Recurrent miscarriages: Caution regarding development

of clinical trials using low molecular weight heparin and

pregnancy

To the editor: Over the past few years, several large negative randomized

control trials have been published showing no benefit with prophylactic

doses of low molecular weight heparin (LMWH) in women with recurrent

miscarriages [1,2]. Furthermore, among pregnant women with antiphospholi-

pid syndrome, there has been conflicting data to show benefit of LMWH to

prevent recurrent miscarriages [3,4]. We are concerned that these protocols

do not provide a standardized assessment of the activity of prophylactic

doses of LMWH in pregnancy using anti-Xa levels.

Pharmacokinetics studies of fixed doses of LMWH have shown that anti-

Xa levels vary during the course of pregnancy [5–8]. Lebaudy et al. studied

anti-Xa levels in 75 pregnant women and 38 nonpregnant women who were

receiving enoxaparin. Pregnant women demonstrated significantly higher

drug clearance and had dramatic changes in their volume of distribution.

Among women receiving enoxaparin 40 mg every day, there was a 24%

decrease in peak anti-Xa levels during the course of their pregnancy (Fig.

1). These changes in pharmacokinetics have been attributed to the normal

physiologic changes that occur in pregnancy, specifically an increase in

weight, increase in plasma volume, changes in renal function, and possible

changes in protein expression within the placenta that may have an affinity

to bind with LMWH [7]. Although these pharmacokinetic studies are small

with no clear evidence of how these differences in peak anti-Xa levels may

affect a women’s risk of developing a recurrent thrombosis, the 2008 Ameri-

can College of Chest Physicians guidelines have offered a grade 1C recom-

mendation for monitoring anti-Xa levels in pregnant women receiving thera-

peutic doses of LMWH [9].

We present a case to further support anti-Xa testing for pregnant women

on LMWH therapy. Recently, a 21-year-old female (G1P0) presented to our

institution at 30 weeks of pregnancy with worsening lower extremity pain af-

ter being starting 1 week prior on enoxaparin 60 mg (1 mg/kg) for a right

lower extremity DVT. Despite treatment for 1 week with appropriate weight

based dosing of LMWH, a repeat Doppler revealed further extension of the

common femoral vein thrombus. Blood levels of antifactor-Xa that were

assessed 4 hours after enoxaparin injection were found to be 0.32 (thera-

peutic dosing 0.5-1.0). We empirically increased her enoxaparin dosage by

70% to obtain a peak anti-Xa level of 0.60, and we suspect that the patient’s

clot extended because her enoxaparin dose was subtherapeutic, despite

receiving the appropriate weight based dose.

Our patient’s history and the above pharmacokinetic studies we have pre-

sented strongly suggest we need to re-evaluate how we design and evaluate

clinical trial protocols that involve the use of LMWH in pregnancy.

DIPIKA MISRA

LOUIS ALEDORT

Division of Hematology and Oncology, Mount Sinai School of Medicine,

One Gustave L. Levy, New York, New York

Published online 27 October 2010 in Wiley Online Library

(wileyonlinelibrary.com).

DOI: 10.1002/ajh.21909

Conflict of interest: Nothing to report.

References1. Kaandorp SP, Goddijn M, van der Post JA, et al. Aspirin plus heparin or aspirin

alone in women with recurrent miscarriage. N Engl J Med 2010;362:1586–1596.

2. Laskin CA, Spitzer KA, Clark CA, et al. Low molecular weight heparin and aspi-rin for recurrent pregnancy loss: Results from the randomized, controlled Hep-ASA Trial. J Rheumatol 2009:36:279–287.

3. Empson M, Lassere M, Craig J, Scott J. Prevention of recurrent miscarriage forwomen with antiphospholipid antibody or lupus anticoagulant. Cochrane Data-base Syst Rev 2005:CD002859.

4. Ziakas PD, Pavlou M, Voulgarelis M. Heparin treatment in antiphospholipid syn-drome with recurrent pregnancy loss: a systematic review and meta-analysis.Obstet Gynecol 2010:115:1256–1262.

5. Friedrich E, Hameed AB. Fluctuations in anti-factor Xa levels with therapeuticenoxaparin anticoagulation in pregnancy. J Perinatol 2010:30:253–257.

6. Laifer SA, Stiller RJ, Dunston-Boone G, Whetham JC. Low-molecular weightheparin for treatment of pulmonary embolism in a pregnant woman. ThrombHaemost 1999:82:1361–1362.

