Does addition of erythropoiesis stimulating agents improve the outcome of higher-risk myelodysplastic syndromes treated with azacitidine?

Leukemia Research 36 (2012) 397–400

Contents lists available at SciVerse ScienceDirect

Leukemia Research journal homepage: www.elsevier.com/locate/leukres

Does addition of erythropoiesis stimulating agents improve the outcome of higher-risk myelodysplastic syndromes treated with azacitidine?夽 R. Itzykson a , S. Thépot a , O. Beyne-Rauzy b , S. Ame c , F. Isnard d , F. Dreyfus e , C. Salanoubat f , A.L. Taksin g , Y. Chelgoum h , C. Berthon i , J.V. Malfuson j , L. Legros k , N. Vey l , P. Turlure m , C. Gardin a , S. Boehrer a , L. Ades a , P. Fenaux a,∗ , on behalf of the Groupe Francophone des Myelodysplasies (GFM) a

Service d’Hématogie Clinique Hôpital Avicenne, Assistance Publique-Hôpitaux de Paris (AP-HP), and Paris 13 University, France Service de Médecine Interne, Centre Hospitalier Universitaire Toulouse, France c Service d’Hémato-oncologie, Centre Hospitalier Universitaire Strasbourg, France d Service d’Hématologie, Hôpital Saint-Antoine, AP-HP and Paris 6 University, Paris, France e Service d’Hématologie Clinique, Hôpital Cochin, AP-HP and Paris 5 University, France f Service d’Hématologie, Centre Hospitalier du Sud Francilien, Corbeil, France g Service d’Hémato-oncologie, Hôpital André Mignot, Le Chesnay, France h Service d’Hématologie, Hôpital Edouard Herriot, Lyon, France i Service des Maladies du Sang, Centre Hospitalier Universitaire (CHU) Lille, France j Service d’Hématologie, Hôpital d’Instructions des Armées Percy, Clamart, France k Service d’Hématologie, Centre Hospitalier Universitaire Nice, France l Département d’Hématologie, Institut Paoli Calmettes, Marseille, France m Service d’Hématologie Clinique, Centre Hospitalier Universitaire Limoges, France b

a r t i c l e

i n f o

Article history: Received 27 September 2011 Received in revised form 10 November 2011 Accepted 24 November 2011 Available online 15 December 2011 Keywords: Myelodysplastic syndromes Hypomethylating agents Erythropoiesis stimulating agents

a b s t r a c t We studied a retrospective cohort of 282 higher-risk MDS treated with azacitidine, including 32 patients who concomitantly received an ESA for a median of 5.8 months after azacitidine onset. Forty-four percent of ESA and 29% of no-ESA patients reached HI-E (p = 0.07); 48% and 20% achieved transfusion independence (p = 0.01). Median OS was 19.6 months in the ESA and 11.9 months in the no-ESA groups (p = 0.04). Addition of an ESA significantly improved OS (p = 0.03) independently of azacitidine schedule and duration, and of our proposed azacitidine risk score (Blood 2011;117:403–11). Adding an ESA to azacitidine in higher-risk MDS should be studied prospectively. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Erythropoiesis stimulating agents (ESAs) yield erythroid responses (HI-E according to International Working Group, IWG) [1,2] in 30–50% of lower-risk (IPSS low or intermediate-1) myelodysplastic syndromes (MDS) and may improve overall survival (OS) compared to red blood cell (RBC) transfusion therapy alone [3–6]. On the other hand, higher-risk (IPSS int-2 or high) MDS have poorer responses to ESAs (24% according to IWG 2006

