure 3.

Thalidomide interconverts between (R)- and (S)-enantiomers with protein binding of 55% and 65%, respectively. The (R)-form is responsible for sedative effects and the (S)-form is responsible for immunomodulatory effects.

No formal dose escalation studies have been conducted; however, single-dose thalidomide studies indicate that bioavailability, absorption, distribution, and elimination are dependent on dose and concurrent illness. The mean time to peak plasma concentrations range from 2.9 to 5.7 hours and peak levels are approximately 1 μg/ml. The mean half-life of elimination is approximately 5–7 hours [Teo et al. 2004; von Moos et al. 2003]. The exact metabolic route of thalidomide is not fully understood. Nonenzymatic hydrolysis leads to the formation of hydrolysis products which are primarily excreted through the urine; in total less than 1% is excreted as unchanged drug [Teo et al. 2004]. There appears to be minimal hepatic metabolism and pharmacokinetics properties in hepatic dysfunction have not been determined. No dosage modifications are needed for patients with renal dysfunction. Single clinical doses administered have ranged from 50 mg to up to 1200 mg; most clinical utility, however, appears to occur in the 50-400 mg per day range.

Mechanism of action

The molecular mechanisms of action for the effects of thalidomide are not fully understood. Thalidomide possesses immunomodulatory, anti-inflammatory, and anti-angiogenic properties related to complex modulation of inflammatory cytokines (Figure 4). One of the primary effects of thalidomide is selective inhibition of TNF-α production in human monocytes, presumably via enhanced degradation of TNF-α mRNA [Moreira et al. 1993; Sampaio et al. 1991]. In addition, thalidomide has been shown to inhibit nuclear factor (NF)-KB activity through a process involving inhibition of I-κB kinase activity [Keifer et al. 2001]. NF-κB is a DNA binding transcription factor involved in regulating the expression of genes that contribute to the immune response, including TNF-α, interleukin (IL)-8, and IL-12 [Barnes and Karin, 1997]. Thalidomide also inhibits IL-6 and IL-12 production [Vacca et al. 2005]. Other anti-inflammatory properties include inactivation of Caspase-1, an activator of pro-inflammatory cytokines including IL-1 [Vacca et al. 2005].

Figure 4.

Proposed mechanisms of action of thalidomide upon tumorous plasma cell and surrounding microenvironment. In addition to direct inhibitory effects on the malignant plasma cell, thalidomide enhances the immune system effector cells, suppresses angiogenesis, inhibits various cytokine mediators and interferes with the bone marrow stromal cell-malignant plasma cell interaction (adhesion molecule interference, yellow circle). NK, natural killer; TNF, tumor necrosis factor; IL-6, interleukin-6; VEGF, vascular endothelial growth factor. The circular ‘plus sign’ indicates stimulation and the circular ‘minus sign’ indicates inhibition. Illustration courtesy of Alessandro Baliani © [2011].

Thalidomide potently stimulates T cells with preferential proliferation in the CD8 ^{T-cell subset [}Haslett et al. 1998]. In addition, thalidomide is effective in T helper cell (Th) immunoregulation by preferentially inducing Th2 cytokine production and inhibiting Th1 cytokine production in peripheral mononuclear cells. This activity shifts the nature of T lymphocyte-dependent immune responses whereby enhancing IL-4 production and inhibiting interferon (IFN)-γ production [McHugh et al. 1995].

In 1971, Dr Judah Folkman proposed that tumors required angiogenesis for survival; hence, strategies designed to interfere with growth of new blood vessels could be effective cancer treatment [Folkman, 1971]. After this discovery, the knowledge regarding the different features of tumor vasculature compared with normal vasculature has proliferated dramatically. Identification of various signaling pathways involved in tumor angiogenesis such as Hypoxia-inducible factor-1, vascular endothelial growth factor, platelet-derived growth factor, and basic fibroblastic growth factor (bFGF) have led to the development of novel therapeutic agents [Cook and Figg, 2010; Eichholz et al. 2010]. The potent anti-angiogenic properties of thalidomide were first described using a rabbit cornea model of FGF-induced neovascularization. D'Amato and colleagues postulated that thalidomide-induced birth defects were due to inhibited blood vessel growth in the developing fetal limb bud and perhaps a similar mechanism may inhibit vasculogenesis in the tumor microenvironment [D'Amato et al. 1994]. Kenyon and colleagues described thalidomide as an inhibitor of angiogenesis in a mouse cornea model, hypothesizing that thalidomide significantly inhibits bFGF and vascular endothelial growth factor (VEGF)-induced corneal neovascularization [Kenyon et al. 1997]. They further characterized the S(-) enantiomer as responsible for the anti-angiogenic properties and identified phthaloylglutamic acid as one metabolite which retains anti-angiogenic activity in the cornea model [Kenyon et al. 1997].

