FormalPara Key Points

Thalidomide, the first immunomodulatory drug approved for multiple myeloma, has been infrequently used in the US, given its adverse effects (e.g., somnolence, constipation, and peripheral neuropathy) and the advent of more potent therapeutic options.

Numerous clinical studies have demonstrated thalidomide’s effectiveness and safety in newly diagnosed and relapsed/refractory multiple myeloma, as well as its low myelotoxicity and negligible renal clearance.

Salvage regimens combining thalidomide and corticosteroids (with/without selected cytotoxic or targeted agents) remain useful in heavily pretreated patients who are ineligible for novel anti-myeloma therapies (e.g., due to poor bone marrow reserve or severe renal dysfunction).

1 Introduction

Multiple myeloma (MM) is a hematologic malignancy characterized by abnormal plasma cell (PC) proliferation, increased production of monoclonal protein, and end-organ damage [1, 2]. Diagnosis requires ≥ 10% clonal PCs on bone marrow (BM) examination or a biopsy-proven plasmacytoma plus at least one myeloma-defining event, such as hypercalcemia, renal dysfunction, anemia, lytic bone lesions, or a biomarker of malignancy (e.g., ≥ 60% clonal PCs within the BM, involved/uninvolved free light chain ratio ≥ 100, or more than one focal lesion on magnetic resonance imaging) [3]. In 2020, according to the World Health Organization (WHO), a total of 176,404 new patients fulfilled such criteria worldwide, making MM the second most common blood cancer after lymphoma [4]. Among affected patients, 63% were older than 65 years of age—a repercussion of the age association of MM’s precursor stage, monoclonal gammopathy of undetermined significance (MGUS) [5, 6].

From the first description of MM in the literature as part of Samuel Solly’s case series in 1844 [7] to the landmark trial that confirmed the safety and efficacy of the chimeric antigen receptor (CAR) T-cell idecabtagene vicleucel (ide-cel) in 2021 [8], remarkable advances have been made in elucidating the clinicopathological aspects and potential therapeutic targets of MM [9]. As a result, many treatment strategies are currently available. Choosing a frontline regimen depends on efficacy, local availability, risk stratification, autologous stem cell transplant (ASCT) eligibility, and patient-specific factors (e.g., performance status [PS], comorbidities and preferences) [10]. However, due to the eventual emergence of genetically heterogeneous subclones of myeloma cells, even individuals who achieved deep responses will eventually relapse and need subsequent lines of therapy [11, 12].

The list of drug classes approved for MM includes corticosteroids, alkylating agents, anthracyclines, topoisomerase II inhibitors, proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), histone deacetylase inhibitors (iHDACs), selective inhibitors of nuclear export (SINEs), monoclonal antibodies (mAbs), antibody-drug conjugates (ADCs), bispecific antibodies (BsAbs), and CAR T cells [13,14,15,16]. Moving from bench to bedside, most of those therapies were developed over the last 20 years, leading to significant prognostic implications [9, 17]. The 5-year relative survival rate among US patients, for example, increased from 28.6% in 1986–1993 to 55.6% in 2011–2017 [18, 19]. One of the pivotal discoveries within this period was the anti-myeloma activity of thalidomide, an old drug infamous for its teratogenic effects [20].

In Europe, thalidomide is still widely studied and incorporated into clinical practice, especially after the phase III CASSIOPEIA trial supported the use of bortezomib, thalidomide, and dexamethasone (VTD) plus daratumumab (Dara-VTD) as a first-line regimen for newly diagnosed MM (NDMM) [21, 22]. However, in the US and Canada, the drug has largely been replaced by newer generations of IMiDs (lenalidomide and pomalidomide)—a direct consequence of thalidomide’s association with multiple adverse events (AEs), including somnolence, constipation, peripheral neuropathy (PN), and venous thromboembolism (VTE) [23]. Moreover, cereblon E3 ligase modulators (CELMoDs), such as iberdomide (CC-220) and CC-92480, are now slowly making their way into US Food and Drug Administration (FDA) approval [24]. As a result, the current role of thalidomide in both the upfront and salvage settings is less clearly defined; yet, this older agent has the advantage of providing significant anti-myeloma activity even in patients with severely compromised kidney and/or marrow function owing to its minimal renal clearance and relative lack of myelotoxicity [25].

From this perspective, the present review summarizes the available data on thalidomide’s history, mechanisms and applications in MM, with a particular focus on its role in relapsed/refractory MM (RRMM). Our ultimate goal is to provide general guidance for physicians who do not currently use this drug in clinical practice, especially considering that some heavily pretreated patients may not be eligible for some of the novel targeted or immune-based therapies.

2 Historical Background

Thalidomide (α-N-phthalimido-glutarimide) was first introduced in 1956 by the West German pharmaceutical company Chemie Grünenthal [26]. Initially marketed as a well-tolerated sedative and antiemetic agent to reduce pregnancy-related morning sickness, thalidomide rapidly gained widespread popularity—by 1960, the drug was being sold in over 40 countries [27, 28]. Early preclinical safety evaluation using pregnant rats and mice showed no interference with embryonic development [29]; however, approval in the US was not obtained at that time, mainly due to FDA concerns about neurotoxicity [30]. Shortly afterwards, two independent clinicians (Lenz in Germany and McBride in Australia) reported the association between human prenatal exposure to thalidomide and congenital limb abnormalities, including phocomelia [31, 32]. By 1962, thalidomide had been withdrawn from most commercial markets, leaving a legacy of approximately 10,000 affected infants [30, 32]. Despite the resistance of some rodent species to thalidomide embryopathy, ensuing studies with rabbits, primates, and fish species were able to demonstrate the drug’s teratogenicity [30,31,32]. Thalidomide’s newly discovered property prompted its investigation as an antineoplastic agent during the early 1960s. Nevertheless, two clinical studies including patients with advanced tumors of various types failed to demonstrate evidence of objective response attributable to the drug [33, 34].

