`
`55. Feinstein, A. R. Clinical Epidemiology: The Architecture of
`Clinical Research (WB Saunders, Philadelphia, 1985).
`56. Hennekens, C. H. & Buring, J. E. Epidemiology in
`Medicine (Little, Brown and Company, Boston, 1987).
`57. Freiman, J. A., Chalmers, T. C., Smith, H. Jr & Kuebler, R. R.
`The importance of β, the type II error and sample size in
`the design and interpretation of the randomized control
`trial. Survey of 71 ‘negative’ trials. N. Engl. J. Med. 299,
`690–694 (1978).
`58. Ransohoff, D. F. Discovery-based research and fishing.
`Gastroenterology 125, 290 (2003).
`
`Competing interests statement
`The author declares that he has no competing financial interests.
`
`Online links
`
`DATABASES
`The following terms in this article are linked online to:
`Cancer.gov: http://cancer.gov/
`breast cancer | colorectal cancer | ovarian cancer |
`pancreatic cancer
`
`Acknowledgements
`Thanks to many colleagues at the National Cancer Institute, The
`University of North Carolina at Chapel Hill and elsewhere for
`reviewing and commenting on earlier versions of the manuscript.
`
`FURTHER INFORMATION
`Research about cancer molecular markers:
`http://www3.cancer.gov/prevention/cbrg/edrn
`Access to this interactive links box is free online.
`
`T I M E L I N E
`
`The evolution of thalidomide
`and its IMiD derivatives as
`anticancer agents
`
`J. Blake Bartlett, Keith Dredge and Angus G. Dalgleish
`
`Thalidomide was originally used to treat
`morning sickness, but was banned in the
`1960s for causing serious congenital birth
`defects. Remarkably, thalidomide was
`subsequently discovered to have anti-
`inflammatory and anti-angiogenic properties,
`and was identified as an effective treatment
`for multiple myeloma. A series of
`immunomodulatory drugs — created by
`chemical modification of thalidomide — have
`been developed to overcome the original
`devastating side effects. Their powerful
`anticancer properties mean that these drugs
`are now emerging from thalidomide’s
`shadow as useful anticancer agents.
`
`Thalidomide (α-(N-phthalimido)glutarimide)
`— a synthetic glutamic-acid derivative — was
`manufactured and marketed by the German
`pharmaceutical company Chemie Grunenthal
`during the mid-1950s (BOX 1; TIMELINE). It is a
`non-barbiturate drug with sedative and anti-
`emetic activity and was found to be useful
`because of an apparent lack of toxicity in
`human volunteers. These properties led to it
`being marketed as the safest available sedative
`of its time. It rapidly became popular as a drug
`to counter the effects of morning sickness in
`Europe, Australia, Asia and South America,
`although it did not receive Food and Drug
`Administration (FDA) approval in the United
`States because of concerns about neuropathy
`— tingling hands and feet after long-term
`administration — that were associated with its
`
`use. It was withdrawn from the other markets
`in early 1961 after two clinicians — William
`McBride in Australia and Widukind Lenz
`in Germany — reported independently that
`thalidomide use was associated with birth
`defects1,2. A report associating thalidomide use
`with neuropathies was also reported at around
`this time3. Unfortunately, this withdrawal was
`too late to prevent the birth of between 8,000
`and 12,000 babies with severe developmental
`deformities, which include the stunted-
`limb development that is characteristic of
`‘thalidomide babies’.
`In 1965, following a serendipitous discov-
`ery by Israeli dermatologist Jacob Sheskin, it
`was reported that thalidomide was remark-
`ably effective at improving lesions, fever and
`night sweats in patients with erythema
`nodosum leprosum (ENL) — a potentially
`life-threatening inflammatory complication
`lepromatous leprosy 4. After finding
`of
`thalidomide in the clinic and remembering
`that it was a sedative, Sheskin administered it
`to a patient who was having trouble sleeping
`and — remarkably — the next morning the
`patient’s inflammation was significantly
`reduced. This discovery was investigated in a
`study that was coordinated by the World
`Health Organization in thousands of men
`who had ENL and showed that a vast major-
`ity had complete remission within a couple of
`weeks of starting thalidomide treatment5.
`This was the catalyst that eventually led to the
`use of thalidomide as an immunomodulatory
`
`and anti-inflammatory drug 6–8. However,
`thalidomide was only given FDA approval for
`the treatment of acute ENL in 1998, after fur-
`ther investigations found an immunological
`basis for this effect9. Even then, its use was
`limited by very strict guidelines.
