The Role of cfDNA Biomarkers and Patient Data in the Early Prediction of Preeclampsia: Artificial Intelligence Model.

OBJECTIVE: Accurate individualized assessment of preeclampsia risk enables the identification of patients most likely to benefit from initiation of low-dose aspirin at 12-16 weeks' gestation when there is evidence for its effectiveness, as well as guiding appropriate pregnancy care pathways and surveillance. The primary objective of this study was to evaluate the performance of artificial neural network models for the prediction of preterm preeclampsia (<37 weeks' gestation) using patient characteristics available at the first antenatal visit and data from prenatal cell-free DNA (cfDNA) screening. Secondary outcomes were prediction of early onset preeclampsia (<34 weeks' gestation) and term preeclampsia (≥37 weeks' gestation). METHODS: This secondary analysis of a prospective, multicenter, observational prenatal cfDNA screening study (SMART) included singleton pregnancies with known pregnancy outcomes. Thirteen patient characteristics that are routinely collected at the first prenatal visit and two characteristics of cfDNA, total cfDNA and fetal fraction (FF), were used to develop predictive models for early-onset (<34 weeks), preterm (<37 weeks), and term (≥37 weeks) preeclampsia. For the models, the 'reference' classifier was a shallow logistic regression (LR) model. We also explored several feedforward (non-linear) neural network (NN) architectures with one or more hidden layers and compared their performance with the LR model. We selected a simple NN model built with one hidden layer and made up of 15 units. RESULTS: Of 17,520 participants included in the final analysis, 72 (0.4%) developed early onset, 251 (1.4%) preterm, and 420 (2.4%) term preeclampsia. Median gestational age at cfDNA measurement was 12.6 weeks and 2,155 (12.3%) had their cfDNA measurement at 16 weeks' gestation or greater. Preeclampsia was associated with higher total cfDNA (median 362.3 versus 339.0 copies/ml cfDNA; p<0.001) and lower FF (median 7.5% versus 9.4%; p<0.001). The expected, cross-validated area under the curve (AUC) scores for early onset, preterm, and term preeclampsia were 0.782, 0.801, and 0.712, respectively for the LR model, and 0.797, 0.800, and 0.713, respectively for the NN model. At a screen-positive rate of 15%, sensitivity for preterm preeclampsia was 58.4% (95% CI 0.569, 0.599) for the LR model and 59.3% (95% CI 0.578, 0.608) for the NN model.The contribution of both total cfDNA and FF to the prediction of term and preterm preeclampsia was negligible. For early-onset preeclampsia, removal of the total cfDNA and FF features from the NN model was associated with a 6.9% decrease in sensitivity at a 15% screen positive rate, from 54.9% (95% CI 52.9-56.9) to 48.0% (95% CI 45.0-51.0). CONCLUSION: Routinely available patient characteristics and cfDNA markers can be used to predict preeclampsia with performance comparable to other patient characteristic models for the prediction of preterm preeclampsia. Both LR and NN models showed similar performance.

Randomized clinical trial of a novel donor-derived cfDNA test to detect rejection in CPV-simulated renal transplant patients.

PURPOSE: Five-year kidney graft loss currently stands at about 30%. We evaluate the clinical utility of a blood test measuring donor-derived cell-free DNA that detects rejection earlier and, potentially, improves diagnostic and therapeutic accuracy. METHODS: In a randomized controlled experiment, we measured the clinical practice of 175 practicing nephrologists, both with and without the use of dd-cfDNA testing. Providers cared for six simulated post-renal transplant patient cases whose ages ranged from 30 to 75 years and were 3-24 months post-transplant with typical presentations. RESULTS: 154 nephrologists completed two rounds of simulated cases. At baseline, the study arms performed similarly, demonstrating no significant differences either in primary diagnosis (p = 0.853), decisions to biopsy or refer to transplant center (p = 1.000), or therapeutic management (p = 0.488). After introduction of the dd-cfDNA test, intervention nephrologists were more likely to arrive at the diagnosis of rejection (OR 4.00, 95% CI 1.93-8.30), make a correct decision on biopsy/transplant center referral (OR 11.07, 95% CI 4.87-25.16), and properly adjust therapeutic management (OR 2.37, 95% CI 1.07-5.24). CONCLUSION: A sample of nationally representative, practicing nephrologists given dd-cfDNA to evaluate post-transplant patients were more likely to correctly diagnose early and subclinical allograft rejection, to send for biopsy or refer to transplant center, and to appropriately change treatment than those nephrologists without dd-cfDNA access.

Variation in Assessing Renal Allograft Rejection: A National Assessment of Nephrology Practice.

