Sie sind auf Seite 1von 6

European Journal of Haematology 86 (512516)

ORIGINAL ARTICLE

Lack of efficacy of pyridoxine (vitamin B6) treatment in


acquired idiopathic sideroblastic anaemia, including
refractory anaemia with ring sideroblasts
Lisa M. Baumann Kreuziger1,2, Alexandra P. Wolanskyj1, Curtis A. Hanson1, David P. Steensma3
1
Mayo Clinic, Rochester, MN; 2Department of Medicine, University of Minnesota, Minneapolis, MN; 3Dana-Farber Cancer Institute, Boston, MA,
USA

Abstract
Pyridoxine, or vitamin B6, is commonly used to treat acquired idiopathic sideroblastic anaemia (AISA,
including refractory anaemia with ring sideroblasts), but the efficacy of this therapy in an unselected AISA
population (i.e. patients without confirmed ALAS2 or other pyridoxine-responsive germline mutations) has
not been established. We reviewed clinical data from 231 patients with AISA and found that 42% of 203
evaluable patients had been treated with pyridoxine. Only 6.8% of pyridoxine-treated patients experienced
a haemoglobin improvement (1.5 g dL) meeting 2006 International Working Group for Myelodysplastic
Syndromes standardised response criteria. As some patients received combination therapy with erythro-
poietin or other agents, improvement could be attributed to pyridoxine monotherapy in only one patient
(1.4%). Smaller, less meaningful increments in haemoglobin levels of 0.5 g dL were observed in 13.5% of
patients. Response to therapy did not correlate with International Prognostic Scoring System (IPSS) risk
group or multilineage vs unilineage dysplasia. New symptomatic peripheral neuropathy was noted in 2.3%
of patients treated with pyridoxine. In this large series of unselected patients with sideroblastic anaemia,
pyridoxine therapy was ineffective and was associated with a risk of adverse effects. Pyridoxine therapy
should be reserved for patients with known or suspected pyridoxine-responsive mutations.

Key words pyridoxine; vitamin B6; sideroblastic anaemia; myelodysplastic syndrome; refractory anaemia with ring sideroblasts

Correspondence David P. Steensma, MD FACP, Associate Professor of Medicine, Harvard Medical School, Division of Hematologic
Oncology, Dana-Farber Cancer Institute, 44 Binney St, Suite D1B30, Boston, MA 02115, USA. Tel: +1 617 582 6410; Fax: +1 617 582
7840; e-mail: David_Steensma@dfci.harvard.edu

Accepted for publication 4 March 2011 doi:10.1111/j.1600-0609.2011.01604.x

Sideroblastic anaemias are a heterogeneous group of dis- The 2001 World Health Organization (WHO) classifi-
orders that have impaired haem synthesis and the pres- cation of MDS separated sideroblastic anaemias (defined
ence of pathognomonic bone marrow ring sideroblasts in by 15% ring sideroblasts and <5% marrow blasts) into
common (1). Sideroblastic anaemias can be either two categories, depending on whether dysplastic changes
acquired or hereditary due to congenital (germline) are limited to the erythroid lineage or whether multilin-
mutations in factors critical to iron processing and haem eage dysplasia is present. The former were designated as
biosynthesis. The molecular pathobiology of acquired refractory anaemia with ringed sideroblasts a term also
sideroblastic anaemias is poorly understood. Most cases used by the French-American-British classification of
of hereditary sideroblastic anaemias are X-linked, but MDS in 1982 (3) while the latter were termed refrac-
autosomal dominant and recessive forms have been tory cytopenia with multilineage dysplasia associated
described (2). Acquired sideroblastic anaemias include with ring sideroblasts (RCMD-RS). In 2008, RCMD-RS
toxin- or medication-mediated forms, vitamin deficien- was merged with RCMD due to a lack of any prognostic
cies, or acquired idiopathic sideroblastic anaemia value to the presence or absence of ring sideroblasts
(AISA), which is commonly considered a form of myelo- when multilineage dysplasia is present (4, 5). It remains
dysplastic syndrome (MDS). unclear the extent to which similar pathophysiological