`It is now clear that despite its teratogenicity
`(BOX 1), which caused the birth defects, thalido-
`mide is useful in treating several clinical condi-
`tions for which there are few or no alternative
`treatment options. An early appreciation of the
`immunosuppressive properties of thalidomide
`in several animal models led to its use in vari-
`ous conditions that are associated with
`immune activation. Initial, but mainly anecdo-
`tal, reports from the early 1980s onwards indi-
`cated that thalidomide was effective in the
`treatment of several autoimmune disorders.
`However, because the use of thalidomide was
`necessarily restricted, large-scale studies were
`not undertaken until much later. Instead, the
`results of various small uncontrolled studies
`were published and these seemed to demon-
`strate the efficacy of thalidomide in the treat-
`ment of patients with autoimmune disorders
`such as rheumatoid arthritis10, cutaneous
`lesions of systemic lupus erythematosus and
`Behcet’s disease11,12. The immunosuppressive
`properties of thalidomide also led to its use in
`the treatment of chronic graft-versus-host dis-
`ease associated with allogeneic bone-marrow
`transplantation13–15.
`As thalidomide initially seemed to show
`promise for the treatment of these conditions,
`it was quickly used in further studies in small
`cohorts of patients with various untreatable
`ailments. From these investigations, it has
`become apparent that thalidomide is not
`merely an immunosuppressant, but that it has
`other clinically useful properties. Each new
`property that has been discovered has led to
`thalidomide being used in different spectra of
`disease. As a result, thalidomide is now an
`option for a diverse range of clinical applica-
`tions and is again a profitable drug, with sales
`that amount to over $200 million per year in
`the United States and rising.
`
`Mechanisms of thalidomide action
`Thalidomide inhibits monocyte-derived
`TNF-α. The key finding that explained, at
`least in part, the potent anti-inflammatory
`activity of thalidomide came in 1991, when it
`was discovered that thalidomide inhibited
`the synthesis of tumour-necrosis factor-α
`(TNF-α) by activated monocytes16 — the
`mRNA becomes less stable. TNF-α is a pro-
`inflammatory cytokine that is an important
`regulator of the inflammatory cascade and is
`a useful therapeutic target in inflammatory
`disease, particularly if activated monocytes
`
`314 | APRIL 2004 | VOLUME 4
`
`www.nature.com/reviews/cancer
`
`Apotex Ex. 1011, p. 1
`
`
`
`have an important role in pathogenesis (FIG. 1).
`There is also evidence that thalidomide might
`inhibit TNF-α that is derived from other cel-
`lular sources that have been activated by
`inflammatory stimuli, such as microglia and
`Langerhans cells17,18. The fact that thalidomide
`inhibits TNF-α explains its therapeutic effect
`in patients with ENL, as they have extremely
`high levels of TNF-α in their blood and in
`dermatological lesions. Most importantly, this
`finding led to the initial use of thalidomide in
`several, small open-label studies in which
`increased TNF-α production is associated
`with disease19, such as AIDS-related Kaposi’s
`sarcoma and cachexia, rheumatological dis-
`ease, Crohn’s disease, cerebral malaria, multi-
`ple sclerosis, psoriasis, sepsis, tuberculosis and
`some cancers6,20.
`
`Thalidomide inhibits angiogenesis. The next
`crucial discovery that uncovered the clinical
`potential of thalidomide came in 1994, when
`thalidomide was found to inhibit angiogenesis
`— the formation of new blood vessels, which is
`a crucial process in the growth and metastasis
`of solid tumours. Judah Folkman was one of
`the first researchers to associate angiogenesis
`with tumour development in the early 1970s
`and it was from his laboratory that the
`inhibitory effect of thalidomide on angiogene-
`sis was demonstrated. He believed that the
`classical congenital defects that are caused by
`thalidomide treatment — abnormal limb
`development — were caused by the inhibition
`of blood-vessel growth in the developing fetal
`limb bud. Using a rabbit cornea micropocket
`assay, it was demonstrated that thalido-
`mide could, in fact, inhibit basic fibroblast
`growth factor (bFGF)-induced angiogenesis21.
