Screening for sickle cell disease: focus on newborn investigations

Pre-analytical phase

In newborn screening, several factors contribute to variability, including the type of sample collection instrument, sample quality, transportation methods, and the duration before analysis. Assessing process performance may involve indicators such as the percentage of inadequate samples received and the percentage of samples arriving too late for processing.

The starting point involves selecting the type of collection, which may entail obtaining cord blood at the time of delivery, or capillary blood taken through a heel stick from a newborn within 36–72 h of birth. In many situations, cord blood is chosen due to the woman typically being discharged on the second day after giving birth. This would also allow this collection to be queued up with the others scheduled for metabolic disease screening. However, ideally, blood from the newborn’s capillary collection should be used, as cord blood carries the risk of contamination with maternal blood.

In the case of blood examination, it should be collected in EDTA and processed using the techniques outlined below. For collection on filter paper (DBS), a drop of blood (approximately 50 μL) is applied to Guthrie cards, allowed to dry thoroughly, and then sent to the laboratory. It must be processed within two days upon arrival. Alternatively, capillary collection systems in EDTA can be used. In our experience, although the ability to elute hemoglobins from DBSs remains feasible even weeks later, hemoglobins become oxidized and degraded, making interpretation of the result increasingly problematic as days pass. This results in a progressive rise in baseline with the presence of degradation peaks, with the result that major components present at birth tend to decrease over time, and that samples 3 weeks after collection provide quantifications of hemoglobin components that are no longer reliable [23, 24].

Analytical phase

This phase includes all factors related to laboratory analysis. However, an appropriate quality assurance program must be used, as with other laboratory tests. This program must include, at a minimum, the components listed below:

1. A standard operating procedure describing the method, how to interpret the results, and how to solve potential problems.

2. A flowchart illustrating the role of each person involved in the analytical process.

3. Internal quality control materials to oversee the method over time.

4. Documentation of any corrective action taken.

5. Participation in at least one external quality assessment program.

The preferred control materials are those with the same matrix as the samples being analyzed. These controls should include both physiological samples and carriers of hemoglobinopathies. Given the limited availability of cord blood, only two external quality assessment programs can be used, to our knowledge. In the United States, a program organized by the Centers for Disease Control and Prevention (CDC, Atlanta, Georgia) is in operation. In Europe, a program organized by the United Kingdom National External Quality Assessment (UKNEQAS) is available.

In the initial examination, both HPLC and capillary electrophoresis are suitable for detecting the presence of HbS. Other methods, such as isoelectric focusing and mass spectrometry (MS/MS), are less commonly utilized due to limitations in availability and cost. Ideally, any method that can accurately detect HbS, quantify it alongside HbA₂ and HbF at the decision levels specified for the differential diagnosis of thalassemic forms [6], could be used.

The Table 1 presents a concise list of validated separative methods for the separation and quantification of HbS, applicable not only to neonatal blood. The commonly used instrumentations, including both HPLCs and CE, detect the presence of a peak potentially corresponding to HbS when it falls within a “window” or “zone” as defined by the instrumentation manufacturer, respectively. However, the various separation systems may also identify certain variants as HbS, which may share similar chemical and physical characteristics with HbS but do not exhibit the sickling phenomenon, despite having the same or different amino acid substitutions. Dedicated databases [25] can be consulted for an overview of potentially interfering variants. Variants with such features are rarer than HbS, and even at birth could still be observed. It should also be noted that alpha variants possibly co-eluting or co-migrating with HbS are not a problem in analytical confirmation at birth, since they can be recognized as defects in the alpha chains by the unequivocal presence of HbF^X (α^X ₂γ₂).

The use of two separative methods with different principles could be a valuable aid in confirming some preliminary hypotheses. However, considering that at birth it is mainly important to recognize SCD conditions, whether homozygous or compound heterozygous, with varying degrees of clinical phenotypes, molecular methods should also be undertaken to confirm the diagnosis.

