ORIGINAL ARTICLE
Cell-Surface CD200 May Predict Efficacy of Paternal
Mononuclear Leukocyte Immunotherapy in Treatment of
Human Recurrent Pregnancy Loss
David A. Clark
Departments of Medicine, Molecular Medicine & Pathology, Obstetrics and Gynecology, McMaster University, Hamilton, ON, Canada
Keywords
CD200, immunotherapy, infertility, recurrent
spontaneous abortion, tolerance, transfusion-
related immunomodulation
Correspondence
David A. Clark, Departments of Medicine,
Molecular Medicine & Pathology, Obstetrics
and Gynecology, McMaster University, 1200
Main St. West Rm 3V39, Hamilton, ON,
Canada L8N 3Z5.
E-mail: clarkd@mcmaster.ca
Submitted July 13, 2008;
accepted October 8, 2008.
Citation
Clark DA. Cell-surface CD200 may predict
efficacy of paternal mononuclear leukocyte
immunotherapy in treatment of human
recurrent pregnancy loss. Am J Reprod
Immunol 2009; 61: 75–84
doi:10.1111/j.1600-0897.2008.00665.x
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
Journal compilation ª 2009 Blackwell Munksgaard
Problem
The allogeneic leukocytes in transfused blood can modulate the recipi-
ent’s immune system so as to induce TGF-b-producing suppressor cells,
and the cell-surface CD200 tolerance-signaling molecule on mono-
nuclear dendritic cells is required for this effect. A subset of couples
with unexplained recurrent pregnancy loss appears to benefit from
transfusion of allogeneic paternal blood leukocytes (LIT), and consider-
able effort has been devoted to characterizing those who may benefit.
Some data has been accumulated for LIT as sole therapy in patients
with classical spontaneous abortions with respect to dose–response,
duration of protection, need for boosting, excluding patients with auto-
immunity, and inefficacy of paternal mononuclear cells stored at 4C
overnight before use which causes loss of cell-surface CD200. Recent
data emphasize an important role of expression of the CD200
tolerance-signaling molecule on cells used to prevent abortions both in
mice and humans.
Method of study
An observational study of outcome as a function of the number of
CD200+ paternal mononuclear cells was performed. Fourteen patients
constituted the pilot group. Patients with autoimmunity who had failed
inspite of treatment with IVIG + Heparin + Aspirin ± Prednisone were
allowed to have paternal mononuclear cells added to their therapy.
CD200 on purified paternal blood mononuclear cells was measured by
flow cytometry.
Results
The number of CD200+ cells administered was significantly greater in
women achieving pregnancy (39.2 · 106 versus 20.8 · 106, P < 0.025)
and in those who achieved a live birth (50.2 · 106 versus 20.8 · 106,
P < 0.005) compared to those who did not achieve pregnancy, and % of
paternal cells that were CD200+ was greater (11–12.5% versus 5.6%,
P < 0.01). Amongst those achieving pregnancy which failed, the CD200+
cell dose was not significantly different from the non-pregnant group
(30.5 · 106 versus 20.8 · 106).
Conclusion
The number of CD200+ paternal mononuclear leukocytes may be an
important determinant of subsequent reproductive outcome in a subset
75
 CLARK

of patients. A lower % CD200+ cell number may also reflect hitherto

unappreciated paternal factors bearing on reproductive success. It is fea-
sible to recruit women to enter observational studies and to obtain use-
ful data as a foundation for further studies. More complete patient
characterization in a larger study is needed.

