From bench to bedside in
reproduction
Siobhan Quenby
Professor Of Obstetrics
University of Warwick
What is recurrent miscarriage?
•3 consecutive miscarriages
•Very distressing
•Occurs over 1-2 years
•Increasingly desperate for
baby
•3% couples trying for a baby
•600,000 births in UK per year
•18,000 couple in UK per year
•50% cases no known cause in
30 blood tests
Try again
Happy pregnant
Recover
Few week later
Pain, bleeding
loss
Historical perspective
• Definition RM: 3 consecutive pregnancy
losses before the 20th weeks
• Sporadic miscarriage rate is 15%
• RM rate 0.153 = 0.3-0.4%.
• The actual prevalence of RM is 1-3%
Karyotypical abnormality
• High (29-57%) in RM population
•
•
•
•
Stern et al., 1996,
Ogasawara et al., 2000,
Carp et al., 2001,
Stephenson et al., 2002
• Same rate recurrent and spontaneous
miscarriage
Three treatments
• Anti-thrombotic
– Aspirin
– Heparin
• Immunotherapy
-IVIG
• Hormones
Progesterone
One treatable cause APS
• ACA
– IgG, IgM
• Lupus anticoagulant
– DRVVT
– Platelet neutralisation
– 2 +ve tests six weeks apart
Antithrombotic
– APS
– Cochrane review Empson et al., 2010
Pathophysiology APS
• First trimester
– Placental histology
– not thrombosis or infarcts
• (Sebier et al., 2003)
– Lack of trophoblast invasion
• (Sebier et al., 2002)
• Second trimester
– Placental thrombosis is identified
Idiopathic first trimester miscarriage
– Canada, Laskin et al 2008 N=88
• included APS, thrombophilia, antinuclear antibodies
• Asp: 78% V asp+LMWH 79%
– Metanalysis Cochrane -2009 N=189, Idiopathic
• Asp 81% v placebo 81%
• Asp 82% v LMWH 84%
– Netherlands ALIFE N=299, Kaandorp et al 2010
• Included idiopathic, thrombophilia, excluded APS, 2 miscarriages
• Asp: 62% V asp+ LMWH 69% v placebo 67%
– Scotland, SPIN, N=294, Clark et al., 2010
• Included idiopathic, thrombophilia, excluded APS, 2 miscarriages
• Standard care 80%V asp+LMWH 78%
– HABENOX N=207, Visser et al., 2010 n=207
• Idiopathic, 3 miscarriages
• LMWH +Placebo 71% V LMWH and aspirin 65% V Aspirin 61%
Aspirin
• Netherlands study
– Placebo v aspirin
– Absolute risk that causes miscarriage is 5.2
– CI (-6.1-16.6)
– Kaandorp et al 2010
Mechanism miscarriage ?
•Excessive oxidative stress
–(Burton and Jauniaux J Soc Gynecol Investig 2004;11:342–5)
Remaining questions
• Heparin and aspirin work second trimester
loss?
• Type and dose?
• Recurrent IVF failure?
Progesterone
Sub clinical hyopthyroidism
Reid et al., 2010
Immunotherapy
• Porter et al 2010 Cochrane
Cochrane Porter 2010
• “A specific assay to diagnose immunemediated early pregnancy loss and a
reliable method to determine which
women might benefit from manipulation of
the maternal immune system are urgently
needed”
Secondary RM
Stephenson et al., 2010
Uterine Natural Killer cells in womb lining
patient with two normal deliveries
Patient who had ten miscarriages
uNK cells more numerous in RM
Quenby et al, 1999,2005; Clifford et al, 1999, Tuckerman et al., 2007
Case History
–17 consecutive miscarriages
–No Cause found
–Most NK cells in study n=40
–Preconceptual prednisolone 5mgs
–Two further miscarriages
–Higher dose (prednisolone 20mg)
–Live Birth aged 42
•(IUGR 32/40)
–Alive and well age 4 years
•Quenby et al., 2004
Quenby et al.,
2005
Fert & Steril
% uNK
cells
P=0.0009
Normal
range
Before Prednisolone
After prednisolone
Uterine NK cells: studies finding more cells in RM
Study
N
Lachapelle 20
et al., 1996
Inclusion Method
criteria
3 or more Flow cytometry
idiopathic D18-25
Results
↑%CD16 D56dim
↓% of CD16-CD56bright cells
Quenby et
al., 1999
22
3 or more Immunohistochemistry ↑ % of CD16+& D56+ cells
idiopathic D19-22
c.f. controls
Manual counting
Clifford et
al., 1999
29
3 or more Immunohistochemistry ↑ CD56 +cells in RM <13 wk
idiopathic D20-23
c.f. controls
Manual counting
Emmer et
al., 2002
9
2 or more Immunohistochemistry
idiopathic Tissue collected after
miscarriage
Manual counting
Dosiou and Giudice 2005
↑ expression of CD56 & CD16
Uterine NK cells: studies finding no difference in RM
Study
N
Inclusion
criteria
Method
Results
Shimada et
al., 2004
17
2 or more
idiopathic
Flow cytometry
Midluteal phase
No difference in % of
CD56+, CD56+CD16or CD56+CD16+ cells
Michimata et 20
al., 2002
2 or more
idiopathic
Immunohistochemistry No difference in numbers
D18-21
of CD56+or CD16+ cells
Manual counting
“In summary, there is likely no difference in the percentage of total uNK
cells in endometrial leukocytes of women with RPL compared with women
without PL.” Dosiou and Giudice 2005
“histological evaluation of endometrium is limited”
Kim-Kwak and Gilman-Sachs 2008
Predict outcome?
