Abstract
Myelodysplastic syndrome (MDS) with del(5q) is a unique hematopoietic stem cell disease that typically follows an indolent course and demonstrates particular sensitivity to lenalidomide, a second-generation immunomodulatory agent. Early trials demonstrated rapid and durable responses leading to US Food and Drug Administration (FDA) approval in 2005. Definitive confirmatory evidence from a large phase III trial was recently published. Other recent advances include a better understanding of the pathogenesis of disease including haplodeficiency of several candidate genes, and elucidation of the lenalidomide-specific effect on two phosphatases ultimately leading to p53 degradation in the erythroid progenitors and cell cycle arrest in earlier myeloid progenitors. In this review, we describe the pathogenesis of MDS with del(5q), summarize the major clinical studies establishing the activity of lenalidomide in this population, discuss commonly encountered adverse events, and shed light on practical uses of this agent in the clinic.
Introduction
The myelodysplastic syndromes (MDSs) are a heterogeneous group of clonal hematopoietic stem cell disorders characterized by cytologic dysplasia and ineffective hematopoiesis leading to peripheral blood cytopenias. Patients with MDS have a variably increased incidence of progression to acute myeloid leukemia (AML); however, subsets of patients enjoy prolonged survival with a relatively low risk of disease progression. Multiple features offer prognostic utility and are incorporated into risk-stratifying tools including the International Prognostic Scoring System (IPSS) [Greenberg et al. 1997], the World Health Organization (WHO) classification-based Prognostic Scoring System (WPSS) [Malcovati et al. 2007], and other prognostic models [Kantarjian et al. 2008]. Common to all of these scoring systems is the importance of cytogenetics.
Originally described by Van den Berghe and colleagues over 35 years ago as a cytogenetic abnormality and cornerstone of a syndrome characterized by distinct clinical and pathologic features, i.e. the ‘5q-syndrome’ [Van den Berghe et al. 1974], deletion of chromosome 5q (del(5q)) is the most common cytogenetic abnormality in MDS, accounting for approximately 10–15% of all cases [Bernasconi et al. 2005, 2007; Haase et al. 2007]. The WHO recognizes this distinction even now as ‘MDS with isolated del(5q)’, one of seven morphologic categories of MDS. The WHO characterizes it by the following features: isolated del(5q), normal to increased megakaryocytes with hypolobulated nuclei, <5% bone marrow myeloblasts, and the absence of Auer rods. In the peripheral blood, anemia is the hallmark accompanied by a normal or increased platelet count with rare blasts [Vardiman et al. 2009]. With time, almost all patients eventually require transfusion support for the progressive anemia [Giagounidis et al. 2004]. Distinct from other forms of MDS, del(5q) has a female predominance and a perceived low rate of progression to AML in registry data with a corresponding prolonged overall survival [Giagounidis et al. 2004; Greenberg et al. 1997; Patnaik et al. 2010]. More recent prospective observational studies, however, indicate that AML potential in this cytogenetic subset is higher than originally estimated, exceeding 20% at 5 years in some series [Mallo et al. 2011].
First-line management of anemia in patients with lower-risk non-del(5q) MDS generally begins with erythropoietic growth factors, provided that patients have favorable response features based on endogenous serum erythropoietin level and red blood cell transfusion requirements [NCCN, 2011]. Patients with del(5q) MDS have a similar rate of response to recombinant erythropoietins; however, responses are less durable (<12 months) and most patients have excessive elevated endogenous erythropoietin levels [Kelaidi et al. 2008]. Transfusion-dependent patients with del(5q) MDS who are either not candidates for treatment with recombinant erythropoietin or fail to respond to such treatment show particular sensitivity to lenalidomide (CC 5013, Revlimid), an orally bioavailable, structural and functional analog of thalidomide. While its exact mechanism of action in various malignancies may vary, lenalidomide is known to have broad biological effects including immunomodulatory, anti-angiogenic, and cytotoxic properties. It is an active compound in MDS [Fenaux et al. 2011; Le Bras et al. 2011; List et al. 2006a; Sekeres et al. 2011], multiple myeloma [Dimopoulos et al. 2007; Weber et al. 2007], non-Hodgkin lymphoma [Wiernik et al. 2008; Witzig et al. 2009], and various solid tumors [Choueiri et al. 2006; Dahut et al. 2009]. Lenalidomide is much more potent than thalidomide and with a better toxicity profile, especially in regards to neurotoxicity. The drug is 30% bound to plasma proteins and two thirds of the drug is excreted unchanged in the urine within 24 hours with a half-life of elimination of approximately 3 hours [Celgene, 2010]. It is not extensively metabolized by the P450 liver enzymes and thus has a low propensity for important drug interactions [Saloura and Grivas, 2010].