of BALB ⁄ c mononuclear cells bearing paternal

Introduction
Paternal mononuclear leukocyte immunotherapy
(LIT) appears to improve the live birth rate in a sub-
set of women with unexplained recurrent mis-
carriages, specifically women with primary recurrent
miscarriages, no evidence of autoantibodies such as
ANA or ACL, and loss of normal karyotype
embryos.1–6 The mechanism of pregnancy failure is
thought to be similar to the pathogenesis of recur-
rent resorptions in the CBA · DBA ⁄ 2 mouse model
where there is an excess of Th1 > Th2,3 cytokine
activation at the feto–maternal interface, activation
of fgl2 prothrombinase, thrombin, complement, and
infiltration by polymorphonuclear leukocytes.5 An
activation of maternal immune effector T and NKT
cells, and lack of Treg-mediated tolerance is anteced-
ent to Th1 cytokine excess.5 These events are coun-
tered by the CD200 tolerance signaling molecule (an
antithesis for CD80 ⁄ 86 and CD40 costimulatory mol-
ecules).5 Whilst some place considerable emphasis
on rationale for a treatment, empirical clinical trial
data represents the key evidence with respect to
whether or not a therapy works.5,6 In this respect, it
is critical to know the characteristics of patients
who respond and those who have no benefit, or a
negative effect of treatment.
CD200 is expressed on trophoblasts and by cells in
maternal decidua, and acts by binding to CD200
receptors.7–9 Blocking CD200 in vivo in the
CBA · DBA ⁄ 2 model augments the rate of abortions,
whereas CD200Fc administration blocks abortions;7
similarly, in LPS-driven abortions in B6 background
mice, up-regulating CD200 prevents losses, consis-
tent with its anti-inflammatory effects.10 CD200 is
also expressed on lymphomyeloid cells, and expres-
sion on allogeneic dendritic cells is proposed to
account for the phenomenon of transfusion-related
immunomodulation (TRIM).11 TRIM can suppress
the inflammatory response and via CD200, activate
suppressor cells in the recipient that make trans-
forming growth factor-b (TGF-b).11 Administration
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
76
DBA ⁄ 2-type paternal antigens to CBA ⁄ J female mice
prevents most of the abortions in the CBA · DBA ⁄ 2
model, but this protection is abrogated by treating
the immunizing cells with anti-CD200 or by storing
the cells overnight at 4C which depletes cell surface
CD200.4,12 Similarly, in recurrent miscarriage
patients, paternal mononuclear leukocytes stored at
4C lost surface CD200 and therapy proved ineffec-
tive, whereas cells stored at 37C, which increases
surface CD200 expression, have been reported to be
highly effective in a recent randomized controlled
clinical trial.5,13,14
Most of the literature on factors determining effi-
cacy of LIT have focused on patient selection. In
most RCTs, the time of onset of the pregnancy loss
has not been strictly delimited. Based on data from
Mowbray, LIT appears effective if given before the
6th week of gestation, and this data has suggested
that losses prior to the 6th week, including chemical
pregnancies, merit an alternative treatment.5,15–17
However, similar immune abnormalities are reported
in women with classical recurrent miscarriages and
losses prior to 6 weeks including ‘implantation fail-
ure’ (which can be detected as chemical pregnancies
or late ‘heavy’ menses),18–26 and Check et al.
reported an improved pregnancy rate and live birth
rate in IVF failure patients who were given LIT.27–29
There has been some data on properties of the treat-
ment maneuver, including dose–response and dura-
tion of protection,15,24,25 but no information with
respect to the dose of CD200+ cells and outcome has
been generated. It was hypothesized that higher
dosed of CD200+ paternal mononuclear leukocytes
would be associated with a higher rate of reproduc-
tive success. Conversely, reproductive failure would
be associated with a lower dose of CD200+ cells, and
a loss where a high dose of CD200+ paternal cells
had been administered would be due to either a
karyotype abnormal embryo or untreated auto-
immunity. An observational study was performed to
test these predictions about CD200+ cell dose.
Journal compilation ª 2009 Blackwell Munksgaard

Materials and methods
Patients
Women with unexplained recurrent pregnancy loss
were referred by their specialist Obstetrician Gyne-
cologist to the author’s Reproductive Medicine Clinic
in the Department of Medicine, McMaster University
Medical Center, 1200 Main St. West, Hamilton,
Ontario. Patients with three or more spontaneous
abortions and no live births with their current part-
ner, and negative tests for ANA and ACL antibodies,
with both husband and wife negative for VDRL,
Hepatitis B antigen, anti-Hepatitis C, HIV and HTLV
were offered a trial of paternal mononuclear leuko-
cyte therapy (without charge) in a study approved
by the Hamilton Health Sciences ⁄ McMaster Univer-
sity Research Ethics Board, 1290 Main St. West,
Hamilton, Ontario. Anti-CMV was measured in all
couples. Informed consent was required, and the
consent document provided full details of back-
ground data, controversies, rationale for the study,
possible risks both real and theoretical, and alterna-
tives. Patients with autoimmune abnormalities and
patients with secondary recurrent miscarriages who
had re-aborted in spite of intravenous immunoglob-
ulin infusions (IVIG) + heparin (H) and aspirin
(A) ± prednisone (P)5,17 as described in detail below
were also eligible to have paternal mononuclear cell
LIT added to their current therapy. All patients were
tested for LAC, anti-thyroid antibodies, ANA, ENA,
anti-smooth muscle ⁄ anti-parietal cell ⁄ anti-mitochon-
drial antibodies, and anti-thyroid and anti-thyroglob-
ulin antibodies, and anti-reagin (VDRL) antibodies.
Patients with autoantibodies were initially offered
IVIG + HA in a separate trial. Some patients had a
panel of APL antibodies measured at Millenova Lab-
oratories (Chicago, IL, USA) if they could afford the
cost. This more extensive test panel was not
available in Canada. All patients had ABO and Rh
typing of both partners and a check for anti-husband
erythrocyte antibodies in the woman.
Immunotherapy
The Mowbray method was used.15 The McMaster
clinic program was conducted under the Immuno-
therapy Oversight Committee chaired by the Head of
Transfusion Medicine at Hamilton Health Sciences
and Director of the Canadian Blood Services Pro-
gram in Hamilton. Standard Operating Procedures
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
Journal compilation ª 2009 Blackwell Munksgaard
CD200 AND PREGNANCY SUCCESS
(SOPs) were in place for all procedures. In accor-
dance with the Mowbray protocol, one unit (400–
450 mL) of blood was collected from the husband in
ACD anticoagulant, centrifuged to isolate the buffy
coat, and the buffy coat was purified by centrifuga-
tion over a cushion of sterile Ficoll-Paque. The puri-
fied cells were washed, counted, and an aliquot was
taken for staining for cell surface CD200 by flow
cytometry as described elsewhere.12 The purified
cells were held at 4C on ice in 4–5 mL phosphate-
buffered saline, and at time of use, a second aliquot
was taken for a repeat CD200 determination by flow
cytometry.8,12 Prior to attempting conception, two-
third of the cells were given i.v. and 1 ⁄ 3 were
divided among two intradermal and two subcutane-
ous sites using both forearms. The cell dose was
limited to 400 million cells based on previous
dose–response data.30–32 All Rh) patients received
anti-Rh globulin (120 lg) i.v. immediately prior to
paternal cells. For patients with autoimmunity who
were continuing to receive IVIG + HA ± P, IVIG rep-
resented intravenous immunoglobulins (Gamagard
or Gammunex, 0.4 g ⁄ kg i.v. every 4 weeks or 40 g
i.v. every 3 weeks for IVFET patients), A represented
aspirin 81 mg p.o. daily, H represented 5000 units of
standard heparin s.c. b.i.d., and P represented 5 mg
of prednisone taken bid until a positive pregnancy
test and then 10 mg b.i.d. until 12 weeks of gesta-
tion when the prednisone was rapidly tapered to
zero to avoid complications.5 All treatments were
commenced prior to conception and with the excep-
tion of P, were continued to term. Patients were fol-
lowed by thrice weekly bhCG values, a one weekly
estradiol + progesterone level, and by serial ultra-
sound examinations to 12 weeks gestation. If the
serial bhCG trajectory fell below the expected
curve33 and ultrasound showed absent fetal heart,
the uterus was emptied and the products of concep-
tion were sent for karyotyping by culture in
accordance with RCOG guidelines and the recom-
mendation from Christiansen et al. for investiga-
tional studies,34,35 and for FISH staining for common
trisomies was performed where possible.36,37
Statistics
Pregnancy was defined as one or more bhCG values
>5. The endpoint was live birth. Patients were classi-
fied based on outcome. The absolute number of
CD200+ cells was used to determine the treatment
dose. The mean and S.E.M. was calculated. The null
77
 CLARK