7. Lebaudy C, Hulot JS, Amoura Z, et al. Changes in enoxaparin pharmacoki-netics during pregnancy and implications for antithrombotic therapeutic strategy.Clin Pharmacol Ther 2008;84:370–377.

8. Casele HL, Laifer SA, Woelkers DA, Venkataramanan R. Changes in the phar-macokinetics of the low-molecular-weight heparin enoxaparin sodium duringpregnancy. Am J Obstet Gynecol 1999:181:1113–1117.

9. Hirsh J, Bauer KA, Donati MB, Gould M, et al. Parenteral anticoagulants: Amer-ican College of Chest Physicians Evidence-Based Clinical Practice Guidelines(8th Edition). Chest 2008:133:141S–159S.

Variability of bone marrow morphology in G6PC3

mutations: Is there a genotype–phenotype correlation or

age-dependent relationship?

To the editor: Mutations in G6PC3 cause severe congenital neutropenia

Type 4 (SCN4, MIM612541) [1] and the allelic condition, Dursun syndrome

(MIM613034) [2]. McDermott et al. recently described two siblings with

homozygous G6PC3 mutations and increased expression of CXCR4 with

myelokathexis, a contrasting presentation from the previously described

patients with SCN4 [3]. We identified four individuals with SCN4, of whom

two had bone marrow profiles similar to their description [4]. McDermott

et al. raised the possibility of phenotypic variability due to differences in ge-

notype or age. To test this hypothesis, we undertook a genotype–phenotype

analysis of the 25 patients with mutations in G6PC3 reported to date and

include two previously unreported patients from our center (Table I).

Of the 27 patients, 15 are of Middle Eastern ancestry, six are Caucasians,

two are Pakistani and in four cases the ethnicity is unreported. Only two

patients have compound heterozygous mutations, whereas 25 have homozy-

gous mutations. A total of 16 distinct G6PC3 mutations have now been

reported, of which nine are missense, three nonsense, two small deletions,

one single-base insertion, and one is a splice-site mutation. Interestingly,

five mutations in exon 6 account for 32 mutated alleles in 17 SCN4 patients.

This is due to the effect of two more common mutations, p.G260R in Cauca-

sians and p.R253H in patients with Middle Eastern ancestry, suggesting that

there are founder mutations in these populations.

Bone marrow morphology has been reported in 15 of the 27 patients. Of

these, myeloid hypoplasia and/or maturation arrest was reported in eight

patients, one with a frameshift mutation (patient 5 in Table I, p.86Efs, age 22

years at the time of description) and seven with missense mutations

(patients 6 and 7, p.M116V, both 1 year and patients 12, 15, 16, 18, and 19,

p.R253H, aged 26, 3, 4, 6, and 6 years). In contrast, like the patients

described by McDermott et al. (patients 22 and 23, p.G260R, aged nine andFigure 1. Typical anti-Xa kinetics profile with repeated administration of enoxaparinat 40 mg/day, in relation to pregnant state and gestational age [7].

Correspondence

VVC 2010 Wiley-Liss, Inc.

American Journal of Hematology 235 http://wileyonlinelibrary.com/cgi-bin/jhome/35105

TABLEI.

Summary

ofage,bone-m

arrow

examinationresult,ethnicity,

classofmutation,exon,nucleotideandprotein

changein

allcasesofbi-allelicmutationsin

G6PC3

Patient

Age

(years)

Sexb

Bone-m

arrow

examinationb

Ethnicity

Classofmutation

Exon

Mutationa

Protein

change

Reference

18

FNorm

alcellularity

and

nomaturationarrest.

Pakistani

Missense

1c.[130C>T]1

[130C>T]

p.P44S

New

213

FNA

Caucasian,Greece

Nonsense

1c.[141C>G]1

[141C>G]

p.Y47X

(1)

35

MNA

Arab,Lebanon

Nonsense

1c.[144C>A]1

[144C>A]

p.Y48X

(1)

4NA

MNA

NA

Deletion,frameshift

1NA

P.70Ifs

(5)

522

MPaucityofgranulocyte

series

beyo

ndpromye

locyte

stage

Moroccan

Deletion,frameshift

2* c.[257delA]1

[257delA]c

p.86Efs

(6)

61.5

FHypoplasia,dysplasia

ofalllines.

Turkish

Missense

3c.[346A>G]1

[346A>G]

p.M

116V

(7)

71.5

MHypoplasia,dysplasia

ofalllines.

