To the Editor: Placenta accreta spectrum (PAS) disorders refer to a spectrum of abnormal placentation. PAS subtypes include placenta accreta (creta or adherenta, PA), placenta increta (PI), and placenta percreta (PP). The incidence of PAS has dramatically increased over the past four decades. A study using a nationally representative sample in the United States showed that the prevalence of PAS was 0.29% among women who underwent cesarean delivery with live birth.[1] Women with PAS are at a considerably higher risk of concurrent hemorrhage, bladder and urinary tract injury, and hysterectomy during delivery.[1]

Accurate prenatal confirmation of PAS is crucial to conducting timely multidisciplinary management and preoperative planning to improve maternal outcomes. As a primary tool for the diagnosis and risk stratification of PAS, ultrasound evaluation has been operator-dependent and limited in the detection of the extent of posterior placentation invasion and parametrial extension. Magnetic resonance imaging is not advocated as a routine diagnostic approach because of its higher cost and limited clinical value. Additional factors need to be explored to further improve the PAS diagnosis, assessment of PAS severity and prediction of adverse perioperative outcomes. The detection process of maternal circulating biomarkers is objective, noninvasive, and cost-effective. Previous studies have shown that biomarkers may have potential value in PAS diagnosis. However, to date, reported PAS biomarkers remain inconclusive and are not used in clinical practice. We aimed to identify biomarkers with potential clinical value related to PAS-associated biological processes and to explore the underlying heterogeneity in the research design. We searched for studies describing PAS biomarkers in eight Chinese and English electronic databases (Pubmed, MEDLINE, EMBASE, Cumulative Index to Nursing and Allied Health Literature [CINAHL], Cochrane, ClinicalTrials.gov, China National Knowledge Internet and Wanfang database) from 2010 to 2021. We summarized the potential pathologic processes associated with the changes in biomarkers in maternal circulation. We also compared and summarized the research design of 29 studies included in this article [Supplementary Table 1, https://links.lww.com/CM9/B477], including the biomarker, aim of research, testing time, number of total cases, trend in PAS, and area under the curve (AUC).

Biomarkers reported in the PAS studies could be divided into five categories: (1) peptides and proteins, (2) vascular endothelial factors, (3) DNA, (4) RNA, and (5) cells [Supplementary Table 1, https://links.lww.com/CM9/B477]. These biomarkers could also be classified on the basis of five abnormal biological processes of PAS. (1) Excessive extravillous trophoblast (EVT) invasion: Studies have shown that interleukin-8 (IL-8) and hyperglycosylated human chorionic gonadotropin (H-hCG) stimulate EVT invasion in PAS, whereas decorin, a transforming growth factor beta-binding proteoglycan in decidual cells, suppresses EVT invasion. Downregulation of micro RNA (miR)-139-3p and miR-196a-5p was found to promote the epithelial–mesenchymal transition of EVT, thereby maintaining EVT invasion in PAS. Upregulation of cell-free fetal DNA, troponin I, creatine kinase (CK)-MB (CK-MB), CK, and circulating cytotrophoblasts (cTBs) might result from excessive EVT invasion. Anti-thrombin III is a key player in the coagulation cascade and inflammation, and its upregulation is caused by the inflammatory state of the placenta induced by excessive EVT invasion in PAS. (2) Enhanced angiogenesis: Vascular endothelial growth factor (VEGF), placental growth factor (PLGF), and soluble vascular endothelial growth factor receptor 2 (sVEGFR2) were overexpressed by EVT, promoting angiogenesis in PAS. Anti-angiogenic soluble Fms-like tyrosine kinase 1 (sFlt-1) declined in PAS. In contrast, sTie2 (a receptor tyrosine kinase expressed by endothelial cells with a role in angiogenesis) was elevated in PAS. Plasminogen activator inhibitor 1 (PAI-1) blocks the activation of plasminogen to plasmin, resulting in fibrin-activating signaling pathways that promote angiogenesis. Downregulation of PAI-1 in PAS may reflect a compensatory change in PAS. (3) Abnormal placental function: Pregnancy-associated plasma protein A (PAPP-A), β-human chorionic gonadotropin (β-hCG), alpha-fetoprotein (AFP), and markers of placental development in the maternal circulatory system have been shown to be related to pregnancy complications and reflect abnormalities in placental function. (4) Enhanced necrosis/apoptosis: Nucleic acids in the maternal circulation system are cell-free and can be released from deported trophoblast knots, breakdown of apoptotic/necrotic cells, and blebbing of microvesicles from trophoblast membranes. (5) Imbalance of oxidative stress: Uyanikoglu et al[2] reported an imbalance in the oxidant/antioxidant status characterized by decreased native and total thiols.