hypothesis based on previous published results was

60
patients who became pregnant and had a live birth
had not received a larger dose of CD200+ paternal 50

positive cells x10–6

Number of CD200
cells.4,5,12 The significance of differences between
40
groups of patients was determined using Student’s
t-test. The data was plotted using Prism Graphpad. 30

20
Results
10
The characteristics of the 14 patients in this study
are provided in Table I. Spontaneous losses with the 0
P PS PF NP
current partner are shown. Remote elective termina-
tions are not listed. There were no ectopic pregnan- Fig. 1 Number of CD200+ cells administered and patient outcome.
cies. Results of immune testing are also given, along P, achieved pregnancy (n = 9); PS, pregnancy resulted in a livebirth
(n = 4); PF, pregnancy failed (n = 5); NP, no pregnancy ⁄ implantation
with the total dose of CD200+ cells administered.
occurred and hence, ‘infertile’ (n = 5). ####Significantly greater than
There was no difference in % CD200+ cells immedi- NP, P < 0.025; ***Significantly greater than NP, P < 0.005; +++Not
ately after purification and at the time the cells were significantly greater than NP, P > 0.10.
administered (after a short <2-hr period of storage
on ice.) (data not shown).
Fig. 1 summarizes the CD200+ cell dose in differ- (39.22 ± 6.79 · 106 versus 20.82 ± 3.90 · 106,
ent categories of patients. Patients achieving preg- P < 0.025). In pregnant patients who had a live
nancy had received a significantly larger number of birth, the CD200+ cell dose was also significantly
CD200+ cells than those not achieving a pregnancy greater than in the no pregnancy (infertile) group

Table I Characteristics of the Individual Patients in the Study

Immune PreLIT CD200+

No. Age History testing Rx Result PBL dose cell dose Result

1 32 AAAA Skin rash in Neg. 400 · 106 66.9 · 106 A46XY

Cuban sun
2 36 AAAAA infert ⁄ IVF Neg. (aPA+) IVIGHA A46XX 299 · 106 42.4 · 106 A46XX
3 30 AAAA ANA+RF+ AT3lo IVIGHA A46XX 304 · 106 23.1 · 106 Infert
4 38 AALaAa Neg. 194 · 106 16.3 · 106 A <6 weeks
5 41 AAALb Neg. 400 · 106 14.4 · 106 A >6 weeks
6 39 AAAAALc AALAAAA Neg.(Graves) IVIGHA A 400 · 106 59.4 · 106 L
7 33 AAAAA infert ⁄ IVF ACL+ IVIGHAP A 400 · 106 54.6 · 106 L
8 39 AAAAAALd ANA+ACL+ ATA+ IVIGHAP A 400 · 106 24.9 · 106 Infert
9 42 AAA infert ›NKe IVIGHAP A 400 · 106 40.9 · 106 L
10 27 LAAAAA AAA ACL+ATA+ IVIGHAP A 200 · 106 12.4 · 106 A
11 34 AAAA Neg. 400 · 106 45.7 · 106 L
12 29 A+f ENA+ ›NKe IVIGHAP Inf 400 · 106 19.8 · 106 Infert
infert ⁄ IVF
13 31 AAAA infert Neg. 400 · 106 29.7 · 106 Infert
14 31 AA+g infert ⁄ IVF ANA+ATA+ IVIGHA A 400 · 106 6.6 · 106 Infert
a
Patient had a live birth after LIT and a second pregnancy without a boost that miscarried (A).
b
Patient had ha live birth with LIT >3 years previously. Note that pre-LIT therapy was continued in all patients.
c
Twins – one stillborn and one died immediately after birth. A subsequent pregnancy with live birth occurred after a metroplasty.
d
Previous live birth on IVIG + A >3 years previously.
e
Elevated NK activity at Millenova Labs. Where patient had arranged her own testing.
f
At least one miscarriage + history of late heavy menses (occult losses).
g
Two early losses (chemical pregnancies after IVFET) + six IVFET failures (negative bhCG).

