Introduction

Being asked to write this opinion piece offers the opportunity to reflect on what has changed in the preeclampsia (PE) field over a 40-year career. One could be quite cynical and say not much. Forty years on, we still cannot predict, prevent or treat preeclampsia despite enormous research spending and gathering of data. Over this time specific areas of study have waxed and waned but there is still too much obedience to the dogma and a lack of innovative thinking.

We have progressed from the host of small disparate cross-sectional studies, performed in the 1980’s and 90’s, which described changes in almost every organ system and pathway studied in women with established preeclampsia. These provided the basis for predictive biomarker studies from which we learned that these biomarkers are not specific for preeclampsia per se, nor for all phenotypes of preeclampsia.  The ability to genetically manipulate mice gave us animals with “preeclampsia” defined on the basis of hypertension and proteinuria but did not model the underlying (unknown) pathophysiology of PE.  The basic issue still is if we don’t understand the underlying pathophysiology(ies) then how can we define predictive or preventative strategies?

My involvement in the prediction of preeclampsia dates back to the late 1990’s when I was a faculty member at the University of Cincinnati which was then part of the NIH Maternal-Fetal Medicine Units (MFMU) Network. In 1999 I wrote a review in Seminars of Perinatology (1) on the prediction of preeclampsia highlighting the concept of different underlying subtypes or etiologies that all lead to the PE syndrome and those different combinations of biomarkers might identify these subtypes if used in a longitudinal approach. Jim Roberts had presented the idea to the NIH network of an antioxidant trial to prevent preeclampsia following Lucy Chappell’s Vitamin C/E study (2) and I came along with the idea of a prediction study.

Making Decisions

In 2001 the decision was made to combine the studies into the NIH CAPPS trial. Interestingly there was a lot of discussion on whether to perform the study in high-risk women, with a greater yield for the development of PE, or in low-risk nulliparous women who contribute the highest number of PE patients in clinical practice.  The latter won the day. There followed a frustratingly long time in developing this “Cadillac” of a study, to hopefully not repeat the mistakes of earlier studies such as the NIH low dose aspirin studies where, for example, gestational age at onset of PE was not recorded! As a placental biologist, I was piqued that collection of placentas from the entire cohort (10 years before the human placenta study was even thought of) was deemed too expensive so only a limited number were actually collected. What a waste of a resource! At the same time the SCOPE study was being developed and we shared the CAPPS protocol.

Alas, neither CAPPS (3) nor SCOPE (4) was able to show clinical utility of biomarkers for first-trimester prediction of PE in low-risk women.  This was attributed to the low incidence of PE in the low-risk population, despite over 7500 women being enrolled, together with the idea of heterogeneity of the underlying etiology in these subjects. Perhaps we should have anticipated this based on my 1999 paper! The recent transcriptomic studies of placentas from women with PE (5) have perhaps reinforced the concept of heterogeneous phenotypes, however, it is worth noting that application of high-level bioinformatic analyses to the CAPPS dataset and to the 44 biomarkers in the SCOPE dataset have failed to provide even a hint of different phenotypic clusters. One cannot help but wonder if a study of high-risk women (previous PE, pregestational diabetics, chronic hypertensives, multifetal gestation) each with a more a priori “defined” clinical phenotype might have been more fruitful.

The Future of Pre-Eclampsia Research

Where does the future lie then? If we believe in the concept of heterogeneity of preeclampsia and have failed to reveal it with CAPPS or SCOPE, then even larger cohorts are needed. This is beyond the capabilities of a single center or even a single network and needs international collaboration. To create larger cohorts needs standardization of study design, of clinical definitions and sample collection together with standardized data collection. I have been privileged to work with some remarkable individuals in the Global Pregnancy Collaboration (CoLab) towards this goal. What has been rewarding has been the enthusiasm for this endeavor and the tremendous progress that has been made in development and rollout of the COLLECT database, which is a low-cost solution to the idea of a universal easy-to-use database, often the most expensive piece of a study.

The COLLECT database will be available free of charge for investigators in low and middle-income countries and for a nominal charge ($100 per month) in high-income countries. The data belong to the investigator but COLLECT offers the facility to combine data in a common format into large studies. The harmonization and standardization of studies offers an opportunity to gather large numbers of subjects, to clearly define clinical presentation, clinical phenotypes and link clinical data to tissue biobanks collected in a standardized manner. It may be another 40 years before we can predict, prevent or treat preeclampsia, but gathering high-quality data defining the patients and the problem is an essential step.

References

1. Myatt L, Miodovnik M. Prediction of preeclampsia. Semin Perinatol. 1999 23(1):45-57. PMID: 10102170
2. Chappell LC, Seed PT, Briley AL, Kelly FJ, Lee R, Hunt BJ, Parmar K, Bewley SJ, Shennan AH, Steer PJ, Poston L. Effect of antioxidants on the occurrence of pre-eclampsia in women at increased risk: a randomised trial. 1999 354(9181):810-6. PMID: 10485722
3. Myatt L, Clifton RG, Roberts JM, Spong CY, Hauth JC, Varner MW, Thorp JM Jr, Mercer BM, Peaceman AM, Ramin SM, Carpenter MW, Iams JD, Sciscione A, Harper M, Tolosa JE, Saade G, Sorokin Y, Anderson GD; Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network. First-trimester prediction of preeclampsia in nulliparous women at low risk. Obstet Gynecol. 2012 119(6):1234-42. PMID: 22617589
4. Kenny LC, Black MA, Poston L, Taylor R, Myers JE, Baker PN, McCowan LM, Simpson NA, Dekker GA, Roberts CT, Rodems K, Noland B, Raymundo M, Walker JJ, North RA. Early pregnancy prediction of preeclampsia in nulliparous women, combining clinical risk and biomarkers: the Screening for Pregnancy Endpoints (SCOPE) international cohort study. 2014 64(3):644-52. PMID: 25122928
5. Leavey K, Benton SJ, Grynspan D, Kingdom JC, Bainbridge SA, Cox BJ. Unsupervised Placental Gene Expression Profiling Identifies Clinically Relevant Subclasses of Human Preeclampsia. 2016 68(1):137-47. PMID: 27160201