Implementing non-invasive RHD genotyping on cell-free foetal DNA from maternal plasma: the Pavia experience

Abstract

Background

The occurrence of cell-free foetal DNA in maternal circulation opens new possibilities in non-invasive antenatal diagnosis. Real-time polymerase chain reaction (PCR) analysis is an useful approach to foetal RhD blood group determination, thus being important in the prevention of haemolytic disease of foetus and new-born (HDFN).

Study design and methods

Using real-time PCR assays we typed 20 samples of plasma, provided in a blinded fashion, from the International Reference Laboratory, two plasma samples sent by the “2010 Workshop on Molecular Blood Group Genotyping”; seven samples from D-negative mothers at the 16th week of gestation provided by our Hospital as prospective validation cases, and two plasma samples received from the “1^st Collaborative study establishing the sensitivity standard for non-invasive prenatal determination of foetal RHD genotype”. To confirm the RHD typing of the seven prospective samples, PCR with sequence specific primers (PCR-SSP) was applied on genomic DNA from amniocytes (5 cases) and paternal peripheral blood (2 cases).

Results

The results for the 31 investigated samples showed 100% concordance. Our measurable benefits were: confidence with a new technology, awareness of having gained the European standard level and increased self-assurance regarding the introduction of this typing technique in prenatal diagnostics.

Discussion

These results demonstrate the feasibility and accuracy of our validation protocol. RHD typing on cell-free foetal DNA is a procedure which requires particular care and a great degree of expertise for diagnostic use. International collaborations are essential for monitoring the performance of laboratories in the absence of specific quality control programmes.

Introduction

The advent of the genomic era and knowledge of the molecular backgrounds of blood group polymorphisms provide a means to predict blood group phenotypes from genomic DNA, thus improving the strategies used in Transfusion Medicine¹. One of the most promising application of these approaches is the determination of foetal blood group to assess whether a foetus is at risk of haemolytic disease². The incidence of haemolytic disease of foetus and newborn (HDFN) declined sharply after the introduction of anti-D immunoglobulin prophylaxis³; nevertheless, despite the widespread availability of both post-natal and antenatal anti-D immunoprophylaxis, anti-D immunisation still occurs, at a rate of 0.1–0.3%^4,5, and this can result in foetal and neonatal morbidity and mortality. There is a clear benefit in determining foetal D phenotype if a D-negative pregnant woman has anti-D. In fact, if the foetus is D-positive, the pregnancy should be strictly monitored, while, if D-negative, unnecessary intervention could be avoided⁶. In this setting, amniocentesis and chorionic villous sampling can be used to obtain foetal DNA, but both procedures are invasive and associated with an increased risk of transplacental haemorrhage and spontaneous abortion⁷. It was demonstrated that, starting from the 16^th week of pregnancy, cell-free foetal DNA is present in the mother’s plasma in sufficient amounts to perform real-time polymerase chain reaction (PCR) testing to determine foetal RHD genotype⁸. In a recent paper, Akolekar R. et al. reported the feasibility of foetal RHD genotyping in maternal plasma at 11–13 weeks of gestation using a high-throughput technique⁹.

The present work aims to illustrate the setting up and validation procedures of real-time PCR technique in our Department of Transfusion Medicine to provide a good quality service. This test would help obstetricians in managing the pregnancies of alloimmunised women and could be considered advantageous in prenatal care of all RhD-negative pregnant women.

Materials and methods

A non-invasive foetal blood group genotyping service has been provided at the International Blood Group Reference Laboratory (IBGRL) in Bristol, UK since 2001. To validate this specific diagnostic test in our Transfusion Service we collaborated with the IBGRL. Their protocol¹⁰ was custom-made and we had to adapt the procedures to our instruments and reagents.

Maternal samples

At first, the test was performed on plasma samples, from 20 pregnant women, provided by the Reference Laboratory in a blinded fashion. Subsequently, in a second validation step, EDTA blood samples (16 mL) were collected from seven D-negative pregnant women, attending the Department of Obstetrics and Gynaecology of our Hospital. Five of them were submitted to amniocentesis or chorionic villous sampling for cytogenetic testing at the Medical Genetics Department of the University of Pavia. Cell cultures were used for foetal DNA extraction, while blood samples were sent, within 48 hours, to the Immunogenetics Laboratory of IRCCS Policlinico San Matteo. Informed consent was obtained from all the patients providing the samples. Plasma separation and DNA extraction were performed as described elsewhere¹¹.