Among lenalidomide’s immunologic actions is altered cytokine production, including suppression of the production of tumor necrosis factor alpha (TNF-α) and interleukin 6 (IL-6) [Muller et al. 1999]. With MDS in particular, stromal abnormalities and immune changes lead to the production of inflammatory and inhibitory cytokines, making this an attractive agent. Lenalidomide also stimulates both CD4+ and CD8+ cells and augments natural killer (NK) cell cytotoxicity. Angiogenesis is thought to contribute to the progression of MDS to higher-risk disease, and lenalidomide decreases the expression of vascular endothelial growth factor (VEGF), a key regulator in this process. Interestingly, while VEGF is overexpressed and secreted by bone marrow blasts in patients with MDS, lenalidomide treatment results in a marked decrease in vascularity with a corresponding reduction in serum VEGF concentration in responding patients [Bellamy et al. 2001].
In this review, we briefly explain the pathogenesis of MDS with del(5q), provide an overview of the major clinical trials establishing the activity of lenalidomide in this population, discuss associated adverse events, and shed insight on practical uses of this agent in the clinic.
Pathogenesis of the 5q-syndrome
While the precise mechanism of disease pathogenesis in the 5q-syndrome is incompletely understood, mounting evidence supports the notion of haploinsufficiency for one or more of the genes located in the common deleted region (CDR). First identified by Boultwood and colleagues, the CDR spans 1.5 megabases at 5q31-32, flanked by the marker D5S413 and GLRA1 genes [Boultwood et al. 2002]. It is distinct from the more proximal CDR located at 5q31 associated with AML or therapy-related MDS. No point mutations or biallelic deletions have been identified among these 44 genes to account for homozygous deletion or gene silencing [Pellagatti et al. 2007]. Expression of the disease phenotype is likely a result of the haplodeficiency of candidate genes, including
MicroRNAs (miRNAs) are small noncoding RNA sequences that bind to specific messenger RNA targets, resulting in repression and gene silencing.
The RPS14 protein cleaves the 18SE/18S rRNA molecule, which is essential for the assembly of 40S ribosomal subunit. Using an RNA interference screen for 40 of the 44 genes contained in the CDR region, Ebert and colleagues showed that partial loss of function of only
Lenalidomide: mechanism of action
In a series of elegant experiments, Pellagatti and colleagues demonstrated that lenalidomide selectively inhibits growth of del(5q) erythroid progenitors
Cell division cycle 25 C (
Mechanism of action of lenalidomide. (A) Lenalidomide directly inhibits
Early efficacy studies
The MDS-001 trial was the first clinical trial evaluating the use of lenalidomide in patients with MDS (Table 1) [List et al. 2005]. A total of 43 patients with MDS and symptomatic anemia, defined as a either a hemoglobin less than 10 g/dl or transfusion dependence requiring at least 4 units of red cells within 8 weeks before enrollment were given one of three treatment schedules of lenalidomide: 25 mg daily, 10 mg daily, or 10 mg daily for 21 out of 28 days. Patients had either failed recombinant erythropoietin or were unlikely to respond to erythropoietin based on high endogenous levels and transfusion frequency. The study excluded patients with neutropenia less than 500 per mm3 or platelets less than 10,000 per mm3. A total of 77% of the patients had either refractory anemia (RA) or refractory anemia with ringed sideroblasts (RARS) and 88% had low or intermediate-1 risk MDS by IPSS; 30% of patients had previously failed treatment with thalidomide.