512 2011 John Wiley & Sons A/S


Baumann Kreuziger et al. Lack of efficacy of pyridoxine (vitamin B6) treatment

mechanisms exist between the hereditary and various bin increased by >0.5 mg dL on at least two
acquired sideroblastic anaemias; mutations described in measurements 4 weeks apart. The relationship between
congenital cases have been found only rarely in AISA, receiving and response to pyridoxine and disease sub-
including refractory anaemia with ring sideroblasts type or International Prognostic Scoring System (IPSS)
(RARS) and RCMD-RS (6). stratification was assessed using chi-squared test, using
Pyridoxine, a form of vitamin B6, plays a critical role a P-value limit of <0.05 for statistical significance. A
in haem synthesis as a cofactor for d-aminolevulinic acid Fischers exact and two sample t-test were used to com-
synthetase (ALAS). ALAS catalyses the first step in pare receiving pyridoxine and gender and age at diag-
haem biosynthesis when succinyl CoA and glycine are nosis respectively. Statistical analysis was completed
combined into d-aminoleuvulinic acid. The gene encoding using JMP 8.0 (SAS Institute, Cary, North Carolina,
the form of ALAS expressed in erythroid cells, ALAS2, USA). The study was approved by the Mayo Clinic
is localised to chromosome Xp11.21. Mutations of Institutional Review Board, and all patients had con-
ALAS2 have been found in patients with X-linked side- sented to review of their records.
roblastic anaemia. Some ALAS2 mutations confer
responsiveness to pyridoxine therapy at doses ranging
Results
from 5 to 500 mg day (7). Pyridoxine deficiency can be
seen in patients who abuse alcohol, which may be due to Of the 231 patients with AISA, treatment and follow-up
decreased intake or inhibited absorption of the vitamin information was available from 203 (111 patients with
(8). Treatment with pyridoxine antagonists such as isoni- RARS and 92 with RCMD-RS). Baseline demographics
azid has also resulted in sideroblastic anaemia (9). are noted in Table 1.
Lastly, patients with AISA have been shown to have Forty two percent of patients (n = 86) were treated
reduced levels of ALAS activity, and treatment with pyr- with pyridoxine for an average of 19 months (range 1
idoxine has increased the ALAS activity and ameliorated 114 months) at a mean dose of 167 mg day (range 50
the haematological abnormalities in some cases (10). 600 mg day). Of the 86 patients who received pyridox-
Despite anecdotal reports of improvement of AISA ine, data regarding response to therapy were available
cases with pyridoxine (11, 12, 13) that have led to for 86% of patients (n = 74). Only five patients (6.8%)
widespread clinical use of this agent for patients with met IWG response criteria for haematological improve-
RARS and RCMD-RS, to our knowledge, there are no ment (Table 2). The average haemoglobin increase was
systematic studies on the effectiveness of pyridoxine in 2.6 mg dL (1.63.9). An additional cohort of five
AISA, including RARS and RCMD-RS. Moreover, patients [total n = 10 (13.5%)] experienced a minimal
considerable time is often expended while waiting for a
response and some patients experience adverse events
such as neuropathy (14). To assess the effectiveness of Table 1 Baseline characteristics of patients with acquired idiopathic
sideroblastic anaemia according to pyridoxine therapy exposure
pyridoxine to treat AISA, we performed a retrospective
case review of pyridoxine use and outcomes at our Did not
institution. Received receive
pyridoxine pyridoxine
N = 86 (42%) N = 117 (58%) P-value

Patients and methods Mean age at diagnosis 69.9 (4489) 73.0 (4490) 0.01
in years (range)
We reviewed clinical data from 231 adult patients with Male n (%) 54 (63) 82 (70) 0.29
biopsy-proven acquired sideroblastic anaemia evaluated WHO classification
at Mayo Clinic between 1994 and 2007. Cases possibly RARS n (%) 59 (69) 52 (44) 0.0006
related to drug toxin exposure and pyridoxine defi- RCMD-RS n (%) 27 (31) 65 (56)
ciency were excluded. Diseases were classified as RARS IPSS stratification*
or RCMD-RS based on WHO criteria (15, 16). Data Low n (%) 55 (64) 58 (50) 0.17
Intermediate-1 n (%) 28 (33) 40 (34)
collected included haematological parameters, the dose
Intermediate-2 n (%) 3 (3) 18 (15)
and duration of pyridoxine therapy, and use of con- High n (%) 0 (0) 0 (0)
comitant medications. Responses were assessed using
2006 International Working Group (IWG) standardised WHO, World Health Organization; RARS, refractory anaemia with ring
criteria for MDS (17). As a limited number of patients sideroblasts; RCMD-RS, refractory cytopenia with multilineage dyspla-
sia and with ring sideroblasts; IPSS, International Prognosis Scoring
met response criteria for haematological improvement
System; N or n, number; %, percentage of patients within the
during pyridoxine therapy, a less stringent standard was column.
also applied: patients were considered to have a mini- *IPSS classification could not be completed on one patient who did
mal response to pyridoxine treatment if their haemoglo- not receive pyridoxine.

