Cystic fibrosis screening laboratory handbook

Question 1

Screening for CF

Accepted Answer

All babies across the UK are offered a blood test to identify CF as part of the NHS NBS screening programme. Screening for CF aims to identify babies with a classical CF diagnosis before they develop symptoms, to enable the early initiation of high-quality care.

About CF

CF is a heterogeneous disorder with a large number of different gene variants affecting the putative gene (the CF Transmembrane Conductance Regulator (CFTR) gene) and has a range of different clinical phenotypes. For further information, see the Cystic Fibrosis Trust website and research on the future of CF care.

Whilst babies with a clear CF diagnosis represent the majority of true positives, identification of babies with an inconclusive diagnosis after a positive NBS result is an increasingly recognised outcome. This leads to uncertainty for both families and healthcare professionals.

The approach to the care of these babies is evolving with increased experience and reporting of outcomes. Babies with an unclear diagnosis after NBS screening are designated as ‘CF Screen Positive, Inconclusive Diagnosis’ (CFSPID). There is international guidance on the care of these babies. It is important that strategies are based, whenever possible, on evidence of benefit for the child and family. To default CFSPID babies to a full CF care pathway is not appropriate, but it is equally important that the potential risks that these babies face are recognised and addressed.

General organisation of NBS screening

CF screening is fully integrated within the existing NBS screening programme and based on the same screening laboratory populations. The initial screening test (the assay of immunoreactive trypsinogen (IRT)) uses blood collected on the standard NBS screening sample collection card (see section 9.1). Quality assurance (QA) and performance management arrangements follow the same general principles as those for other newborn screening programmes.

With CF, as for other blood spot screening conditions, the screening laboratory is a major communication hub. The laboratory sends screening results to child health records departments (CHRDs)/child health information services (CHISs), which send negative results to the parents/carers. Over 99% of results are ‘CF not suspected’ and are generated promptly. However, in some cases, screening may not be completed until the baby is over a month old.

The CF screening programme requires partnership working between the screening laboratory and molecular genetics laboratory.

Some parents/carers may decide that they do not want their baby to be screened for some or all of the conditions covered by NBS. Screening can be declined for CF, sickle cell disease and congenital hypothyroidism individually, but the 6 inherited metabolic disorders (IMDs) can only be declined as a group. Details of these 6 inherited metabolic disorders are available.

The sample taker completes the blood spot card to indicate which conditions have been accepted or declined, as outlined in the NBS screening sampling guidelines.

Question 2

The screening protocol

Accepted Answer

The UK screening protocol is intended to:

maximise the early detection of CF across all populations so that pre-symptomatic treatment can be initiated to reduce the long-term morbidity associated with the condition if untreated
minimise the need for second heel pricks, and any other causes of diagnostic delay
minimise detection of unaffected heterozygotes
minimise recognition of babies with CFSPID
allow for the fact that a clear diagnosis is not always possible

The standard protocol

IRT analysis is performed on a single spot from an initial dried blood sample followed by a one or 2-stage genetic analysis of the CFTR gene on all samples with confirmed IRT concentrations above a defined cut-off. Subsequent action depends on the results of the genetic analysis.

This protocol is summarised diagrammatically in Figure 1 below. No alternative to this protocol is recommended. A text description of Figure 1 is available.

Definition of the cut-off values

As of January 2023, the national cut-off values for the AutoDELFIA analyser are:

cut-off A = 55 µg/L
cut-off B = 65 µg/L

As of April 2022, the national cut-off values for the Genetic Screening Processor (GSP) analyser are:

cut-off A = 46 µg/L
cut-off B = 56 µg/L

These cut-off values are subject to regular review based upon continued data collection from the laboratories.

The value used for cut-off A is based on selecting an average of 1.5% of all results for repeat IRT analysis. It is set to 10µg/L below cut-off B.

Cut-off B, used to select samples for further genetic analysis, has been defined based upon experience in East Anglia using a 2-stage IRT-IRT screen over a 17 year period that showed a 99.5th centile cut-off for the initial sample gave an overall sensitivity of 97% (Heeley and others, 1999)^{[footnote 1]}.

For further information about the rationale of the cut-off values, see section 5.1 below.

The first screening specimen

Samples are initially assayed in singleton.