BACKGROUND: The clinical utility of early detection and treatment of allograft rejection is well-established. Despite frequent testing called for by standard of care protocols, the five-year kidney allograft survival rate is estimated to be as low as 71%. Herein, we report on posttransplant care provided to kidney allograft recipients by board-certified nephrologists in the United States. METHODS: We measured clinical practice in a representative sample of 175 practicing nephrologists. All providers cared for simulated patients' status after renal transplant ranging from 30-75 years in age and 3-24 months after transplant. Our sample of nephrologists cared for a total of 525 allograft cases. Provider responses to the cases were reviewed by trained clinicians, and care was compared to evidence-based care standards and accepted standard of care protocols. RESULTS: Among nephrologists, practicing in settings ranging from transplant centers to community practice, we found that the clinical workup of kidney injury in posttransplant patients is highly variable and frequently deviates from evidence-based care. In cases with pathologic evidence of rejection, only 29.1% (102/350) received an appropriate, evidence-based biopsy, whereas, in cases with no pathological evidence of rejection, 41.3% (45/109) received low-value, unnecessary biopsies. CONCLUSION: Clinical care in the posttransplant setting is highly variable. Biopsies are often ordered in cases where their results do not alter treatment. Additionally, we found that misdiagnosis was common as were opportunities for earlier biopsy and detection of rejection. This evidence suggests that better diagnostic tools may be helpful to determine which transplant patients should be biopsied and which should not. This study suggests that nephrologists and transplant patients need better tests than creatinine and proteinuria and less invasive approaches than routine biopsies to determine when transplant patients should be investigated for rejection and additional treatment.

Are blastocyst aneuploidy rates different between fertile and infertile populations?

PURPOSE: This study aimed to determine if patients with infertility or recurrent pregnancy loss have higher rates of embryo aneuploidy than fertile controls. METHODS: This was a retrospective review of all pre-implantation genetic screening (PGS) cases processed by a single reference lab prior to March 2014 after a blastocyst biopsy. Cases were excluded if no indication for PGS was designated or patients were translocation carriers. The fertile control group consisted of patients undergoing IVF with PGS for sex selection only. The comparison cohorts included those with recurrent pregnancy loss, male factor infertility, unexplained infertility, prior failed IVF, or previous aneuploid conceptions. A quasi-binomial regression model was used to assess the relationship between the dependent variable, aneuploidy rate and the independent variables, maternal age and reason for PGS. A quasi-Poisson regression model was used to evaluate the relationship between similar independent variables and the number of blastocyst biopsies per case. RESULTS: The initial study population consisted of 3378 IVF-PGS cycles and 18,387 analyzed trophectoderm samples. Controlling for maternal age, we observed an increased rate of aneuploidy among patients with recurrent pregnancy loss (OR 1.330, p < 0.001), prior aneuploid pregnancy (OR 1.439, p < 0.001), or previous failed IVF cycles (OR 1.356, p = 0.0012) compared to fertile controls. Patients with unexplained and male factor infertility did not have a significantly different aneuploidy rate than controls (p > 0.05). The increase in aneuploidy in patients with RPL and prior IVF failure was driven by both an increase in meiotic (OR 1.488 and 1.508, p < 0.05) and mitotic errors (1.269 and 1.393, p < 0.05) relative to fertile controls, while patients with prior aneuploid pregnancies had only an increased risk of meiotic error aneuploidies (OR 1.650, p < 0.05). CONCLUSIONS: Patients with recurrent pregnancy loss, previous IVF failures, and prior aneuploid pregnancies have a significantly higher, age-independent, aneuploidy rate compared to patients without infertility.

Reliability of 46,XX results on miscarriage specimens: a review of 1,222 first-trimester miscarriage specimens.

OBJECTIVE: To examine the rate of maternal contamination in miscarriage specimens. DESIGN: Retrospective review of 1,222 miscarriage specimens submitted for chromosome testing with detection of maternal cell contamination (MCC). SETTING: Referral centers requesting genetic testing of miscarriage specimens at a single reference laboratory. PATIENT(S): Women with pregnancy loss who desire complete chromosome analysis of the pregnancy tissue. INTERVENTION(S): Analysis of miscarriage specimens using single-nucleotide polymorphism (SNP) microarray technology with bioinformatics program to detect maternal cell contamination. MAIN OUTCOME MEASURE(S): Chromosome content of miscarriages and incidence of 46,XX results due to MCC. RESULT(S): Of the 1,222 samples analyzed, 592 had numeric chromosomal abnormalities, and 630 were normal 46,XX or 46,XY (456 and 187, respectively). In 269 of the 46,XX specimens, MCC with no embryonic component was found. With the exclusion of maternal 46,XX results, the chromosomal abnormality rate increased from 48% to 62%, and the ratio for XX to XY results dropped from 2.6 to 1.0. CONCLUSION(S): Over half of the normal 46,XX results in miscarriage specimens were due to MCC. The use of SNPs in MCC testing allows for precise identification of chromosomal abnormalities in miscarriage as well as MCC, improving the accuracy of products of conception testing.