810

FNorm

alcellularity

andno

maturationarrest.

Pakistani

Missense

3c.[347T>C]1

[347T>C]

p.M

116T

New

9NA

NA

NA

NA

Missense

3* c.[347T>A]1

[347T>A]c

p.M

116K

(8)

10

12

FNA

Turkish

Missense

5c.[554T>C]1

[554T>C]

p.L185P

(1)

11

29

FHypercellularmarrow

withmye

loid

hyperplasia.Increasein

blast-likeform

s.

Arab,Israel

Missense

6c.[758G>A]1

[758G>A]

p.R253H

(4)

12

26

MSlightlydecreasedmye

loid

cells.

13

25

FHypercellularmarrow

withmild

dysmye

lopoieticchanges

(reducedgranules).

14

2M

Norm

alcellularity

andno

maturationarrest.

15

6M

Paucityofmature

neutrophils.

Aramean,Turkey

Missense

6c.[758G>A]1

[758G>A]

p.R253H

(1)

16

3F

Paucityofmature

neutrophils

17

11

FNA

18

6M

Paucityofmature

neutrophils

19

4M

Paucityofmature

neutrophils.

Aramean,Turkey

Missense

6c.[758G>A]1

[758G>A]

p.R253H

(1)

20

17

MNA

Caucasian,Germ

any

Missense

6c.[778G>C]1

[778G>C]

p.G

260R

(1)

21

NA

MNA

NA

Missense

6c.[778G>C]1

[778G>C]

p.G

260R

(8)

22

13

MNomaturationarrest.Hyperplasia

ofmye

loid

cells

withva

cuolization

ofpro,meta

andmye

locytes

Caucasian

Missense

6c.[778G>C]1

[778G>C]

p.G

260R

(3)

23

9F

Nomaturationarrest.Hyperplasia

ofmye

loid

cells

withva

cuolization

ofpro,meta

andmye

locytes

24

7F

NA

Caucasian,Germ

any

Missense

6c.[784G>C]1

[784G>C]

p.G

262R

(1)

25

10

MNA

Persian

Insertion,

Frame-shift

6c.[935dupT]1

[935dupT]

p.N313fs

(1)

26

NA

FNA

NA

Missense

3and6

* c.[353C>G]1

[778G>C]c

p.[T118R]1

[G260R]

(5)

27

13

FNA

Caucasian,France

Spliceand

Nonsense

intron5andexon6

c.[67711G>A]1

[829C>T]

p.[?]1

[G277X]

(1)

aThetable

hasbeenarrangedasperascendingorderoftheresidueaffectedbytheprotein

change.Twocasesofcompoundheterozygosity

have

beenlistedattheendofthetable.Resultsofpatients

belongingto

thesamefamily

have

beenmerged.

bF,

female;NA,notava

ilable

andM,male.

cIndicatedwhere

precisenucleotidechangewasnotgivenin

theoriginalpublicationbuthasbeenderive

dfrom

theava

ilable

inform

ation.

correspondence

236 American Journal of Hematology

13 years), two of our patients (patients 11 and 13, p.R253H, aged 29 and

25 years) also had myeloid hyperplasia. Additionally, the bone marrow of our

two new patients (patients 1 and 8, p.P44S and p.M116T, aged 8 and 10

years) and patient number 14 (p.R253H and age 2 years) were of normal

cellularity without maturation arrest.

From this it is clear that the same mutation can cause either maturation

arrest or hypercellular or normocellular bone marrow (e.g., p.R253H). How-

ever, it is notable that all four patients reported with myeloid hypercellularity

have missense mutations and therefore it is possible that patients with non-

sense or frameshift mutations give rise to a different phenotype. There is also

no consistent age-dependent relationship for the bone marrow morphology.

In summary, our findings do not support a genotype–phenotype or age-

dependent relationship for bone marrow phenotype in patients with G6PC3

mutations. We have also reported two new patients with this rare disorder

and demonstrated the possibility of founder mutations in patients with Cau-

casian and Middle Eastern ancestry. At present, the reason for variability of

the bone marrow phenotype in patients with mutations in G6PC3 is unknown

and further characterisation is warranted.

ACKNOWLEDGMENTS

We acknowledge the support of NIHR Manchester Biomedical Research

Centre.

