I started this article 3-4 days ago and was about to post it to the site last night and realized it looked more like a scientific review article for the New England Journal of Medicine than anything most of you would care to read. So, I have pared it down (although it’s still a bit long), hopefully, simplifying enough that the points will come through without compromising the scientific integrity of the work being done in this area. The general concepts are what is important here and not the scientific jargon. I want you to know where the current research is taking us and offer some thoughts of my own regarding risks and opportunities of therapeutic advances in this area.
I really could go on for many more posts related to hypertensive disorders in pregnancy, but I think I will call it quits (for awhile) after this one, lest I run the risk of boring to tears all of you who are not interested in the subject. If you haven’t figured it out by now, I am VERY interested in it; and, if you haven’t noticed already, much of what we have discussed related to preeclampsia overlaps with the subject of recurrent early pregnancy loss upon which we have spent lots of time! Indeed, I often look, rather simplistically, at the issues related to abnormalities of placentation that comprise the most common source of preeclampsia as pregnancies that almost miscarried, but didn’t quite; and, the ones that came closest to being lost are also the ones most likely to result in the more severe forms of preeclampsia.
I think the greatest advances that will be made in the management of preeclampsia over the next decade will be the early identification of the biological ‘markers/mediators’ involved in its pathogenesis, applying this knowledge to identifying patients truly at risk, and implementing specific interventions designed to counteract their effects as a means of preventing, reducing the risk of developing, or decreasing the severity of preeclampsia that eventually develops. It just makes sense, if most preeclampsia is the end result of abnormalities of placentation that occur in early pregnancy, then there MUST be factors that can be detectable during the first and early second trimester of pregnancy that reliably reflect the pathologic processes in progress.
Indeed, many studies are now being conducted around the world to identify and characterize just such factors. One group of factors, under intense investigation, is antiangiogenic substances. If you recall in our 4th post on this subject, we pointed out that angiogenic (blood vessel growth promoting) and other growth factors are probably critical to the success of early placentation in terms of trophoblast invasion, migration to, and remodeling of, the spiral arterioles, and arborization and vascularization of the placental villi. Two ANTIangiogenic substances currently in the news are soluble fms-like tyrosine kinase-1 (sFlt-1) and soluble endoglin (sEng). Bear with me while I give you an overview of this fascinating area of research…
In 2003, Maynard and colleagues (J Clin Invest 2003;111:649-58) reported that the soluble (circulating) placental substance sFlt-1 is produced in greater amounts in women with preeclampsia. We now know this upregulation, or increase in sFlt-1 production, are the consequence, at least in part, of a placental environment with suboptimal levels of oxygen. sFlt-1 is an antagonist of vascular endothelial growth factor (VEGF) and placental growth factor (PlGF). Indeed, sFlt-1 actually binds to these growth factors (and is now sometimes referred to as soluble VEGF receptor-1, or sVEGFR-1), thereby, preventing them from interacting with endothelial cells. In a series of elegant experiments, Maynard and colleagues demonstrated that serum from preeclamptic women with increased levels of sFlt-1 caused endothelial cell (cells on the inner lining of blood vessels) dysfunction and arteriolar constriction in vitro and that these deleterious effects could be overcome with addition of exogenous VEGF and PlGF. Furthermore, they showed that the “administration of sFlt-1 to pregnant rats induces hypertension, proteinuria, and glomerular endotheliosis, the classic lesion of preeclampsia” found in the kidneys.
While these observations (and many others I will not mention today) suggested that an excess of sFlt-1 in preeclamptic women (now known to arise from both placental and maternal sources) might contribute to the final pathogenesis of the end-stage preeclamptic syndrome, it did not answer the questions related to the actual cause of the abnormalities of placentation in early pregnancy that set the stage for placental hypoxia and sFlt-1 upregulation. Subsequent studies by other investigators have suggested a role for sFlt-1 in the pathogenesis of abnormal placental growth and development. As examples of the work in this area, Ahmad and Ahmed (Circ Res 2004;95:884-91) have shown that serum from preeclamptic women inhibits endothelial cell migration and angiogenesis and that this could be restored by specific removal of sFlt-1 from the serum to levels comparable to that found in the serum of nonpreeclamptic women. Nagamatsu, et al. (Endocrinology 2004;145:4838-45) have shown in tissue culture experiments that the cytotrophoblasts (placental cells) involved in the uterine invasion and vascular remodeling of maternal vessels not only can make VEGF, but also produce sFlt-1 in response to a reduced oxygen environment, thereby, countering the ‘angiogenic’ properties of VEGF. This suggests that the cytotrophoblasts themselves have an innate mechanism for regulating the extent and limits of their invasion into maternal tissues and that this control depends in part on sFlt-1.