Ito and coworkers proposed that the mechanism of thalidomide anti-angiogenesis, and thus its teratogenic properties, is related to direct thalidomide binding and inhibition of cereblon which is a component of an E3 ubiquitin ligase complex [Ito et al. 2010]. The ligase activity is important in the expression of FGF and therefore implicated in angiogenesis regulation, cell signaling and limb outgrowth [Toffalini and Demoulin, 2010; Hideshima et al. 2000]. In addition to anti-angiogenic properties, thalidomide has several actions against plasma cell myeloma. Thalidomide inhibits tumor cell adhesion and cell-cell communication important for proliferation and survival. Cytokines important for plasma cell growth in the bone marrow microenvironment include IL-6, IL-10, and TNF-α which are inhibited by thalidomide. Lastly, thalidomide may directly prevent myeloma cell growth by inducing free radical-mediated DNA damage [Segarra et al. 2010; Mitsiades et al. 2002; Hideshima et al. 2000]. These varied properties have justified use of this agent for anti-myeloma therapy.

Toxicity

In addition to the well-known teratogenicity, adverse effects of thalidomide include peripheral neuropathies and sedation occurring in up to 30% and 50% of patients, respectively [Chaudhry et al. 2008, 2002]. Using electrophysiologic studies, thalidomide-associated neuropathies tend to manifest as axonal sensory injuries more so than motor. Polyneuropathies often present as painful paresthesias or numbness; such correlate with total cumulative exposure to thalidomide [Chaudhry et al. 2008]. Johnson and colleagues reported the identification of a series of single nucleotide polymorphisms (SNPs) which were associated with thalidomide-related neuropathy. These findings suggest that genetic factors may mediate the risk for developing neuropathies [Johnson et al. 2011].

Another serious reported complication is toxic epidermal necrolysis occurring in 3-4% of patients [Colagrande et al. 2009; Rajkumar et al. 2000]. Other more common but less severe toxicities include rash, constipation, tremor, and interstitial pneumonia. Metabolic abnormalities include hypothyroidism, hypocalcemia, and hyperkalemia, particularly in myeloma patients with pre-existing kidney dysfunction [Pretz and Medeiros, 2009; de Savary et al. 2004; Fakhouri et al. 2004; Ghobrial and Rajkumar, 2003; Harris et al. 2003; Badros et al. 2002] (see Table 1 for reported incidence of adverse events).

Thrombotic complications

Plasma cell myeloma is associated with increased thrombotic events. Proposed mechanisms are coagulation factor abnormalities including higher concentrations of von Willebrand factor and factor VIII, activated protein C resistance, and hypofibrinolysis [Auwerda et al. 2007; Elice et al. 2006; van Marion et al. 2005; Zangari et al. 2002a]. Thalidomide, particularly as part of combination therapy in plasma cell myeloma, further increases risk for thrombotic events [Carrier et al. 2011; Zangari et al. 2009]. This hazard is further increased with the addition of more traditional cytotoxic agents, particularly anthracycline chemotherapy [Zangari et al. 2002b, 2001]. Both venous and arterial thrombotic events have been described [Libourel et al. 2010; Martin and Vij, 2009; Scarpace et al. 2005]. While thalidomide monotherapy minimally increases the likelihood of thrombosis, in large randomized trials evaluating thalidomide plus dexamethasone the reported rate of deep venous thrombosis is increased to 17-19% [Rajkumar et al. 2008, 2006].

A suggested mechanism by which thalidomide may promote thrombosis is modification of thrombomodulin levels in the first month of therapy [Corso et al. 2004]. Thrombomodulin complexes with thrombin in the anticoagulant interactions with protein C; the transient decrease in thrombomodulin plasma concentration may promote a thrombophilic state. In addition, thalidomide-derived immunomodulatory drugs downregulate PU.1 which is a transcription factor involved in granulocyte differentiation. This action leads to promyelocyte accumulation and resultant high levels of cathepsin G, a platelet aggregation agonist stored in promyelocyte granules, which may increase the risk for thrombotic events [Pal et al. 2010].