2011 John Wiley & Sons A/S 513


Lack of efficacy of pyridoxine (vitamin B6) treatment Baumann Kreuziger et al.

Table 2 Baseline characteristics of patients with acquired idiopathic sideroblastic anaemia according to response to pyridoxine therapy

Patients with Hbg increase >1.5 mg dL


(n = 5)
Patients Pyridoxine + Pyridoxine
Patients without with Hbg increase additional therapy monotherapy
response (n = 64) >0.5 mg dL (n = 10) (n = 4) (n = 1)

Mean age at diagnosis in years (range) 71.7 (4689) 68.2 (5076) 65 (5073) 67
Male n (%) 39 (61) 5 (50) 3 (75) 1 (100)
WHO classification
RARS n (%) 43 (67) 6 (60) 2 (50) 1 (100)
RCMD-RS n (%) 21 (33) 4 (40) 2 (50) 0
IPSS stratification
Low n (%) 40 (63) 6 (60) 2 (50) 1 (100)
Intermediate-1 n (%) 20 (31) 4 (40) 2 (50) 0
Intermediate-2 n (%) 4 (6) 0 0 0

WHO, World Health Organization; RARS, refractory anaemia with ring sideroblasts; RCMD-RS, refractory cytopenia with multilineage dysplasia
and with ring sideroblasts; IPSS, International Prognosis Scoring System; %, percentage of patients within the column.
Comparison between patients who did not respond and patients in either response group did not yield any statistically significant P-values.

response to therapy (increase in haemoglobin by group compared to 16.7% (4 24) of patients in the Inter-
>0.5 mg dL). mediate-1 risk group (P = 0.66) responded to pyridox-
Half of the patients who had increases in haemoglobin ine. Of the four patients with Intermediate-2 risk disease,
by >0.5 mg dL with pyridoxine therapy were female. none responded to pyridoxine therapy.
The average mean corpuscular volume (MCV) of Adverse effects associated with pyridoxine administra-
patients with minimal response to pyridoxine was tion were also assessed. Two patients treated with pyri-
105.8 fL (range 96.1113.6) and IWG response was doxine developed new onset peripheral neuropathy while
104.0 fL (96.1111.6). Of the five patients meeting IWG on therapy (2.3%, 2 86 patients). Extensive evaluation
response criteria, three were started on erythropoietin for alternative causes of these patients neuropathy was
and one was started on prednisone concomitantly with unrevealing. Neither patient responded to pyridoxine
pyridoxine therapy. Therefore, only 1 (1.4%) patients therapy with an increase in haemoglobin. The first
improvement in haemoglobin could be attributed to pyri- patient took 200 mg of pyridoxine daily, whereas dosage
doxine therapy alone. This patient was male and had information was not available for the second patient.
macrocytosis (MCV 111 fL) at diagnosis. He responded The patients received pyridoxine for 18 and 60 months
with an increase in haemoglobin within 1 month of ther- respectively at which point the diagnosis of peripheral
apy and maintained response for 2.5 years until his death neuropathy was made and therapy was discontinued.
from unrelated causes. The four other patients meeting The first patient expired within 1 month of diagnosis of
IWG response criteria responded within 26 months of peripheral neuropathy due to acute respiratory distress
starting combination therapy. The dose of pyridoxine syndrome and the second patients neuropathy had not
was not associated with response to therapy as the mean improved after 5 years of follow-up. No other adverse
dose of patients who responded was 187.5 mg daily com- effects were recognised and attributable to pyridoxine
pared to 157 mg daily in patients who did not respond therapy.
(P = 0.60).
Patients with RARS were more likely to be treated
with pyridoxine compared with patients with RCMD-RS Discussion
(59 111 vs. 27 92, RARS vs RCMD-RS, P = 0.0006), Pyridoxine is commonly used to treat patients with
but response to pyridoxine did not differ significantly AISA. Pyridoxine therapy is unique to sideroblastic
between subtypes (3 49 vs. 2 25, IWG response in anaemia; a review of 120 patients with other subtypes of
RARS vs. RCMD-RS; P = 0.76) (6 49 vs. 4 25, mini- MDS seen at our institution found that none was treated
mal response in RARS vs. RCMD-RS; P = 0.66). Of with pyridoxine (unpublished work). Advantages to pyri-
patients in the low IPSS risk group, 6.5% (3 46) doxine therapy include that it is inexpensive and treat-
responded to pyridoxine compared to 8.3% (2 24) of ment is generally well tolerated. Our data demonstrate
patients in the Intermediate-1 risk group (P = 0.82). that pyridoxine is an ineffective treatment for AISA as
When considering a minimal haemoglobin increase of only a small minority of patients (1.4%) responded to
>0.5 mg dL, 13% (6 46) of patients in low-risk IPSS pyridoxine monotherapy.