Consider babies in whom the IRT concentration is < cut-off A in the initial screening sample to have a negative screening result for CF. Report them as ‘CF not suspected’.

Re-assay samples with results ≥ cut-off A in duplicate from the same card but on a different spot(s) to give a more definitive result. This is to minimise effects of volumetric variability of the punched discs, day-to-day variation in IRT assay calibration, and to detect possible sample misidentification.

Take the average of the 3 results.

Review the set of triplicate results for consistency as poor analytical performance can produce different results. Spuriously high results can occur with faecal contamination or blood spot layering and spuriously low results can occur if there is a missing spot or poor sample.

Further action therefore requires an assessment of agreement between results. The following sections provide guidance only and laboratories should use discretion for each individual set of results.

Review of triplicate mean results

For babies in whom the triplicate mean IRT concentration in the initial screening sample is < cut-off B, and provided there is no suspicion of a ‘missing spot’, issue a ‘CF not suspected’ result.

If the average of the 3 results is ≥ cut-off B *, assess the results for consistency before proceeding further.

The pathway for handling IRT results from the first screening specimen is summarised diagrammatically in Figure 2 below. No alternative pathway is recommended. A text description of Figure 2 is available.

Figure 2: handling of IRT results from the first screening specimen

Regard a set of 3 results as discrepant if they meet any one of the following criteria:

a coefficient of variation (standard deviation/mean) ≥ 0.25
one or more individual results being ≥ 30% above or below the average of all 3
the ratio of the highest result to the average of the other 2 results ≥ 1.5

For sets of 3 results considered to be discrepant:

if there is a single high result (≥ cut-off B *) with the other 2 results < cut-off B *, and provided that there is no uncertainty about sample identification, ignore the single high result and issue a ‘CF not suspected’ report
if 2 results are ≥ cut-off B * and the average of all 3 results ≥ cut-off B* but < 120 µg/L, proceed to genetic analysis as per the screening protocol
if the average of all 3 results is ≥ 120 µg/L, request a repeat sample on the grounds that this is likely to be sample contamination (only if a repeat sample can be taken before the baby is 21 days old, otherwise an immediate referral is indicated)

*If the sample was taken on day 21 or later then apply cut-off A.

Assaying further samples from a card with discrepant results is unlikely to completely resolve the issue.

Insufficient sample: no re-assay possible

If it is not possible to punch any further discs from the card (for example, for a single result ≥ cut-off A), request that a repeat specimen is taken as soon as possible due to insufficient sample.

Only one re-assay possible

If only 2 IRT results can be obtained (so after the initial analysis, only one further spot from the card is possible) then proceed as follows. If the average of both results is:

< cut-off B *, report as ‘CF not suspected’
‘≥ cut-off B *, proceed to genetic analysis; widely discrepant results may require a repeat specimen taken as soon as possible– laboratory staff should use judgement on this.

*If the sample was taken on day 21 or later then apply cut-off A.

The second specimen (repeat IRT test)

Requests for second blood samples for IRT testing are via locally agreed pathways defined as part of the newborn screening responsibilities. Laboratories should request that the repeat blood spot sample is collected on day 21 (by day 24; this allows for day 21 to fall on a weekend when a special visit is not warranted).

The second dried blood spot specimen is for IRT testing only. It is recommended practice not to repeat other screening tests on this specimen.

The sample taker should explain to parents/carers that further tests need to be done for cystic fibrosis. An information sheet for parents about the repeat blood spot test for CF is available on GOV.UK. The results of genetic analysis are not given out at this stage as this may result in premature disclosure to the parents/carers before the definitive screening result is known. Confirm the repeat request in writing to the appropriate health professional(s).

If there is a delay and a second sample has not been taken by 8 weeks of age, it is too late for IRT testing to be carried out reliably (see section 2.3 below).

Actions to be taken on the repeat IRT test

The following actions are undertaken on the repeat sample depending on the reason for its request.

If one CFTR variant is detected – assay the repeat sample for IRT in duplicate. Report babies with an average of results ≥ cut-off A as ‘CF suspected’ and refer to a CF specialist. Report babies with an average of results < cut-off A as ‘CF carrier’. No referral is required. A health professional contacts the parents/carers to discuss the screening results.