Summary of relevant clinical trials for lenalidomide in del(5q) myelodysplastic syndrome.
IPSS, International Prognostic Scoring System; FAB, French–American–British; Int-1, intermediate-1; Int-2, intermediate-2; RBC, red blood cell; TI, transfusion independence; DVT, deep vein thrombosis; RA, refractory anemia; RARS, refractory anemia with ringed sideroblasts; RAEB, refractory anemia with excess blasts; RAEB-t, refractory anemia with excess blasts in transformation; CMML, chronic myelomonocytic leukemia.
In the entire cohort, 24 patients (56%) had an erythroid response, including durable transfusion independence (TI) in 20 of 32 patients who were previously transfusion-dependent. The median time to response was 9 weeks in the 25 mg daily group. Most importantly, however, was the finding that 10 of the 12 patients (83%) who had a chromosome 5q deletion had a response compared with 57% of those with normal cytogenetics and 12% of those with other cytogenetic abnormalities (
Building on these results, the MDS-003 registration trial evaluated lenalidomide exclusively in patients with transfusion-dependent, lower-risk del(5q) MDS (Table 1) [List et al. 2006a]. A total of 148 patients were enrolled, all of whom had transfusion-dependent anemia defined as requiring at least 2 units within the 8 weeks prior to enrollment. Overall, 74% of patients had an isolated 5q deletion, but only 27% had the 5q-syndrome. Ninety-five patients (64%) had either RA or RARS, and 120 (81%) had lower-risk MDS by IPSS. Patients were originally given either 10 mg of lenalidomide for 21 days out of a 28 day cycle, but shortly after activation the schedule was amended to 10 mg daily given the apparent faster time to response observed in the MDS-001 trial.
A total of 112 patients (76%) had a response to treatment, including 99 (67%) that became transfusion independent and 13 (9%) that had more than a 50% reduction in transfusion needs. There were no significant differences in responses between the two treatment schedules. Median time to response was 4.6 weeks and responses were durable with maintenance of TI for at least 1 year in 62%. Median increase in hemoglobin in the responders was 5.4 g/dl. A total of 85 patients were evaluable for cytogenetic response, of which 45% had a complete cytogenetic remission. Of those with an isolated 5q deletion, 77% had a cytogenetic response compared with 67% of patients with one additional cytogenetic abnormality and 50% of patients with at least two additional abnormalities, a finding that was not statistically significant [List et al. 2006a]. Among patients with del(5q) treated on both the MDS-001 and MDS-003 trials, complete and partial cytogenetic responders had a significant survival advantage compared with nonresponders and nonevaluable patients (median not reached
In contrast, the European Medicines Agency (EMA) denied approval mostly due to concerns regarding the frequency of AML evolution (as discussed later) and the nonrandomized trial design. Lenalidomide was therefore unavailable to patients in Europe except in the context of a clinical trial or a compassionate use program. The Groupe Francophone des Myélodysplasies (GFM) recently reported the results of their compassionate use program utilizing lenalidomide in 95 patients with a similar profile as those in the MDS-003 study. Patients had IPSS low and Intermediate-1 risk MDS with del(5q) and transfusion-dependent anemia. They found a similar transfusion independence rate of 65% by International Working Group (IWG) 2006 criteria with a median time to TI of 16 weeks. While this appears longer than the median time to TI of 4.6 weeks in the MDS-003 study, hematologic improvement was assessed at 8 week intervals only and started with the absence of transfusion requirement during the previous 8 weeks on lenalidomide. This contrasts with the MDS-003 study, which defined TI starting from the first dose on lenalidomide in MDS-003. Among the 75 patients with isolated del(5q), 49 (65%) achieved TI compared with 7 of 13 patients (54%) with one additional abnormality and 4 of 6 (66%) patients with more than one additional abnormality. Those with baseline platelet counts greater than 150 g/l were also more likely to respond as were those whose platelet counts decreased by at least 50% during the first weeks of treatment, whereas absolute neutrophil count (ANC) decrease had no impact on the probability TI achievement. Although median follow up was relatively short (18.5 months), the median duration of TI was not reached and only 6.3% of patients had progressed to AML [Le Bras et al. 2011].