Author ContributionsS.B. performed experiments; R.W. diagnosed the patients and described

the bone marrow findings; S.B. and W.G. designed the research, analyzed

results, and wrote the paper.

SIDDHARTH BANKA1

ROBERT WYNN2

WILLIAM G. NEWMAN1

1Genetic Medicine, St Mary’s Hospital, Manchester Academic Health

Sciences Centre (MAHSC), University of Manchester, Manchester, United

Kingdom; 2Department of Paediatric Haematology, Royal Manchester

Children’s Hospital, Manchester, United Kingdom

Conflict of interest: Nothing to report

Published online 15 November 2010 in Wiley Online Library

(wileyonlinelibrary.com).

DOI: 10.1002/ajh.21930

References1. Boztug K, Appaswamy G, Ashikov A, et al. A syndrome with congenital neutro-

penia and mutations in G6PC3. N Engl J Med 2009;360:32–43.2. Banka S, Newman WG, Ozgul RK, Dursun A. Mutations in the G6PC3 gene

cause Dursun syndrome. Am J Med Genet A 2010;152A:2609–2611.

3. McDermott DH, De Ravin SS, Jun HS, et al. Severe congenital neutropenia

due to G6PC3 deficiency with increased neutrophil CXCR4 expression and

myelokathexis. Blood, 2010;116:2793–2802.

4. Banka S, Chervinsky E, Newman WG, et al. Further delineation of the pheno-type of severe congenital neutropenia type 4 due to mutations in G6PC3. Eur JHum Genet, in press; doi: 10.1038/ejhg.2010.136

5. Xia J, Bolyard AA, Rodger E, et al. Prevalence of mutations in ELANE, GFI1,HAX1, SBDS, WAS and G6PC3 in patients with severe congenital neutrope-nia. Br J Haematol 2009;147:535–542.

6. Arostegui JI, de Toledo JS, Pascal M, et al. A novel G6PC3 homozygous 1-bp de-letion as a cause of severe congenital neutropenia. Blood 2009;114:1718–1719.

7. Dursun A, Ozgul RK, Soydas A, et al. Familial pulmonary arterial hypertension,leucopenia, and atrial septal defect: a probable new familial syndrome with mul-tisystem involvement. Clin Dysmorphol 2009;18:19–23.

8. Germeshausen M, Zeidler C, Stuhrmann M, et al. Digenic mutations in severecongenital neutropenia. Haematologica 2010;95:1207–1210.

Levels of miR-29b do not predict for response in patients

with acute myelogenous leukemia treated with the

combination of 5-azacytidine, valproic acid, and ATRA

To the editor: At the present time, we do not have access to predictors

of response to hypomethylating agents. MicroRNA (miRNAs) are noncod-

ing RNAs of 19–25 nucleotides in length that regulate gene expression

by inducing translational inhibition and cleavage of their target mRNAs

through base pairing to partially or fully complementary sites. Altered

miRNA expression has been documented in multiples types of leukemia

[1]. Acute myeloid leukemia (AML) is a cytogenetically and molecularly

heterogeneous disorder characterized by differentiation arrest and malig-

nant proliferation of clonal myeloid precursors. Specific miRNA signatures

are associated with cytogenetics and prognosis in AML [2]. Recently it

has been reported that pretreatment expression levels of miR-29b are

associated with clinical response in patients with AML treated with a 10-

day schedule of decitabine [3]. miR-29b is involved in the regulation of

DNA methylation and is down regulated in AML [3,4]. The hypothesis is

that patients with higher miR-29b levels would have lower levels of DNA

methyltransferase, less hypomethylation, and potentially increased sensitiv-

ity to decitabine [3]. This hypothesis is in line with previous observations

associating lower levels of p15 methylation with response to decitabine

based therapy [5].The findings of Blum et al. [3] are therefore very signifi-

cant as detection of miR-29b, if validated, could be used to select hypo-

methylating-based therapy in AML.

To follow on these observations, we have analyzed miR-29b and miR-101

expression levels in a cohort of patients with AML or high-risk myelodysplas-

tic syndrome treated in a phase I/II study of the combination of 5-azacitidine

(5-AZA), valproic acid, and ATRA [6]. We analyzed 45 patients, 28 of which

had not received prior therapy. The characteristics of these patients have

been previously reported [6]. Eight (29%) patients achieved a complete

remission and 1 (4%) partial remission. Total cellular RNA (extracted from

unselected peripheral blood mononuclear cells collected before the first

cycle of treatment) was used for reverse transcription reactions using Taq-

Man miRNA reverse transcription kit (Applied Biosystems). As a control, we

also analyzed levels of miR-101. For real-time PCR, miR-29b and miR-101

TaqMan - miRNA assays were purchased from Applied Biosystems and ana-

lyzed with TaqMan Universal PCR Master Mix (Applied Biosystems) using

an Applied Biosystems Prism 7500 Sequencing detection system. The U6

small nuclear RNA was used as internal control. Expression levels of miR-

29b and miR-101 from 45 patients were compared to 6 healthy donors.