In 1965, Jacob Sheskin reported thalidomide’s ability to control erythema nodosum leprosum (ENL) [35]. This triggered a perspective shift on the drug from a teratogenic antiemetic to a potent anti-inflammatory and immunomodulatory agent [36]. In 1971, a WHO-coordinated, male-only clinical trial confirmed thalidomide’s effectiveness in treating acute leprosy reactions [37]. However, the drug only became FDA-approved for ENL in 1998, after studies identified an immunological basis for its clinical effects [28]. The FDA subsequently supported thalidomide’s experimental use in other inflammatory disorders (e.g., sarcoidosis, cutaneous lupus erythematosus, Behcet’s syndrome, ankylosing spondylitis, rheumatoid arthritis, inflammatory bowel diseases, recurrent aphthous stomatitis, and chronic graft-versus-host disease), with the achievement of promising results [29, 38, 39].

Throughout the 1990s, the number of publications related to thalidomide increased exponentially, as well as the off-label uses of the drug [40, 41]. In this scenario, efforts on elucidating the teratogenic mechanisms of thalidomide eventually renewed the interest in its potential use for cancer treatment [42]. In 1994, thalidomide was shown to inhibit angiogenesis, a key process in fetal limb development [43]. Already a well-known hallmark of solid tumors, angiogenesis was also correlated to blood cancer progression by that time [44,45,46]. Soon, multiple preclinical and clinical studies evaluating the antineoplastic activity of thalidomide were initiated [27]. Despite varying degrees of success in solid tumors, remarkable benefits were observed in some hematologic malignancies, especially MM [27, 47]. In 2006, after mounting evidence of its efficacy, thalidomide became the first new agent in over a decade to gain FDA approval for MM. In 2008, a similar approval was obtained from the European Medicines Agency (EMA) [48, 49].

3 Mechanisms of Action

The abnormal PC proliferation inherent to MM occurs predominantly within the BM, suggesting the importance of the BM microenvironment in supporting disease activity [50]. BM angiogenesis, for example, progressively increases along the spectrum of PC disorders—from the more benign MGUS stage, throughout the smoldering phase, and up to overt MM [51]. This is secondary to the direct production of angiogenic molecules by malignant PCs, as well as their induction in BM stromal cells (BMSCs) [52, 53]. Moreover, tumor cell expression of basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF), and their respective receptors creates autocrine loops of growth, survival, and migration [53,54,55]. Accordingly, BM microvascular density (MVD) has been shown to have a direct correlation with the PC labeling index and inverse correlation with the overall survival (OS) of MM patients [45, 56,57,58,59]. In addition, BM angiogenesis has a similar prognostic value in patients with solitary plasmacytomas [60].

In 1994, using a rabbit cornea micropocket assay, D’Amato et al. demonstrated that thalidomide inhibited bFGF-induced angiogenesis [43]. Using a model of murine cornea, his group then showed the drug’s ability to suppress VEGF-induced angiogenesis [61]. Later, thalidomide was found to attenuate nitric oxide-driven angiogenesis [62, 63] and downregulate other key angiogenic genes (e.g., angiopoietin-1, insulin-like growth factor [IGF]-1, and IGF binding protein-3) [64]. Since BM neovascularization is a critical event for MM progression, thalidomide attains disease control by simultaneously targeting multiple angiogenesis pathways, ultimately decreasing the BM MVD in responders [65,66,67]. In contrast, anti-VEGF drugs targeting fewer angiogenesis pathways (e.g., bevacizumab, sorafenib, and vandetanib) failed to demonstrate a significant improvement in patient outcomes [65].

Besides their antiangiogenic effect, thalidomide and its analogs induce tumor cell apoptosis via activation of caspase-8, enhanced sensitivity to Fas and downregulation of cellular inhibitor of apoptosis protein-2 (cIAP-2) [68]. These agents can also suppress BMSC production of important mediators of myeloma cell proliferation, such as tumor-necrosis factor (TNF)-α and interleukin (IL)-6 [27,28,29]. Contrary to expectations, IL-6 blockade with siltuximab failed to improve OS or progression-free survival (PFS) in previous trials [69]. Thalidomide further hinders tumor growth via downregulation of surface adhesion molecules mediating myeloma cell interactions with the surrounding extracellular matrix and BMSCs [70, 71]. Moreover, IMiDs can potentiate T-cell proliferation, differentiation, and survival by augmenting B7-CD28 co-stimulatory signals. Apart from enhancing tumor antigen presentation by dendritic cells, thalidomide directly induces tyrosine phosphorylation of CD28, promoting activation of phosphoinositide-3-kinase (PI3K) and nuclear factor-kappa B (NF-κB) [70,71,72,73]. The resultant stimulation of interferon (IFN)-γ and IL-2 production boosts the number and function of natural killer (NK) cells, further improving the anti-myeloma immune response in IMiD-treated patients [70, 74]. Accordingly, published data on IFN use as an anti-myeloma agent demonstrated a significant, although limited, improvement in clinical outcomes [75].

Despite in vivo and in vitro demonstration of a wide range of pharmacological effects, thalidomide’s primary molecular target remained uncertain until 2010, when binding to cereblon (CRBN) was identified as the key mechanism of its teratogenicity [76, 77]. Shortly afterwards, CRBN expression was also found to be an essential requirement for the anti-myeloma activity of IMiDs [78]. CRBN serves as a substrate receptor of the Cullin-Ring ligase 4 E3 ubiquitin ligase complex (CRL4CRBN), which recognizes substrates for ubiquitination and proteasomal degradation [79]. The binding of thalidomide to CRBN causes an allosteric modification of the CRL4CRBN complex, changing its substrate specificity [80]. This can lead to several downstream effects, depending on the proteins subsequently ubiquitinated [81]. Thalidomide-induced degradation of spalt-like transcription factor 4 (SALL4) and tumor protein 63 (TP63), for example, causes fetal malformations [82, 83]. Meanwhile, degradation of the zinc-finger transcription factors Ikaros (IKZF1) and Aiolos (IKZF3) exerts cytotoxic effects in myeloma cells, mainly through downregulation of IFN regulatory factor 4 (IRF4) and MYC [84, 85]. Thalidomide can also trigger the degradation of casein kinase 1α (CK1α), a pro-growth and anti-apoptotic enzyme in B-cell malignancies [86]; however, CK1α ubiquitination is substantially more extensive with lenalidomide [79]. Additionally, pomalidomide is more effective in promoting degradation of AT-rich interactive domain-containing protein 2 (ARID2), a critical protein for MYC expression [87]. Thus, substrate recognition by CRBN differs depending on the ligand structure, creating potential for sequential use of different IMiDs as a way to overcome myeloma cell resistance [79, 87]. Figures 1 and 2 illustrate the mechanisms of action of thalidomide and its analogs.