American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author

78 Journal compilation ª 2009 Blackwell Munksgaard

(50.16 ± 4.19 · 106 versus 20.82 ± 3.90 · 106, P <
0.0005). In the group that became pregnant and
aborted, the CD200+ cell number was intermediate
and was not significantly greater than in the
infertile group in contrast to pregnant patients who
achieved a live birth (30.46 ± 10.60 · 106 versus
20.82 ± 3.90 · 106).
The pregnant patients with success had received a
larger number of CD200+ paternal cells than those
who became pregnant and failed, but this difference
did not achieve statistical significance in part due to
insufficient numbers for adequate statistical power;
the pregnant and fail group also showed consider-
able heterogeneity in CD200+ cell dose (as reflected
in the larger S.E.M. value compared to the pregnant
and success group), and in timing of the pregnancy
failure. Three patients had classical abortions begin-
ning after 6 weeks gestation, and two had received
>50 million CD200+ cells. One lost a karyotype nor-
mal male embryo, and a subsequent pregnancy suc-
ceeded with IVIG. A second lost a karyotype normal
female embryo. Both had received >50 million cells.
Karyotyping of one other patient was not successful
for technical reasons: she had received <20 million
CD200+ cells. Two other patients receiving similar
low doses had losses before 6 weeks.
Fig. 2 shows that the difference in CD200+ cell
dose among the patients set out in Table I was lar-
gely determined by the husband. In some cases, the
% CD200+ cells was low, and in other cases, total
cell recovery from 1 unit of blood was <400 million.
Patient 12 had a history of one definite miscarriage
15
% CD200 positive cells
10
5
0
P PS PF NP
Fig. 2 Percentage of paternal mononuclear leukocytes that were
CD200+ in patient groups with different outcomes. P, achieved
pregnancy; PS, pregnancy resulted in a livebirth; PF, pregnancy failed;
NL, no pregnancy ⁄ implantation occurred and hence, ‘infertile’.
###Significantly greater than NP, P < 0.013; +++Significantly greater
than NP, P < 0.0025; +++Not significantly greater than NP, P > 0.05.
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
Journal compilation ª 2009 Blackwell Munksgaard
CD200 AND PREGNANCY SUCCESS
and a number of late heavy menses suggestive of
occult losses. If she is excluded from the calculations
in Fig. 1, the patients achieving pregnancy still
received significantly more CD200+ paternal cells
(P < 0.05) and the % CD200+ cells in their husbands
blood (Fig. 2) was still significantly greater than for
their husbands of the infertile outcomes (P < 0.025).
Discussion
The data in this study show that patients achieving a
successful pregnancy had received a larger number
of CD200+ cells than patients not achieving preg-
nancy. It was an a priori prediction of this study that
women achieving successful pregnancy would have
received more CD200+ cells than women aborting
their pregnancies (or conversely, that a higher dose
of CD200+ cells would be associated with a better
reproductive outcome). This was, in fact, the trend
but the difference did not reach P < 0.05 due to
small sample size (inadequate power),7 and the lar-
ger degree of variability in those who re-aborted as
set out in the Results.
Pregnancy failure in treated patients can have sev-
eral causes, and this explains the increased variabil-
ity in the CD200+ doses in this group. Indeed, as
noted, three patients had classical losses >6 weeks
gestation (one putatively autoimmune as the loss
was normal XY karyotype and IVIG led to a subse-
quent success) and two had losses <6 weeks. What
was unexpected was that those achieving pregnancy
had received a larger number of CD200+ cells than
those who did not achieve pregnancy (or conversely,
that a larger dose of CD200+ cells was associated
with achieving pregnancy). The effect of LIT in
increasing the chance of achieving pregnancy was
similar to that of Check et al. who reported
improved pregnancy rates in IVF-failure patients
receiving LIT with their next cycle (38.3%, 70.3%
and 90% with LIT versus 28.7%, 45.9% and 60%
for untreated controls in three different reports)27–29
but a significant increase had not been seen in RSA
patients in the RMITG study1 where pregnancy rates
were only slightly greater (78%) in the LIT-treated
patients compared to the controls (75%). However,
autoimmune testing in this study was quite limited,
and infertility (requiring IVF) was not the major
problem in these patients. In the Ober et al. study,13
79.1% of immunized women achieved pregnancy
compared to 74.1% in the control group; this study
used only 250 million cells and these cells were
79
 CLARK
stored overnight (which reduces cell surface CD200
levels and efficacy in preventing miscarriages,5
although perhaps not for the improvement in preg-
nancy rate). In comparison to the RMITG and Ober
et al. reports, the patients in this study were a select