514 2011 John Wiley & Sons A/S


Baumann Kreuziger et al. Lack of efficacy of pyridoxine (vitamin B6) treatment

The response rates seen in our study are significantly responding patients. Determining if pyridoxine delayed
less than previously published studies. MacGibbon and the onset of transfusion dependence or altered transfu-
Mollin reported a series of 35 patients with primary side- sion requirements would require a prospective study,
roblastic anaemia in which 27 (77%) were treated with which is unlikely to be cost-effective given the observed
pyridoxine (12). A 37% haematological response was minimal response rate in our study.
reported (haemoglobin increase not defined). Four of the Overall, this large case series demonstrates that pyri-
patients included in this series were infants with likely doxine is an ineffective therapy for AISA. The genetics
X-linked sideroblastic anaemia, but it remains unclear if of AISA need to be more clearly defined, and it is not
these patients were included in the efficacy analysis. Time evident that every AISA is MDS (20). Until molecular
to response was not noted, whereas other case reports analysis can further alter our classification schemes, pyri-
have described haematological improvement within doxine therapy should be limited to patients with known
25 months, similar to our study (11, 13). or suspected mutations that confer responsiveness, or
Most patients with X-linked sideroblastic anaemia due alternatively can be considered in low-risk MDS patients
to ALAS2 mutations have microcytic erythrocyte indices for a brief therapeutic trial.
and present in childhood. Patients up to 83 years of age
have been identified with the disease, presumably due to
Acknowledgement
unfavorable X-inactivation and perhaps restricted clonal-
ity of haematopoiesis with ageing (18). The variable age We thank Darrell Schroeder for advice regarding statisti-
of onset could lead to misdiagnosis as acquired siderob- cal analysis. This study was funded by a grant from the
lastic anaemia. The ALAS2 gene mutation status was Robert A. Kyle Hematologic Malignancies Fund.
not known for most patients in our series because few
underwent molecular analysis for hereditary sideroblastic Conflict of interest
anaemia for clinical indications; however, molecular
analysis of DNA samples from 60 patients from this The authors have no potential conflicts of interest to
cohort did not disclose any mutations [(6) and unpub- report.
lished work]. Unfortunately, DNA from the patient who
responded to pyridoxine monotherapy was not available References
for ALAS 2 mutation testing. However, the patients who 1. Alcindor T, Bridges KR. Sideroblastic anaemias. British
responded to pyridoxine combined or as monotherapyin Journal of Haematology 2002;116:73343.
our series had normocytic or macrocytic anaemia, which 2. Camaschella C. Hereditary sideroblastic anemias: patho-
would argue against misclassification. physiology, diagnosis, and treatment. Semin Hematol
Pyridoxine therapy not only has limited efficacy, but it 2009;46:3717.
can also be associated with harm, as seen in two patients 3. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton
in this series. Use of pyridoxine may delay initiation of DA, Gralnick HR, Sultan C. Proposals for the classifica-
more effective therapies such as recombinant erythropoi- tion of the myelodysplastic syndromes. British Journal of
etin, to which 1525% of patients with RARS will Haematology 1982;51:18999.
respond (19), or a hypomethylating agent. In patients 4. Vardiman JW, Thiele J, Arber DA, Brunning RD, Boro-
with a low IPSS sideroblastic anaemia, however, a lim- witz MJ, Porwit A, Harris NL, Le Beau MM, Hellstrom-
ited trial of pyridoxine (26 months) could be considered Lindberg E, Tefferi A, Bloomfield CD. The 2008 revision
of the World Health Organization (WHO) classification of
if treatment with other therapies would not otherwise be
myeloid neoplasms and acute leukemia: rationale and
started. In addition to delaying effective therapy, pyri-
important changes. Blood 2009;114:93751.
doxine use confers a small risk of neuropathy, which
5. Steensma DP, Hanson CA, Tefferi A. Indepedent Valida-
may be higher in patients with underlying medical com-
tion of the 2008 World Health Organization (WHO)
orbidities like diabetes. As peripheral neuropathy second-
Reclassification of Myelodysplastic Syndromes (MDS)
ary to pyridoxine has been reported in the literature Associated with Ring Sideroblasts. SanFrancisco, CA:
including our study in patients on therapy >1 year, a American Society of Hematology, 2008.
limited trial in low-risk patients would have a decreased 6. Steensma DP, Hecksel KA, Porcher JC, Lasho TL. Candi-
risk of harm. date gene mutation analysis in idiopathic acquired side-
The retrospective nature of the study meant that treat- roblastic anemia (refractory anemia with ringed
ment was not standardised and limited the outcomes that sideroblasts). Leuk Res 2007;31:6238.
could be assessed. Patients were prescribed a variable 7. Cotter PD, Rucknagel DL, Bishop DF. X-linked siderob-
dose and duration of pyridoxine. However, the dose of lastic anemia: identification of the mutation in the ery-
pyridoxine did not significantly differ between those few throid-specific delta-aminolevulinate synthase gene
patients whose haemoglobin level improved and non-