夽 Results presented in part at The 11th International Symposium on Myelodysplastic Syndromes, Edinburgh, UK, May 18–21, 2011. ∗ Corresponding author at: Service d’Hématogie Clinique Hôpital Avicenne, Assistance-Publique des Hôpitaux de Paris, Université Paris 13, INSERM U848, 125 route de Stalingrad, 93009 Bobigny, France. Tel.: +33 1489 57050; fax: +33 1489 55499. E-mail address: [email protected] (P. Fenaux). 0145-2126/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.leukres.2011.11.019

criteria [2] in our experience [6]), but obtain significant survival improvement with the hypomethylating agent azacitidine (AZA) [7]. Recently, we presented interim results of a randomized phase II study suggesting that combining an ESA to azacitidine in lower-risk MDS resistant to an ESA alone may improve HI-E rates compared to azacitidine alone [8]. The impact of ESA addition to azacitidine has not to our knowledge been addressed in higher-risk MDS. We tried to analyze it retrospectively in our recently reported patient named program of azacitidine in higher risk MDS, where a fraction of patients concomitantly received an ESA [9]. 2. Patients and methods We recently reported a retrospective cohort of 282 higher-risk MDS patients who received azacitidine in a patient-named compassionate program [9]. This cohort included patients with IPSS

0.4

0.6

0.8

ESA no−ESA

0.2 0.0

intermediate-2 or high risk MDS (including AML with 20–30% of bone marrow blasts, previously considered as refractory anemia with blast excess in transformation [RAEB-t]), not previously treated by intensive chemotherapy or allogeneic stem cell transplantation. In this cohort, concomitant administration of an ESA during azacitidine therapy was at the clinician’s discretion, without any specific recommendations (including regarding ESA duration) other than standard recommendations regarding ESA dose in MDS. Responses, including HI-E, were assessed according to IWG 2006 criteria [2]. Comparisons between patients receiving an ESA (“ESA” group) or not (“no-ESA” group) were performed with the Mann–Whitney and Fisher’s exact tests for continuous and dichotomic variables, respectively. Overall survival was plotted with the Kaplan–Meier method and compared with the log-rank test. Multivariate analyses used logistic regression and Cox models for prognostic factors of response and OS, respectively. All analyses were performed with the Statview (SAS, Cary, NC, USA) and R 2.10.1 softwares.

1.0

R. Itzykson et al. / Leukemia Research 36 (2012) 397–400

Cumulative Probability of Survival

398

32

29

25

14

6

2

2

239

181

115

69

35

21

14

24

30

36

0

3. Results Among the 282 higher-risk MDS patients, 43 received an ESA during azacitidine treatment; 11 patients started ESA after azacitidine onset, because of worsening anemia, and were excluded from the present analysis, the remaining 239 patients constituting the “no ESA” group. Of note, including those 11 patients in the “no-ESA” group did not affect our conclusions. The “ESA” group (n = 32) included 10 patients having started ESA concomitantly to azacitidine onset, and 22 having started ESA before azacitidine. Onset of ESA was made less than 3 months before azacitidine in 13 cases and more than 3 months in 9 cases (median 6.3 months, range 3.5–18 months). ESA consisted of darbepoietin (n = 21), epoietin beta (n = 9) and epoietin alpha (n = 2), using doses adapted to MDS (i.e. darbepoitin ≥ 150 ␮g/week or epoietin ≥ 30,000 U/week) in all 32 cases. There was no statistically significant difference in outcome measures between patients receiving darbepoitin and those receiving epoietins. Median duration of ESA treatment after onset of AZA was 5.8 months (range 1.6–19 months). Only 4 of the 32 patients received less than 12 weeks of ESA. Baseline characteristics at onset of azacitidine including age, gender, WHO diagnosis, BM blast percentage, proportion of therapy-related MDS, prior MDS duration, IPSS cytogenetic risk, IPSS at AZA onset, and azacitidine risk group (defined according to our recent “score” based on red blood cell transfusion dependence [RBC TD], performance status, peripheral blood blasts and cytogenetics [9]) were all comparable in the two groups (Table 1; all p > 0.2). IPSS at diagnosis was low or intermediate1 in 32% of ESA patients and 20% of no-ESA patients (p = 0.23). At the onset of azacitidine, the frequency and severity of anemia was similar in both groups: 27 patients (84%) in the ESA group and 207 patients (87%) in the no-ESA group, respectively, had Hb < 10 g/dL and/or RBC TD and were thus evaluable for HI-E at AZA onset (p = 0.19). Five ESA patients with Hb > 10 g/dL and no RBC TD had responded to ESA prior to azacitidine introduction and were not evaluable for HI-E. Twenty-one and 157 patients (both 65%) of patients had RBC TD, respectively (p = 0.9), and TD was ≥4 RBC U/8 weeks in 50% of ESA pts versus 45% of no-ESA patients (p = 0.70). Fifty-three percent of ESA patients and 69% of no-ESA patients received the FDA/EMEA AZA schedule (75 mg/m2 , 7 days/cycle), while the remaining received reduced azacitidine schedules, mostly 75 mg/m2 for 5 days/cycle (p = 0.10 between ESA and no-ESA groups). The median number of azacitidine cycles was 7 (2–21) and 6 (1–39) in ESA and no-ESA patients, respectively (p = 0.05).