`However, despite this study, it is worth noting
`that the link between the teratogenic properties
`of thalidomide and its anti-angiogenic activity
`
`P E R S P E C T I V E S
`
`Box 1 | The chemistry of thalidomide
`
`OO
`
`N
`
`H
`
`O
`
`NH
`
`HN
`
`O
`
`O
`
`O O
`
`N
`
`H
`
`O
`
`Thalidomide consists of a racemic mixture of S(–) and R(+) enantiomers (isoforms) —
`molecules with identical chemical composition that are mirror images of one another and that
`can not be superimposed (see figure). In nature, compounds often exist as enantiomers,
`although generally only one form is physiologically useful. In the case of thalidomide, there
`seems to be a segregation of activities between these different forms. Of particular interest in
`terms of potential clinical application has been the association of the S(–) enantiomer with the
`teratogenic effects of thalidomide, which are responsible for the abnormalities that occur during
`embryonic development, whereas the R(+) isoform seems to be responsible for sedation. This
`indicated that purification of the R(+) isoform, although less effective as a TNF-α inhibitor and
`anti-angiogenic agent, could provide a safer drug and could have prevented the earlier tragic
`events. However, the rapid interconversion of the two isomers under physiological conditions, as
`demonstrated in humans in vivo, proved that
`purification is not a feasible option.
`Furthermore, although the two isoforms of
`thalidomide were initially shown to have
`different teratogenic effects in rodent
`models, differences were not observed in the
`New Zealand rabbit model that is
`traditionally used to measure drug toxicity.
`
`R-(+)-Thalidomide
`
`S-(–)-Thalidomide
`
`remains unproven. Other groups have more
`recently demonstrated that thalidomide
`mediates inhibitory effects on mesenchymal
`proliferation in the limb bud22 and induces
`embryonic oxidative stress23. Irrespective of
`these findings, the anti-angiogenic properties
`of thalidomide sparked a huge interest in its
`use for the treatment of cancer.
`
`T-cell co-stimulatory activity of thalidomide.
`Yet another activity of thalidomide was
`demonstrated in 1998, when it was shown
`that thalidomide is able to co-stimulate
`T cells that have been partially activated by
`the T-cell receptor (TCR; FIG. 2)24. Co-stimu-
`lation is the crucial process by which a sec-
`ond signal is delivered to naive T cells, which
`facilitates their activation and the subsequent
`generation of an antigen-specific effector
`
`response. It is mediated by interactions
`between members of the B7 family of pro-
`teins on antigen-presenting cells and the
`CD28 co-stimulatory molecule that is
`expressed on the surface of T cells. This
`interaction, in conjunction with the primary
`TCR-mediated signal, prevents the induction
`of immunological tolerance (or anergy),
`which would occur in the presence of the
`TCR alone.
`The co-stimulatory activity of thalido-
`mide is important as it could be used as an
`immunological adjuvant to promote an oth-
`erwise ineffective immune response. For
`example, it could provide an alternative
`approach for treating patients with cancer by
`enhancing their response to tumour anti-
`gens. However, it should be noted that the
`immunomodulatory effects of thalidomide
`
`Timeline | Chronology of thalidomide and IMiD development
`
`Thalidomide is
`synthesized by
`Chemie Grunenthal.
`
`Thalidomide use is associated
`with neuropathy and birth defects
`and is subsequently withdrawn.
`
`Use in graft-versus-
`host disease.
`
`Thalidomide shown to
`possess anti-angiogenic
`properties.
`
`Thalidomide shown to co-stimulate
`T cells. Food and Drug Administration
`(FDA) approval in the United States for
`thalidomide use in patients with ENL.
`
`The IMiD CC-5013 shown to be
`effective in treating MM. IMiDs
`shown to enhance cancer
`vaccine responses in vivo.
`
`1954
`
`1956
`
`1961
`
`1965
`
`1988
`
`1991
`
`1994
`
`1996
`
`1998
`
`2000
`
`2002
`
`2003
`
`Introduced in
`Germany as
`a sedative.
`
`First report showing effectiveness
`in patients with erythema
`nodosum leprosum (ENL).
`
`Thalidomide shown to inhibit
`lipopolysaccharide-induced
`tumour-necrosis factor-α
`(TNF-α) expression.
`
`Design of thalidomide
`analogues with improved anti-
`TNF-α properties — the birth
`of the immunomodulatory
`drugs (IMiDs).
`
`Reports of the effectiveness
`of thalidomide in multiple
`myeloma (MM). First clinical
`IMiD programme initiated.