Fig. 1
figure 1

CRBN interacts with DDB1, Cul4A or Cul4B, and RoC1 to form the CRL4CRBN. CRBN functions as a receptor with a conserved tryptophan pocket in the complex, which leads to protein targeting for degradation via the ubiquitin-proteasome pathway. a Under normal, physiologic circumstances, binding of endogenous ligands to CRBN regulates cell metabolism, proliferation, and differentiation. Known substrates include GltS, AMPKγ, and Meis2. b Exposure to an IMiD leads to allosteric modification, changing the substrate specificity of CRBN and increasing degradation of Aiolos, Ikaros, and CK1α, among other proteins. Depletion of these neosubstrates leads to pleiotropic anti-myeloma effects, such as IRF4 and MYC downregulation in myeloma cells, increased IL-2 production, and enhanced immune response against the tumor. CRBN cereblon, DDB1 DNA damage-binding protein-1, Cul4 Cullin 4, RoC1 regulator of Cullins 1, CRL4CRBN Cullin-Ring ligase 4 E3 ubiquitin ligase complex, GltS glutamine synthase, AMPKγ adenosine monophosphate-activated protein kinase gamma subunit, Meis2 Meis homeobox 2, IMiD immunomodulatory drug, CK1α casein kinase 1α, IRF4 interferon regulatory factor 4, IL interleukin

Full size image
Fig. 2
figure 2

IMiDs target not only the MM cell but also the BM microenvironment. By suppressing BMSC production of IL-6 and TNFα, these agents decrease MM cell proliferation. This is potentiated via downregulation of surface molecules that mediate BMSC–MM cell interactions, such as ICAM-1 and VCAM-1. Anti-angiogenesis occurs via modulation of chemotactic factors involved in endothelial cell migration, such as VEGF and bFGF. By inducing tumor antigen presentation by dendritic cells and tyrosine phosphorylation of CD28, IMiDs stimulate T-cell activation. The resultant increase in IFN-γ and IL-2 levels also enhances NK cell activity. PRF/GzmB is the main pathway of target cell apoptosis, but granzyme-independent mechanisms can also occur. IMiDs immunomodulatory drugs, MM multiple myeloma, BM bone marrow, BMSC bone marrow stem cell, IL interleukin, TNF tumor necrosis factor, ICAM-1 intercellular adhesion molecule-1, VCAM-1 vascular cell adhesion molecule-1, VEGF vascular endothelial growth factor, bFGF basic fibroblast growth factor, IFN interferon, NK natural killer, PRF perforin, GzmB granzyme B, MHC I major histocompatibility complex I

Full size image

4 Therapeutic Applications in Multiple Myeloma

4.1 Use in Relapsed/Refractory Multiple Myeloma

After decades of ineffective therapies primarily based on traditional empiricism (e.g., orange peel infusions, rhubarb pills, and urethane), successful anti-myeloma treatment using prednisone and lower doses of melphalan (80–100 mg/m2) was introduced in the late 1960s [88, 89]. Various dosing regimens and combinations of alkylating agents, with or without corticosteroids, were investigated during the following years [90]. In 1983, monotherapy with higher doses of melphalan (140 mg/m2) was demonstrated to improve complete remission rates, with the drawback of profound myelosuppression [91]. In the late 1980s, Barlogie et al. suggested ASCT as a measure to hasten patient hematologic recovery after marrow-ablative chemotherapy and total body irradiation [92, 93]. In 1996, a randomized controlled trial (RCT) of 200 untreated adults under age 65 years showed that high-dose chemotherapy followed by ASCT (HDC/ASCT) could improve overall response rate (ORR), event-free survival (EFS), and OS [94]. This strategy was then established as a standard frontline treatment for transplant-eligible (TE) NDMM patients [88]. Later, further studies supported the alternative of reserving ASCT for the first disease relapse [95] and the possibility of a second peripheral blood stem cell (PBSC) infusion as part of salvage therapy in selected cases [96]. Even so, therapeutic options for RRMM remained scarce and of limited clinical benefit [97]. Some of the available salvage regimens, such as vincristine/doxorubicin/dexamethasone (VAD) and etoposide/dexamethasone/cytarabine/cisplatin (EDAP), could lead to unacceptable AEs in a patient population mainly composed of elderly adults with numerous comorbidities and cumulative toxicity from previous therapies [97, 98].

As higher BM angiogenesis favored MM progression and thalidomide possessed antiangiogenic properties, the drug emerged as an investigational therapy for compassionate use in patients with advanced disease [25]. Phase I data were first described by Singhal et al. in 1999. In that landmark study, 84 patients with RRMM received oral thalidomide monotherapy for a median of 80 days. If tolerated, an initial dose of 200 mg nightly was increased by 200 mg every 2 weeks until reaching 800 mg/day. Even though study participants had been heavily pretreated (e.g., 90% received at least one cycle of HDC/ASCT), 32% of them achieved a ≥ 25% decline in serum or urine paraprotein levels, and 10% had nearly complete or complete response (CR) [99, 100]. Of the 27 patients with a paraprotein response, 21 could have a follow-up BM examination, which showed a concurrent BM response (defined as < 5% PCs) in 81% of cases [100]. Such results stimulated multiple subsequent trials, further evaluating thalidomide’s efficacy for RRMM, either as single-agent or as part of combination regimens [99]. Table 1 synthesizes data from clinical studies in which at least 25 patients received a thalidomide-based regimen for RRMM [100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157]. The ORR, defined as the percentage of patients who had at least a partial response (PR), and the PFS were reported according to International Myeloma Working Group (IMWG) uniform response criteria [158].