6

12

18

ESA 6

42

1

no−ESA

48

Months Fig. 1. Kaplan–Meier overall survival plots of patients from the ESA (full line) and no-ESA (dotted line) groups.

The overall response rate was 17/32 (53%) in ESA patients, including 6 CR (19%), 4 marrow CR (13%, and 7 stable disease with HI (22%), compared to 103/239 (43%), including 31 CR (13%), 9 PR (4%), 28 marrow CR (12%), and 35 stable disease with HI (14%) in patients not receiving ESA (p = 0.34). Among patients evaluable for HI-E, 12 of 27 ESA patients (44%) and 55 of 207 no-ESA patients (27%) reached HI-E (p = 0.07). Among patients with RBC transfusion dependence, 10 of 21 ESA patients (48%) achieved RBC TI, compared to 33 of 157 no-ESA patients (21%) (p = 0.01, Table 2). Of note, the ESA group included 18 patients with a previous history of ESA resistance, who were evaluable for HI-E at the onset of azacitidine and were concomitantly rechallenged with ESA; 8 of those 18 patients achieved HI-E, and 4 reached RBC TI. Since there was a trend for more frequent use of reduced azacitidine schedules and longer exposure to azacitidine in ESA patients compared to no-ESA patients, we performed a multivariate logistic regression including ESA treatment, azacitidine schedule and number of cycles, to check that the higher improvement rates of ESA patients was not attributable to a longer exposure to azacitidine or to the use of reduced schedules. In such a model, the trend for superior HI-E and the significant improvement in RBC TI in ESA patients were conserved, with an odds ratio (OR) of 2.23 (95% confidence interval [CI]: 0.91–5.62, p = 0.077) for the achievement of HI-E, and of 3.14 (95% CI: 1.11–8.91, p = 0.03) for RBC TI. Finally, in a multivariate model including normal karyotype and BM blasts >15% (according to our previous report [9], the trend for superior RBC TI was still found in patients of the ESA group (OR = 2.70, 95% CI: 0.97–7.56, p = 0.056). Of note, IPSS at diagnosis (low/int-1 versus int-2/high) did not significantly influence the rate of HI-E or RBC TI with azacitidine (p > 0.2). With a median follow-up of 26 months, median durations of HI-E and of RBC TI were 9.8 and 10.1 months, respectively, not influenced by addition of ESA (p = 0.35 and p = 0.38, respectively). Time to progression (according to IWG 2006) was similar in ESA and non-ESA patients (p = 0.65). However, overall survival (OS) was significantly more prolonged in ESA patients (median 19.6 months) compared to no-ESA patients (11.9 months, log-rank test: p = 0.04, Fig. 1). In a Cox model including azacitidine schedule, number of azacitidine cycles received, risk group according to our recently published azacitidine risk score [9] and ESA use or not, addition of ESA significantly improved OS (hazard ratio: 0.60

R. Itzykson et al. / Leukemia Research 36 (2012) 397–400

399

Table 1 Baseline patient characteristics. ESA

a

no ESA

32 72 (35–87) 10 (31%) 5 (16%) 5.2 (0–45.9)

