Oral 9-cis retinoic acid (Alitretinoin) in the treatment of myelodysplastic syndromes: results from a pilot study

Abstract

A multicenter phase II study was initiated to investigate the efficacy, toxicity and tolerability of an oral regimen of 9-cis retinoic acid (9CRA) as a differentiation-inducing agent stimulating both retinoic acid receptor (RAR) and retinoic X receptor (RXR). Thirty patients with myelodysplastic syndromes (MDS) were enrolled into the study. The MDS subtypes were distributed as follows: 14 refractory anaemia (RA), four refractory anaemia with ringed sideroblasts (RARS), and 12 refractory anaemia with excess blasts (RAEB). The age ranged from 40 to 81 years (median 70). None of these had previously received treatment for MDS other than supportive therapy. 9CRA (Alitretinoin capsules, kindly provided by Allergan-Ligand Retinoid Therapeutics) was given daily at 60 mg/m² p.o. for 1 week, followed by an intra-patient escalation to 100 mg/m² during the second week, up to a maximum of 140 mg/m². The planned treatment duration was 48 weeks. Twenty-five were available for assessment. One patient (4%) with RA achieved complete hematological remission. Four (16%), two with RA, two with RAEB, had minor responses resulting in decreased transfusion requirements or increased neutrophils. Thus, the overall response rate was 20% in evaluable patients with MDS and 17% in the study group on an intention-to-treat basis. The most frequent side-effects included headache (77%), dry skin (57%), arthralgias (30%), and rash (23%). In conclusion, although modest responses were noted in this study, the treatment tolerability was suboptimal. It is conceivable that a lower dosage schedule may be efficacious and better tolerated so enabling prolonged exposure which may be required to induce a differentiation effect. Leukemia (2000) 14, 1583–1588.

Introduction

Retinoic acids are non-specific differentiation inducing agents which have profound effects on normal and leukemic cells in vitro.1234 Retinoic acid enhances the cytokine-induced colony formation of both granulocyte–macrophage and erythroid progenitor cells. Furthermore, in vitro studies showed that retinoids inhibit clonal growth of human leukemic cells and induce their differentiation.56 The retinoid response is mediated by nuclear receptors, the retinoic acid receptor (RAR) and the retinoic X receptor (RXR).5 RAR was originally described as the receptor for all-trans retinoic acid (ATRA),7 and RARα is one of the genes disrupted by the t(15;17) translocation in acute promyelocytic leukemia (APL).89 In 1990, a new receptor, RXR, was described.10 This receptor forms heterodimers that increase the DNA-binding affinity of RAR and other nuclear receptors.11

Myelodysplastic syndromes (MDS) are clonal disorders characterized initially by ineffective hematopoiesis and the frequent development of acute leukemia.121314 The underlying mechanisms of ineffective hematopoiesis in primary MDS have not yet been established, but may involve suppression by inhibitory cytokines,15 enhanced spontaneous apoptosis,1216 and defective responses to growth factors due to aberrations of growth factor receptors17 or post-receptor signal transduction pathways.18 In addition, there is a block in cell differentiation of all three hematopoietic lineages.192021

The clinical effects of some retinoic acid compounds in MDS patients are known from studies with 13-cis retinoic acid (13CRA) and ATRA (as reviewed by Morosetti and Koeffler).22 In several studies 10% to 25% of the patients showed response to treatment with retinoids as single agent. To improve and extend the treatment results, a number of studies using 13CRA and ATRA in combination with hematopoietic growth factors, or other differentiation inducing agents, such as vitamin D₃ and cytoreductive agents, such as low-dose cytarabine, were initiated,2324252627 and the results have shown response rates varying from 15 to 45%. The experiences using combination treatment have been reviewed by Ganser and Karthaus28 and Santini and Ferrini.29

In 1992, 9-cis retinoic acid (9CRA) was identified as being the natural ligand of RXR.3031 Unlike ATRA, which only binds RAR, 9CRA binds and activates both RAR and RXR.3233 Preclinical studies have suggested that 9CRA may have a greater potency than ATRA in its ability to induce differentiation and to inhibit proliferation of leukemic cells.3435

Recently, two clinical phase I studies investigated the optimal dose scheduling of 9CRA in patients with advanced cancer.3637 Based on the analysis of toxicity and side-effects of 9CRA, doses from 140 to 230 mg/m²/day were recommended for single agent trials. In patients with APL, 9CRA, used as single agent therapy at dosages from 30 to 230 mg/m²/day, was able to induce complete remissions (CR) in four of 12 relapsed patients.38

On the basis of these in vitro and in vivo findings, the main goal of the present study was to assess 9CRA as a potential component of treatment of MDS patients, with particular respect to hematopoiesis, transfusion requirements and toxicity.