High peripheral NK
Normal peripheral NK
Study or Subgroup
Events
Total
Events
Total Weight
5.1.1 Recurrent Miscarriage
Emmer et al
7
14
0
8 35.9%
Subtotal (95% CI)
14
8 35.9%
Total events
7
0
Heterogeneity: Not applicable
Test for overall effect: Z = 1.83 (P = 0.07)
5.1.2 infertility
Thum et al
6
Subtotal (95% CI)
Total events
6
Heterogeneity: Not applicable
Test for overall effect: Z = 0.42 (P = 0.67)
21
21
7
30
30
Odds Ratio
M-H, Random, 95% CI
17.00 [0.82, 350.60]
17.00 [0.82, 350.60]
64.1%
64.1%
1.31 [0.37, 4.68]
1.31 [0.37, 4.68]
38 100.0%
3.29 [0.28, 39.34]
7
Total (95% CI)
35
Total events
13
7
Heterogeneity: Tau² = 2.08; Chi² = 2.48, df = 1 (P = 0.12); I² = 60%
Test for overall effect: Z = 0.94 (P = 0.35)
Test for subgroup differences: Not applicable
0.01
0.1
1
10
100
Favours experimental Favours control
High peripheral NK
Normal peripheral NK
Study or Subgroup
Events
Total
Events
Total Weight
8.2.1 Recurrent Miscarriage
Aoki et al
17
24
9
44 39.7%
Emmer et al
5
17
3
7 31.7%
Subtotal (95% CI)
41
51 71.3%
Total events
22
12
Heterogeneity: Tau² = 3.42; Chi² = 6.67, df = 1 (P = 0.010); I² = 85%
Test for overall effect: Z = 0.65 (P = 0.52)
8.2.2 Infertility
Matsubayashi et al
3
Subtotal (95% CI)
Total events
3
Heterogeneity: Not applicable
Test for overall effect: Z = 1.84 (P = 0.07)
Odds Ratio
M-H, Random, 95% CI
7
7
2
21
21
Odds Ratio
M-H, Random, 95% CI
Odds Ratio
M-H, Random, 95% CI
9.44 [3.01, 29.68]
0.56 [0.09, 3.44]
2.51 [0.16, 40.29]
28.7%
28.7%
7.13 [0.88, 57.55]
7.13 [0.88, 57.55]
72 100.0%
3.55 [0.60, 21.03]
2
Total (95% CI)
48
Total events
25
14
Heterogeneity: Tau² = 1.73; Chi² = 6.87, df = 2 (P = 0.03); I² = 71%
Test for overall effect: Z = 1.40 (P = 0.16)
Test for subgroup differences: Not applicable
0.01
0.1
1
10
100
Favours experimental Favours control
Predict outcome?