`
`CC-5013 gets FDA fast-track approval
`for MM and myelodysplastic
`syndromes. CC-5013 is shown to
`have antitumour activity in patients with
`solid tumours. CC-4047 has activity in
`patients with MM.
`
`NATURE REVIEWS | CANCER
`
`VOLUME 4 | APRIL 2004 | 3 1 5
`
`Apotex Ex. 1011, p. 2
`
`
`
`P E R S P E C T I V E S
`
`Bone resorption
`
`Stimulus
`
`TNF-α, IL-1,
`IL-6
`
`Proteolysis
`
`Osteoclast
`
`Monocyte/
`macrophage
`
`Monocyte
`
`Cytokines, adhesion molecules,
`coagulation factors, iNOS
`
`Myocyte
`
`IFN-β,
`collagenase
`
`Fibroblast
`
`TNF-α
`
`Endothelium
`
`B cell
`
`Antibodies
`
`Adipocyte
`
`T cell
`
`Liver
`
`Brain
`
`Inhibition of
`lipoprotein lipase
`
`IL-2, IFN-γ,
`other cytokines
`
`Acute-phase
`proteins
`
`Fever, sleep
`
`Figure 1 | Tumour-necrosis factor-α has numerous targets. Tumour-necrosis factor-α (TNF-α) is mainly produced by monocytes and macrophages, but is
`also produced by other cell types, in response to a large number of stimuli and physiological conditions. TNF receptors are expressed on most cell types, which
`respond to TNF-α by activating a range of transcription factors and gene products. Effects on TNF-α bioactivity, therefore, directly influence a diverse range of cell
`activities. IFN, interferon; IL, interleukin; iNOS, inducible nitric-oxide synthase. Figure adapted with permission from Ref. 57 (1997) Elsevier Science Publishers.
`
`are variable, and depend on the type of
`immune cell that is activated, as well as the
`type of stimulus that the cell receives.
`Therefore, the effects of thalidomide on a par-
`ticular cohort of patients are likely to depend
`on their disease state and immunological
`status. For example, thalidomide-mediated
`inhibition of the key pro-inflammatory and
`regulatory cytokines TNF-α16 and inter-
`leukin-12 (IL-12; REF. 25) during microbial
`stimulation of monocytes could be countered
`by thalidomide-mediated augmentation of
`the same cytokines during T-cell activation.
`This differential response might explain
`the clinically diverse effects of thalidomide,
`which include beneficial activity in some
`autoimmune conditions that are associated
`with increased T helper 1 (TH1)-type cellular
`immunity and some cancers that are associ-
`ated with lack of tumour-specific TH1-type
`cellular immunity. Therefore, thalidomide
`can no longer be referred to simply as a
`TNF-α inhibitor, as T-cell co-stimulation is
`likely to explain the unexpected increase in
`TNF-α production that is observed in certain
`clinical settings26.
`
`Thalidomide: an anticancer agent
`The main impetus for using thalidomide to
`treat patients with cancer came with the
`discovery of its anti-angiogenic potential.
`This also happened to coincide with the
`emerging concept that treatment could
`be aimed at the infrastructure that supports
`the growth of the tumour, rather than
`targeting tumour cells directly. Similarities
`between the angiogenic process in the
`promotion of
`tumour growth and in
`chronic inflammation also lent further
`support for a possible role for thalidomide
`as an anti-inflammatory agent in the treat-
`ment of cancers. In particular, the anti-
`TNF-α effects of thalidomide were thought
`to be relevant, as TNF-α seems to have a
`role in angiogenesis by upregulating
`the expression of endothelial integrin,
`which is crucial for this process27. Finally,
`it is well established that the increase of
`TNF-α in the serum of patients with cancer
`is often associated with advanced disease, so
`using thalidomide to reduce these levels
`might prove to be beneficial in the treatment
`of patients.
`
`Thalidomide and multiple myeloma
`In the past few years, thalidomide has begun to
`impact on the treatment of multiple myeloma
`(MM; BOX 2). This is an incurable B-cell malig-
`nancy in which increased bone-marrow
`microvessel density (MVD) is associated with
`poor prognostic outcome, providing the ratio-
`nale
`for treatment with thalidomide.
`Remarkably, an initial report published in
`1999 indicated that thalidomide was an effec-
`tive treatment in 30–40% of patients with
`advanced and refractory MM28 and showed
`that, of the 84 patients treated, there was an
`overall clinical response rate of 32%.