Patient number Age, years (median, range) Gender, female Therapy-related MDS Prior MDS duration, months (median, range) WHO diagnosis RA/RARS/RCMD RAEB1 RAEB2 AML (RAEB-t) Marrow blast % (median, range) Cytogenetic risk (IPSS) Good Intermediate Poor Not available IPPS risk at AZA onset Intermediate High NA (int-2 or high) Performance status ≥ 2 RBC TD ≥ 4 U/8 weeks ANC < 1.0 g/L Platelets < 100 g/L Prior ESA therapy Azacitidine risk groupa Low Intermediate High Not available Azacidine schedule FDA/EMEA approved (75 mg/m2 /day, 7 days/cycle) Reduced doses Number of AZA cycles (median, range)

p

239 71 (20–89) 96 (40%) 62 (26%) 4.9 (0–142.6)

1 (3%) 8 (25%) 18 (56%) 5 (16%) 14 (1–30)

11 (5%) 44 (18%) 127 (53%) 57 (24%) 15 (1–30)

10 (31%) 8 (25%) 12 (37%) 0 (0%)

75 (31%) 36 (15%) 115 (48%) 13 (6%)

20 (62%) 12 (38%) 0 (0%) 6 (19%) 16 (50%) 19 (59%) 25 (78%) 22 (69%)

127 (53%) 105 (44%) 7 (3%) 47 (20%) 111 (46%) 125 (52%) 179 (75%) 64 (27%)

2 (6%) 26 (81%) 4 (13%) 0

27 (11%) 156 (65%) 43 (17%) 13 (5%)

17 (53%) 15 (47%) 7 (2–21)

165 (69%) 74 (31%) 6 (1–39)

0.97 0.44 0.27 0.60 0.63

0.30 0.52

0.44

0.99 0.85 0.57 0.83 <0.0001 0.33

0.10

0.05

According to [9].

Table 2 Response to azacitidine.

Overall response to azacitidinea Complete response (CR) Partial response (PR) Marrow CR (mCR) Stable disease with HI HI-E (IWG 2006 criteria)b Median HI-E duration, months RBC TIc Median RBC TI duration, months

ESA

no ESA

p

17/32 (53%) 6/32 (19%) 0/32 (0%) 4/32 (13%) 7 (22%) 12/27 (44%) 8.3 (3.0–24.2+) 10/21 (48%) 8.3 (3.0–24.2+)

103/239 (43%) 31/239 (13%) 9/239 (4%) 28/239 (12%) 35/239 (14%) 55/207 (27%) 9.8 (2.0–283+) 33/157 (21%) 10.3 (2.4–28.3+)

0.34

0.07 0.35 0.01 0.38

HI, hematological improvement. a CR + PR + mCR + SD with HI. b Eligible patients: Hb < 10 g/dL and/or RBC transfusion dependence. c Eligible patients: RBC transfusion dependence.

[95% confidence interval: 0.37–0.96], p = 0.03) independently of azacitidine risk group, schedule and duration of exposure. Of note, 14 of the 32 patients in the ESA group concomitantly received GCSF, but addition of G-CSF did not seem to improve overall response rate, HI-E, RBC-TI, response duration or overall survival (all p > 0.2). 4. Discussion Our results suggest that patients receiving an ESA concomitantly to AZA tend to have higher rates of erythroid hematological improvement, may achieve significantly more often RBC transfusion independence, and may have significantly improved overall survival compared to patients receiving azacitidine alone. Our analysis obviously warrants caution, as it is retrospective, in a patient cohort where treatment with ESA was made at the clinician’s discretion. However, in order to limit biases, the 11 patients in our

cohort who initiated ESA therapy after onset of azacitidine were excluded from our analysis, as these patients received ESA because they had incomplete response to azacitidine. Furthermore, comparison of baseline (prior to azacitidine) characteristics of patients from the remaining ESA and no-ESA groups did not reveal any major difference. In particular, the incidence and severity of anemia and RBC transfusion dependence, cytogenetics and BM blast percentage, the azacidine risk score we previously reported to predict overall survival in those patients (and that takes into account cytogenetic risk, RBC transfusion dependence, performance status and peripheral blasts [9]) were similar in both groups. In addition, high (≥4 U/8 weeks) RBC transfusion dependence before onset of azacitidine, an independent prognostic factor of survival in higher-risk MDS patients treated with azacitidine [9], but also an important predictive factor of response to ESAs [3,4,6] was found at similar incidence in both the ESA and no ESA groups. Serum EPO level,