The phase III trial
Owing to its FDA approval for lower-risk MDS patients with del(5q), lenalidomide became the treatment of choice in the United States for the last several years in this population, but the definitive MDS-004 phase III trial was only recently published (Table 1) [Fenaux et al. 2011]. In this three-arm study conducted throughout Europe, all patients had lower-risk MDS with del(5q) with or without additional cytogenetic abnormalities, and red bold cell (RBC) transfusion-dependent anemia. A total of 205 patients were randomized to lenalidomide 10 mg on days 1–21, lenalidomide 5 mg on days 1–28, or placebo on days 1–28 for each 4-week cycle. Crossover was allowed at 16 weeks if at least a minor erythroid response was not achieved, and all but 11 patients on the placebo arm crossed over to lenalidomide 5 mg. The primary endpoint was RBC transfusion independence for ≥26 weeks and was reached in 56.1%, 42.6%, and 5.9% of patients, respectively (compared with placebo, both
Lenalidomide in other MDS populations
Patients with transfusion dependent, lower-risk MDS without del(5q) was studied in the MDS-002 trial [Raza et al. 2008]. A total of 214 patients received therapy, of which major erythroid response was achieved in 33% of patients by IWG 2006 criteria. There were few cytogenetic responders, suggesting that the effect of lenalidomide was to restore erythropoiesis in the existing MDS clone [Raza et al. 2008]. Use of erythropoiesis-stimulating agents (ESAs) prior to initiation of lenalidomide was not well-documented, but another small study in patients who were refractory to ESAs found that 42% of patients had an erythroid response with lenalidomide treatment, and 42% of the transfusion-dependent patients became transfusion independent [Sibon et al. 2010]. The frequency of major erythroid response using lenalidomide in combination with epoetin alpha is the subject of an ongoing phase III clinical trial.
As of now, the use of lenalidomide in higher-risk MDS patients remains investigational, however, preliminary data indicate that response rates are inferior using the approved dose and schedule. A multicenter phase II trial evaluated the safety and efficacy of lenalidomide in 47 higher-risk MDS patients with 5q deletion, reporting a 27% response rate, including 7 complete remissions (CRs). Most responses were rapid, but duration of response was only 6.5 months. Patients with an isolated del(5q) were more likely to respond compared with those with one additional abnormality and those with two or more additional abnormalities. Of the 47 patients enrolled, 18 (38%) had RAEB-t (now considered AML by WHO criteria), and while 25 patients had previously received therapy for MDS, none had received azanucleosides [Ades et al. 2009]. Further dose escalation may improve response rate and investigation continues in this population [Mollgard et al. 2011].
Combination therapy with azacitidine has been studied in phase I trials. One study treated 18 patients with higher-risk MDS patients, including one who had del(5q) MDS. The maximum tolerated dose of combined azacitidine and lenalidomide treatment was not reached and 12 (67%) patients responded to therapy, including 8 (44%) CRs [Sekeres et al. 2010]. At the American Society of Hematology meeting in 2010, a similar phase I study was presented, but limited the patient population to those with chromosome 5 abnormalities and allowed patients with both high-risk MDS as well as AML. Thus far, only results for the first eight patients are available, of which four had MDS. No maximum tolerated dose has been reached and after a median of two cycles, one patient had rapidly progressive disease, one achieved a marrow CR, and the other six patients had stable disease [Platzbecker et al. 2010].
Adverse events
The most common adverse events seen in the clinical studies were neutropenia and thrombocytopenia, with reported frequencies as high as 75% and 44%, respectively [Fenaux et al. 2011; Le Bras et al. 2011; List et al. 2005, 2006a]. These generally occur within the first two cycles and often necessitate interruption of treatment or dose reduction, but subsequently subside. Early development of thrombocytopenia is associated with a higher response rate in lower-risk del(5q) MDS and may act as a surrogate for suppression of the malignant clone. In a multivariate analysis of all patients enrolled in the MDS-002 and MDS-003 studies, the development of thrombocytopenia (≥50% decrease from baseline) and neutropenia (≥75% decrease from baseline) was associated with a higher probability of transfusion independence in del(5q) patients, but not in non-del(5q) patients [Sekeres et al. 2008]. The correlation between the development of thrombocytopenia and response was confirmed in the GFM experience, but this was not seen with neutropenia, possibly owing to the effect of granulocyte colony stimulating factor (G-CSF) [Le Bras et al. 2011].