Coagulation prevention strategies include use of aspirin, low or therapeutic doses of warfarin, and low molecular weight heparin (LMWH). An Italian randomized, phase III trial evaluated prophylactic use of aspirin (100 mg/day), warfarin (1.25 mg/day), or LMWH (40 mg/day) in previously untreated plasma cell myeloma patients who received regimens containing thalidomide. Serious thrombotic event rates in the aspirin, warfarin, and LMWH group were 6.4%, 8.2%, and 5%, respectively [Palumbo et al. 2011].

Thalidomide revival

Despite being banned from commercial markets, thalidomide remained available primarily in developing countries. Clinical interest in thalidomide resurfaced after Dr Jacob Sheskin first prescribed thalidomide in 1964 as a sedative in a patient with erythema nodosum leprosum (ENL). Sheskin was at the Hadassah University Hospital and Hansen Leper Hospital in Jerusalem. He observed an unexpected and dramatic resolution of the patient's lepra skin eruption within 48 hours of administration. This activity later was reproduced and reported by Sheskin in 1965 [Sheskin, 1965a, 1965b]. In 1971 the World Health Organization published the results of a randomized trial that confirmed the efficacy of thalidomide in ENL [Silverman, 2002; Iyer et al. 1971].

After the reported success in treating ENL, thalidomide was supported on an experimental basis by the FDA in the 1970s primarily in other dermatologic and inflammatory conditions. The drug also became available through various compassionate use programs for its sedative properties. Thalidomide use was subsequently researched in a number of inflammatory conditions including sarcoidosis, cutaneous lupus, Behçet's syndrome, inflammatory bowel diseases, ankylosing spondylitis, and rheumatoid arthritis [Lazzerini et al. 2007; Lehman et al. 2004; Wei et al. 2003; Bauditz et al. 2002; Vasiliauskas et al. 1999; Hamuryudan et al. 1998].

Thalidomide also was evaluated in refractory or high-risk chronic graft-versus- host disease (GVHD). Using escalating doses ranging from 200-1600 mg per day, response rates varied from 20% to 79% [Arora et al. 2001; Parker et al. 1995; Vogelsang et al. 1992]. In one trial in which 59 subjects were randomized to receive either thalidomide or placebo as GVHD prophylaxis, however, the thalidomide group was associated with higher incidence of GVHD and worse survival [Chao et al. 1996]. While thalidomide has been shown to be an effective agent for the therapy of chronic GVHD, prophylactic use resulted in a paradoxical outcome with a higher incidence of chronic GVHD and a lower overall survival. The authors concluded that early thalidomide use resulted in a shift in the balance between GVHD and induction of tolerance.

A novel application of thalidomide's anti-emetic, sedative, and analgesic properties has been in the palliative care setting, particularly cancer cachexia and HIV wasting syndrome [von Moos et al. 2003; Peuckmann et al. 2000]. Furthering the clinical revival of this agent, thalidomide emerged as an effective drug to treat complications of HIV disease such as the HIV wasting syndrome, Kaposi's sarcoma, and aphthous ulcers [Little et al. 2000; Jacobson et al. 1997; Reyes-Teran et al. 1996].

These findings led to continued thalidomide availability in controlled clinical trials and on a compassionate use basis. By this time ‘drug buyer’ clubs were being established in major cities. These groups were channeling sometimes impure forms of thalidomide through the growing black market in South America, most notably Brazil, where thalidomide had remained largely available since its inception. The primarily unregulated distribution of thalidomide pressed the FDA into soliciting drug applications from several companies that were investigating the drug. Celgene Corporation gained approval to formally market the drug. In 1998 the FDA approved thalidomide for the treatment of ENL, an action that facilitated the continued availability of the drug for other conditions [Rajkumar, 2004] (Figure 5).

Figure 5.

Timeline of events in thalidomide and immunomodulatory drug (IMiD) development.