group (Table I), and most were receiving additional
treatments for autoimmune problems when given
LIT. Interestingly, Winger et al. recently reported
improved live birth rates in an observational study of
women with primary RSA where LIT was added to
IVIG + HA.38 Similar effects were noted where anti-
TNF-a was substituted for LIT.39,40 In TRIM studies in
mice, allogeneic blood cells treated with anti-CD200
led to an enhanced TNF-a response in the recipient
[see ref. (11) and D. A. Clark, unpublished data].
A live birth rate of 4 ⁄ 14 may not seem particularly
impressive, given that for the average number of
prior losses, had this been a population of primary
recurrent miscarriage patients without a positive
ANA or ACL, one would have expected this number
of live births without LIT based on an intention-
to-treat analysis.3 However, the patients in this study
differ by having evidence of autoimmunity in 8 ⁄ 14,
and such patients may have a worse prognosis
untreated. Indeed, a worse prognosis has been docu-
mented for women with an ANA or ACL autoanti-
body by Nielsen and Christiansen.41 In other words,
the patients reported in this study represented a
highly selected group with a worse prognosis. In
patients with autoimmunity, it has been argued that
LIT alone may be inappropriate,1,5 but LIT + auto-
immune therapy has not been formally addressed in
a full paper.5,38 Further, only 9 ⁄ 14 (64%) patients in
this study achieved a pregnancy, compared to 91%
in the study mentioned above.3 If one corrects for
the infertility in this study population, a 40% live
birth rate would have been expected, which com-
pares favorably with a 53% success rate in primary
recurrent miscarriage patients and no ANA or ACL.
As the sample size is small in this study, there is
insufficient power to determine if 40% and 53% are
different; such a comparison should be properly
made using groups of patients with the same pheno-
types, and that would require a much larger study.
Additionally, Clifford et al. reported a high rate of
successful pregnancy in a group of women treated
with supportive care, but women with autoimmunity
(and a putatively worse prognosis) had been
excluded.41,42 An additional explanation for a low
success rate suggested by this study is inefficacy of
CD200+ cell doses <40 · 106 but in an observational
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
80
study, one can only demonstrate associations. It is
not possible to say that a larger CD200+ cell dose
would have improved the chance of a pregnancy
without randomizing to high or low CD200+ cell
dose, a randomized phase II trial.43 Where the %
CD200+ cells was low in the donor blood, a mirror
increase in the non-CD200+ population may have
been a factor in the infertility. Indeed, in TRIM stud-
ies in mice, treatment of allogeneic blood cells with
anti-CD200 to neutralize the subpopulation of den-
dritic cells with CD200 not only abolished the TRIM
effect (detected a stimulation of tumor nodule
growth via TGF-b effects),11 it led to inhibition of
nodule formation and an increase in TNF-a+ cells in
the recipients (DA Clark, unpublished data). As not
all of the patients who received a low dose of
CD200+ cells had received a total of 400 · 106 cells,
it seems unlikely that deleterious effects of CD200)
blood mononuclear cells was responsible to the
reduced pregnancy success rate in patients receiving
lower dosed of the CD200+ cells. Alternately, low %
CD200+ cell numbers may be telling us something
about the husbands of pregnancy failure patients.
These husbands may be less likely to father pregnan-
cies that succeed for non-immunological reasons.
Recurrent unexplained miscarriages have been pro-
posed to be partner specific, and a rationale for LIT
has been a more efficient induction of an
anti-abortive response (in women whose innate
immune system is poised to reject embryos) and the
husbands antigens have been viewed as insuffi-
ciently immunogenic when delivered by the normal
route.1,4–6,30 However, a deficient tolerance signal
rather than lack of immunogenicity could explain
partner specificity, and little attention has been given
to reproductive performance of the male partner in
previous relationships. In this study, none of the
women had had miscarriages with another partner.
These hypotheses are testable in a randomized
phase II design, where CD200+ cell dose is the vari-
able. For example, if all women received a total of
50 · 106 CD200+ paternal cells, would there be a dif-
ferent outcome in those who had more non-CD200+
cells rather than fewer? If there were not enough
CD200+ cells recovered from 400 mL of the partner’s
blood, would a repeat treatment to achieve the
50 · 106 total be more effective? Would the out-
come differ for all types of patients if 50 · 106
CD200+ cells were set as the target? (Recurrent preg-
nancy failure is a syndrome that can have many
causes, and to suggest that a particular therapy such
Journal compilation ª 2009 Blackwell Munksgaard