Real-time polymerase chain reaction assay

The IBGRL protocol was tailored to carry out the experiments with our reagents, primers and labelled probes, and with the instrument platform LightCycler 480 (Roche Italia SpA, Milan, Italy). The procedure involves the amplification of exons 4, 5 and 10 of the RHD gene with probes specifically designed to prevent the amplification of the homologous region of the RHCE gene¹². Moreover, the reaction of exon 10 enables the presence of confounding pseudogene RHDΨ¹³ and hybrid RHCE-D-CE genes, peculiar to Africans, to be established¹⁴. Reactions for detecting the Y-linked gene SRY are included in the test and provide an internal control if the foetus is male¹⁵.

First NIBSC Collaborative Study and 2010 ISBT Workshop on Molecular Blood Group Genotyping

We took part in the 1^st Collaborative Study to establish the sensitivity standard for non-invasive prenatal determination of foetal RHD genotype, organised by the National Institute for Biological Standards and Control (NIBSC), UK. This was an international collaborative study to assess the suitability of a frozen dried plasma preparation, to be tested at increasing dilutions, as a WHO Reference Reagent for the detection of RHD and SRY genes using PCR. Furthermore, we attended the 2010 Workshop on Molecular Blood Group Genotyping within the XXXI^st International Congress of the International Society of Blood Transfusion (ISBT).

Results

The various steps of the purposed validation protocol are schematically represented in the flow diagram shown in Figure 1.

Figure 1.

Validation process of non-invasive prenatal RHD genotyping protocol.

Internal validation process

The suitability of the test, performed with our reagents and instruments, was validated in terms of specificity, sensitivity, assay variability and robustness. Setting up our protocol we ran real-time PCR to determine RHD foetal genotype in 20 mothers’ plasma provided as blinded samples by the IBGRL. Our results were in agreement with IBGRL reports, as shown in Table I.

Table I.

Comparison of real-time PCR for non-invasive foetal RHD genotyping results between the Immunogenetics Laboratory and the Reference Laboratory.

Sample	Pavia Typing		Reference Laboratory Typing		Amount of DNA (ng/well)
	Rh	Sex	Rh	Sex	Reference Laboratory	Pavia
1	POS	Male	POS	Male	1 ng	1.84 ng
2	POS	Male	POS	Male	0.35 ng	0.203 ng
3	POS	Female	POS	Female	0.35 ng	0.265 ng
4	POS	Male	POS	Male	0.7 ng	0.1675 ng
5	NEG	Male	NEG	Male	0.4 ng	1.498 ng
6	POS	Female	POS	Female	1.8 ng	1.605 ng
7*	(POS)	(Male)	(POS)	(Male)	36 ng	40.85 ng
8	NEG	Male	NEG	Male	0.8 ng	1.565 ng
9	POS	Female	POS	Female	2.5 ng	2.285 ng
10	POS	Male	POS	Male	1.5 ng	1.67 ng
11	NEG	Female	NEG	Female	6.8 ng	6.62 ng
12	NEG	Male	NEG	Male	0.5 ng	0.7295 ng
13	POS	Female	POS	Female	0.6 ng	0.786 ng
14	NEG (3/12)	Female	NEG	Female	0.3 ng	0.617 ng
15	NEG	Female	NEG	Female	0.14 ng	0.194 ng
16	POS	Female	POS	Female	6.9 ng	6.95 ng
17*	(POS)	(Male)	(POS)	(Male)	46 ng	67.5 ng
18	NEG	Female	NEG	Female	0.55 ng	0.946 ng
19	POS	Male (1/4)	POS	Male (2/4)	0.5 ng	0.441 ng
20	POS	Female	POS	Female	1.1 ng	1.25 ng

^*sample was contaminated with excessive amounts of maternal DNA.

We processed five more samples of both plasma and genomic DNA obtained by amniocentesis from mothers in the 16^th week of gestation who were referred to our Hospital for prenatal diagnosis of chromosomal pathologies. The experiment was carried out double-blind using real-time PCR on cell-free foetal DNA from plasma, and PCR with sequence specific primers (PCR-SSP) for genomic foetal DNA testing. The latter was performed using READY GENE CDE and READY GENE Dneg commercial kits (INNO-TRAIN Diagnostik GmbH, Kronberg/Taunus, Germany). The results of these tests were also in agreement, as reported in Table II.

Table II.