Anticancer properties

The potential benefit of thalidomide in cancer treatment first was hypothesized in the 1960s shortly after the teratogenic properties were reported. Investigators believed that the effects responsible for teratogenicity also represented potential mechanisms for cancer treatment. For a variety of reasons, the early clinical trials exploring thalidomide as an antitumor agent, however, failed to show significant activity [Rajkumar and Kyle, 2001; Grabstald and Golbey, 1965; Olson et al. 1965]. More recently, promising results have been noted in castrate-resistant prostate cancer. A phase II trial of 60 patients using thalidomide as an alternative anti-angiogenic agent added to standard chemotherapy and bevacizumab demonstrated a prostate-specific antigen (PSA) decline of at least 50% in 90% of patients [Ning et al. 2010]. Mixed results have been reported in other solid tumors where angiogenesis is implicated including renal cell, hepatocellular, and ovarian cancers [Ch'ang et al. 2011; Hurteau et al. 2010; Motzer et al. 2002]. Currently multiple phase II and phase III trials are exploring thalidomide as an adjunctive agent in various solid tumors. To minimize the risk of teratogenicity, thalidomide access is restricted through the System for Thalidomide Education and Prescribing Safety (S.T.E.P.S.) program. All prescribing physicians, pharmacies, and patients must register with the program and comply with safety guidelines in order to control the prescribing, dispensing, and use of thalidomide [Zeldis et al. 1999]. In addition to minimizing the potential risk for fetal harm associated with thalidomide therapy, this strategy provided a model for future cases in which drug therapy offers compelling benefits but poses profound risks unless its distribution is carefully controlled.

Hematologic malignancies

Plasma cell myeloma

The discovery of potent anti-angiogenic properties renewed interest in the use of thalidomide as an antitumor agent. Plasma cell myeloma is associated with increased bone marrow microvascular density and elevated serum VEGF concentration. These features led to the recognition that angiogenesis is an important pathogenic mechanism in this disorder and prompted clinical use [Kotla et al. 2009]. The investigators at the University of Arkansas first reported a 32% overall response rate with single-agent thalidomide in a series of 84 refractory plasma cell myeloma patients [Singhal et al. 1999]. A follow-up study of escalated doses resulted in a reported overall survival of 48% at 2 years [Barlogie et al. 2001]. Updated results at a median follow up of 9.2 years suggested better overall survival in patients receiving higher cumulative doses of thalidomide, suggesting a dose-dependent effect [van Rhee et al. 2008]. Additional prospective studies using single-agent thalidomide further confirmed the benefit of thalidomide monotherapy [van Rhee et al. 2008; Richardson et al. 2004; Kumar et al. 2003; Mileshkin et al. 2003]. Systematic reviews of phase II trial data demonstrate response rates of 30% with overall median survival of 14 months in refractory disease patients treated with thalidomide monotherapy [Prince et al. 2007; Glasmacher et al. 2006]. Subsequently, the thalidomide and dexamethasone combination therapy was found to enhance response rates [Rajkumar, 2011; Murakami et al. 2007; Anagnostopoulos et al. 2003]. A literature review by von Lilienfeld-Toal and colleagues of 12 studies in 451 patients of the thalidomide and dexamethasone combination therapy demonstrated an increased overall response rate of 46% in relapsed/refractory plasma cell myeloma patients, but with a toxicity rate comparable to thalidomide monotherapy [von Lilienfeld-Toal et al. 2008].

Thalidomide combination therapy for newly diagnosed plasma cell myeloma has resulted in an improved overall response rate as well as a lengthening of time to disease progression; such combinations have become the standard [Hussein et al. 2009; Weber et al. 2003; Rajkumar et al. 2002]. Complete response rates and 5-year event-free survival rates of 62% and 56%, respectively, were noted in a study incorporating thalidomide into high-dose therapy and autologous hematopoietic cell transplant [Barlogie et al. 2006]. Two randomized controlled trials have compared the thalidomide plus dexamethasone combination versus dexamethasone alone as induction therapy in newly diagnosed untreated patients. Combination treatment resulted in significantly better overall response rates of 63% and extended time to progression from 6.5 to 22.6 months [Rajkumar et al. 2008, 2006] (Table 2).

