Could statins be a novel therapeutic for the prevention of preterm delivery? Evidence from an animal model and Phase II feasibility trial

Abstract

Preterm delivery affects 15 million pregnancies annually, and is the leading cause of mortality in the under 5s globally. Spontaneous preterm delivery accounts for 70% of these, and novel preventative strategies are urgently needed. Statins have both anti-inflammatory and anti-contractile properties. Recent evidence suggests statins reduce inflammation-induced preterm delivery in a mouse model. Pravastatin has an increasing safety record in pregnancy, and is used in a research setting for the prevention of pre-eclampsia. However, the feasibility of a trial of pravastatin for the prevention of preterm birth is unknown. The PIPIN trial ran over 15 months, recruiting women with threatened or confirmed preterm labour to a randomised trial of pravastatin 40mg or placebo to be taken once daily for 7 days or until delivery. Recruitment of 7 participants over 15 months indicated that the trial design was not feasible. However, recruitment of 39% of all eligible participants approached indicates that this intervention is acceptable to women, and could be further investigated in an alternative trial design. One such design could include low dose pravastatin given prophylactically to at risk women from second trimester onwards. This thesis also hypothesised that antenatal statin treatment may prevent preterm-birth associated brain injury. A mouse model of inflammation-induced preterm birth and neurological injury was generated using a dose response curve of lipopolysaccharide (LPS) injected between two gestational sacs on day 17 gestation. At 62.5ug/dam LPS, delivery of pups occurred at 27.6 +/- 24 compared with 58.0 +/- 6.3 hours after LPS injection (p<0.05). In surviving pups, brain tissue collected at P0 showed significantly increased il-6 and gfap mRNA, suggesting neuro-inflammation and injury. The addition of antenatal intraperitoneal (IP) simvastatin treatment pre- and post-LPS did not replicate these findings. In the LPS-alone group, time to delivery from LPS did not differ from control, and dams treated with LPS + simvastatin delivered significantly sooner than control alone (36.9+/-19.5 vs 62.0 +/- 12.0 hours). There was no neuro-inflammation or injury found in surviving pups from any group, P0 to P6. Further studies are required to validate the reproducibility of the model prior to drawing conclusions on the therapeutic potential of statins for the prevention of preterm delivery and neurological injury. The solubilisation of simvastatin in strong acid and administration IP may have influenced the results. Comparison of several statins concurrently also requires an alternative method of preparation. A method using flavoured jellies were prepared containing pravastatin or simvastatin was validated and mice entrained to voluntarily eat these after a single overnight fast. Both pravastatin and metabolite were detected peripherally after 7 days oral dosing. The hydroxy-simvastatin metabolite was detected peripherally, but not the parent drug. This fits with known pharmacokinetics, and indicates that oral dosing with flavoured jellies is a feasible route of administration of comparative doses of pravastatin or simvastatin to mice. This could be used in a validated model of inflammation induced preterm delivery e.g. ascending infection with E.coli during pregnancy, to assess maternal and neonatal outcomes.