`Moreover, 10% of patients had complete, or
`near complete, remissions. Partial remission
`— defined by a >50% decrease in serum or
`urine monoclonal protein, an established
`prognostic indicator — was achieved in 25%
`of patients. The authors of this study were
`unable to show an association between the
`clinical response to thalidomide and a
`decrease in bone-marrow MVD. However,
`very recent data showing decreased MVD only
`in patients who responded to thalidomide
`does support the theory that angiogenesis is a
`
`316 | APRIL 2004 | VOLUME 4
`
`www.nature.com/reviews/cancer
`
`Apotex Ex. 1011, p. 3
`
`
`
`therapeutic target in MM29. Subsequent
`studies have confirmed these initial clinical
`findings and indicate that thalidomide
`treatment leads to a 25–33% response rate
`in patients with refractory MM, and also
`has significant response rates in patients at
`other stages of disease30. More recently,
`thalidomide treatment in combination with
`the chemotherapeutic agent dexametha-
`sone has been shown to act synergistically
`and induce even greater partial response
`rates of 60–70%, even when patients have
`been unresponsive to either agent alone31.
`Addition of cyclophosphamide seems to
`improve the response rates still further32.
`Since this discovery, thalidomide has
`also been evaluated in clinical trials as a
`treatment for various solid tumours with a
`varying degree of success. There are pub-
`lished reports of efficacy in the treatment
`of patients with solid tumours such as
`advanced renal cancer33, metastatic prostate
`cancer34, high-grade glioma35 and metasta-
`tic melanoma36. More complete overviews
`of thalidomide use in patients with solid
`tumours can be found elsewhere37,38.
`
`Development of IMiDS
`Clearly, even in patients with advanced can-
`cer, the use of thalidomide could present sig-
`nificant problems due to its teratogenic side
`effects. This requires intense patient moni-
`toring during thalidomide administration.
`Therefore, it is hardly surprising that not
`long after the discovery of the anti-angio-
`genic properties of thalidomide, and given its
`obvious clinical benefits, attempts were made
`to synthesize thalidomide analogues that had
`fewer side effects than the parent compound.
`Immunomodulatory drugs (IMiDs) are
`a series of compounds that were developed
`by using the first-generation IMiD thalido-
`mide as the lead compound in a drug-
`discovery programme. The thalidomide
`structural backbone was used as a template
`by chemists to design and synthesize com-
`pounds with increased immunological and
`anticancer properties, but lacking the toxic-
`ity associated with the parent compound39.
`Initially, the rationale for developing the
`second-generation IMiDs in the mid 1990s
`was to improve the inhibition of TNF-α40,41
`and, with this aim, a series of amino-
`phthaloyl-substituted thalidomide ana-
`logues were generated42. The 4-amino
`analogues — in which an amino group is
`added to the fourth carbon of the pthaloyl
`ring of thalidomide — were found to be up
`to 50,000 times more potent at inhibiting
`TNF-α than the parent compound in vitro.
`Extensive preclinical testing, involving
`
`P E R S P E C T I V E S
`
`a Activation of naive T cells requires co-stimulation
`
`APC
`
`APC
`
`MHC
`
`TCR
`
`Peptide
`fragment
`
`T cell
`
`B7
`
`CD28
`
`Cytokine
`production
`
`b IMiDs overcome the requirement for co-stimulation
`
`T-cell proliferation
`
`APC
`
`APC
`
`Enhanced T-cell proliferation
`
`IMiD
`
`T-cell stimulation
`
`IMiD
`
`Cytokine production
`
`Enhanced cytokine
`production
`
`Figure 2 | Co-stimulatory activity of thalidomide. a | Antigen-presenting cells (APCs) activate T cells by
`presenting major histocompatibility complex (MHC)-bound peptides to the T-cell receptor (TCR). Effective
`T-cell activation also requires interaction between accessory molecules, such as B7 on the APC and
`CD28 on the T cell, which provides the secondary signals that are necessary for activation to occur and
`prevents T-cell anergy (non-responsiveness). b | Thalidomide and immunomodulatory drugs (IMiDs) seem
`to enhance TCR-mediated signalling both in the absence and presence of these secondary signals,
`thereby enhancing immune responses.
`
`pharmacology, pharmacokinetics and toxi-
`city, has led to the identification of CC-
`5013 (Revimid) and CC-4047 (Actimid) for
`testing in clinical trials (FIG. 3).
