Sickle cell testing: High throughput mass spectrometry method could lead to cheap, universal neonate screening

 Sickle screening

In an ideal world, every newborn child will be screened for a host of diseases to enable early diagnosis and treatment. In fact, screening could even begin before that, with the child still in the womb, as is the case for some disorders such as Down's syndrome.

However, as desirable as they are, any aspirations towards universal screening are invariably shot down by real world practicalities, namely time and cost. If tests take too long to perform and interpret, they will not be able to compete with the sheer numbers of children born each year. It they are too expensive, health authorities and governments will be unable to support them with funding.

Currently, some countries do carry out universal screening of newborns, but only for specific diseases. For instance, the UK has adopted an official neonatal screening programme for sickle cell disease, choosing universal screening rather than ethnically targeted screening.

In France, where sickle cell disease is the most frequently detected disease in newborns, the current policy is targeted screening. This is possible due to the distribution of sickle cell disease which is commonly found in people whose families come from Africa, South or Central America, the Caribbean, India, Saudi Arabia and Mediterranean countries such as Turkey, Greece, and Italy.

National policy was established by the French Association for Screening and Prevention of Infant Handicaps (AFDPHE) but universal screening is due to be evaluated as part of the National Rare Diseases Plan. If they are to change policy, then high-throughput, fixed-cost tests will be required so that the volume and cost of the programmes can be predetermined.

A team of French scientists has acknowledged that the current standard sickle cell profiling technique is relatively time consuming for multiple samples. It combines isoelectric focusing followed by HPLC with a cation exchange column. The latter step is automated and quantitative but the former step is slow and creates the bottleneck.

So, the team has designed a new procedure built around matrix-assisted laser desorption/ionisation mass spectrometry (MALDI MS). Jean-Marc Perini and co-researchers from the University Hospital of Lille, the University of Artois and AFDPHE published details in Proteomics - Clinical Applications.

Made for MALDI

The new MALDI MS approach was based on a retrospective analysis of blood spots on Guthrie cards, which had been drawn from the heels of newborns. The samples covered several types of haemoglobin disorder, including sickle cell disease, which were associated with different mutant forms of haemoglobin.

Sickle cell disease corresponds to the occurrence of haemoglobin S, which contains a variant of the beta-chain and may be associated with other variants such as haemoglobin C, D-Punjab and E.

Haemoglobin disorders such as alpha- and beta-thalassaemia, in which formation of the alpha- and beta-chains is restricted or prevented, were also present in the samples.

In theory, the presence of sickle disease and the other disorders should be detectable by mass spectrometry, because they are each associated with globin chains of different composition and/or length, which would give different peaks in the mass spectra.

The blood was removed from the cards by shaking them in water and the solutions were analysed by MALDI MS in positive-ion mode. Mass spectra from 1000 laser shots were summed over the m/z range in which the haemoglobin chains appear, m/z 15,000-16,200, and each sample was tested four times.

The two strongest signals appeared at m/z 15,138 and 16,010, corresponding to the complete alpha and gamma chains. They were useful in framing the mass spectral region of interest and permitted spectral alignment.

Sickle cell success at high speed and low cost

One sample set contained 844 Guthrie cards from France and some French territories for babies with and without the haemoglobin S variant. The MALDI MS analysis was successful on all but 20 samples, with those from premature babies tending to fail more often. No poor spectra were obtained for full term babies.

The predictive capability of the average mass spectra for classifying the three main sickle cell categories compared with the standard IEF/HPLC protocol was 100%. Every case of healthy, sickle cell carrier and sickle cell anaemia was correctly identified.

The good quality spectra produced no false negatives. Haemoglobin S carriers were also identified in some of the pre-term babies, indicating that they should still be included in the screening system.

There were some difficulties with other haemoglobin variants and with thalassaemia, with several of the variant chains being classified as adult haemoglobin due to the poor mass resolution of the linear time-of-flight mass spectrometer.

However, the researchers argued that many of these variant types could be excluded from a sickle cell disease screening programme because they are asymptomatic or weakly symptomatic.

The highly accurate predictive performance of the MALDI MS system is supported by low cost implications. The equipment cost for the standard IEF/HPLC test is relatively low but when the cost of consumables is included, the method becomes less attractive for large-scale screening programmes.

The alternative MALDI procedure has a higher initial outlay for the equipment but is capable of analysing 1000 samples a day, Perini estimated that the unit cost per sample, including lab work and data validation time, is as little as 1.15 Euro (about $1.69) based on 200,000 samples a year on a machine that will run for 5 years with its own sample prep robot and informatics system.

Although more research needs to be performed to support these conclusions, this inexpensive and rapid procedure could lead to a "culture change" in neonatal screening for sickle cell disease in many countries, with every newborn baby being screened at birth.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.