Patients and methods

Patients

Thirty patients with MDS from six centers (15 men, 15 women) with a median age of 70 years (range 40–81) were enrolled into the study. According to the French–American–British cooperative group (FAB) criteria for myelodysplastic syndromes,39 14 patients had refractory anaemia (RA), four patients RA with ringed sideroblasts (RARS), and 12 patients RA with excess blasts (RAEB). One patient had secondary RA to previous radiation. The patients with RAEB were not candidates for bone marrow transplantation due to advanced age. Patients were also classified according to the International Prognostic Scoring System (IPSS).40 Five patients had low risk disease, 17 were in the intermediate-1 group and eight in the intermediate-2 group. There were no high risk group patients. Eligibility criteria included a performance status of more than 50% on the Karnofsky scale, life expectancy of more than 3 months, normal hepatic, renal, cardiac function, normal coagulation function, and the absence of clinically apparent systemic illness. No women of child-bearing potential were included in the trial. None of the patients had previously received any treatment other than supportive therapy. Details of patient characteristics are given in Table 1. The study was approved by the local ethics committees, and informed written consent was obtained from the patients prior to treatment with 9CRA.

Table 1  Patient characteristics prior to treatment with 9-cis retinoic, clinical responses and side-effects during treatment with 9-cis retinoic acid

Study design

9-Cis retinoic acid (Alitretinoin) was provided by Allergan-Ligand Retinoid Therapeutics (Bucks, UK) and was supplied as 25 mg or 60 mg soft gelatine capsules. The capsules were given daily at a dose of 60 mg/m² p.o. for 1 week, followed by an intra-patient escalation to 100 mg/m² during the second week if tolerability permitted, up to a maximum of 140 mg/m². Patients who did not tolerate the initial protocol dose of 60 mg/m²/day despite prophylaxis for headache had a dose reduction to their individual maximum tolerated dose. The planned treatment duration was 48 weeks. During the study, patients were monitored weekly for the first 4 weeks and 4-weekly thereafter. Each visit included a complete medical history, physical examination, and laboratory tests, including a complete blood count, differential count, a chemistry profile, coagulation profile, and urinalysis. An electrocardiogram and chest X-ray were performed before the study. Bone marrow aspirate and trephine biopsy were obtained before, after 6 months of therapy, and at the end of treatment. The bone marrow was evaluated morphologically and cytogenetically.

In patients who received treatment with 9CRA at least for 6 weeks the response was assessed on the basis of peripheral blood parameters. In addition, examination of bone marrow was performed at the end of treatment. Dose limiting toxicity was defined as a toxicity of grade 3 or higher by National Cancer Institute (NCI) criteria (Table 2). Red blood cell (RBC) concentrates were administered when the hemoglobin level dropped below 8 g/dl or if anemia-related symptoms occurred. Platelet transfusions were only given in cases of bleeding. Antibiotics and antipyretics were prescribed when the body temperature rose to more than 38°C. Emollients were prescribed if patients complained of dry skin, and the patients were instructed to drink more than 2 liters water a day.

Table 2  Adverse events observed with 9-cis retinoic acid in MDS patients

Results

Treatment compliance

The median duration of treatment in this study was 13 weeks (range 1–48). Twenty-five patients (83%) received 9CRA for 6 weeks or more and were assessable for hematological response. Four patients received the study medication for the planned time of 48 weeks. The therapy was stopped early in three patients (one RA, one RARS, one RAEB) following transformation of MDS to acute myeloid leukemia (AML). The median time to transformation to AML from start of the study medication was 21 weeks. Two patients died during the study treatment period, one due to myocardial infarction, and the other from intestinal obstruction (Table 1).

Treatment response

Complete remissions:

One patient with RA (UIN304, IPSS Int-1), who was transfusion dependent with a frequency of 2 units of RBC per week, achieved complete remission with normalization of peripheral blood cells counts. After 4 weeks of treatment no transfusions were necessary, and after 12 weeks the hemoglobin was 12.1 g/dl without blood product support. The platelet count increased from 79 × 10⁹/l to 200 × 10⁹/l after a treatment period of 12 weeks. The bone marrow appearance before any treatment with study medication was characteristic for RA with hypercellularity, dysplastic erythropoiesis and micromegakaryocytes, whereas the results after 44 weeks of treatment with 9CRA showed a normal cellularity, with only minor dysplastic changes. The CR was ongoing for the duration of 32 weeks. Unfortunately, in week 44 of treatment with 9CRA and during the phase of continuous hematological complete remission the patient developed acute pancreatitis and died.

