a

L

CP

hptpSmtp

D(c1dDa

pr

cwncv‘Pfc

nwfmepg

p1ErUr

www.AJOG.org Prematurity Oral Concurrent Session 1

weeks. The only SNP protective against PTB was found in the IL3 gene(rs3091336).CONCLUSION: We identified 8 novel SNPs in genes critical to inflam-mation, cell signaling, and angiogenesis that are associated with pre-term delivery in black women. These genetic variants may be impor-tant in the pathogenesis of preterm birth, however further studies areneeded to elucidate their precise role.

13 Prevention of preterm birth with progestationalgents: revealing molecular mechanisms

Christopher Nold1, Monique Maubert1,auren Anton1, Michal Elovitz1

1University of Pennsylvania Perelman School of Medicine, Maternal andhild Health Research Program, Department of Obstetrics and Gynecology,hiladelphia, PA

OBJECTIVE: Clinically, vaginal progesterone (Vag P) and 17 alpha-ydroxyprogestreone caproate (17P) have been shown to preventreterm birth (PTB) in high risk populations. We hypothesizehese agents may be preventing PTB by altering signal transductionathways that would promote premature cervical remodeling.pecifically, we hypothesize that progestational agents modify theucosal immune response in the cervix through altered produc-

ion of TH17 cytokines and increased expression of anti-microbialroteins.

STUDY DESIGN: Using a mouse model, on days E14-E17 CD-1 preg-nant mice were treated with either 0.1cc of 10mg/ml of 17P subcuta-neously, 0.1cc of castor oil (CO) subcutaneously, 0.1 cc of 25 mg/ml ofprogesterone in Replens vaginally, or 0.1cc of Replens vaginally, withfour dams per treatment group. Mice were sacrificed six hours aftertreatment on E17.5. Cervices were collected and quantitative real-time polymerase chain reaction (qPCR) was performed on the follow-ing targets: IL17, IL22, IL22R, Defensin 1, Defensin 3, Defensin 4,SLP1, and Claudin-2.RESULTS: Exposure to Vag P significantly increased the expression of

efensin 1 compared to Replens (p�0.01), CO (p�0.001), and 17Pp�0.001). Vag P also significantly increased expression of IL22Rompared to CO (p�0.05) and 17P (p�0.05). Treatment with either7P or Vag P did not have a statistically significant effect on the pro-uction of SLP1, or Claudin-2. Baseline mRNA values of IL17, IL22,efensin 3, and Defensin 4 levels were very low; expression was not

ltered by any of the treatments.CONCLUSION: These studies demonstrate that progesterone adminis-tered vaginally can alter the musical immune response. Notably, VagP, through an increase in Defensin 1 and IL22R expression, served toboost the host mucosal immune response in the cervix. Whether thesemolecular changes from Vag P result in a functional effect and are akey mechanism by which Vag P prevents PTB requires further study.

†All P � .001.

Importantly, 17P had no effect on these pathways in the cervix.

Suppl

14 Extravillous trophoblasts within the placental basallate serve as a protected niche for intracellular bacterialeservoirs: potential new etiology for preterm birth

Bin Cao1, Indira Mysorekar1

1Washington University School of Medicine, Ob/Gyn, St. Louis, MOOBJECTIVE: Infection and inflammation are important mechanismsontributing to Preterm birth (PTB). However, the mechanism byhich pathogens infect and induce pathological change during preg-ancy remains unknown as does the contribution of latent andhronic infection. Antibiotic treatment is relatively ineffective for pre-enting PTB suggesting that the putative infectious organisms may behidden’ within the tissue. Our central hypothesis is that one cause ofTB is that bacterial pathogens establish clinically occult reservoirs in

etal extravillous trophoblasts (EVTs) on the maternal side of the pla-enta (basal plate).

STUDY DESIGN: Placental basal plate biopsies were collected from�200 women with PTB �37 weeks and controls �37 weeks, stainedith the Brown-Hopps modification of the Gram-stain and examined

or intracellular bacteria. To explore the mechanisms underlying for-ation of intracellular bacterial reservoirs, ex vivo human basal plate

xplants were developed and infected with Gram-negative. (Uro-athogenic E. coli, UPEC) and Gram-positive (Listeria monocyto-enes) pathogens.