The addition of bortezomib to thalidomide and dexamethasone induction treatment also improves complete response rates. In a phase III trial evaluating 474 patients bortezomib, thalidomide and dexamethasone (VTD) induction therapy yielded a complete or near complete response rate of 31% [Cavo et al. 2010]. Similarly, bortezomib, thalidomide, and prednisone use provided a median progression-free survival of 25 months and 65% overall 3-year survival [Mateos et al. 2010].

Among elderly patients or those who are not candidates for hematopoietic cell transplantation, thalidomide in combination with melphalan and prednisone significantly improves event-free and overall survival. Median progression-free survival rates range from 15 to 28 months and overall survival ranges from 40 to 52 months [Kapoor et al. 2011; Waage et al. 2010; Wijermans et al. 2010; Hulin et al. 2009; Facon et al. 2007]. These patients typically were over 65-75 years of age.

Recent studies have evaluated the role for thalidomide maintenance therapy but the data have shown conflicting results as the clinical benefit is unclear yet the toxic effects generally are increased. One study evaluated thalidomide combined with dexamethasone maintenance. The two year progression-free survival improved from 32% to 63% and 2-year overall survival improved from 68% to 84% [Offidani et al. 2009]. In another study, however, thalidomide plus IFN maintenance therapy improved progression-free survival from 13.2 to 27.7 months but without a significant difference in overall survival [Ludwig et al. 2010].

Thalidomide maintenance treatment also has been studied in the post-transplant setting. One trial randomized 556 patients to receive vincristine, doxorubicin and dexamethasone induction (VAD) followed by interferon maintenance versus thalidomide, doxorubicin, and dexamethasone induction (TAD) followed by thalidomide post-transplant. The thalidomide group had an improved progression-free survival from 25 to 34 months; however, there was no statistically significant difference in overall survival [Lokhorst et al. 2010]. A second trial assigned 269 patients to receive thalidomide plus prednisolone, versus prednisolone alone, after autologous transplant. At 3-year follow up, patients assigned to the thalidomide group had improved progression-free survival (42% versus 23%) and overall survival (86% versus 75%) [Spencer et al. 2009]. Thalidomide was continued for 12 months duration and therefore can be considered as consolidation as opposed to maintenance effect. Neurologic toxicities led to discontinuation of thalidomide in up to 60% of patients. Hence, it is unclear which population benefits the most from maintenance therapy.

Thalidomide therapy also has been evaluated in combination with conventional chemotherapeutic agents including anthracyclines and alkylators such as cyclophosphamide. Despite reports of higher response rates in some studies, however, the lack of convincing survival benefit along with higher toxic effects including thrombotic and infectious complications make these regimens less attractive [Ciolli et al. 2008; Offidani et al. 2007; Kyriakou et al. 2005]. In May of 2006, thalidomide was approved for the treatment of newly diagnosed plasma cell myeloma in combination with dexamethasone.

Myelofibrosis

Thalidomide has been used as an effective agent in the treatment of myelofibrosis. Preliminary data suggest improvement in cytopenias and splenomegaly. In a study of 36 patients with median follow-up of 25 months, 28% of patients had a durable response to thalidomide [Thomas et al. 2006; Marchetti et al. 2004; Mesa et al. 2004].

Myelodysplastic syndromes

Based on anticytokine immunomodulatory and anti-angiogenic properties, thalidomide has been used in the treatment of myelodysplastic syndromes. Erythroid response in transfusion-dependent patients has been good; however, patients poorly tolerate effective (higher) doses. Thalidomide doses used in clinical trials were in the range 200-1000 mg per day. Such dosing variability may explain the inconsistent study results as well as the various toxicities in this setting [Moreno-Aspitia et al. 2006; Bouscary et al. 2005; Musto, 2004].

Waldenstrom macroglobulinemia

Phase II studies have shown excellent activity of thalidomide alone or in combination with other agents as therapy in Waldenstrom macroglobulinemia [Dimopoulos et al. 2009; Dimopoulos et al. 2001]. Treon and coworkers reported a 72% overall response rate in primarily untreated patients using thalidomide and rituximab for symptomatic disease [Treon et al. 2008]. Median serum IgM significantly decreased from 3670 to 1590 mg/dl and median hematocrit rose from 33.0% to 37.6% (p = 0.004) at time of best response; for responders, median time to progression was 38 months. Thalidomide antitumor activity has been modest in non-Hodgkin lymphoma, acute myeloid leukemia, and chronic lymphocytic leukemia [Kay et al. 2009; Smith et al. 2008; Barr et al. 2007; Laurenti et al. 2007].