If no CFTR variants are detected and the initial IRT ≥ 120 µg/L – assay the repeat sample for IRT in duplicate. Report babies with an average result ≥ cut-off A as ‘CF suspected’ and refer to a CF specialist. Otherwise, issue a ‘CF not suspected’ result.

For information about the reporting and communication of results, see section 7.3 below.

Late sampling

Blood spot IRT concentration is increased in most babies with CF in the first few weeks of life. However, early research on blood spot screening and on age-related decline in IRT indicates IRT becomes unreliable as an indicator of CF at around 8 weeks. This decline in IRT has implications for the reliability of CF screening if samples are taken outside the specified time windows. It is therefore important that appropriate cut-off values are applied to the times at which samples are taken.

This means that CF screening cannot be undertaken or completed for babies who:

have arrived in the UK after 56 days (8 weeks) of age
have not, for whatever reason, had their first screening sample collected by 56 days (8 weeks) of age
require a second (day 21) blood spot sample for IRT assay but do not have it taken at the correct time – for example, because the sample collection was delayed or the sample was lost in transit

The following guidance applies in these situations.

Late first samples

For samples taken before day 21, apply cut-off values as for samples that have not been delayed.

For samples taken when the baby is aged between 21 days (3 weeks) and 56 days (8 weeks) inclusive, apply cut-off A as the decision point for genetic analysis.

Do not test any samples taken after 56 days (8 weeks) for IRT. Procedures should be in place for laboratories to identify these samples.

If a first (routine screening) sample is taken after day 56 (8 weeks) and is unavoidably tested, then the laboratory staff should proceed as follows. If the IRT is:

< cut-off A, report as ‘CF – not screened (baby too old >8 weeks age)’
≥ cut-off A (average result), report as ‘CF suspected’ and refer the baby to a CF clinician; send a blood spot sample at the same time for genetic analysis, and give advice to the family health visitor and/or midwife about the reason for referral so that this can be explained to the parents/carers

The pathway for handling late first IRT samples is summarised diagrammatically in Figure 3 below. No alternative pathway is recommended. A text description of Figure 3 is available.

Figure 3: handling of late first IRT samples

Late or absent second samples

For second samples taken on or before day 56 (8 weeks) of age, apply cut-off A per the CF screening protocol.

Babies aged over 56 days (8 weeks) are too old for CF screening to be carried out reliably. Do not collect the second sample.

If a second sample is not received because the baby aged over 56 days (8 weeks), report as ‘CF suspected’ and refer the baby to the CF clinical team for follow-up.

If a second sample is taken after day 56 (8 weeks) of age, do not analyse it for IRT. Report as ‘CF suspected’ and refer the baby to the CF clinical team for follow-up.

If a second sample is taken after day 56 (8 weeks) of age and has been unavoidably analysed, report as ‘CF suspected’ and refer the baby to the CF clinical team for follow-up regardless of the IRT concentration.

In these situations, laboratory staff should provide information to the CF clinical team about the reason for referral so an explanation can be given to the parents/carers.

Report babies that have transferred out of the UK and are still awaiting the collection of a repeat first or second IRT sample as ‘screening incomplete’. A ‘cystic fibrosis screening not complete’ template letter is available on www.GOV.UK

Action following genetic analysis

Babies with 2 CFTR variants detected

Babies with CFTR variants detected in both copies of the gene (either a homozygote or compound heterozygote) have a presumptive positive diagnosis of CF. Report them as ‘CF suspected’ and refer to a CF paediatric service for immediate evaluation (clinical assessment and sweat test).

Babies with one CFTR variant detected

Most babies with only one CFTR variant detected will be unaffected carriers but there is a risk that they carry a second CFTR variant not detected by the panel of CFTR variants tested. They are therefore tested again (second IRT) on a repeat dried blood spot sample. Laboratories should request that this second IRT sample is collected on day 21 (by day 24; this allows for day 21 to fall on a weekend when a special visit is not warranted). For further information, see section 2.2 above.

Babies with no detected CFTR variants

Babies with no detected CFTR variants (by the initial panel of 4 CFTR variants) are divided into 2 groups depending on the initial IRT result:

those with first IRT < 120 µg/L – report as ‘CF not suspected’
those with first IRT ≥ 120 µg/L – a second IRT test is undertaken on a repeat dried blood spot specimen; laboratories should request that this second IRT sample is collected on day 21 (by day 24, this allows for day 21 to fall on a weekend when a special visit is not warranted) – for further information see section 2.2 above

For more information on genetic analysis, see section 6 below.