Table 1 Prospective trials evaluating the efficacy of thalidomide-containing regimens for relapsed/refractory multiple myeloma
Full size table

In a systematic review of 42 trials of single-agent thalidomide (50–800 mg/day) for RRMM, ≥ PR was seen in 479 of 1629 patients (29.4%, 95% confidence interval [CI] 27–32%). Among the 17 studies that assessed median time to response (TTR), 65% reported a period of between 1 and 2 months. The median EFS and OS were reported at 12 and 14 months, respectively. Severe AEs (grade 3–4) included somnolence (11%), constipation (16%), PN (6%), rash (3%), VTE (3%), and cardiac complications (2%) [159]. Meanwhile, a multicentric RCT coordinated by the Intergroupe Francophone du Myélome (IFM) compared the efficacy and safety of two doses of thalidomide (100 and 400 mg/day). Patients receiving 100 mg/day had significantly lower rates of most AEs, while 1-year OS was similar in both study arms after intention-to-treat (ITT) analysis [143]. The OPTIMUM trial, which compared dexamethasone against three different doses of thalidomide (100, 200, or 400 mg/day), also demonstrated that it is difficult for patients to tolerate a higher dose for the whole treatment duration—in the 400 mg/day arm, average dose intensity was 256 mg/day [144]. Noteworthy, treatment response appears to be higher if thalidomide is started after the first relapse or progression than later in the disease course [130].

Thalidomide’s clinical efficacy in advanced MM and synergistic activity with dexamethasone in myeloma cell lines prompted extensive evaluation of the thalidomide/dexamethasone (TD) combination as salvage therapy [25, 160, 161]. In a systematic review of 12 trials, among 451 RRMM patients treated with TD, 209 (46%, 95% CI 42–51%) achieved at least a PR. Treatment-related toxicity was comparable with thalidomide monotherapy. Dexamethasone’s median starting dose was 200 mg/day and the median target dose was 350 mg/day. While six studies reported an EFS with a weighted median value of 8 months, five studies reported an OS with a weighted median value of 27 months [161]. In 2012, Zamagni et al. investigated TD as a therapy for first relapse in 100 patients, achieving an ORR of 46%, a median PFS of 21 months, and a median OS of 43 months [142]. Subsequently, Hjorth et al. compared the efficacy of TD and bortezomib/dexamethasone (VD) in 131 patients relapsing after or refractory to initial melphalan-based treatment. Even though both study arms had a similar ORR (55% for TD, 63% for VD) and median PFS (9.0 months for TD, 7.2 for VD), rates of AEs were significantly higher in the VD arm [145].

Since thalidomide was found to lack myelosuppressive potential, the addition of conventional cytotoxic agents to salvage regimens became an attractive approach. Worldwide, many groups started to evaluate the association of TD or thalidomide/prednisolone (TP) with different alkylating agents and anthracyclines [25, 162]. A German protocol consisting of hyperfractionated intravenous cyclophosphamide (HyperCy; 300 mg/m2 every 12 h on days 1–3), pulsed oral dexamethasone (20 mg/m2 once daily on days 1–4, 9–12, 17–20), and escalating-dose thalidomide (100–400 mg/day) promoted CR, PR, and minor response (MR) rates of 4, 68, and 12%, respectively. However, there was high toxicity in study participants, including grade 3–4 neutropenia (86%), severe PN (16%) and VTE (8%) [111]. Later, a Spanish group evaluated the efficacy of an oral combination of thalidomide (200–800 mg/day), cyclophosphamide (50 mg/day) and pulsed dexamethasone (40 mg/day for 4 days every 3 weeks) in 71 RRMM patients. After 3 months of therapy, CR, PR and MR rates were 2%, 55% and 26%, respectively. After 6 months of therapy, most responses were maintained and the CR rate increased to 10%, although a small group of patients (n = 6) progressed. The replacement of HyperCy by continuous oral cyclophosphamide allowed a considerable decline in grade 3–4 neutropenia rates (from 86 to 10%) [116]. In a Korean real-world data (RWD) study, patients treated with TD plus cyclophosphamide (n = 236) and TP plus melphalan (n = 140) had a similar ORR (72 vs. 65%, p = 0.121). Analysis of all study subjects revealed a median PFS of 10.4 months and median OS of 28.0 months, with the drawback of a 10% rate of infections warranting intensive supportive care [152]. Due to its established role for relapsed lymphomas and better toxicity profile compared with other alkylating agents, bendamustine has also been tested in the RRMM setting [135, 149, 154, 163,164,165,166]. After phase I data showed ≥ PR in 24 of 28 patients who received TP plus bendamustine 60 mg/m2 (B60) [135], Schey et al. and Mian et al. evaluated TD plus B60, obtaining an ORR of 41.5 and 37%, respectively [149, 154]. TD has also been tested in association with doxorubicin, doxorubicin/vincristine, cyclophosphamide/etoposide, and cisplatin/doxorubicin/cyclophosphamide/etoposide (TD-PACE), providing an ORR of 40–84% [105, 112, 124, 125]. Nowadays, these protocols are rarely indicated due to the high risk of profound cytopenias and infectious complications, with the important exception of TD-PACE [116, 167].

Similar to dexamethasone/cyclophosphamide/etoposide/cisplatin (DCEP), TD-PACE is considered a useful regimen in early, aggressive relapses [23]. This protocol, generally administered every 4–6 weeks for 1–4 cycles, combines continuous thalidomide (50–200 mg/day) and a 4-day course of dexamethasone (40 mg/day), cisplatin (10 mg/m2/day), doxorubicin (10 mg/m2/day), cyclophosphamide (400 mg/m2/day), and etoposide (40 mg/m2/day) [112, 168]. Multiple retrospective cohort studies have reported the benefit of consolidating TD-PACE-like regimens with transplantation, highlighting the continued importance of HDC/ASCT in RRMM patients who are not eligible for, do not have access to, or did not respond to novel agents [136, 168,169,170,171,172]. For instance, Gerrie et al. analyzed the outcomes of 75 heavily pretreated patients at two tertiary care centers who received TD-PACE as salvage therapy. Overall, despite a reasonable ORR (49%), there was a short median PFS (5.5 months) and OS (14.0 months). However, when compared with non-transplant candidates, patients who proceeded to ASCT or achieved sufficient disease control to allow clinical trial enrollment had a non-significant trend towards improved PFS (13.4 vs. 2.9 months) and OS (20.5 vs. 7.5 months) [168].