Study or Subgroup
Quenby et al
Tuckerman et al
High endometrial NK
Normal endometrial NK
Events
Total
Events
Total Weight
5
10
2
11 44.1%
4
15
15
36 55.9%
Total (95% CI)
25
Total events
9
17
Heterogeneity: Tau² = 1.64; Chi² = 3.24, df = 1 (P = 0.07); I² = 69%
Test for overall effect: Z = 0.27 (P = 0.79)
47 100.0%
Odds Ratio
M-H, Random, 95% CI
4.50 [0.63, 32.29]
0.51 [0.14, 1.91]
Odds Ratio
M-H, Random, 95% CI
1.33 [0.16, 11.11]
0.01
0.1
1
10
100
Favours experimental Favours control
Cell counting
•Manual hand counting
under microscope
•Manual counting of
digital images
•Digital Image analysis
Manual counting
•Time consuming
•Different fields analysed
•Retrospective checking of
counts
•Observer dependent:
fatigue, distractions
•Poor intra and inter
observer agreement
•Risk to observer: RSI
Digital image analysis
•Semi- automated
•Manual thresholding
•PC + Freeware
•Constants: size,
circularity parameters
Results
Image J
Tally count
600
400
200
0
-200
0
200
400
600
800
1000
1200
1400
1600
-400
-600
Observer 1 - observer 2
Observer 1- observer 2
600
400
200
0
-200
0
200
400
600
800
1000
1200
1400
1600
-400
-600
Mean cell count
Mean cell count
Point picker
Tally
Point picker Image J
Observer 1 - observer 2
600
Mean CV (SD)
400
11.3 (8.6) 8.2 (5.1) 4.9 (3.1)
200
0
-200
0
200
400
600
800
1000
-400
1200
1400
1600
Mean difference
-105
-57
39
Upper limits agreement
239
138
146
Lower limits agreement
-449
-253
-68
-600
Mean cell count
New developments -
Study
N
Inclusion Method
Results
criteria
Quenby et
22 3 or more Immunohistochemistry ↑ % of CD16+& D56+ cells
al., 1999
idiopathic D19-22
c.f. controls
Manual , Frozen
Liverpool patients
Quenby et
85 3 or more Immunohistochemistry ↑ CD56 +, CD16 -, CD3 – cells
al., 2005
idiopathic D19-23
c.f. controls
Manual, Frozen
UK wide
Drury et al., 120 3 or more Immunohistochemistry ↑ CD56 +, CD16-, CD3- cells
2007
idiopathic LH + 7
Paraffin, Image analysis
UK wide
Tuckerman 87 3 or more Immunohistochemistry ↑ CD56 +,
et al.,
idiopathic LH + 7
Paraffin
NATURE MEDICINE
2006
More blood UNK cells more blood flow
Recurrent miscarriage
Recurrent IVF failure
Quenby et al., 2009
Prednisolone treatment reduces
endometrial angiogenic growth factor
expression at LH+7
0.08
0.07
delta delta Ct
delta delta Ct
0.005
0.004
0.003
0.002
0.04
0.03
0.02
0.01
0
0
BAI1 CCL11 CXCL6 EGF
PlGF
Lash 2008
After Prednisolone Treatment
0.05
0.001
Ang1
Before Prednisolone Treatment
0.06
STAB1
TEK
TNFalpha
Figure 2
A
30
Total Factor VIII vessels
(Mean)
25
20
15
10
Pre
During
5
0
0
5
Case Number
10
15
B
C
14.0
10.0
8.0
6.0
4.0
Pre
2.0
During
0.0
0
5
Case Number
10
15
%SMA complete (Mean)
Total SMA Vessels (Mean)
12.0
D
Pre
0
5
Case Number
10
15
E
25.0
Pre
15.0
10.0
5.0
0.0
0
5
Case Number
10
15
%MyHC complete (Mean)
20.0
Total MyHC Vessels
(Mean)
90
80
70
60
50
40
30
20
10
0
90
80
70
60
50
40
30
20
10
0
Pre
0
5
Case Number
10
During
15
Observational study
Prednisolone in early pregnancy n=22
• Included
–
–
–
–
>5 miscarriages,
high uNK cells preconceptually
Age > 35
Expect 40% live birth rate
• Outcome to date
– 16 term deliveries
– 6 miscarried
– 73% live birth rate
A randomised controlled
trial of prednisolone for
women with recurrent
miscarriage and high levels
of uNK cells in the
endometrium.
•
Funded by The Moulton Charitable Foundation
Inclusion
• >2 consecutive idiopathic first trimester
miscarriages
• >5% of endometrial cells CD56+,
• Age 20 - 40.