Sibling testing

Older siblings (of a baby with confirmed CF diagnosed from newborn screening) may be at risk of CF. Any testing is at the discretion of the clinician, with the family’s consent.

For any subsequent siblings of the index newborn, the clinical team offers expedited genetic analysis of cord blood sample together with the mother’s blood sample.

Newborn screening should be undertaken as normal. ’Early’ screening for siblings (prior to day 5, counting day of birth as day 0) is not recommended. The blood spot sample should be collected on day 5 for all babies regardless of medical condition, milk feeding and prematurity. This is to enable detection of abnormal results and initiation of appropriate treatment.

Family history and other risk factors

Babies are regarded as high risk if they:

present with meconium ileus
have shown echogenic bowel in-utero
are known to be at risk due to family history

They should be investigated independently according to clinical circumstances as well as being screened in the normal way.

If a genetics centre has carried out genetic testing before routine screening, it is helpful if they advise the screening laboratory of the results. This may avoid unnecessary duplication of follow-up or diagnostic testing. It will also help avoid any confusion arising from a ‘CF not suspected’ screening result (from IRT analysis) which might cause concern for the family.

Unscreened babies

These include babies who were never part of the CF screening process, for example:

babies born abroad
babies not tested because screening was offered too late (> 8 weeks)
babies for whom screening was declined (by the parents/carers)

For any subsequent requests for CF screening from the parents/carers, the family health visitor should explain to the family why their baby has not been screened for CF.

Screening laboratories should inform the baby’s GP if screening for CF is incomplete. A ‘cystic fibrosis screening not complete’ template letter is available.

The family health visitor should inform the GP of any subsequent requests for screening from the parents/carers.

Such babies should not routinely be offered any testing and should only be referred on a clinical basis if required. If the family or GP has any concerns, a referral for assessment (which may include sweat testing) by the local designated CF team is appropriate.

Previously screened babies where subsequent results differ

A raised IRT ≥ cut-off A on a sample taken after day 21 where there is a previous CF not suspected result should be dealt with as described in the following situations.

Discrepant IRT result

Suspect contamination and request a further repeat if there is sufficient time to obtain a specimen up to and including day 56. See section 2.1 above.

Non-discrepant IRT result ≥ cut-off A with previous IRT < cut-off B and ‘CF not suspected’ reported

Send for genetic analysis. Subsequent action depends on the results of the genetic analysis. If:

one or more CFTR variant is detected, refer and report as ‘CF suspected’
no CFTR variants are detected (sample taken at ≤ 56 days), request a repeat sample
no CFTR variants are detected (sample taken at > 56 days), refer and report as ‘CF suspected’

Non-discrepant IRT result ≥ cut-off A with previous IRT ≥ cut-off B, no CFTR variants detected, and ‘CF not suspected’ reported

This shows a persistently raised IRT. Refer and report as ‘CF suspected’.

Question 3

Laboratory quality and performance

Accepted Answer

Screening laboratory

Laboratories undertaking NBS screening must use processes accredited in accordance with ISO 15189 for Medical Testing Laboratories by a competent accreditation testing service, for example UKAS.

NBS screening is provided within the organisational structure of the NBS screening programme and undertaken by specialist newborn screening laboratories already providing screening programmes.

There should be written agreed procedures describing the working arrangements between the screening laboratory and their referral laboratory.

There should be documented local policies and standard operating procedures describing the whole screening process including pre-analytical, analytical and post-analytical processes. Where appropriate these will include reporting results, and referral and follow-up arrangements for presumptive positive cases and carriers, as specified in laboratory handbooks. Provide processes in line with relevant NBS national standards and CF guidance. Processes should be reviewed periodically taking into account audit data, accumulating results, technical developments and local changes in healthcare provision.

Molecular genetics laboratory

In general, each screening laboratory should send samples to a single molecular genetics laboratory. Molecular genetics laboratories may receive samples from more than one screening laboratory.