400

R. Itzykson et al. / Leukemia Research 36 (2012) 397–400

which also predicts ESA response [3,4,6], was not available in our patients. However, in the published GFM experience of ESAs in MDS, only 11% of higher-risk MDS patients had low endogeneous ESA levels [6], and it is unlikely that an imbalance in serum EPO levels could have been sufficient to explain the response differences observed between the two groups. The overall response rate of 53% achieved by ESA patients appeared higher than the 43% rate of the “no ESA” patients, but this difference did not reach statistical significance. In our cohort, addition of ESA tended to improve the rate of HI-E and, more importantly, significantly improved the rate of transfusion independence, but as expected did not affect other responses, nor delay time to progression or prolong erythroid response. Yet, addition of ESA appeared to prolong overall survival independently of azaciditine risk group, schedule and duration of exposure. Though the differences were not statistically significant, a greater proportion of ESA patients received a reduced azacitidine schedule, and the duration of exposure to azacitidine was longer in those patients. The latter probably reflected the trend for superior overall response and survival in those patients. We nonetheless verified that the trend for superior HI-E and improved RBC TI rates and OS were independent of these features, by performing multivariate analyses. Several studies have demonstrated that G-CSF synergizes with ESAs to achieve HI-E in lower risk MDS [10]. Though clearly limited by small numbers, our analysis did not find any obvious synergism in terms of response achievement, duration, or overall survival by adding G-CSF to an ESA in higher risk MDS patients receiving azacitidine. An important issue is whether addition of azacitidine can revert ESA resistance. Our cohort included a limited number of patients in this situation (n = 18), all having progressed to higher risk MDS at AZA onset, but this question is being prospectively addressed in an ongoing clinical trial of the Groupe Francophone des Myelodysplasies (GFM) in lower risk MDS [8]. Correction of anemia appears to be associated to improve overall survival in higher-risk MDS patients treated with azacitidine [9,11]. However, given the overall poor prognosis of those patients, it is unlikely that correction of anemia by itself can directly improve survival, for instance by reducing cardiovascular morbidity or iron overload, as proposed for lower risk patients treated with ESA [4,6]. Rather, it has been suggested that achievement of hematological improvement with azacitidine reflects the regression of MDS to a less advanced stage. In this hypothesis, erythroid response would represent a marker of better prognosis, rather than a mechanism of survival improvement per se. Our findings of improved erythroid response and survival with addition of ESA in higher risk MDS treated with azacitidine are therefore somewhat unexpected, and raise the provocative hypothesis that improvement of erythroid responses by adding an ESA may be sufficient to prolong survival in higher risk MDS. The mechanisms underlying this survival improvement are elusive, and could involve reduction of nonleukemic death through improvement of anemia, or even effects of ESAs other than on the erythroid lineage [12], which however remain poorly characterized in MDS. Since our results are retrospective, based on a non-randomized comparison, and because the number of patients receiving the ESA–AZA combination was relatively limited, it would be of interest to verify these findings in other cohorts, before testing them in prospective clinical studies in higher risk MDS. Until then, use of ESAs in MDS should be restricted to patients with lower risk disease where they have been shown to be safe [4,6], and according to current recommendations [13].

Conflict of interest statement PF: Celgene: Honoraria and Research Funding. All other authors have no conflict of interest to disclose.