The licensing of bortezomib as an anti-myeloma agent rapidly led to extensive investigation of its addition to thalidomide-based salvage regimens. In the MMVAR/IFM 2005-04 trial, 269 patients were randomly assigned to receive thalidomide (200 mg/day) and dexamethasone (40 mg/day) either with or without bortezomib (1.3 mg/m2 intravenous bolus on days 1, 4, 8, and 11 for eight cycles, and then on days 1, 8, 15, and 22 for four cycles). The VTD arm achieved higher ORR (45% vs. 21%, p = 0.001) and median PFS (18.3 vs. 13.6 months, p = 0.001); however, 24-month OS rates were not significantly different (71% for VTD vs. 65% for TD; p = 0.093) [146]. The Total Therapy 3 trial demonstrated significantly improved CR rates and EFS, with an upfront association of bortezomib to TD-PACE (VTD-PACE) for induction prior to and consolidation after high-dose melphalan-based tandem autotransplants [173]. This prompted empirical use of VTD-PACE to rescue induction failures or for PBSC mobilization in RRMM patients [170]. In 2017, a single-center, retrospective cohort used an ITT model to evaluate the outcomes of 141 patients (median age 59.7 years) who received salvage therapy with VTD-PACE-like regimens (VTD-PACE in 67.4%). Despite a median of four previous lines of treatment, attainment of ≥ PR and ≥ very good PR (VGPR) occurred in 54.4 and 10.3% of patients, respectively. Median PFS was 3.1 months (95% CI 1.9–3.9 months), while median OS was 8.1 months (95% CI 6.2–9.9 months). However, in the subgroup who received ASCT consolidation, median OS was 15.1 months [172]. As a result, the 2022 National Comprehensive Cancer Network (NCCN) guidelines have listed both TD-PACE and VTD-PACE as salvage therapy options for patients with aggressive MM [23]. Of note, in a retrospective cohort, 32% of patients who received TD-PACE had previously been exposed to or were refractory to bortezomib. Thus, the addition of this proteasome inhibitor (PI) in the RRMM setting may not yield as significant improvements as in the NDMM setting [168]. Moreover, the cumulative neurotoxicity associated with bortezomib (particular its intravenous formulation) may be a significant dose-limiting factor in clinical practice, hindering some individuals from taking advantage of the added therapeutic benefit [174]. Approaches to optimize bortezomib tolerance include subcutaneous administration, dose reduction, and weekly scheduling [175].

In this scenario, combination regimens based on alternative PIs with a decreased neurotoxicity profile and to which fewer patients have previously been exposed (e.g., carfilzomib) could potentially improve clinical outcomes in RRMM patients [23, 168]. In 2015, a phase Ib/II study investigated the efficacy and safety of CYKLONE, a four-drug protocol administered in 28-day cycles (cyclophosphamide 300 mg/m2 on days 1, 8, 15; carfilzomib 300 mg/m2 on days 1, 2, 8, 9, 15, 16; thalidomide 100 mg on days 1–28; and dexamethasone 40 mg on days 1, 8, 15, 22). Among the 64 previously untreated, TE patients enrolled in the trial, 2 achieved stringent CR (sCR), 3 achieved CR, 39 achieved VGPR, and 14 achieved PR, totaling an ORR of 91%. Besides being highly efficacious, the regimen was associated with manageable toxicities (e.g., all PN was grade 1, while grade 3–4 neutropenia only occurred in 23% of cases) [176]. These results led the CYKLONE protocol to be included as a salvage therapy option in the 2022 NCCN guidelines, under the list of regimens “useful in certain circumstances for early relapses (1–3 prior therapies)” [23]. Although significant cardiotoxicity was not observed with the use of carfilzomib in the CYKLONE trial, cardiovascular AEs can be an important limitation of using this PI for treating MM patients [177].

The development of other targeted anti-myeloma agents also created interest for their combination with IMiDs as a way to overcome resistance in heavily pretreated patients [167].

In recent phase II studies, TD plus ixazomib (a novel PI with oral administration and an improved safety profile) led to an ORR of 51–56% [155, 157], while TD plus elotuzumab (an anti-SLAMF7 monoclonal antibody [mAb]) provided an ORR of 38% [150]. In the MUK-six trial, 91% of patients achieved at least a PR after receiving a salvage regimen composed of TP, bortezomib, and the iHDAC panobinostat [151]. An ongoing phase II study by the Asian Myeloma Network is investigating the efficacy of TD plus daratumumab (an anti-CD38 mAb) plus TD (Dara-TD) in RRMM. Among 36 patients included in an interim analysis, 3 achieved sCR, 3 achieved CR, 10 achieved VGPR, and 10 achieved PR, totaling an ORR of 72% [178]. Despite such encouraging results, phase III RCTs are still needed to determine if these regimens are associated with a survival benefit in the RRMM setting.

Thalidomide doses of up to 800 mg/day were administered in early clinical trials, with some evidence of a dose–response relationship [100, 159]. Nevertheless, in a trial of thalidomide 100 versus 400 mg⁄day, Yakoub-Agha et al. demonstrated that the dose–response relationship of single-agent thalidomide is abrogated once corticosteroids are added [143]. In addition, doses exceeding 100 mg/day were found to cause significantly higher rates of toxicity-related symptoms (e.g., PN, constipation, somnolence), affecting health-related quality of life (HRQoL) and hindering long-term treatment continuation [143, 159, 179, 180]. Thus, in current practice, thalidomide doses of 50–200 mg/day are typically used [179]. In fact, most recent trials have employed a standard dose of 100 mg/day for RRMM patients [148, 149, 151, 152, 154, 155, 157]. Even so, 200–400 mg/day can be considered in those with good tolerance and aggressive disease that is not responding to lower doses (especially when newer therapies are unavailable, contraindicated, or cannot be tolerated) [179, 180]. A reasonable approach is to initiate therapy at 50–100 mg nightly and escalate the dose by 50 mg every 1–2 weeks as tolerated [179, 181]. The optimal duration of thalidomide-based salvage therapy has not yet been defined. In most clinical trials, RRMM patients received treatment until disease progression or AEs requiring discontinuation [180].

As previously discussed, one of the possible AEs associated with IMiDs is VTE. When administered as a single agent, thalidomide does not raise the risk of VTE compared with dexamethasone alone or melphalan/prednisone (MP). However, the risk of VTE increases from 3–4% to 10–34% when thalidomide is combined with other agents, such as high-dose dexamethasone and/or cytotoxic chemotherapy [182]. As a result, current guidelines recommend that MM patients being treated with an IMiD-containing combination regimen should receive pharmacological thromboprophylaxis (aspirin if standard risk, low-molecular-weight heparin or direct oral anticoagulants if high risk) for as long as the IMiD is being administered [1, 2, 182].