The use of G-CSF to prevent dose reduction can be considered when the neutrophil count falls below 1000 × 109/l, but this did not reduce the proportion of patients that required dose reduction in the GFM study [Le Bras et al. 2011] and has yet to be investigated systematically. Temporary interruption of therapy can also be considered until neutrophils increase to above 750 × 109/l, at which time a lower dose can be reinitiated. For patients who develop thrombocytopenia less than 25 × 109/l while on lenalidomide, therapy should be held and reinitiated at a lower dose when the platelets rise above 50 × 109/l. However, some experts disagree and do not recommend lenalidomide with a platelet count of less than 50 × 109/l owing to the risk of sustained thrombocytopenia [Kurtin and Sokol, 2006]. Alternatively, there are limited data to support the use of alternate day dosing with 10 mg on days 1–21 to reduce hematologic toxicity, but this may delay time to transfusion independence and response rate, given that the comparable cumulative dose using 5 mg daily was inferior in the MDS-004 trial [Defina et al. 2010]. Regardless of baseline values, blood counts should be monitored weekly for the first 8 weeks, then monthly thereafter.
Aside from the hematologic events, lenalidomide is generally well-tolerated. Common nonhematologic adverse events include pruritus (42%), rash (36%), diarrhea (49%), muscle cramps (21.6%) and fatigue (31%) [Celgene, 2010]. Rare but important side effects include hypothyroidism and hypogonadism, both of which require hormone replacement [List et al. 2006a].
Dry skin, rash, and pruritus are common and are often self-limiting during early weeks of treatment, requiring no specific intervention. When severely symptomatic, they can generally be managed with emolients and antihistamines, and on rare occasions, a short course of local or systemic corticosteroids. Itching of the scalp is common in the first weeks of therapy with subsequent spontaneous resolution. Diarrhea can affect nearly half of all patients treated with lenalidomide [Celgene, 2010] and can have an impact on quality of life. This is often managed with antimotility agents such as loperamide, but may require brief interruption of therapy in more severe cases. Patients with known lactose deficiency should add lactase to their diet, as lenalidomide capsules contain small amounts of lactose.
While deep venous thromboses (DVTs) are a well-known adverse event associated with the use of lenalidomide with dexamethasone in multiple myeloma [Dimopoulos et al. 2007; Rajkumar et al. 2010; Weber et al. 2007; Zonder et al. 2010], its incidence appears to be relatively low in patients with MDS. Among the 491 patients treated in the GFM, MDS-001, MDS-003, and MDS-004 study, a total of 17 (3.4%) DVTs or pulmonary emboli were reported [Fenaux et al. 2011; Le Bras et al. 2011; List et al. 2005, 2006a]. These generally occur in the first few months of therapy [Fenaux et al. 2011; Le Bras et al. 2011], and lenalidomide treatment should be interrupted until stable anticoagulation is achieved, then reinitiated cautiously. Prophylaxis with low-molecular-weight heparin should only be used for patients with a history of DVTs, and aspirin is not recommended due to the high incidence of thrombocytopenia [Kurtin and Sokol, 2006].