To this regard, dedicated kits in Reverse Dot Blot (RDB) or direct nucleotide sequencing [26] can be utilized. However, it is crucial to specify diagnostic limitations and nucleotide variations tested in the result report (e.g., “the result does not, however, exclude the possible presence of mutations not included in the panel”).

The need to confirm the HbS variant hypothesized at birth using methods different than those used in adults should also be considered. In the newborn, it is not possible to rely on the solubility test or the in vitro sickling test due to the presence of high concentrations of HbF. This may easily lead to false-positive results using the solubility test, and false-negative results with the Na metabisulfite sickling test. Some laboratories perform the sickling test on cord blood, aware that they may result in false negatives. Instead, a positive result confirms the diagnosis of SCD, obviating the need for molecular analysis. The molecular study will still be necessary for the exact definition of the defects contributing to SCD.

Confirmation of heterozygous conditions observed at birth can also be considered later, in adulthood or otherwise upon completion of the HbF switch. In a newborn screening setting, particularly when mandatory as in Italy, managing the diagnosis of heterozygosity is not straightforward. Ideally, this information should always be communicated to every mother who has been adequately informed about potential negative implications (such as non-paternity, iatrogenic effects, etc.) and has given consent [27].

Not all methods have established protocols for extracting hemoglobins from DBSs, but these can be easily clarified locally, often with guidance from manufacturers. The emergence of methods conducive to point-of-care testing has prompted pilot studies, particularly in countries with limited resources [28]. A lateral-flow immunochemical method capable of detecting the presence of HbS and HbC should be noted [29].

Post-analytical phase, results interpretation and reporting

Various pre-analytical information is important for the interpretation of neonatal analysis data. In particular, it is advisable to try to understand:

1. Gestational age

2. Whether transfusions were performed at birth, or otherwise prior to collection

3. Origin/ethnicity of both parents

4. Whether there was neonatal jaundice

5. Whether there were twins

6. Test results for hemoglobinopathies in both parents

7. Consanguinity status of the parents, if any

The information in the first two points is considered essential, because hemoglobin status changes with the gestational period and because possible transfusion makes it impossible to characterize the phenotype of the newborn. From the experience of centers currently practicing neonatal screening on an ongoing basis, information regarding parents can often be difficult to find or may be deficient.

Table 2 displays the threshold values for various hemoglobins typically present at birth in individuals with non-pathologic hemoglobin status, as well as in the most common forms of SCD. The large variability in HbA levels in a subject without hemoglobinopathies can be explained by variability due to gestational age and the type of analytical technique used. HbA₂ values are not reported due to their very low expression in the first months of life, nor are other fractions of limited clinical significance (e.g., acetylated HbF). However, it should be kept in mind that the presence of HbA₂ at birth in concentrations >1 % may be an indication of maternal blood contamination. Once the presence of HbS is confirmed in a compound defect condition, all down-stream procedures for identifying an affected individual should be initiated. The timing of definitive diagnosis for SCD (HbS/HbS, HbS/βthal, HbS/other hemoglobinopathy) should be extended in order to be able to study the parents and/or perform second level molecular investigations.

Once the analytical process is concluded, the data interpretation and the related report should highlight the following conditions:

1. non-pathological hemoglobin status (AA)

2. carrier subject (AS)

3. affected subject (SCD)

It should be noted that the absence of hemoglobin A at birth, together with the appearance of non-physiological hemoglobin fractions, already points with certainty to some of the most frequent genotypes of drepanocytosis (HbSS, HbSC, HbSD, HbSE). Identification of any other variants will be flagged for later investigation, but is not the goal of screening.

Some possible formulations of the report, along with an interpretive overview of the laboratory results and accompanying comments, have been published [30]. A facsimile parental information sheet for carrier cases, accompanying the report, can be downloaded from the website of the Italian Thalassemia and Hemoglobinopathy Society (SITE) [30] and could be used as a template to be eventually adapted to various local needs. For affected patients, referral to the reference center should follow, as shown in the lower part of Figure 1.