Although evidence is limited, MM patients with disease progression while taking lenalidomide may obtain clinical benefit from thalidomide. In a small Italian trial, among the 20 lenalidomide-refractory subjects who received thalidomide-based salvage therapy, 40% had a paraprotein response [183]. For patients with quadruple refractory disease (i.e., resistance to lenalidomide, pomalidomide, bortezomib, and carfilzomib), clinical trial enrollment is highly recommended if their PS allows. Additionally, some patients may still benefit from ASCT, especially if they previously had a durable response to high-dose melphalan [168, 184]. In this scenario, according to Mayo Stratification for Myeloma and Risk-Adapted Therapy guidelines, one or two cycles of VDT-PACE may be used as salvage therapy if rapid disease control is needed or as a bridge to either clinical trial enrollment or ASCT for suitable candidates [184].

4.2 Other Therapeutic Applications

Besides its role in RRMM, thalidomide is also approved for use in NDMM. During the early 2000s, VAD was long considered a standard induction therapy before ASCT [182]. However, in 2006, an Eastern Cooperative Oncology Group (ECOG)-coordinated study turned TD into the most commonly used induction regimen for TE patients worldwide [185, 186]. With the advent of PIs, the backbone of pre-ASCT induction therapy evolved into a three-drug combination of dexamethasone, a PI, and either cyclophosphamide, doxorubicin, or an IMiD [187]. While VTD became the standard induction regimen in Europe, VD plus lenalidomide (VRD), cyclophosphamide (VCD), or doxorubicin (PAD) have been more commonly used in the US [21, 23]. In phase III RCTs, PAD and VCD had similar efficacy [188], while VTD led to higher CR rates and longer PFS than VCD [189]. A meta-analysis suggested that such greater response is associated with no extra burden of toxicity [190]. Despite limited evidence suggesting a better benefit–risk profile of VRD over that of VTD [191], there is no RCT directly comparing these induction regimens [23].

Recently, the CASSIOPEIA trial reinforced thalidomide use in Europe as part of induction therapy for NDMM. In this landmark study, 1085 patients received four pre-transplant induction and two post-transplant consolidation cycles of either VTD or DaraVTD. Superiority of DaraVTD was demonstrated by the significantly increased post-consolidation sCR rates (29% vs. 20%, p = 0.001) and PFS at 18 months (93% vs. 85%, p < 0.001) [22]. The TOTAL therapy trials established TD-PACE and VTD-PACE as additional primary therapy options for transplant candidates [173, 192]; however, these regimens are generally reserved for patients with clinically aggressive disease [23, 173]. As suggested by recent phase II trials, carfilzomib/thalidomide/dexamethasone alone (CARTHADEX) or in combination with cyclophosphamide (CYKLONE) also appear to be well tolerated, rapidly effective, and well-tolerated induction regimens for TE patients [176, 193]. Importantly, as these regimens do not use bortezomib or lenalidomide, patients may still be treated with such agents if needed during consolidation, maintenance, or salvage therapy [176].

Although single-agent lenalidomide is the first-line choice for maintenance therapy after ASCT, thalidomide is an off-label alternative often used in low-resource healthcare systems [194, 195]. In 2012, an IMWG-coordinated meta-analysis found that employing thalidomide maintenance over an observation-only approach was associated with a 35% reduction in the risk for disease progression [196]. Given the ability of bisphosphonates to hinder tumor cell proliferation and migration, induce apoptosis, block angiogenesis, and stimulate γδ T cells, these agents may synergize with IMiDs to enhance anti-myeloma effects in patients receiving maintenance [197]. Accordingly, one of the studies included in the above-cited meta-analysis was the IFM 99 02 trial, in which 597 patients were randomly assigned to receive no maintenance (arm A), pamidronate (arm B), or thalidomide/pamidronate (arm C) after double ASCT. In arm C, 65% of patients had a CR or VGPR, compared with 55% in arm A and 57% in arm B (p = 0.03). Meanwhile, 3-year EFS estimates were 36% in arm A, 37% in arm B, and 52% in arm C (p < 0.009) [198]. Later, RWD from 14 Korean university hospitals (n = 258) demonstrated a significantly higher 3-year PFS with thalidomide maintenance over observation (55.4 vs. 37.2%; p = 0.005) [195].

Despite the above data, thalidomide has not received FDA or EMA approval as maintenance therapy, since less consistent findings have been reported regarding survival benefits [199, 200]. For instance, the improved OS initially observed in the IFM 99 02 study was not maintained after long-term follow-up of patients with cytogenetics available, with an estimated 5-year OS rate of 74% in the thalidomide/pamidronate arm and 70% in both control groups (p = 0.53) [201]. In 2012, the Medical Research Council (MRC) Myeloma IX trial also showed a lack of survival benefit with thalidomide maintenance, although median PFS was significantly longer in the maintenance arm (23 vs. 15 months; hazard ratio [HR] 1.45, 95% CI 1.22–1.73, p < 0.001). In addition, subgroup analysis revealed a significantly worse OS in patients with high-risk cytogenetic abnormalities [200]. Noteworthy, approximately half of patients randomized to thalidomide maintenance had received the drug during induction, making them prone to the selection of thalidomide-resistant subclones [196, 200]. A subsequent meta-analysis with survival data from MRC Myeloma IX and four other trials demonstrated a significant effect of thalidomide maintenance therapy on OS (p = 0.047) [200]. Furthermore, modeling the survival benefit with effective salvage therapy removed heterogeneity between studies (p = 0.24) and thalidomide maintenance was shown to reduce the risk of death by 25% (p < 0.001) [199, 200].