LIT should benefit everyone is not unlike demanding
that a single antibiotic should cure all types of pneu-
monia.) Up-regulating CD200 expression by cultur-
ing paternal cells at 37C overnight has been
reported in a RCT to achieve a high livebirth rate,
but other effects of cell culture cannot be excluded,
and culturing at 37C risks bacterial contamina-
tion.5,14 A similar problem could arise if one
attempted to purify large numbers of CD200+ blood
cells by cell-sorting procedures. In a mouse model of
partner-specific recurrent miscarriage, large doses of
exogenously administered soluble CD200 (OX-2Fc)
was able to prevent losses, but similar preparations
are not available for humans at present.7 In the
TRIM model, syngeneic lymphoid cells which
express CD200 but no alloantigens are inactive in
comparison to allogeneic blood cells;11 whilst the
explanation for this observation is speculative, it
could pertain to a role of alloantigen in bringing
antigen + physiological levels of CD200 together in
the encounter with cells of the recipient’s immune
system. Another conclusion from this pilot study is
that patients in future studies will need more exten-
sive laboratory testing to adequately characterize
them. It is important to have better diagnosis of sub-
sets that benefit from LIT alone or added to other
treatments as in this study, and subsets that do not.
Some of the patients in the study had an APL anti-
body or elevated NK cell activity documented by
testing at a US laboratory. In this study, the full
panel of APL, and NK activity testing, was not possi-
ble due to cost to the patient. As the patients are
clearly heterogeneous, there is merit in broad inclu-
sion criteria with more complete immunological
characterization as possible so that the effect
of CD200+ cell dose in definable subsets can be
elucidated.
Previous observational studies have suggested LIT
that suppressed circulating NK activity was likely to
be associated with successful pregnancy.44 In a
recent IVIG study we did, the importance of measur-
ing the in vivo effect of treatment on blood NKT cells
as a predictor of success or failure was emphasized.45
Allogeneic leukocyte administration may also induce
immunoregulatory cells that suppress NK ⁄ NKT and
Th1 cytokine production.11,44 In the present observa-
tional study of CD200+ cells in LIT, we did not have
a surrogate biological marker to assess effect on
physiology of the recipient.11,14,44 Such endpoints
need to be included in future studies. There may be
patients who fail to respond to CD200+ cells, or,
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
Journal compilation ª 2009 Blackwell Munksgaard
CD200 AND PREGNANCY SUCCESS
who have a subnormal response. To define what
represents a clinically relevant response in itself
requires an observational approach. However, only
an observational study with a panel of immune
assays can define responses that correlate with suc-
cess or failure. Such assays will be an essential part
of any phase III trial.
There are at least five conclusions one can draw
from the present small ‘pilot’ observational study:
(a) is it feasible to do such studies, and women will
agree to participate (which is not the situation
where there is randomization to a placebo); (b) is it
worth investigating the relationship between dose of
CD200+ cells and outcome further; (c) more detailed
measurements (laboratory and ⁄ or clinical) should be
done including more extensive autoantibody testing,
NK and NKT analysis, Th1 ⁄ Th2 ⁄ Th3 cytokine
responses of maternal PBL and salivary cortisol levels
among other measures of ‘stress level’, both before
and after therapy,46 particularly given the impor-
tance of stress as a putative trigger of pregnancy
failure and infertility and the claim by Clifford
and others that stress relieve may prevent mis-
carriages;5,31,42 (d) the treatment design should be
altered or expanded to examine effects of altering
total cell dose and dose of CD200+ cells; (e) a much
larger sample size should be used to increase statisti-
cal power and to allow comparison of groups of phe-
notypically similar groups of patients (e.g. it would
have required twice as many patients to achieve
P < 0.05 for success versus failure for CD200+ cell
dose in Fig. 1, and an even larger sample to show a
difference in % of donor cells that were CD200+).
However, the data in this study prove that the
approach is feasible and can yield potentially valu-
able information. Indeed, some striking differences
were noted which points to a way forward. The most
important step is to independently repeat the obser-
vations made in this study using a larger number of
patients. A pilot study is useful in that it provides an
effect estimate that can be used to decide sample size
required for a new study. Those centers who con-
tinue to use LIT may have a sufficient patient vol-
ume to repeat this study using the needed larger
sample size and may have a more extensively char-
acterized patient population. Phase III studies will