Exclusion
• Known cause for pregnancy losses;
• APS
• parental balanced translocation,
• uterine anomaly
• subseptate uterus, cervical weakness
• known thrombophilia
• Contraindications to steroid therapy;
• hypertension, diabetes, mental health problems, obesity BMI>230
Outcome measures
• Primary
• Live birth rate
• Secondary
• Miscarriages
• First/second trimester losses
• Karyotype of miscarried pregnancies
• Still births
• Obstetric complications
• IUGR, Pre-eclampsia, abruption, gestation at delivery
• Fetal abnormality
• Side effects of steroids
Appendix A: Flow diagram of trial procedures- clinical part
Referral letter from GP/Miscarriage Clinic
Letter to patient explaining trial
Screening
See patient mid-luteal phase
Take history, and standard RM investigations
(If not already performed)
If known cause of pregnancy
losses - ineligible
usual care
If uNK <5% ineligible
Usual care
Explain trial and take consent (1)
Take endometrial biopsy
If uNK >5%
Advise woman to ring clinic when she conceives
4-6 weeks gestation
Ultrasound scans to confirm intrauterine pregnancy
consent (2),
given randomisation number,
collect medication from pharmacy who
randomise to prednisolone or placebo according to
randomly generated code.
Alternate week ultrasound scan 4-12 weeks
12 weeks refer to obstetric care
14 weeks collect side effect diary and proforma
20 week anomaly scan
28 and 34 weeks growth scan
6 weeks postnatal, baby outcome proforma
Feasibility trial
• Aims
– See if recruitment possible
– Test trial procedures
– Recruit 40 women
Demographics
Placebo
Prednisone
Age (mean)
33
34
BMI (mean)
26
26
No who had LB
3
4
NO Miscarriages
(median)
4
4
uNK (median)
5.7%
7.2%
Screened 160 women
>2 miscarriages <40
Had timed biopsy
68 (43%) Screen +ve
40 women conceived
randomized
Placebo 20 women
miscarriages
6 fetus, 3 sac
3 ectopic
Karyotype
1/2 trisomy 22
1/2 normal
Live Birth
8 women
40%
Prednisolone 20 women
Live birth
12 women
60%
Relative risk 1.5; 95% CI 0.8-2.9;
absolute risk difference 20%;
95% CI 11%-48%
Miscarriages
4 fetus, 3 sacs
1 molar
Karyotype
1/3 trisomy 22
2/3 normal
Power
calculation
90% power
5% α
Screened 850 women
>2 miscarriages <40
timed biopsy
340 (40%) Screen +ve
214 women conceived
randomized
Placebo 107 women
Live Birth
43 women
40%
Prednisolone 107 women
Live birth
64 women
60%
Success in control group
Author
date
patients
control
Live birth
rate
Kutteh
1996
APS
aspirin
44%
Rai
1997
APS
aspirin
42%
Farquharson
2002
APS
aspirin
72%
Laskin
2009
RM -all
aspirin
78%
Cochrane
heparin
2009
idiopathic
Aspirin
placebo
82%
81%
Kaandorp
2010
idiopathic
Aspirin
placebo
67%
62%
Clark
2010
idiopathic
Intensive
care
80%
El-Zibdeh
2005
idiopathic
placebo
70%
Cochrane IVIG
2010
Idiopathic
control
60%
Stephenson
2010
secondary
placebo
62%
Visser
2010
idiopthic
Aspirin
61%
Quenby
2010
Endometrial
raised NK
cell
placebo
40%
Live birth rates
weeks
general
pop
Biochemical
<5
75%
Clinical
5-10
88%
First
trimester
10-12
97%
reference
Bottomley
2009
0
previous
miscarriages
1
previous
miscarriage
2 previous
miscarriages
3 previous
miscarriages
4
Previous
miscarriages
previous
live birth
94%
86%
77%
72%
58%
95%
Bhattacharya
et al.,
2010
Bhattacharya
et al.,
2010
Bhattacharya
et al.,
2010
Bhattacharya
et al.,
2010
Bhattacharya
et al.,
2010
Bhattacharya
et al.,
2008
What next
• EME grant trial with clinical and science
outcomes?
• Warwick CTU
• Clinical live birth rate
• Science
– Protien damage- Thornalley
– Endometrial differentiation – Brosens
– Angiogensis- Bulmer
– Geneotyping ? Correlated outcome