RESULTS: Intracellular bacteria were documented in over a third ofreterm placentas and were exclusively localized to fetal EVTs (FigureA-B). We show that both UPEC and Listeria invade into HLA-G�VTs embedded within the basal plate. Bacteria appear to hone di-ectly to fetal EVTs and not maternal stromal cells. Interestingly,PEC appear to form biofilm-like clusters in EVTs whereas Listeria

emain as single organisms (Figure 1C-D).CONCLUSION: Our work has thus demonstrated that EVTs are highlysusceptible to bacterial invasion and colonization likely due to theirunique immune-privileged status. We show that Gram positive andnegative bacteria may use different mechanisms to persist. We positthat such reservoirs persisting within EVTs in the basal plate couldre-emerge and predispose to adverse pregnancy outcomes such as

PTB.

ement to JANUARY 2013 American Journal of Obstetrics & Gynecology S9

s

S

K

nbak

t(rstpon

to

K

KS

dmpcaftlbt

twsiea

stm2tc

ad

RLRV

l

BO

pp

mwi

s

Oral Concurrent Session 1 Prematurity www.AJOG.org

15 The influence of prenatal smokingtatus on cervical cytokine distribution

Kristin Ashford1, John O’Brien2, Andrea McCubbin1,usan Westneat1, Janine Barnett1

1University of Kentucky, College of Nursing, Lexington, KY, 2University ofentucky, Deparment of Obstetrics and Gynecology, Lexington, KY

OBJECTIVE: Maternal tobacco use and cervical immune responsive-ess are individually associated with PTB, and the interaction of to-acco use on cervical cytokines requires further study. Our primaryim is to determine whether prenatal smoking impacts cervical cyto-ine distribution.

STUDY DESIGN: This was a planned secondary analysis of a prospective,longitudinal multi-center trial, pregnant women with singleton ges-tation underwent cervicovaginal fluid (CVF) sampling during eachtrimester. Cytokines (IL-1a, IL-1b, IL-2, IL-6, IL-8, IL-10, TNFa) andCRP were measured using a multiplex beadlyte assay on a LuminexIS-100. At each timepoint, smoking status was collected via self report(survey) and validated by preset urine cotinine limits. Cytokine datawere log transformed.RESULTS: Eighty-one women underwent evaluation in the first trimes-er, 76 in the second and 69 in the third. Thirty-nine of the participants48%) were smokers and self-report smoking status was strongly cor-elated with urine cotinine (� � .68). First trimester cervical IL -10 wasignificantly decreased (p � .01) in smoking women compared tohose who did not admit to smoking (2.3 � 1.2 pg/mL vs. 3.8 � 2.1g/mL, respectfully). No significant differences were observed inther cytokine concentration of cervical mucus between smoking andonsmoking women at any other gestational age.

CONCLUSION: Prenatal tobacco use decreases IL -10 concentration incervical mucus potentially increasing susceptibility to pro-inflamma-tory mediators. Deregulation of this anti-inflammatory cytokine mayalter cervical epithelial structure and immune responsive; potentiallyimpacting cervical length and preterm birth risk.

A and B, Intracellular bacteria in EVTs in placental basal plate. C and D, UPEC and Listeria invading andurviving intracellularly within EVTs in placental basal plate explants (HLA-G�, red).