Expression levels of miR-29b were significantly down-regulated in patients

compared to normal controls (P 5 0.001) (Fig. 1A). There were no differ-

ence in terms of miR-101 expression between patients and health donors.

Figure 1. Difference in baseline miR-29b expression levels. (A) Difference in baselinemiR-29b expression levels between AML patients and healthy controls. (B) Differencein baseline miR-29b expression levels between responders and nonresponders. ExactP values are shown over the columns; unpaired, two-tailed t test were performed forstatistical analysis.

correspondence

American Journal of Hematology 237

We then compared the miR-29b and miR-101 expression levels between

patients who responded to the combination therapy versus nonresponders:

no differences were observed for both miR-29b and miR-101 expression

(P 5 0.56) (Fig.1B).

Our results fail to detect a relationship between miR-29b levels and

response to 5-AZA based therapy. This difference could be explained by

the relative small sample size, the source of RNA (peripheral blood ver-

sus bone marrow) and potentially the fact that miR-29b may predict for

response to decitabine but not to 5-AZA in combination. In regards to the

source of RNA, it should be noted that Blum et al. used unsorted bone

marrow specimens whereas we used peripheral blood. The median blast

percent in the blood of our patients was 6% (range 0–86%) whereas it

was 35% (range 8–93%) in the bone marrow. In our experience, it is

unlikely that the differences in blast percentage would explain the poten-

tial differences in gene expression levels. Despite these limitations, these

results indicate that larger studies need to be performed to confirm the

role of miRNAs in response to hypomethylating agent based therapy.

Author ContributionsH.Y. designed and performed research, analyzed data and wrote the manu-

script; Z.F., Y.W., Y.H., G.C., and H. K. performed research, G.G.-M. designed

experiment, analyzed data, wrote manuscript and funded the study.

HUI YANG1

ZHIHONG FANG1

YUE WEI1

YUMIN HU1

GEORGE A. CALIN2

HAGOP M. KANTARJIAN1

GUILLERMO GARCIA-MANERO1

1Department of Leukemia, University of Texas MD Anderson Cancer Center,

Houston, Texas; 2Experimental Therapeutics, University of Texas MD

Anderson Cancer Center, Houston, Texas

Conflict of interest: Nothing to report.

Published online 22 November 2010 in Wiley Online Library

(wileyonlinelibrary.com).

DOI: 10.1002/ajh.21937

References1. Fabbri M, Garzon R, Andreeff M, et al. MicroRNAs and noncoding RNAs in

hematological malignancies: Molecular, clinical and therapeutic implications.Leukemia 2008;22:1095–1105.

2. Garzon R, Volinia S, Liu CG, et al. MicroRNA signatures associated with cyto-genetics and prognosis in acute myeloid leukemia. Blood 2008;111:3183–3189.

3. Blum W, Garzon R, Klisovic RB, et al. Clinical response and miR-29b predictivesignificance in older AML patients treated with a 10-day schedule of decitabine.Proc Natl Acad Sci USA 2010;107:7473–7478.

4. Garzon R, Liu S, Fabbri M, et al. MicroRNA-29b induces global DNA hypome-thylation and tumor suppressor gene reexpression in acute myeloid leukemia bytargeting directly DNMT3A and 3B and indirectly DNMT1. Blood 2009;113:6411–6418.

5. Garcia-Manero G, Kantarjian HM, Sanchez-Gonzalez B, et al. Phase 1/2 studyof the combination of 5-aza-20-deoxycytidine with valproic acid in patients withleukemia. Blood 2006;108:3271–3279.

6. Soriano AO, Yang H, Faderl S, et al. Safety and clinical activity of the combina-tion of 5-azacytidine, valproic acid, and all-trans retinoic acid in acute myeloidleukemia and myelodysplastic syndrome. Blood 2007;110:2302–2308.

correspondence

238 American Journal of Hematology