Screening laboratories may contract for this service any molecular genetics laboratory that:

is capable of meeting the performance standards (specified below)
uses ISO 15189 accredited processes
is part of one of the 7 Genomic Laboratory Hubs (a nationally commissioned genomics laboratory in England)

Overall performance and timeliness

Laboratory services should be configured to enable CF newborn screening to be completed in time for all babies with positive screening results to have their first clinic appointment by:

day 28 for babies in whom 2 CFTR variants have been detected
day 35 for babies who have needed a second sample IRT measurement

Analysis for IRT should be performed frequently enough to generate a screening test result (including any retest results, where required, to be confirmed in duplicate), no later than 4 working days from receipt of an adequate sample. Definitive results from genetic analysis must be available on or before the 4th working day from receipt of the sample within the genetic laboratory.

Report CF suspected results to the appropriate clinical team within 1 working day of becoming available. Do not issue intermediate reports (those showing increased IRT in the initial sample without follow-up results).

Ideally, parents/carers should not be informed of a positive screening result on a Friday or Saturday, to make sure that the clinical assessment and sweat test can be arranged for the day after the results.

Internal quality and performance monitoring

Laboratories should participate in audit at local, regional and national levels, to assess the effectiveness of the national screening programme. They should publish the results and performance of their newborn blood spot screening programme within an annual report.

There must be a documented risk management policy for the laboratory aspects of the CF screening programme as part of an overall pathway risk assessment. This should describe the steps in the testing protocol where failures could occur and the procedures that have been implemented to minimise the risk of their occurrence.

External quality assessment

Laboratories should participate in an approved external quality assurance (EQA) scheme (for example The United Kingdom National External Quality Assessment Service (UK NEQAS)). The EQA scheme assesses laboratories on the precision and accuracy of analytical steps.

Following agreement from the NBS screening programme, the laboratory must release reports on screening performance (including external quality assurance and accreditation assessments) to agencies with a legitimate interest in the quality and safety of the programme on behalf of the public.

Question 4

Pre-analytical factors

Accepted Answer

Specimen requirements

Blood spot sampling should be undertaken according to the Guidelines for newborn blood spot sampling. Specimens should be dispatched to the laboratory within 24 hours of taking the sample. They must be kept in a dry environment at room temperature or refrigerated at 4°C before analysis. Storage after analysis should follow the guidelines provided by the code of practice for the retention and storage of residual newborn blood spots.

Venepuncture or venous/arterial sampling from an existing line is an alternative method to collect the blood spot sample. This is providing the sample is not contaminated with ethylenediaminetetraacetic acid (EDTA)/heparin and the line is cleared of infusate. Anticoagulants may affect the assay. Capillary tubes (plain or heparinised) must not be used to collect blood samples.

Blood spot stability

IRT was shown to be relatively unstable in early studies (Heeley and others, 1982; Kirby and others, 1981) and there are few published studies using modern methods of IRT analysis. In 2006, the Centres for Disease Control and Prevention (CDC) showed that IRT-enriched blood added to a filter paper matrix was stable when stored with desiccant for 1 year at either -20°C or 4°C, and 30 days at ambient temperature. Only 75% of IRT remained in dried blood spots stored at ambient temperature for up to one year (Li and others, 2006). However, dried blood spots in this study were prepared by lysing blood before the addition of IRT and recovery studies demonstrated that IRT degrades more quickly in whole blood. A further CDC study using whole blood with intact cells (thus more closely mimicking newborn DBS samples) showed that samples stored at 37°C in low-humidity environments retained 83% of IRT for 30 days, whereas in a high humidity environment only 47% was retained (Adam and others, 2011).

Samples received in the laboratory more than 14 days after the sample was taken are therefore deemed unsuitable for testing. Request a repeat sample as soon as possible in line with the blood sampling guidelines. Test the original sample following the standard algorithm and refer the baby if a screen positive result is obtained.

For CF, it is important to be aware that samples taken when the baby is aged over 56 days (8 weeks) should not be analysed (see section 2.3 above).

Non-analytical factors affecting the screening result

Potential for false negative results

Several factors (in addition to late testing – see section 2.3 above) are known to lower blood IRT concentration in babies with CF, leading to false negative screening results.

CFTR modulator therapy

Affected infants born to mothers with CF, who are on CFTR modulator therapy (tradenames Kaftrio TM or Kalydeko TM) may have false negative NBS results, with a day 5 IRT measurement below the cut-off. The CF team caring for the mother should arrange alternative forms of excluding CF in the baby in these circumstances.