not become ethical without the necessary observa-
tional phase II data, and even then, women will
only be willing to enter phase III trials if there is
equipoise between assigned treatment arms. Phase
III trial designs that lack a solid basis obtained from
81
 CLARK
phase II studies and which lack adequate
laboratory and clinical endpoints and follow-up as
discussed above are unlikely to meet with the favor
of regulatory bodies.
This study was designed to test the a priori hypoth-
esis that higher doses of CD200+ cells were more
likely to be associated with a successful pregnancy,
and the result was positive. It should not be taken as
proof that LIT works, as clearly under certain circum-
stances it did not. There are both patient and treat-
ment variables that require additional study, and a
goal of this study is to stimulate such work. LIT is a
biological therapy, related to transfusion-induced
immunomodulation,44 which belongs in the realm of
‘tissue therapy’.5 The use of complex mixtures has
been replaced by use of the purified active agent,
particularly in the field of endocrinology. It is of
interest that the magnitude of improvement in take-
home-baby rate with LIT may be achieved using
more specific alternatives, such as anti-TNF-a
drugs.5,39,40 It would be an important step forward if
the same patient population that benefits from LIT
were the one that benefits from anti-TNF-a therapy.
If that approach is too expensive, one might consider
evaluating therapy with a better defined ‘tissue’ with
anti-abortive effects commercially available as Intrali-
pid.47 However, there will only be progress if such
potential treatments are part of well designed studies
the results of which are published.
References
1 Recurrent miscarriage immunotherapy Trialists group
(RMITG): Worldwide prospective observational study
and meta-analysis on allogeneic leukocyte
immunotherapy for recurrent spontaneous abortion.
Am J Reprod Immunol 1994; 32:55–72.
2 Daya S, Gunby J, the Recurrent Miscarriage
Immunotherapy Trialists Group: The effectiveness of
allogeneic leukocyte immunization in unexplained
primary recurrent abortion. Am J Reprod Immunol
1994; 32:294–302.
3 Clark DA, Daya S, Coulam CB, Gunby J, the
Recurrent Miscarriage Immunotherapy Trialists
Group: Implications of abnormal human trophoblast
karyotype for the evidence-based approach to the
understanding, investigation, and treatment of
recurrent spontaneous abortion. Am J Reprod Immunol
1996; 35:495–498.
4 Clark DA, Coulam CB, Daya S, Chaouat G:
Unexplained sporadic and recurrent miscarriage in
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
82
the new millennium: a critical analysis of immune
mechanisms and treatment. Hum Reprod Update 2001;
7:501–511.
5 Clark DA: Immunological factors in pregnancy
wastage: fact or fiction. Am J Reprod Immunol 2008;
59:277–300.
6 Clark DA: Shall we properly re-examine the status of
allogeneic lymphocyte immunotherapy for recurrent
pregnancy failure? Am J Reprod Immunol 2004;
51:7–15.
7 Clark DA, Ding JW, Yu G, Levy GA, Gorczynski RM:
Fgl2 prothrombinase expression in mouse trophoblast
and decidua triggers abortion but may be countered
by OX-2. Mol Hum Reprod 2001; 7:185–194.
8 Clark DA, Keil A, Chen Z, Markert U, Manuel J,
Gorczynski RM: Placental trophoblast from successful
human pregnancies expresses the tolerance signaling
molecule CD200 (OX-2). Am J Reprod Immunol 2003;
50:187–195.
9 Gorczynski R, Chen Z, Clark DA, Yu K, Lee L,
Nachman J, Wong S, Marsden P: Structural and
functional heterogeneity in the CD200 R family of
immunoregulatory molecules at the feto-maternal
interface. Am J Reprod Immunol 2004; 52:147–163.
10 Yu G, Sun Y, Foerster K, Manuel J, Molina H, Levy
GA, Gorczynski RM, Clark DA: LPS-induced murine
abortions require C5 but not C3, and are presented by
up-regulating expression of the CD200 tolerance
signaling molecule. Am J Reprod Immunol 2008;
60:135–140 DOI:10.1111 ⁄ J.1600-0897-2008.00605.x
11 Clark DA, Gorczynski RM, Blajchman MA:
Transfusion-related immunomodulation due to
peripheral blood dendritic cells expressing the CD200
tolerance signaling molecule and alloantigen:
stimulation of tumor nodule formation and growth by
TGF-b in a murine model. Transfusion 2008; 48:814–
821.
12 Clark DA, Chaouat G: Loss of surface CD200 on
stored allogeneic leukocytes may impair anti-abortive
effect in vivo. Am J Reprod Immunol 2005; 53:13–20.
13 Ober C, Karrison T, Odem RR, Barnes RB, Branch
DW, Stephenson MD, Baron R, Walker MA, Scott JR,
Schreiber JR: Mononuclear-cell immunization in
prevention of recurrent miscarriages: a randomized
trial. Lancet 1999; 354:365–369.
14 Pandey MK, Agrawal S: Induction of MLR-Bf and
protection of fetal loss: a current double blind
randomized trial of paternal lymphocyte
immunization for women with recurrent spontaneous
abortion. Int Immunopharmacol 2004; 4:289–298.
15 Mowbray JF: Immunology of early pregnancy. Hum
Reprod 1988; 3:79–82.
Journal compilation ª 2009 Blackwell Munksgaard