Finally, thalidomide-based regimens can be used as initial treatment for transplant-ineligible patients [202]. In a meta-analysis of six RCTs comparing MP versus melphalan/prednisone/thalidomide (MPT), the latter regimen provided significant benefits to PFS (HR 0.68, 95% CI 0.61–0.76, p < 0.001) and OS (HR 0.83, 95% CI 0.73–0.94, p = 0.004), extending the median survival time by approximately 20% (6.6 months) [203]. MPT was then established as a standard therapy for non-transplant candidates, despite thalidomide’s unfavorable toxicity profile in elderly individuals receiving the drug for prolonged periods [204]. Such treatment paradigm changed after publication of the FIRST trial, which compared MPT for 12 cycles (n = 547) versus lenalidomide/dexamethasone (RD) for 18 cycles (n = 541) versus RD until disease progression (n = 535). In this phase III study, continuous RD was superior to MPT in regard to PFS (HR 0.72, 95% CI 0.61–0.85, p < 0.001), OS (HR 0.78, 95% CI 0.64–0.96, p = 0.02), and all other secondary endpoints (e.g., ORR, TTR, response duration, time to treatment failure, time to second-line therapy, safety, and HRQoL) [205]. Later, the ECOG EA106 and HOVON87/NMSG18 trials showed no advantages in ORR, PFS, or OS with the use of melphalan/prednisone/lenalidomide induction followed by lenalidomide maintenance (MPR-R) over MPT induction followed by thalidomide maintenance (MPT-T); however, less toxicity was reported in the lenalidomide arm, which translated into better HRQoL [206, 207].

4.3 Role in Patients with Renal Dysfunction

Renal impairment (RI) is present in approximately 20% of MM patients at diagnosis, while 40–50% develop it throughout the disease course [208]. This complication primarily results from the nephrotoxic effects of monoclonal protein, although other factors may contribute (e.g., dehydration, hypercalcemia, hyperuricemia, use of nephrotoxic drugs, hyperviscosity, and PC infiltration) [209, 210]. Currently, the standard management of MM-induced RI combines high-dose dexamethasone with bortezomib, whose efficacy and tolerance are well-established without dose adaptation [211]. However, the addition of thalidomide to bortezomib-based combinations is a commonly used strategy for NDMM with RI, especially in Europe [208, 212]. Thalidomide is primarily metabolized via non-enzymatic hydrolysis and < 1% of unchanged drug is excreted in the urine, causing AE rates to be similar in patients with abnormal or normal kidney function. Therefore, no dose adjustment of thalidomide is needed in the setting of a low creatinine clearance (CrCl) [210, 213, 214]. In contrast, lenalidomide is eliminated predominantly via urinary excretion, such that an abnormal CrCl remarkably impacts its pharmacokinetics [215].

As listed in Table 2, many studies have evaluated the role of thalidomide in the management of MM-related kidney disease [216,217,218,219,220,221,222,223]. Kastritis et al. reported no significant difference in renal function recovery (RFR) rates among NDMM patients treated with non-thalidomide or thalidomide-based regimens (69 vs. 80%, p = 0.453), but median time to RFR was significantly lower in patients receiving thalidomide (2.0 vs. 0.8 months, p = 0.005) [217]. Tosi et al. reported an RFR rate of 55% for patients receiving TD before ASCT, with normal renal function being achieved more frequently by those with a lower degree of RI at baseline (93% if CrCl 30–50 mL/min vs. 19% if CrCl < 30 mL/min, p < 0.001) [218]. Morabito et al. compared bortezomib/melphalan/prednisone (VMP) versus VMP plus thalidomide followed by bortezomib/thalidomide maintenance (VMPT-VT) in renally impaired patients unfit for ASCT, reporting statistically significant improvements in ORR and PFS in those with moderate RI (estimated glomerular filtration rate [eGFR] 31–50 mL/min). Conversely, VMPT-VT had no advantage in terms of RI reversal over VMP (25.4 vs. 40.3%, p = 0.092) [219]. In a retrospective study comparing thalidomide- and lenalidomide-based regimens, the former were associated with a non-significant 2.36-fold increase (95% CI 0.868–6.405, p = 0.092) in the probability of renal PR or CR (as per IMWG criteria for the definition of renal response) [215]. Furthermore, in a secondary cohort analysis of the FIRST trial, the efficacy benefits of RD over MPT could not be demonstrated in patients with severe RI (CrCl < 30 mL/min) [216].

Table 2 Clinical studies evaluating the role of thalidomide-containing regimens on the management of myeloma-related renal impairment
Full size table

Nonetheless, contrasting evidence on thalidomide’s advantages for renally impaired MM patients has been recently published. The OPTIMAL trial demonstrated higher rates of paraprotein response and dialysis independence among newly diagnosed individuals with a CrCl < 30 mL/min who received bortezomib/bendamustine/dexamethasone (VBD) over VTD [217]. In an RWD study, the incidence of RI in MM patients treated with thalidomide was higher than in those treated with lenalidomide [20]. However, this could be due to physician preference for thalidomide use in individuals with a low CrCl at baseline. Thus, thalidomide administration, either as part of frontline bortezomib-based combinations or as a component of second-line therapy following bortezomib refractoriness, remains a useful approach in the setting of MM-related kidney disease [206, 207]. Noteworthy, close monitoring is essential for patients undergoing dialysis, given the possibility of unexpected hyperkalemia [207].

4.4 Role in Patients with Myelotoxicity

In patients with RRMM, the association of heavy PC infiltration in the BM and numerous previous lines of therapy can often pose restrictions to the use of myelosuppressive agents [218]. While thalidomide frequently leads to dose-dependent non-hematologic adverse effects, thalidomide-induced myelotoxicity is unusual, even with doses above 400 mg/day [25]. The rates of neutropenia and thrombocytopenia may be low even when thalidomide is combined with cytotoxic chemotherapy. For instance, in a clinical trial of ThaCyDex for RRMM, 10% of patients developed grade 3–4 neutropenia, which was resolved after reducing the cyclophosphamide dosage. In the same trial, no other hematologic toxicity was attributable to the protocol.

In contrast, myelosuppression is a dose-limiting toxicity for lenalidomide- and pomalidomide-based regimens, despite the significantly lower incidence of PN with these agents [29]. In a retrospective matched-pair analysis comparing TD (n = 183) against RD (n = 228) as initial therapy for NDMM patients, TD led to significantly lower rates of neutropenia (0.6 vs. 14.0%, p < 0.001), thrombocytopenia (0 vs. 4.8%, p = 0.002), and anemia (0 vs. 4.4%, p = 0.003) [219]. In the HOVON87/NMSG18 trial, which included 637 untreated patients not eligible for transplant, grade 3–4 hematologic toxicity was significantly more common in the MPR-R arm (neutropenia: 64 vs. 27%; thrombocytopenia: 30% vs. 8%; anemia: 14 vs. 5%; p < 0.001 for all). Despite leading to higher growth factor support requirements, MPR-R use did not translate into increased infection risk [207].