Funding No funding to declare. Acknowledgements The following contributors enrolled patients in the French azacitidine compassionate patient name program from the mentioned centers (all in France): C. Recher (Toulouse), C. Dartigeas (Tours), J. Delaunay (Nantes), S. Visanica (Metz), A. Stamatoullas (Rouen), A. Marfaing-Koka (Clamart), S. De Botton (Villejuif, Gustave Roussy Institute), I. Plantier (Roubaix), G. Damaj (Amiens), O. De Renzis (Clermont-Ferrand), P. Morel (Lens), E. Wattel (Lyon), A. GuerciBresler (Nancy), L. Sanhes (Perpignan), P. Cony-Makhoul (Annecy), A. Banos (Bayonne), S. Cheze (Caen), K. Ghomari (Beauvais), G. Tertian (Kremlin-Bicêtre), S. Harel (Paris, Necker Hospital), L. Voillat (Chalons), W. Abarah (Meaux), B. Delmas-Marsalet (Villejuif, Paul Brousse Hospital), C. Himberlin (Reims), and C. Sohn (Toulon). The following contributors collected clinical data from all centers: F. Chermat and B. Beve (Bobigny), R. Sapena (Paris, Cochin Hospital), and E. Mortera (Toulouse). Contributions. RI and ST collected the data. RI performed statistical analysis and drafted the manuscript. LA and PF designed the study and wrote the manuscript. The other authors enrolled the patients and collected the data. All authors approved the final version of the manuscript. References [1] Cheson BD, Bennett JM, Kantarjian H, Pinto A, Schiffer CA, Nimer SD, et al. Report of an international working group to standardize response criteria for myelodysplastic syndromes. Blood 2000;96(12):3671–4. [2] Cheson BD, Greenberg PL, Bennett JM, Lowenberg B, Wijermans PW, Nimer SD, et al. Clinical application and proposal for modification of the International Working Group (IWG) response criteria in myelodysplasia. Blood 2006;108(2):419–25. [3] Hellstrom-Lindberg E, Gulbrandsen N, Lindberg G, Ahlgren T, Dahl IM, Dybedal I, et al. A validated decision model for treating the anaemia of myelodysplastic syndromes with erythropoietin + granulocyte colony-stimulating factor: significant effects on quality of life. Br J Haematol 2003;120(6):1037–46. [4] Jadersten M, Malcovati L, Dybedal I, Della Porta MG, Invernizzi R, Montgomery SM, et al. Erythropoietin and granulocyte-colony stimulating factor treatment associated with improved survival in myelodysplastic syndrome. J Clin Oncol 2008;26(21):3607–13. [5] Jadersten M, Montgomery SM, Dybedal I, Porwit-MacDonald A, HellstromLindberg E. Long-term outcome of treatment of anemia in MDS with erythropoietin and G-CSF. Blood 2005;106(3):803–11. [6] Park S, Grabar S, Kelaidi C, Beyne-Rauzy O, Picard F, Bardet V, et al. Predictive factors of response and survival in myelodysplastic syndrome treated with erythropoietin and G-CSF: the GFM experience. Blood 2008;111(2):574–82. [7] Fenaux P, Mufti GJ, Hellstrom-Lindberg E, Santini V, Finelli C, Giagounidis A, et al. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet Oncol 2009;10(3):223–32. [8] Boehrer S, Beyne-Rauzy O, Prebet T, Park S, Guerci A, Stamatoulas A, et al. Interim results of a randomized phase II trial of azacitidine (AZA) +/− EPO in lower risk myelodysplastic syndrome (MDS) resistant to an erythropoietic stimulating agent (ESA) alone. Blood (ASH Annual Meeting Abstracts) 2010;116:1880. [9] Itzykson R, Thepot S, Quesnel B, Dreyfus F, Beyne-Rauzy O, Turlure P, et al. Prognostic factors for response and overall survival in 282 patients with higher-risk myelodysplastic syndromes treated with azacitidine. Blood 2011;117(2):403–11. [10] Hellstrom-Lindberg E, Malcovati L. Supportive care, growth factors, and new therapies in myelodysplastic syndromes. Blood Rev 2008;22(2):75–91. [11] List AF, Fenaux P, Mufti GJ, Hellstrom-Lindberg E, Gore S, Bennett JM, et al. Effect of azacitidine (AZA) on overall survival in higher-risk myelodysplastic syndromes (MDS) without complete remission. In: ASCO Annual Meeting Abstracts. 2008. p. 7006. [12] Arcasoy MO. The non-haematopoietic biological effects of erythropoietin. Br J Haematol 2008;141(1):14–31. [13] Rizzo JD, Brouwers M, Hurley P, Seidenfeld J, Arcasoy MO, Spivak JL, et al. American Society of Hematology/American Society of Clinical Oncology clinical practice guideline update on the use of epoetin and darbepoetin in adult patients with cancer. Blood 2010;116(20):4045–59.