Alternative uses

The vascular effects of thalidomide have led to further innovative uses. Thalidomide has been used successfully to control recurrent, obscure gastrointestinal bleeding or refractory bleeding from angiodysplasia [Almadi et al. 2009; Szilagyi and Ghali, 2006; Bauditz et al. 2004]. These results have extended to patients with von Willebrand's disease [Nomikou et al. 2009]. Lebrin and colleagues reported the successful use of thalidomide therapy to reduce the frequency of epistaxis in a group of hereditary hemorrhagic telangiectasia (HHT) patients [Lebrin et al. 2010]. They postulated that thalidomide bypasses an endothelial cell regulatory pathway, thereby stimulating vessel maturation and preventing vessel wall malformations [Akhurst, 2010; Lebrin et al. 2010].

Immunomodulatory derivatives

Thalidomide analogues, termed immunomodulatory drugs (IMiDs), are a class of agents developed to take advantage of the anti-inflammatory and antitumor properties of thalidomide while maintaining a more favorable side effect profile (Table 3). Lenalidomide (Revlimid, CC-5013) and pomalidomide (CC-4047) are two such analogues [Begna et al. 2011; Lacy et al. 2010]. Lenalidomide is approved as an effective treatment in relapsed plasma cell myeloma and myelodysplastic syndromes associated with the deletion 5q abnormality [Rajkumar et al. 2010; List et al. 2006]. Two randomized clinical trials compared lenalidomide (initial dose 25 mg per day) plus dexamethasone versus dexamethasone therapy alone in myeloma patients who had received at least one prior treatment. Both studies showed improvement in the primary endpoint of time to progression [Dimopoulos et al. 2007; Weber et al. 2007]. Lenalidomide also was evaluated in a study of 148 patients with transfusion-dependant anemia in intermediate-1 risk myelodysplastic syndrome and deletion 5q abnormality. Lenalidomide was shown to decrease transfusion requirements as 67% of patients achieved transfusion independence. A daily 10 mg dose was used and median time to response was 4.6 weeks [List et al. 2006].

Pomalidomide is a potent investigational thalidomide analogue. Encouraging early phase clinical trials have shown activity in relapsed plasma cell myeloma and myelofibrosis. Typical doses range from 2 to 5 mg per day and generally are well-tolerated [Lacy and Rajkumar, 2010].

Myelosuppression is the dose-limiting toxicity for both lenalidomide and pomalidomide. Peripheral neuropathies are generally less severe as compared with thalidomide. The incidence of venous thromboembolism is approximately 7-8% [Dimopoulos et al. 2007; Weber et al. 2007]. Both lenalidomide and pomalidomide are listed as Category X drugs in pregnancy. Other adverse toxicities include rash, constipation, and fatigue [Richardson and Anderson, 2002]. Ongoing development and research of thalidomide analogues, in addition to further identification of thalidomide metabolites and their actions, should lead to promising new treatments in hematologic malignancies.

Conclusions

It is highly likely that no one could have predicted such a clinical resurgence for thalidomide nearly four decades after the recognition of teratogenicity. The thalidomide saga is directly responsible for many changes in the drug development proceedings by highlighting many of the shortfalls in this approval process. In the United States new laws were passed that implemented strict regulations, restructured the FDA, and protected patients and consumers by expanding the informed consent process. At present, manufacturers must prove both effectiveness and safety of a new agent before it is authorized to be tested on humans. Another result was that all drugs must undergo formal testing for teratogenic effects; as a result, the teratology field has developed rapidly.

Beyond the social and legal implications, other lessons have been learned from the scientific advancements made after Dr Sheskin first used thalidomide in ENL. As the complex mechanisms of action and associated toxicities were better understood, thalidomide was applied to a variety of disorders. This previously ill-fated, teratogenic sedative possessed unusual anti-inflammatory and anti-angiogenic properties. Gaining insight into these mechanisms allowed clinicians to utilize thalidomide as an antitumor agent. Under the careful supervision of the S.T.E.P.S. program, continued research and development of thalidomide analogues represent additional potential for new treatments. The new legacy for thalidomide will be judged by these significant advancements. From past tragedies to present triumphs, we now look to a future full of immeasurable promise.