Meconium ileus:

Some babies with CF and meconium ileus are likely to give negative screening results when tested during the first week of life, though IRT concentrations may become clearly abnormal a week or 2 later. It is unclear whether surgery itself is responsible or whether it is the concomitant lack of enteral feeding. In any case, CF should be strongly suspected in any baby with meconium ileus irrespective of the screening result. Neonatal units should have appropriate investigation protocols in place (see Sathe and Houwen, 2017).

Blood transfusion:

The effects of blood transfusion on IRT are unclear but could result in a false negative IRT result. A repeat sample should therefore be taken after a minimum of 72 hours has elapsed. As white cells are removed prior to blood transfusion, misleading results with genetic analysis are unlikely (see Brauner and others, 1997).

Viral infection:

Viral infection leading to acute gastroenteritis or respiratory illness may also be associated with a false negative screening result (see Dunn and others, 2011).

Prematurity:

The performance of the IRT assay for premature babies is less clear and there is potential for both false positive and false negative results (see Cortes and others, 2014; Bourguignon and others, 1986; Kirby and others, 1981; Heidendael and others, 2014).

It is not practicable to adopt alternative diagnostic approaches routinely in these last 2 groups of babies. Always bear in mind in mind that not all cases of CF will be detected on newborn screening and any child showing appropriate symptoms should be investigated accordingly.

Potential for false positive results

Elevated IRT levels have been reported in association with congenital infections, renal failure, bowel atresias and nephrogenic diabetes insipidus. False positives also occur in children who are very sick on paediatric intensive care unit (PICU) who do not specifically have any bowel or renal problems. High IRT levels can occur in babies with a variety of chromosomal abnormalities particularly trisomies 13 and 18 (see Castellani and others, 2009).

Other causes of raised IRT include ΔF508 carriers, hypoxic insult to pancreas, congenital heart disease, spina bifida, gastroschisis, viral infections and galactosaemia.

The standard screening protocol should be followed in all of these situations. If CFTR variants are not found, laboratory staff should explain to the clinician looking after the baby that the raised IRT may be a secondary phenomenon and consider whether or not further investigation is required (for example, repeat IRT or sweat test).

Question 5

The IRT assay

Accepted Answer

Assay IRT in the blood spot using a methodology that has been demonstrated to be fit-for-purpose and approved by the NBS screening programme. Where possible, the reagents and instrumentation should be CE marked. The laboratory should follow the procedures detailed in the manufacturer’s instructions.

Include internal quality control samples at 3 IRT levels with each analysis batch. Each laboratory should assign acceptable ranges for these samples.

Rationale for IRT cut-off values

The sensitivity and specificity of screening are crucially dependent on the performance of the IRT assay. In particular, if cut-off B is set too low, too many samples will be sent for genetic analysis and there will be a disproportionate increase in the number of carriers detected; if it is too high then CF cases may be missed.

Unfortunately, this is a particularly difficult assay to control as human blood contains a variety of trypsinogen species which react differently with the various antibodies used for immunoassay. Additionally, the IRT species increased in neonates with CF has different properties from that present normally in neonatal blood (see Dhondt and Farriaux, 1994).

For these reasons, it is not possible to deliver the quality assurance required for newborn screening solely by means of an EQA scheme with circulated blood spots, particularly as there are marked matrix effects.

The 99.5th centile for IRT is used to select samples for subsequent genetic analysis and consequently to help determine which patients will be referred for clinical investigation as a ‘condition suspected’ case. Comparing data from several UK laboratories reveals significant variation in the population distribution of IRT values with different kit lots (lot variation should not exceed +/-8%) as well as systematic inter-laboratory variations due to differences in instrumentation (see Pollitt and Matthews, 2007) and possibly ethnicity. In order to determine the 99.5th centile in a reliable way, it is estimated that >12,000 samples must be analysed, but even in laboratories with large workloads this is not possible in a reasonable time frame.

National cut-off for IRT

Due to concerns about obtaining sufficient data and varying the cut-off frequently, a new approach came into effect from 1 April 2020 to define a ‘national’ cut-off based on retrospective data analysis from several laboratories over a longer period of time. It was agreed that this would be subject to regular review based upon continued data collection from laboratories on a quarterly basis (for the current values, see section 2.1 above). Note that these cut-offs may not apply in the same way in the Republic of Ireland laboratory (Dublin) who screen at day 3 of life and in the Scottish laboratory (Glasgow) who screen at day 4.