The heterogeneity of the study design included the diagnostic criteria of PAS, aim of research, testing trimester, sample size, control selection, sample types, and testing methods [Supplementary Table 1, https://links.lww.com/CM9/B477]. Specifically, not all the studies used pathological criteria for PAS confirmation. Most studies have focused on the diagnostic value of biomarkers in distinguishing PAS from non-PAS, while a few studies have explored the association between biomarkers and PAS severity as well as maternal outcomes. AFP, PAPP-A, human chorionic gonadotropin (hCG), and CK were targets in the first trimester, while vascular endothelial growth factors and peripheral blood nucleic acids were mostly identified during the second or third trimester. Most studies applied a case–control design to explore and quantify maternal circulating biomarkers of PAS. The total sample size of the included studies varied from 20 to 736. We also noticed that the current studies with large sample sizes were mostly based on reanalysis of the Down's screening database during the first and second trimesters, rather than the PAS-associated cohort. In addition to healthy-term pregnancies, patients with placenta previa alone were included as a control. Plasma and serum were the main research samples. Most studies targeted single biomarkers using enzyme-linked immunosorbent assays. Shainker et al[3] used high-throughput proteomic technology to screen 1305 potential plasma biomarkers. Afshar et al[4] reported a nanostructure-embedded microchip that efficiently enriched both single and clustered cTBs from maternal blood for the detection of PAS. In PAS, elevated biomarkers included AFP, PAPP-A, ischemia modified albumin (IMA), troponin I, pro-brain natriuretic peptide (ProBNP), decorin, VEGFR2, sTie2, antithrombin III (AT III), cell-free fetal RNA (cffRNA) and cTBs. In contrast, H-hCG, thiol, tumor necrosis factor-related apoptosis-inducing ligand receptor 2 (TRAIL-R2), VEGF, PLGF, sFlt-1, and specific microRNAs (miRNAs) were downregulated in PAS. We also observed that β-hCG showed opposite trends in different studies. For diagnostic performance, the distribution of the AUC ranged from 0.573 to 0.976 [Supplementary Table 1, https://links.lww.com/CM9/B477]. We noticed that the combination of biomarkers with other indicators, such as other biomarkers, clinical risk factors and imaging findings, could improve the AUC. Some traditional and nonspecific molecules did not perform well, such as AFP and β-hCG. Among these biomarkers, VEGF and cTBs in the maternal circulation combined with imaging findings provided an excellent diagnostic performance for distinguishing PAS from non-PAS.

There are several considerations and recommendations for future studies of PAS biomarker. First, the current diagnosis of PAS is mainly based on clinical findings rather than histopathology. However, typical histopathological signs always vanish when surgeons attempt to detach the placenta in clinical practice. Therefore, although the gold standard of diagnostic criteria for PAS is histopathological evidence, clinical diagnosis is also possible. Second, cohort studies tracing the dynamic changes in biomarkers along with PAS development throughout pregnancy are lacking. Related studies should also focus on the different subtypes of PAS because of the different preoperative management plans and maternal outcomes. Third, sample size is critical in prediction research and should consist of screening, training, and validating phases. The cohort of PAS patients was limited because of the relatively low prevalence of PAS, especially the placenta percreta subtype. To control for individual variations and confounding factors affecting conclusions, large-scale, normative multicenter studies are needed to validate such findings. Fourth, the identification of biomarkers in PAS patients without placenta previa should not be neglected. Carusi et al[5] showed that PAS patients without placenta previa were less likely to be diagnosed in time, thereby missing the opportunity for well-developed multidisciplinary management. The severe maternal risks of the non-previa group, including reoperation, pulmonary edema, acute kidney injury, and thromboembolism, were similar to those of the placenta previa PAS group. Fifth, the current studies mainly focus on maternal plasma and serum. However, other biological fluids, such as maternal urine, fetal umbilical blood, and amniotic fluid, may also provide valid samples to identify the pathogenesis of PAS. Sixth, for testing methods, high-throughput technologies, including proteomics, genomics, and metabolomics, should be applied to provide more robust and unbiased information. Seventh, the current biomarkers were also applicable to other pregnancy complications, especially for placenta derived disorders, such as preeclampsia and fetal intrauterine growth restriction. Further studies should focus on specific biomarkers based on the pathophysiological mechanisms of PAS. Finally, there were some single biomarker studies reported without the combination of imaging findings. The ideal diagnosis pattern of PAS should combine imaging and biomarker findings. In future studies, integrating clinical high-risk factors, imaging and maternal biomarker findings is warranted.

Antenatal confirmation of PAS diagnosis and severity is critical to improving maternal outcomes. A noninvasive screening pattern integrating biomarkers, imaging findings and clinical risk factors for the early detection of PAS will be valuable to identify PAS patients, especially in low-resource and rural settings lacking subspecialists trained in ultrasound. Standardized procedures involving every aspect of the study design are needed to discover the maternal circulating biomarkers that can distinguish PAS from non-PAS. We believe that our correspondence provides constructive suggestions and recommendations for the future studies involving biomarkers of PAS.

Funding

This study was supported by grants from the Strategic Collaborative Research Program of the Ferring Institute of Reproductive Medicine (No. FIRMA 181104) and the National Key Research and Development Program of China (No. 2021YFC2700700).

Conflicts of interest

None.

References 1. Matsuzaki S, Mandelbaum RS, Sangara RN, McCarthy LE, Vestal NL, Klar M, et al. Trends, characteristics, and outcomes of placenta accreta spectrum: a national study in the United States. Am J Obstet Gynecol 2021;225: 534.e1–534.e38. doi: 10.1016/j.ajog.2021.04.233. 2. Uyanikoglu H, Sak ME, Tatli F, Hilali NG, Sak S, Incebiyik A, et al. Serum ischemia modified albumin level and its relationship with the thiol/disulfide balance in placenta percreta patients. J Obstet Gynaecol 2018;38: 1073–1077. doi: 10.1080/01443615.2018.1450369. 3. Shainker SA, Silver RM, Modest AM, Hacker MR, Hecht JL, Salahuddin S, et al. Placenta accreta spectrum: biomarker discovery using plasma proteomics. Am J Obstet Gynecol 2020;223: 433.e1–433.e14. doi: 10.1016/j.ajog.2020.03.019. 4. Afshar Y, Dong J, Zhao P, Li L, Wang S, Zhang RY, et al. Circulating trophoblast cell clusters for early detection of placenta accreta spectrum disorders. Nat Commun 2021;12: 4408. doi: 10.1038/s41467-021-24627-2. 5. Carusi DA, Fox KA, Lyell DJ, Perlman NC, Aalipour S, Einerson BD, et al. Placenta Accreta Spectrum Without Placenta Previa. Obstet Gynecol 2020;136: 458–465. doi: 10.1097/AOG.0000000000003970.