Minor responses:

Four patients (two RA, two RAEB) had minor responses. Two patients with RA (UIN401, IPSS Low and UIN504, IPSS Int-1) had an increase in hemoglobin from 7.2 g/dl to 9.2 g/dl and 8.5 g/dl to 9.7 g/dl, respectively. In patient 401 RBC substitution could be stopped whereas in patient 504 this was associated with a significant reduction in transfusion requirement from 3 units of RBC monthly before the trial to a mean of 1 unit monthly from week 8 of trial medication. One patient with RAEB (UIN509, Int-1) showed a significant increase in the absolute neutrophil count (ANC), from 1 × 10⁹/l prestudy to 4 × 10⁹/l after 8 weeks treatment. One further patient with RAEB (UIN201, Int-2) had a reduction in the number of blasts in the peripheral blood associated with an increase in the platelet count from 108 × 10⁹/l prestudy to 172 × 10⁹/l after 8 weeks of 9CRA (Table 3). The response to 9CRA was not associated with particular clinical or laboratory findings in any of these patients.

Table 3  Responses to 9-cis retinoic acid in MDS patients

Toxicity

Headache (NCI grade 1–3) was the main side-effect and occurred in 23 out of the 30 MDS patients treated with 9CRA (Table 2). This was the primary reason that treatment was stopped early in four patients. Sixteen patients developed NCI grade 1–2 dermatoses, mainly dryness of the skin. One patient suffered from severe dry skin syndrome NCI grade 3. Arthralgia (NCI grade 1–3) was reported by nine patients and seven patients reported rash. The serum triglyceride levels were elevated in four patients. In two patients the treatment was stopped early due to symptomatic hypercalcaemia which was associated with acute psychotic syndrome in one patient.

Discussion

The aim of the present pilot study was to determine the toxicity, safety and efficacy of the recently available retinoic receptor agonist 9-cis retinoic acid in patients with myelodysplastic syndromes. A number of clinical studies using retinoids such as 13-cis retinoic acid4142 and all-trans retinoic acid4344 in patients with MDS have shown only a limited efficacy of about 10–25% and the treatment was associated with dose-dependent side-effects. In studies of patients with RA and RARS the response to retinoids was enhanced when combined with other differentiation agents or hematopoietic growth factors2324262745 to an overall response rate of about 30%.

Preclinical studies have suggested that 9CRA, which stimulates both retinoic acid receptor and retinoic X receptor, has greater potency than ATRA in its ability to induce differentiation of normal hematopoietic cells and clonal cells from AML cell lines and from patients with AML.3546 Treatment with 9CRA was associated with a number of adverse reactions such as headache and dry skin syndrome which was reflected in previous studies using 9CRA at doses ranging from 5 to 230 mg/m²/day36 which also showed that hypercalcemia was an important limitation to dose escalation beyond 140 mg/m². For these reasons our study initiated treatment at a dosage of 60 mg/m²/day followed by dose escalation to 100 mg/m²/day to a maximum of 140 mg/m²/day.

Only four patients received the study drug for the planned period of 48 weeks, two of whom received reduced dosages of 15 mg/m²/day or 30 mg/m²/day. The highest dose level of 140 mg/m²/day was reached only in five patients for a median treatment period of 8 weeks. In 11 patients the dosage was reduced to a level below the starting dose of 60 mg/m²/day to allow continuous treatment for a median duration of 28 weeks. These results demonstrate that the proposed dose escalation was associated with a high rate of adverse events, which resulted in a decrease in the number of patients who finished the length of treatment planned in the study.

The present study reports an overall response rate of 20% which is in keeping with previously published studies of retinoids in MDS.22 One patient achieved complete remission while receiving a dose of 9CRA lower than originally intended. In addition one patient classified as a minor responder during the 48 weeks study period achieved CR after treatment with 9CRA for 30 months at the dose level of 140 mg/m². Minor responses were also seen at low doses of trial drug while other patients failed to respond even to high doses. It appears that there is no clear relationship between drug dose and response to treatment.

The physiological mechanisms of cell differentiation and the results from a number of clinical studies suggest that differentiation induced by retinoic acid may occur after a long latency period of up to 6 months in vivo.2747 We assume that the assessment of efficacy of 9CRA in MDS in this study does not reflect the potential of this agent to stimulate cell differentiation. It remains an open question whether 9CRA is an effective treatment that improves hematopoiesis in MDS patients. As demonstrated in our pilot study, it is possible to induce complete remission in one patient with RA using 9CRA at a dose tolerable for these individuals. Furthermore, minor responses were observed. All of these patients received the treatment for more than 12 weeks although one patient with RAEB showed benefit after only 8 weeks of treatment suggesting that a continuous exposure to 9CRA for several weeks is necessary for clinical relevant improvement of hematopoiesis. A treatment schedule using tolerable doses for a longer treatment period (more than 12 weeks) may be of more benefit than poorly tolerated schedules using higher dosages for a short duration. Tolerability may be improved with the combination of retinoid with alpha-tocopherol42 allowing higher dosing schedules. Further trials are required to answer these questions.

In conclusion, although modest responses were noted in this study, the treatment tolerability was suboptimal. It is conceivable that a lower dosage schedule may be efficacious and better tolerated so enabling prolonged exposure which may be required to induce a differentiation effect. This approach would also be necessary when 9CRA is combined with other agents.