In other posts, we spoke in broad generalities regarding the presumptive importance of the maternal immune response in both normal and abnormal placental development. As an example of the mother’s role in the regulatory processes under discussion today, Matsubara and colleagues (Reprod Immunol 2004;68:27-37) reported that decidual mononuclear cells can be induced by the lymphokine interleukin-2 (IL-2) (produced by T-lymphocytes as part of the immune response) to become ‘killer’ cells (lymphokine-activated killer, or LAK, cells). The LAK cells were found by these investigators to reduce the angiogenic activity of placental cytotrophoblasts and appear to do so by upregulating the production of sFlt-1 since they did not actually affect the amount of VEGF or PlGF made by the trophoblasts. Recently, Lockwood and colleagues (Am J Pathol 2007;170:1398-405) have demonstrated that the maternal decidual cells themselves (and not just the fetal trophoblasts) express sFlt-1 messenger RNA (mRNA) suggesting that these cells also can produce sFlt-1. Thus, these articles suggest at least two maternal regulatory mechanisms for limiting trophoblast invasion that, when ‘over-expressed’, could prevent normal placentation and lead to preeclampsia. Indeed, as part of the Lockwood report, it was also shown that the presence of thrombin enhanced sFlt-1 production by first trimester (but not term) decidual cells. This might offer a mechanism by which abnormalities of the coagulation system that cause excessive clotting (thrombin generation) in both recurrent pregnancy loss and preeclampsia might mediate their deleterious first trimester effects!
Now, briefly let’s discuss another antiangiogenic factor currently under investigation – soluble endoglin (sEng). In 2006, Venkatesha, et al. (Nat Med 2006;12:642-9) reported that “a novel placenta-derived soluble TGF (transforming growth factor)-beta coreceptor, endoglin (sEng)…is elevated in the sera of preeclamptic individuals (and) correlates with disease severity...” sEng was also found in in vitro experiments to inhibit angiogenesis and in vivo, and significant with regard to end-stage preeclampsia pathogenesis, to increase vascular permeability (leakiness) and induce hypertension. In pregnant rats, its effects were found to be amplified by the coadministration of sFlt-1, “leading to severe preeclampsia including the HELLP (hemolysis, elevated liver enzymes, low platelets) syndrome and restriction of fetal growth.” The suggested mechanism of action on the blood vessels is impairment of TGF-beta binding to its receptors that is required for activation of pathways normally resulting in vasodilation (blood vessel relaxation) and more recent studies have supported this hypothesis (Santibanez, et al., J Cell Physiol 2007;210:456-68).
So, we now have two ‘markers’ for preeclampsia that may directly participate not only in the pathogenesis of the placental abnormalities early in pregnancy, but also in the ‘end-stage’ changes that characterize preeclampsia syndromes. This does give us specific targets for medical intervention. The questions remain, however, can the detection of sFlt-1 and/or sEng help us to diagnose the patient who is going to develop preeclampsia early enough in the pregnancy that medical intervention can be started, what interventions may be possible based on the ‘science’ to date, and will such interventions be safe for BOTH mother and baby?
Several recent studies have shown that these antiangiogenic factors increase and can be detected significantly before the onset of the preeclamptic syndrome. Levine and colleagues (NEJM 2006;355:1056-8) showed that sEng levels increase 2-3 months before the onset of preeclampsia in women who developed either preterm preeclampsia or term preeclampsia. Similarly, Rana, et al. (Hypertension 2007;50:137-42) demonstrated that both sFlt-1 and sEng are elevated by 17-20 weeks in women destined to become preeclamptic. Unfortunately, most of the placental development (or lack thereof) of significance related to preeclampsia is completed by this point in the pregnancy. If these markers are to be truly useful in altering the course of preeclampsia, we are going to have to establish their reliability in predicting it at earlier gestational ages.
With regard to ‘treatment’, several of the investigations cited above have shown that the effects of sFlt-1 can be overcome by the addition of excess VEGF (or the removal of sFlt-1 or sEng). I may be wrong about this but, unless this is proven to work early in pregnancy to improve the placental growth and development, I am afraid that such therapy later in pregnancy may improve the maternal condition at the expense of the baby. Remember, the placenta appears to be driving maternal preeclampsia because it is ‘unhappy’, probably as the result of insufficient oxygen. It does not make the mother sick capriciously; it is trying to improve its own condition by the only mechanisms it has at its disposal. Any therapy that only improves the maternal side of the equation has to impact negatively on the fetal side!
In closing, I would like to mention one other recent study that opens up another possible avenue for therapy. Cudmore and colleagues (Circulation 2007;115:1789-87) pointed out that heme oxygenase-1 (HO-1) and its metabolite carbon monoxide (CO) exert protective effects under conditions of oxygen deprivation and hypothesized that this might be the result of downregulation of sFlt-1 and sEng production and release. In a series of experiments, they found that mice lacking HO-1 produced higher levels of sFlt-1 and sEng than normal mice; enhancement of HO-1 production (using an adenovirus vector) inhibited release of both sFlt-1 and sEng from endothelial cells under conditions that would normally increase their production; the CO-releasing molecule (CORM-2) or CO itself inhibit sFlt-1 release; and, treatment of endothelial cells with statins that upregulate HO-1 also inhibit the release of sFlt-1. To me these findings are very exciting, because they may offer an option for early medical therapy that may have the greater chance for success in the actual prevention of the placental changes that contribute to the pathogenesis of preeclampsia. These studies also may help to explain why cigarette smoking (increased CO) seems to reduce the risk of preeclampsia by 50% or more!?!
I have one other thought and then I promise to bring this series to a close. I firmly believe that to have the greatest impact on preventing preeclampsia (and early pregnancy loss unrelated to aneuploidy), we need to maximize a favorable maternal immune response to pregnancy. Perhaps, someday we will have a ‘pregnancy vaccine’ that will do just that! Thanks for sticking with me on this. Hope some of you had as much fun as I did!