Although these trials are not directly comparable, it is also interesting to note the lower rates of neutropenia and thrombocytopenia with the CYKLONE regimen (23 and 4.7%, respectively) relative to those reported with bortezomib/dexamethasone/cyclophosphamide/lenalidomide (VDCR) in the EVOLUTION study (44 and 14%, respectively) [176, 224]. In addition, a large US population-based cohort study showed a lack of survival benefit with the use of lenalidomide over thalidomide in upfront MM therapy [20]. Thus, despite not being widely used in the US, thalidomide-based regimens (e.g., TD, ThaCyDex and CYKLONE) remain useful alternatives in heavily pretreated patients with a poor BM reserve. This is particularly true for patients with age ≤ 65 years, little to no PN at baseline, and no prior thromboembolic events [116, 177]. Furthermore, thalidomide stands out as an attractive IMiD alternative when providing care for Jehovah’s Witness patients with MM, since the drug’s negligible myelotoxicity could translate into lower requirements for allogeneic blood product support [225].

4.5 Future Perspectives

In the POLLUX trial, the addition of daratumumab to RD (Dara-RD) was associated with significantly higher ORR and PFS in RRMM patients [226]. This led to the establishment of Dara-RD as a category 1 salvage therapy for early relapse [23]. However, common grade 3–4 AEs with the regimen included neutropenia (51.9%), thrombocytopenia (12.7%), and anemia (12.4%) [226]. In ongoing and future trials, the replacement of lenalidomide by thalidomide in daratumumab-containing regimens could potentially result in lower rates of dose-limiting cytopenias [178, 227]. Furthermore, Dara-TD could be a good option of salvage therapy for patients from resource-constrained countries, where concomitant use of two novel agents (e.g., daratumumab and lenalidomide) may lead to unaffordable healthcare costs [23, 178].

B-cell maturation antigen (BCMA)-directed therapies have substantial anti-myeloma activity but can lead to profound myelosuppression and other AEs (e.g., keratopathy with the ADC belantamab mafodotin [belamaf], cytokine release syndrome and neurotoxicity with CAR T cells) [228, 229]. In the KarMMa trial, grade 3–4 AEs caused by ide-cell included neutropenia (89%), thrombocytopenia (52%), anemia (60%), and infections (22%) [8]. In the CARTITUDE-1 trial, grade 3–4 AEs caused by ciltacabtagene autoleucel (cilta-cel) included neutropenia (95%), thrombocytopenia (60%), anemia (68%), and infections (20%) [230]. CAR T-cell-related hematologic toxicity tends to be short-lived, with frequent recovery to grade ≤ 2 by day 30. Nevertheless, prolonged neutropenia and/or thrombocytopenia can occur in patients with inadequate BM reserve at baseline, creating the risk for life-threatening events (e.g., intracranial bleeding or sepsis) [8, 230]. Above that, some patients with a high tumor burden may lack enough BM reserve to allow autologous T-cell collection via apheresis, hindering their eligibility for CAR T-cell therapy in the first place [231]. Despite the promising preliminary results obtained with BsAbs (e.g., teclistamab, elranatamab, AMG 420, AMG 701, and REGN5458), early-phase clinical trials are still ongoing, for which many patients might not qualify [228, 229].

In the DREAMM-2 trial, the rates of grade 3–4 neutropenia, thrombocytopenia, and anemia varied according to belamaf dosing (9, 20, and 20%, respectively, in the 2.5 mg/kg cohort; 15, 33, and 40%, respectively, in the 3.4 mg/kg cohort) [232]. DREAMM-6 is an ongoing trial evaluating the safety, tolerability, and efficacy of belamaf plus RD (arm A) or VD (arm B) for RRMM patients [233, 234]. In arm B, grade 3–4 thrombocytopenia occurred in 67% of patients, leading to dose reduction in half of these patients [234]. Although preliminary data are still pending for arm A, increased rates of thrombocytopenia compared with arm B are expected, given lenalidomide’s higher myelosuppressive effect [233, 235].

Investigations into other combinations of BCMA-directed agents and IMiDs are somewhat limited. In vivo studies combining CAR T cells with lenalidomide found that the latter could enhance antitumor activity and delay onset of functional exhaustion of the former [236,237,238]. Further preclinical data in xenograft models suggested enhanced potency of the BsAb AMG 701 with the addition of lenalidomide and pomalidomide [239, 240]. Besides boosting the anti-myeloma effects of immunotherapy, the anti-inflammatory properties of IMiDs could be potentially helpful in reducing the risk of CRS after CAR T-cell infusion. Recently, Jan et al. demonstrated the ability of lenalidomide and pomalidomide to control degradable CAR T-cell cytokine release in vivo, limiting treatment-related toxicity while maintaining antitumor activity [241]. Although only lenalidomide and pomalidomide were tested in the above-cited studies, comparable results could potentially be obtained with thalidomide, given its analogous mechanisms of action. Moreover, thalidomide’s lack of renal dosing requirements or myelotoxicity could increase the inclusion of patients with severely decreased eGFR in future clinical trials, while allowing avoidance of dose delays/reductions or treatment discontinuation related to significant cytopenias. In the US, given the considerable economic burden of MM treatment to Medicare, thalidomide-based regimens can also be regarded as cost-effective alternatives to protocols containing newer-generation IMiDs or CELMoDs [242, 243]. In low- and middle-income countries, where many of the novel agents are not only costly but are still awaiting approval, thalidomide use becomes even more attractive [194, 195]. Notwithstanding, there is a persisting need for prospective trials directly evaluating thalidomide-containing regimens against novel therapies or in association with them.

5 Conclusion

Thalidomide remains a useful anti-myeloma agent in clinical practice, especially in the setting of high-grade myelosuppression and RI. Salvage regimens combining thalidomide (usually in a dose of 100 mg/day) and corticosteroids with or without selected cytotoxic or targeted agents (e.g., TD, ThaCyDex and CYKLONE) can be considered for RRMM patients who have no access to or are not eligible for novel therapies. In order to avoid severe AEs, frequent neurological monitoring and pharmacological thromboprophylaxis should be offered. By comparing thalidomide-containing regimens with protocols based on new-generation IMiDs or CELMoDs, and by investigating the association of thalidomide with novel immunotherapies (e.g., ADC, BsAbs, and CAR-T cells), ongoing and future trials may forge new ground for the revival of this old drug.