`Third-generation IMiDs developed from
`the ongoing research programme are now in
`preclinical testing and will be investigated in
`
`clinical trials if modifications to the second-
`generation compounds are necessary.
`Furthermore, as the emphasis during pre-
`clinical testing has changed from the
`anti-TNF-α activity of the IMiDs to their
`anti-angiogenic and immunomodulatory
`activities, it is possible that third-generation
`
`Box 2 | Multiple myeloma
`
`Multiple myeloma (MM) is a B-cell malignancy that is incurable at present. It is characterized by
`the clonal proliferation of malignant cells in the bone marrow that leads to the production of a
`monoclonal immunoglobulin. MM accounts for approximately 1–2% of all cancers and cancer
`deaths, and afflicts 14,000–15,000 patients annually in the United States alone. The current
`median survival rate for symptomatic patients is 3–5 years. High-dose chemotherapy —
`typically melphalan and prednisolone — combined with transplantation of haematopoietic
`stem cells increases the rate of complete remission and extends event-free and overall survival.
`However, little progress in developing effective treatment regimens has been made over the past
`few decades; relapse rates are very high and there are few salvage therapies available.
`Thalidomide treatment was initiated in MM because this condition correlates with prominent
`bone-marrow vascularization, which is associated with poor prognosis. In addition, plasma
`levels of various pro-angiogenic molecules, such as basic fibroblast growth factor and vascular
`endothelial growth factor, are increased in patients with active MM. Therefore, anti-angiogenic
`drugs, such as thalidomide, are viable therapeutic options.
`
`NATURE REVIEWS | CANCER
`
`VOLUME 4 | APRIL 2004 | 3 1 7
`
`Apotex Ex. 1011, p. 4
`
`
`
`P E R S P E C T I V E S
`
`a
`
`Thalidomide
`
`NH
`
`O
`
`OO
`
`N
`
`O
`
`4
`
`Pthaloyl ring
`
`b
`
`CC-5013
`OO
`
`CC-4047
`OO
`
`N
`
`NH
`
`O
`
`NH
`
`O
`
`N
`
`O
`
`NH2
`
`NH2
`
`Figure 3 | Structure of thalidomide and the
`IMiDs CC-5013 and CC-4047. The thalidomide
`structure (a) was modified by adding an amino
`(NH2-) group at the 4 position of the phthaloyl
`ring to generate the IMiDs CC-5013 and
`CC-4047 (b). For CC-5013, one of the carbonyls
`(C = O) of the 4-amino-substituted phthaloyl ring
`has been removed.
`
`IMiDs could have greater anticancer activity
`and/or enhance immune responses. Because
`of the structural similarity with thalido-
`mide, the IMiDs possess the same properties
`that are of potential benefit to patients with
`cancer — prevention of angiogenesis and
`co-stimulation of T cells.
`
`IMiD functions
`Angiogenesis. Recent results have confirmed
`that the IMiDs, in particular the clinical lead
`compounds CC-5013 and CC-4047, are anti-
`angiogenic43,44. However, as with thalido-
`mide, the mechanism(s) remains elusive.
`Data from in vitro experiments indicate that
`IMiDs vary in their ability to inhibit
`endothelial-cell proliferation. Indeed, the oral
`administration of CC-5013 is able to inhibit
`tumour growth in a mouse model of colorec-
`tal cancer despite having no effect on
`endothelial-cell proliferation in vitro. Of
`particular interest is the observation that
`CC-5013 seems to be non-teratogenic when
`tested in the sensitive New Zealand rabbit
`preclinical model, which is the only animal
`model in which thalidomide-associated
`teratogenicity can be detected.
`
`T-cell co-stimulation. IMiDs are far more
`potent than thalidomide at co-stimulating
`T-cells that have been partially activated via
`the TCR45. Furthermore, co-stimulation
`applies equally to CD4+ and CD8+ T cells. The
`potency of IMiD-induced co-stimulation
`seems to increase when TNF receptor 2 trans-
`port to the cell membrane is inhibited. The
`implications of this are unclear, although this
`is likely to affect T-cell homeostasis. More
`recently, IMiDs have been shown to trigger
`
`the phosphorylation of CD28 and also to
`enhance the activity of the AP-1 transcription
`factor46,47. However, the precise mechanism(s)
`that is involved in IMiD-mediated T-cell
`co-stimulation remains to be elucidated.