16 Clark DA: Is there any evidence for immunologically
mediated or immunologically modifiable early
pregnancy failure? J Assist Reprod Genet 2003; 20:63–
72.
17 Clark DA, Coulam CB, Stricker RM: Is intravenous
immunoglobulins (IVIG) efficacious in early
pregnancy failure? A critical review and meta-analysis
for patients who fail in vitro fertilization and embryo
transfer (IVF). J Assist Reprod Genet 2006; 23:1–13
(Erratum 23:383).
18 Aoki K, Kajijura S, Matsumoto Y, Ogasawara M,
Okada S, Yamagi Y, Gleicher N: Preconceptional
natural killer cell activity as a predictor of miscarriage.
Lancet 1995; 135:1340–1342.
19 Coulam CB, Goodman C, Roussev RG, Thomasen EJ,
Beaman KD: CD56+ cells can predict pregnancy
outcome. Am J Reprod Immunol 1995; 33:40–46.
20 Coulam CB, Roussev RG: Correlation of NK cell
activation and inhibition markers with NK
cytotoxicity among women experiencing
immunological implantation failure after in vitro
fertilization and embryo transfer. J Assist Reprod Genet
2003; 30:58–62.
21 Kwak-Kim JY, Chung-Bang HS, Ng SC, Ntrivales EI,
Mangubat CP, Beaman KD, Beer AE, Gillman-Sachs
A: Increased T helper 1 cytokine responses in
circulating T cells are present in women with
recurrent pregnancy losses and in infertile women
with multiple implantation failures after IVF. Hum
Reprod 2003; 18:767–773.
22 Coulam CB, Roussev RG: Increasing circulating T-cell
activation markers are linked to subsequent
implantation failure after transfer of in vitro fertilized
embryos. Am J Reprod Immunol 2003; 50:340–345.
23 Thum M-Y, Nhaskaran S, Abdalla HI, Ford B, Susman
N, Shehata N, Bansal AS: An increase in the absolute
count of CD56dimCD16+CD69+ NK cells is associated
with a poorer IVF treatment and pregnancy outcome.
Hum Reprod 2004; 19:2395–2400.
24 Thum M-Y, Bhaskaran S, Abdalla HA, Ford B, Sumar
N, Bansal A: Prednisolone suppresses NK cell
cytotoxicity in women with a history of infertility and
NK cell cytotoxicity. Am J Reprod Immunol 2008;
59:259–265.
25 Yamada H, Morikawa M, Kato EH, Shimada S,
Kobashi G, Minakami H: Preconceptional natural
killer activity and percentage as predictors of
biochemical pregnancy and spontaneous abortion
with a normal karyotype. Am J Reprod Immunol 2003;
50:351–354.
26 Hasegawa I, Yamamoto Y, Suzuki M, Murakawa H,
Kurabayashi T, Takakuwa K, Tanaka K: Prednisolone
plus low-dose aspirin improves the implantation rate
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
Journal compilation ª 2009 Blackwell Munksgaard
CD200 AND PREGNANCY SUCCESS
in women with autoimmune conditions who are
undergoing in vitro fertilization. Fertil Steril 1998;
70:1044–1048.
27 Check JH, Liss JR, Check ML, Diantonio A, Duroseau
M: Lymphocyte immunotherapy can improve
pregnancy outcome following embryo transfer (ET) in
patients failing to conceive after two previous ET. Clin
Exp Obstet Gynecol 2005; 32:21–22.
28 Check JH, Liss J, Check ML, Diantonio A, Choe JK,
Graziano V: Leukocyte immunotherapy improves live
delivery rates following embryo transfer in women
with at least two previous failures: a retrospective
review. Clin Exp Obstet Gynecol 2005; 32:85–88.
29 Check JH, Liss JR, Diantonio A, Check ML, Duroseu
M: Matched control study of efficacy of lymphocyte
immunotherapy in donor oocyte recipients to support
or refute the concept of need for histocompatibility
antigen sharing of the couple for benefit. Clin Exp
Obstet Gynecol 2005; 32:155–157.
30 Smith JB, Cowchock FS, Lata JA, Hankinson BT,
Iftekhar A, Rote NS: Pregnancy outcome after
immunotherapy is influenced by the number of cells
in the innoculum. Am J Reprod Immunol 1990; 22:88.
31 Clark DA, Daya S: Trials and tribulation in the
treatment of recurrent spontaneous abortion. Am J
Reprod Immunol 1991; 25:18–24.
32 Mowbray JF, Underwood JL: Effect of paternal
lymphocyte immunization on birthweight and
pregnancy outcome: In Cellular and Molecular
Biology of the Materno-Fetal Relationship, G
Chaouat, JF Mowbray (eds). Paris, John Libby
Eurotext, INSERM Colloque 212, 1991, pp 295–302.
33 Daya S: Human chorionic gonadotropin increase in
normal early pregnancy. Am J Obstet Gynecol 1987;
156:286–290.
34 Royal College of Obstetricians and Gynaecologists,
Guideline No. 17, The Management of Recurrent
Miscarriage, London, RCOG, 2003.
35 Christiansen OB, Nybo Andersen AM, Bosch E, Daya
S, Delves PJ, Hviid TV, Kutteh WH, Laird SM, Li TC,
van der Ven K: Evidence-based investigations and
treatments of recurrent pregnancy loss. Fertil Steril
2005; 83:821–839.
36 Karaoguz MY, Nas T, Konac E, Ince D, Pala E,
Menevse S: Is cytogenetic diagnosis of 46, XX
karyotype spontaneous abortion specimens
erroneous? Fluorescence in situ hybridization as a
confirmatory technique. J Obstet Gynaecol Res 2005;
31:508–513.
37 Bruno DL, Burgess T, Ren H, Nouri S, Pertile MD,
Frances DI, Norris F, Kenney BK, Schouten JA, Choo
KH, Slater HR: High-throughput analysis of
chromosome abnormalities in spontaneous
83
 CLARK
miscarriage using an MPLA subtelomere assay with
ancilliary FISH test for polyploidy. Am J Med Genet
2006; 140:2786–2793.
38 Winger E, Reed J, Stricker R: Lymphocyte
immunization therapy (LIT) combined with
intravenous immunoglobulin (IVIG) improves live
birth rates in women with primary recurrent
spontaneous abortion. Am J Reprod Immunol 2007;
58:203.
39 Winger EE, Reed JL, Stricker RB: Treatment with
tumor necrosis factor (TNF) inhibitors and
intravenous immunoglobulin (IVIG) improves live
birth rates in women with recurrent spontaneous
abortion (RSA). Am J Reprod Immunol 2007; 58:225.
40 Winger EE, Reed JL: Treatment with tumor necrosis
factor inhibitors and intravenous immunoglobulin
improves live birth rates in women with recurrent
spontaneous abortion. Am J Reprod Immunol 2008;
60:8–16.
41 Nielsen HS, Christiansen OB: Prognostic impact of
anticardiolipin antibodies in women with recurrent
miscarriage negative for the lupus anticoagulant. Hum
Reprod 2005; 20:1720–1728.
American Journal of Reproductive Immunology 61 (2009) 75–84 ª 2009 The Author
84
42 Clifford K, Rai R, Regan L: Future pregnancy
outcome in unexplained first trimester miscarriage.
Hum Reprod 1997; 12:387–389.
43 Wieand HS: Randomized phase II trials: what does
randomization gain? J Clin Oncol 2005; 23:1794–1795.
44 Higuchi K, Aoki K, Kimbara T, Hosoi N, Yamamoto T,
Okada H: Suppression of natural killer cell activity by
monocytes following immunotherapy for recurrent
spontaneous abortion. Am J Reprod Immunol 1995;
33:221–227.
45 Van den Heuvel MJ, Hatta K, Peralta CG, Han VH,
Clark DA: Suppression of elevated blood NKT cells in
recurrent pregnancy failure patients given IVIG
correlates with successful pregnancy. Am J Reprod
Immunol 2007; 58:447–459. (corrigendum 59;547)
46 Nakamura K, Sheps S, Arck PC: Stress and
reproductive failure: past notions, present insights and
future directions. J Assist Reprod Genet 2008; 25:47–62.
47 Coulam CB, Acacio B, Rinehart JS, Rinehart L, Ng S,
Roussev RG: Pregnancy outcome after intralipid
infusion among women experiencing early pregnancy
loss. Am J Reprod Immunol 2008; 59:472.
Journal compilation ª 2009 Blackwell Munksgaard