Benefits for infants with SCD

There is now well-established evidence that prophylactic treatment with penicillin from the third month of life and early vaccination for capsulated bacteria (H. influenzae, S. pneumoniae, N. meningitidis) can greatly reduce the risk of serious infections during the first 5 years of life [30], [31], [32], [33]. In addition, education work within the family, especially aimed at noticing early signs of anemia and infection in time, can save many lives [34], as well as being ready to use blood transfusions if needed. Finally, early detection of the disease always allows the family environment to implement informed choices for disease management and awareness of future critical issues in the reproductive phase [35].

When it comes to therapeutic options, the landscape has evolved significantly in recent years. While hydroxyurea used to be the primary treatment, there has been notable progress with the emergence of new drugs (crizanlizumab, voxelotor, and l-glutamine). This development reflects the growing interest of both academia and industry in addressing rare diseases. Although these products are not a cure for the disease, but primarily serve to prevent and manage sickle cell crises, there have been significant therapeutic developments with the application of stem cells and gene therapy. For all these aspects, see the paper by Piel et al. [3] and the more recent paper by Ceglie et al. [36]. In December 2023 EMA approved the first gene therapy for SCD and β-thalassemia, with a gene editing approach using CRISPR/CAS9 on the BCL11A gene and reactivation of fetal hemoglobin. Finally, an ongoing clinical trial has shown that continuous treatment with hydroxyurea significantly reduces the need for transfusions [37].

Benefits for carriers

It is generally believed that HbS carriers can lead normal lives, and have no health problems except under very special conditions (e.g., high altitude). In fact, upon closer observation, skeletal muscle capillary structures are different in HbS carriers than in controls. Furthermore, observations on military personnel and athletes have revealed a 30-fold increased risk of sudden death. Abnormalities in the kidney are indeed among the most frequent clinical manifestations observed in individuals carrying HbS. This is because hyperosmolality and low pH in the renal medulla predispose to red blood cell sickling.

For all these reasons, awareness of the condition and timely medical information are crucial for improving the quality of life for individuals affected by this condition. For an in-depth discussion of the medical aspects of the HbS carrier, please refer to the recent SITE consensus [38]. It is important that SCD Referral Centers reserve adequate space for counseling, either directly or through caregivers, on the health aspects of HbS carriers.

Economic analysis

The introduction of a new Diagnostic/Therapeutic program cannot ignore a cost-effectiveness or cost-benefit analysis. Because the data available to date do not allow a full assessment of quality of life and disease complications in the long term, let alone mortality, this document limits the analysis to the first 5 years of life, using data from the province of Modena. This is a robust model, given the accuracy of epidemiological and clinical data derived from newborn screening initiated since 2014 and consolidated since 2015. This analysis does not consider costs related to hydroxyurea therapy, possible stem cell transplantation or antibiotic treatment, focusing on the more relevant and fully available data of hospitalizations.

The salient epidemiological data are summarized in Table 3A. We estimated a reference population of 200,000, a number 72 % lower than the total of the regional tables, but a reasonable value counting that immigrants from Nigeria and Ghana (carriers of HbS and HbC, as well as various forms of thalassemia) reside in Modena and the close surrounding area (mainly in Sassuolo and Nonantola, ceramic manufacturing areas). The collected data clearly show that the rate of hospitalizations is significantly lower in patients identified by neonatal screening. Consider also the figure, not easily quantifiable from an economic point of view, that 26 patients (from the pre-screening group) developed episodes of respiratory distress during monitoring, whereas only five patients in the post-screening group had similar events.

Data for the cost analysis are summarized in Table 3B. Because the two groups do not have the same numerosity, the data, where possible, have been normalized by patient and year. The choice of analytical method does not significantly impact costs, as both HPLC and capillary electrophoresis analysis have comparable expenses. A flat-rate cost was estimated for molecular confirmatory analysis.

Overall, the data clearly indicate that performing newborn screening for SCD is also cost-effective.