`There is now clear evidence to indicate
`that the co-stimulatory properties of IMiD
`analogues in vitro can translate to beneficial
`antitumour responses in vivo. It has been
`demonstrated that CC-4047 is able to
`enhance a partially effective cancer vaccine
`and enable the generation of a long-term
`protective antitumour response48. Protection
`seems to be mediated by the induction of
`protective TH1-type cellular immunity as,
`although CD8+ cells were activated, there
`were more CD4+ cells responding to the
`tumour cells that comprised the vaccine.
`Some protection is also seen when CC-4047
`is co-administered with a tumour vaccine
`shortly after live-tumour challenge in mice.
`However, a booster regimen seems to be
`required. This more closely mirrors the
`clinical situation, the aim of which is to
`treat an established tumour and protect
`against the formation of new tumours and
`the regrowth of residual tumour after
`surgical resection.
`There is also emerging evidence that
`IMiDs can activate the innate component
`of
`the immune system. For example,
`CC-5013 seems to augment the cytotoxicity
`of natural-killer cells, leading directly to lysis
`of MM cells49. Furthermore, CC-4047 seems
`to have a potent augmentary effect on CD28-
`negative γδ T cells that have been stimulated
`with their natural bacterial antigen isopen-
`tenyl pyrophosphate, (J.B.B., unpublished
`
`observations). These cells are also able to
`directly lyse tumour cells, augment the early
`expression of cytokines in response to bacter-
`ial infection and help the development of the
`adaptive immune response.
`
`Direct antitumour activity of IMiDs
`Surprisingly, the IMiDs were found to share
`another important anticancer property —
`the ability to directly induce growth arrest
`and caspase-dependent apoptosis of
`tumour cells50. Initial preclinical data
`showed that CC-5013 possesses direct anti-
`myeloma activity in the absence of acces-
`sory immune cells50,51. Primary human MM
`cells derived from the bone marrow of
`patients resistant to chemotherapy were
`shown to be susceptible to IMiD-induced
`growth arrest. This could be overcome by
`the exogenous addition of the pro-inflam-
`matory cytokine IL-6, indicating that inhi-
`bition of IL-6 is likely to be involved in the
`mechanism that regulates this effect.
`Importantly, other mechanistic details have
`begun to emerge, including effects on apop-
`totic pathways52. Furthermore, IMiD activ-
`ity is able to potentiate the effects of TRAIL
`(TNF-related apoptosis-inducing ligand),
`dexamethasone and proteasome inhibitors
`that are used as anti-myeloma therapies at
`present. There is also strong evidence that
`IMiDs can interfere in interactions between
`myeloma cells and bone-marrow stromal
`cells, which seem to be crucial for MM-cell
`growth and survival, and prevent the
`upregulation of IL-6 and vascular endothe-
`lial growth factor, which is involved in
`angiogenesis53 (FIG. 4).
`
`Stromal cells
`
`MM cell
`
`Cell death
`
`Cytokine
`production
`
`IL-6
`TNF-α
`
`Bone
`marrow
`
`NK cell
`
`T cell
`
`Cell growth
`
`IMiD
`
`IL-2
`IFN-γ
`
`T-cell activation
`
`VEGF
`bFGF
`
`Angiogenesis
`
`Figure 4 | Antitumour activity of IMiDs in multiple myeloma. Immunomodulatory drugs (IMiDs)
`induce growth arrest and/or apoptosis in multiple myeloma (MM) cells and inhibit adhesion of MM cells to
`bone-marrow stromal cells. Stromal-cell expression of vascular endothelial growth factor (VEGF) and
`basic fibroblast growth factor (bFGF) is reduced by IMiDs, which decreases angiogenesis. Expression of
`interleukin-6 (IL-6) and tumour-necrosis factor-α (TNF-α) by the stromal cells is also reduced, which
`inhibits growth of MM cells. The IMiDs also enhance T-cell stimulation and proliferation. The activated
`T cells release IL-2 and interferon-γ (IFN-γ), which activate natural-killer (NK) cells (which might also be
`activated directly) and causes MM-cell death.