2

S10 American Journal of Obstetrics & Gynecology Supplement to JANUARY 20

16 Serum derived exosomes evaluated early in secondrimester provide predictive biomarkersf risk for preterm birth before week 34

Alan Ezrin1, Brian Brohman1, Sarah Baxter3,eith Moore2, Mike Luther2

1NX PharmaGen Inc., Maternal Fetal Medicine Program, Louisville, KY,2David H. Murdock Research Institute, Dept. of Immunobiology,

annapolis, NC, 3David H. Murdock Research Institute, Dept. of Analyticalcience, Kannapolis, NC

OBJECTIVE: The purpose of this study was to confirm in an indepen-ent, blinded manner whether a proteomic differentiation exists inaternal blood-derived exosomes obtained from term and preterm

atients early in the second trimester. Current methods indicate thatirculating microparticles such as exosomes shed from tissue can serves stable biomarker-harboring vehicles. Placental-derived exosomesreely circulate in high titer in maternal blood and reflect the status ofhe syncytiotrophoblast maternal-fetal interface. Previous work (Tay-or et al., unpublished) has shown that subtle homeostatic changes cane reflected via a unique set of dysregulated exosome-associated pro-eins when pregnancy complications occur.

STUDY DESIGN: Frozen maternal serum samples (n�48) obtained be-ween 15-17 weeks gestation were analyzed from asymptomaticomen (avg. age 28) bearing live birth (excluded: multiple births;

ystemic disease; known fetal anomalies). Exosomes were isolated us-ng gel filtration, with samples blinded to outcome; proteins werextracted and analyzed using an open proteomic LC-MS differentialnalysis between term (�37wks) and preterm cohorts (�34wks).

RESULTS: Among 213 proteins identified, a unique pattern was ob-erved in serum derived exosomes isolated from asymptomatic pa-ients that subsequently delivered preterm. Using an ANOVA assess-

ent and rigorous reproducibility, accuracy, and confidence criteria,1 proteins were characterized across two study phases differentiatinghe cohorts, with functional analysis implicating inflammatory andell injury pathways.

CONCLUSION: A novel library of statistically valued exosomal biomark-ers has been identified with expected clinical usefulness for the earlyidentification of women at risk for preterm birth, as well as improvedunderstanding of the underlying pathways leading to preterm birth.Priority markers are being further evaluated for commercial deploy-ment in a small-plex assay.

17 Vaginal progesterone as maintenance treatment aftern episode of preterm labor (PROMISE Study): a randomized,ouble blinded, placebo-controlled trial

Montse Palacio1, Teresa Cobo1, Eugenia Antolín2, Maríaamirez3, Francisco Cabrera4, Fernando Mozo de Rosales5, Joséuis Bartha6, Miquel Juan7, Anna Martí8, Daniel Orós9, Àguedaodríguez10, Elena Scazzocchio11, José María Olivares12, Saraarea1, on behalf of the PROMISE Collaborative Group1

1Hospital Clínic Barcelona, Maternal Fetal Medicine, Barcelona, Spain,2Hospital Gregorio Marañón, Obs&Gyn, Madrid, Spain, 3Hospital Virgen dea Macarena, Obs&Gyn, Sevilla, Spain, 4Complejo Hospitalario Universitario

lnsular, Obs&Gyn, Las Palmas de Gran Canaria, Spain, 5Hospital de Basurto,Obs&Gyn, Bilbao, Spain, 6Hospital Puerta del Mar, Obs&Gyn, Cádiz, Spain,7Hospital de Son Llàtzer, Obs&Gyn, Mallorca, Spain, 8Hospital Sant Joan deDéu, Obs&Gyn, Manresa, Spain, 9Hospital Clinico Universitario “Lozano

lesa�, Obs&Gyn, Zaragoza, Spain, 10Corporació Sanitaria Parc Tauli,bs&Gyn, Sabadell, Spain, 11USP lnstitut Universitari Dexeus, Obs&Gyn,

Barcelona, Spain, 12Hospital de Terrassa, CST, Obs&Gyn, Terrassa, SpainOBJECTIVE: To demonstrate if maintenance treatment with vaginal

rogesterone after an arrested preterm labor reduces the incidence ofreterm delivery before 34.0 or 37.0 pregnancy weeks.

STUDY DESIGN: This was a prospective, randomized, double blind,ulticenter clinical trial. 12 centers all over Spain were includingomen with single pregnancy and preterm labor successfully arrested,

n whom decision of discharge was made, gestational age between

4.0-34.0 weeks and cervical length � 25mm. Women were randomly

13