One of the principal concerns that has arisen over the last several years is the incidence of increased clonal evolution and progression to AML in patients treated with lenalidomide [Eclache et al. 2008; Gohring et al. 2008; List et al. 2006a]. In the MDS-003 trial, 16 patients progressed to a more advanced French–American–British (FAB) MDS subtype or AML and 24 developed new chromosomal abnormalities during the course of treatment [List et al. 2006a]. With long-term follow up (median of over 3 years) of 42 patients treated on this trial, 15 patients (36%) progressed to AML and 17 (40%) had karyotypic evolution. With the exception of one patient, all of these patients died within several months of AML diagnosis [Gohring et al. 2010]. Patients with an erythroid or cytogenetic response carried a much lower risk of progression than those without a response. With a cytogenetic response, the 3- and 5-year cumulative incidence of AML were 10% and 21% respectively, compared with 46% and 60% in those without a cytogenetic response. Most of those who progressed accumulated additional chromosomal abnormalities, thus genetic instability and clonal evolution appear to be the driving forces [Gohring et al. 2010]. The selective activity of lenalidomide for deletion 5q clones may inadvertently give a selective advantage to pre-existing clones that were present in low levels [Cazzola, 2008]. TP53 mutations were recently reported to be associated with lower rates of response to lenalidomide and a higher risk of leukemic transformation [Jadersten et al. 2011]. Shorter telomere length was also recently associated with increased risk of AML in a small study [Gohring et al. 2011b].
Historically, patients with <5% bone marrow blasts and isolated del(5q) have a risk of AML transformation approaching only 9%, but this increases to as high as 80% for those with more than 5% bone marrow blasts [Giagounidis et al. 2004]. However, patients in this historical group are not entirely comparable with those in the lenalidomide-treated group, as some patients enrolled in these trials had additional cytogenetic abnormalities and all were transfusion dependent, a known poor prognostic indicator. Placebo-controlled, randomized data with long-term follow up are not available and thus it cannot be excluded that this apparent tendency towards leukemic transformation is actually a reflection of the natural disease course. In the phase III trial, with a median follow up of over 30 months in each group, 52 patients (25.4%) progressed to AML. Only 11 patients were randomized to placebo and never received lenalidomide, and 4 of these patients progressed (36.4%), compared with 30.4% of those who received placebo and crossed over to lenalidomide 5 mg, 23.2% in the lenalidomide 5 mg group, and 21.7% in the 10 mg group [Fenaux et al. 2011]. Given that almost every patient in the placebo group crossed over to the treatment group after just 16 weeks, whether lenalidomide increases the risk of AML progression remains unclear. No formal guidelines currently exist for monitoring cytogenetics in patients on lenalidomide, but it appears that karyotyping is much more sensitive than fluorescent
Other considerations
As thalidomide has a long history of birth defects that led to its withdrawal from the market in the 1960s, lenalidomide use is highly restricted. Teratogenicity has not been reported with lenalidomide, but women in their childbearing years must have two negative pregnancy tests within 14 days prior to the initiation of therapy. They must agree to concomitant use of two forms of birth control at all times during therapy as well as 4 weeks before and after therapy.
Patients with significant renal impairment were excluded from all trials with this agent, and no formal studies have assessed the impact of renal function on the use of lenalidomide. However, risk of toxicity is expected to be greater in patients with renal impairment, and this must be taken into account especially due to the elderly demographic of the disease. The following adjustments in dosing have been proposed and are recommended by a consensus panel: 5 mg daily for patients with a creatinine clearance (CLcr) between 30 and 50 ml/min, 5 mg every other day for those with a CLcr less than 30 ml/min and not required dialysis, and 5 mg three times a week for those requiring dialysis [Kurtin and Sokol, 2006; List et al. 2009].
Summary
Lenalidomide is the treatment of choice for patients with lower-risk MDS with del(5q) and transfusion-dependent anemia, leading to rapid, robust, and durable responses in the majority of patients. Definitive evidence of its superiority was recently confirmed in a large, placebo- controlled phase III trial. Myelosuppression requiring dose reduction/interruption is the most frequent event but correlates with improved response. Future directions include a better understanding of the underlying mechanism of action of lenalidomide, addressing combination strategies, refinement of its role in higher risk MDS and novel potential strategies to prevent emergence of drug resistance.
Footnotes
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
VHD has no conflicts of interest to declare. RSK has received research support from Celgene and serves on the speakers bureau. AFL has received research support from Celgene as a principal investigator. He also serves as a consultant, has received honoraria from speaking, and serves on their scientific advisory board.