`
`318 | APRIL 2004 | VOLUME 4
`
`www.nature.com/reviews/cancer
`
`Apotex Ex. 1011, p. 5
`
`
`
`Table 1 | Current clinical studies of IMiDs
`Drug
`Indication
`Centre
`CC-5013
`Relapsed MM (n = 27)
`Dana–Farber Cancer
`Institute, USA
`
`Phase Stage
`I
`Completed
`
`P E R S P E C T I V E S
`
`Comments and references
`First published report in MM54. A dose-escalating
`study with 24 evaluable patients. Best responses in
`terms of reduction in serum M-protein in evaluated
`patients were >50% in 7/24 (30%), >25–50% in
`10/24 (42%) and <25% in 2/24 (8%) patients. The
`maximum-tolerated dose was 25 mg/day. Grade 3
`myelosuppression was apparent in patients
`treated with 50 mg/day. No somnolence or
`neuropathy observed.
`First published report in solid tumours55. Seventeen
`evaluable patients. One partial response and two clear
`objective responses. Evidence of T-cell activation and
`increased serum IL-12, GM-CSF and TNF-α. No
`serious adverse effects were observed.
`Still recruiting. No data reported.
`
`Completed
`
`Completion
`due April 2005
`Started early Still recruiting. No data reported.
`2002
`Started early Still recruiting. No data reported.
`2002
`Started 2003 Still recruiting. No data reported.
`
`CC-5013
`
`CC-5013
`
`CC-5013
`
`CC-5013
`
`CC-5013
`
`CC-4047
`
`Metastatic malignant
`melanoma and other
`advanced solid tumours
`(n = 20)
`
`Refractory solid tumours
`(n = 24)
`Reccurrent high-grade
`glioma (n = 80)
`Refractory metastatic
`cancer (n = 30)
`Refractory solid tumours
`and/or lymphoma
`(n = 3–30)
`Advanced MM (n = 18)
`
`St George’s Hospital
`Medical School, UK
`
`Wake Forest University,
`USA
`NCI, Bethesda, USA
`
`NCI, Bethesda, USA
`
`I
`
`I
`
`I
`
`I
`
`NCI, Bethesda, USA
`
`I/II
`
`Guy’s and St Thomas’s,
`UK
`
`I/II
`
`CC-5013
`
`Relapsed/refractory MM
`(n = 60)
`
`Multicentre (USA) based II
`at the Dana–Farber
`Institute, USA
`
`CC-5013
`
`CC-5013
`
`CC-4047
`
`CC-5013
`
`CC-5013
`
`CC-5013
`
`Relapsed/refractory MM
`(n = 100)
`Refractory MM (n = 200)
`
`University of Arkansas,
`USA
`Multicentre, USA
`
`Metastatic hormone-
`refractory prostate cancer
`(n = 36)
`MDS with cytogenetic
`abnormality (n = 36)
`
`University of Colorado
`and Baylor College of
`Medicine, Texas, USA
`NCI and Memorial
`Sloan–Kettering Cancer
`Center, USA
`MDS with 5q cytogenetic Multicentre, USA
`abnormality (n = 90)
`MDS (n = 136)
`
`Multicentre, USA
`
`CC-5013
`
`MDS (n = 25)
`
`Multicentre, USA
`
`Completed
`
`Study showed anti-myeloma activity and an
`acceptable safety profile56. CC-4047 was given in a
`dose-escalating regimen (1 mg/day up to
`10 mg/day). All patients improved clinically. The
`M-protein response on trial was <25% reduction in
`8/18 (44%), >25–50% in 7/18 (39%) and >50% in
`3/18 (17%). The maximum-tolerated dose was
`2 mg/day because of neutropaenia at the higher doses.
`Unpublished data presented at the 2003 American
`Completion
`in early 2004 Society of Hematology meeting indicates that so far
`39/46 evaluable patients (85%) with progressive
`disease experienced a reduction or stabilization in
`their M-protein levels.
`Still recruiting. No data reported.
`
`Completion
`in 2004
`Completion
`in Dec. 2005
`Completion
`in mid-2005
`
`Still recruiting. No data reported.
`
`Still recruiting. No data reported.
`
`Unknown
`
`Still recruiting. No data reported.
`
`Completion
`in May 2004
`Completion
`in Sept. 2004
`Completed
`
`Completion
`in June 2004
`Completion
`at end of 2005
`
`Still recruiting. No data reported.
`
`Still recruiting. No data reported.
`
`Completed. 64% of 25 patients needed at
`least 50% fewer blood transfusions after
`CC-5013 treatment. Also, 8/8 patients
`with 5q- syndrome lost all sign of cells with the
`telltale 5q-chromosomal deletion.
`Still recruiting. No