2011 John Wiley & Sons A/S 515


Lack of efficacy of pyridoxine (vitamin B6) treatment Baumann Kreuziger et al.

(ALAS2) in the original family described by Cooley. Blood 15. Bennett JM. World Health Organization classification
1994;84:391524. of the acute leukemias and myelodysplastic syndrome.
8. Middleton HM 3rd. Intestinal hydrolysis of pyridoxal Int J Hematol 2000;72:1313.
5-phosphate in vitro and in vivo in the rat: effect of 16. Jaffe ES, Harris NL, Stein H, Vardiman JW. World
ethanol. Am J Clin Nutr 1986;43:37481. Health Organization Classification of Tumours: Pathology
9. Verwilghen R, Reybrouck G, Callens L, Cosemans J. and Genetics of Tumours of Haematopoietic and Lymphoid
Antituberculous drugs and sideroblastic anaemia. British Tissues. Lyon, France: IARC Press, 2001.
Journal of Haematology 1965;11:928. 17. Cheson BD, Greenberg PL, Bennett JM, Lowenberg B,
10. Meier PJ, Giger U, Brandli O, Fehr J. [Acquired, vitamin Wijermans PW, Nimer SD, et al. Clinical application and
B6-responsive, primary sideroblastic anemia, an enzyme proposal for modification of the International Working
deficiency in heme synthesis]. Schweiz Med Wochenschr Group (IWG) response criteria in myelodysplasia. Blood
1981;111:15335. 2006;108:41925.
11. Vuylsteke J, Verloop MC, Drogendijk AC. Favourable 18. Cotter PD, May A, Fitzsimons EJ, Houston T, Woodcock
effect of pyridoxine and ascorbic acid in a patient with BE, al-Sabah AI, Wong L, Bishop DF. Late-onset
refractory sideroblastic anaemia and haemochromatosis. X-linked sideroblastic anemia. Missense mutations in the
Acta Medica Scandinavica 1961;169:11323. erythroid delta-aminolevulinate synthase (ALAS2) gene in
12. Macgibbon BH, Mollin DL. Sideroblastic anaemia in two pyridoxine-responsive patients initially diagnosed with
man: observations on seventy cases. British Journal of acquired refractory anemia and ringed sideroblasts. J Clin
Haematology 1965;11:5969. Invest 1995;96:20906.
13. Nicholls MD, Atkinson K, Biggs JC, Concannon AJ, 19. Stein RS, Abels RI, Krantz SB. Pharmacologic doses of
Dodds AJ. Late onset pyridoxine-responsive sideroblastic recombinant human erythropoietin in the treatment of
anaemia after allogeneic bone marrow transplantation. myelodysplastic syndromes. Blood 1991;78:165863.
British Journal of Haematology 1984;56:1536. 20. Gattermann N, Aul C, Schneider W. Two types of
14. Parry GJ, Bredesen DE. Sensory neuropathy with low- acquired idiopathic sideroblastic anaemia (AISA). British
dose pyridoxine. Neurology 1985;35:14668. Journal of Haematology 1990;74:4552.

516 2011 John Wiley & Sons A/S


Copyright of European Journal of Haematology is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.

Das könnte Ihnen auch gefallen