Comparison of non-invasive real-time PCR from cell-free foetal DNA with PCR-SSP from genomic foetal DNA.

	real-time PCR							PCR-SSP
Sample	Ct ex 4	Ct ex 10	Ct ex 5	SRY	CCR5	Genotype	Sex	Genotype	Sex (from Kariotype)
PV1	43.51	39.01	37.69	41.07	35.21
	43.66	37.5	39.06	42.04	35.42	RHD positive	Male	CcDdee	Male
	41.2	39.31	39	41.87
	41.1	39.23	41.42	42.16
PV2	43.42	nd	40.19	41.82	37.65
	43.53	39.89	39.26	nd	36.89	RHD positive	Female	CcDdee	Female
	nd	36.92	40.8	nd
	43.67	39.83	39.45	nd
PV3	nd	nd	nd	nd	36.56
	nd	nd	nd	nd	36.41	RHD negative	Female	Ccddee	Female
	nd	nd	nd	nd
	nd	nd	nd	nd
PV4	42.27	39.89	39.16	39.65	37.14
	42.42	39.53	38.59	nd	37.32	RHD positive	Male	CcDdee	Male
	42.11	39.64	38.6	40.09
	42.08	40.21	40.45	40.86
PV5	43.35	38.82	40.25	nd	36.25
	42.03	39.47	39.94	41.2	36.29	RHD positive	Male	CcDdee	Male
	43.11	40.22	40.08	41.38
	nd	40.33	40.41	nd

Each sample was tested in quadruplicate for RHD exon 4, RHD exon 5, RHD exon 10, SRY and twice for CCR5 according to the Reference Laboratory protocol. Ct: Cycle threshold values; nd: not detected (Ct>45); SRY: sex determining region Y; CCR5: chemokine, CC motif, receptor 5.

Finally, using this non-invasive technique, we determined the presence of foetal RHD sequences in two plasma samples from sensitised RhD-negative pregnant women in the 16^th week of gestation referred to our Hospital. In both cases, the foetus was predicted to be RHD positive. The data obtained were subsequently confirmed performing the PCR-SSP D-zygosity test on paternal DNA.

First NIBSC Collaborative Study and 2010 Workshop participation

Our laboratory satisfied the requirements of both studies. Our measurable benefits were confidence with a new technology, awareness of having gained the European standard level and increased self-assurance regarding the introduction of this typing technique in prenatal diagnostics. Our results demonstrated the feasibility and accuracy of our validation protocol. Results gained from international collaborations are essential for monitoring the performance of a laboratory in the absence of specific quality control programmes.

Discussion

Determination of foetal RHD genotype, using a non-invasive approach, is an important challenge because it has crucial implications for the clinical management of sensitised RhD-negative pregnant women. Foetal DNA analysis in maternal plasma offers the opportunity to meet this challenge. Here, we have shown that real-time PCR technology, performed on cell-free foetal DNA obtained from maternal plasma is feasible and very reliable, albeit highly demanding. We demonstrate that accurate procedures for determining foetal RHD genotype, also in HDFN high-risk ethnicities, are practicable in our laboratory, thus allowing early diagnosis of at-risk pregnancies and the correct management of HDFN. This enables unnecessary treatments to be avoided, and prevents wastage of anti-D immunoglobulins. Furthermore, the introduction of a clinical test to determine foetal D blood group from maternal blood would significantly reduce the number of invasive procedures carried out for foetal RhD determination. The rate of accuracy achieved in predicting foetal D status would encourage the implementation of a non-invasive foetal RHD genotyping service in Transfusion Departments. Although this test is currently performed by only a limited number of specialised laboratories, it is anticipated that the demand will increase in the near future. In fact, there would be several benefits from using this test strategy. First, it is likely to prove cost-effective with a substantial reduction of costs of overall immunoprophylaxis associated with a considerable decline in the use of anti-RhD immunoglobulin, an expensive blood product in short supply. Besides, women with an RhD-negative foetus would be spared unnecessary exposure to this pooled human blood product with its associated discomfort and possible risk from contamination. Moreover, the Recommendations for Further Research in the UK Guidelines on HDFN prevention encourage assessment of tests for RH genotyping on foetal DNA present in maternal circulation and refer these tests for review in May 2011¹⁶. The introduction of this test for the screening of all D-negative pregnant women would be highly desirable for clinical, economic and ethical reasons. Consequently, the current “handmade” assays must move towards more automated ones, at least until the availability of commercial tests, in order to ensure robustness and reliability. We trust that RHD typing on cell-free foetal DNA obtained from the plasma of RhD-negative pregnant women will become standard practice in many Laboratories monitored through international quality controls programmes.