Question 6

Genetic analysis

Accepted Answer

Genetic analysis must be performed using an ISO 15189 accredited process.

Sample requirements, identity and transport

There is no requirement for a measured amount of blood, so the residual blood from a spot that has already had a disc punched out is likely to be sufficient.

There must be a tracking system to ensure that dried blood spot samples sent for genetic analysis are identified unequivocally. The card should not normally leave the screening laboratory.

Samples must be securely identified. The simplest method is to cut an irregularly shaped strip from the blood spot card (with the blood at one end), and on the blank section of the strip write patient identifiers including at least 2 from the baby’s date of birth, surname and NHS number. The irregularly shaped strip can then be matched with the card if subsequently required. With increasing automation and an electronic IT system, a bar-code sample identifier is desirable.

Appropriate timely transport arrangements to the genomics laboratory must be organised and the laboratory made aware of the imminent arrival of a screening specimen(s). Sample receipt by the genomic laboratory should be acknowledged back to the screening laboratory.

2-stage analysis

Genetic analysis is carried out in 2 stages to minimise the detection of homozygotes for the ‘milder’ alleles (CFTR variants associated with adult onset CFTR-related disorders rather than classic CF) and the number of carriers detected.

The first stage establishes whether the baby has any of the 4 most common CFTR variants in the English population associated with severe disease. Nomenclature complies with the Human Genome Variation Society Guidelines to ensure consistent and precise nomenclature. These 4 variants are as detailed below.

cDNA	Protein	Traditional
c.1521_1523delCTT	p.Phe508del	ΔF508 (see Note 1)
c.1652G>A	p.Gly551Asp	G551D (see Note 2)
c.1624G>T	p.(Gly542Ter) or p.(Gly542*)	G542X
c.489+1G>T		621+1G>T

CFTR reference sequences: cDNA – LRG_663t1 (NM_000492.3); protein – LRG_663p1 (NP_000483.3)

Note 1: Depending on the technology used, the c.1519_1521delATC, p.(Ile507del), (ΔI507) variant may also be detected.

Note 2: Depending on the technology used, the c.1657C>T, p.(Arg553Ter), (R553X) variant may also be detected.

The first 3 variants are also described according to the protein change so p.Gly551Asp indicates that the glycine at amino acid residue 551 is predicted to change to aspartic acid. c.489+1G>T is an intronic variant in the canonical splice site and thus has no directly associated amino acid change. The numbering indicates the 1st nucleotide (+1) after nucleotide 489 (which is at the last nucleotide of the exon).

This first stage detects over 80% of disease-causing variants in the UK population.

There are several technologies, including a range of commercial kits, available for detection of these 4 CFTR variants. It is essential that in the first stage of genetic analysis, no variants other than the ones specified above are co-incidentally detected.

The second stage covers a wider range of CFTR variants and is applied, using the original sample, in all cases where a single variant is detected in the heterozygous state at the first stage.

The panels used are specified by the CF Screening Advisory Board and reviewed periodically in the light of technical developments and on-going evaluation of the programme.

Reporting results

Samples with no detected CFTR variants may be reported to the screening laboratory by list with appropriate specimen identifiers.

For any sample showing one or more CFTR variants, the genomic laboratory must supply the screening laboratory with a formal report, largely limited to a factual description of the findings (variants tested for and results). Standard advice on family studies, such as referring the baby to a CF centre for follow-up, is inappropriate in the screening context as these options are incorporated in the protocol. Send copies (not transcriptions) of the genetic analysis report with the screening laboratory’s report or referral letter to the baby’s consultant or GP as appropriate.

Next steps

Following the genetic analysis, the screening laboratory should proceed according to the screening protocol (see section 2.4 above).

Question 7

Reporting and communication of results

Accepted Answer

Reporting to CHRDs/CHISs

Screening laboratories and CHRDs/CHISs should use the national status codes and subcodes to record the outcomes of NBS screening. Ideally, the laboratory sends screening results to CHRDs/CHISs and the newborn blood spot failsafe solution (NBSFS) using electronic messaging. Status codes should also be used for reporting to the NBSFS.

Reports of all screening results should have a generic disclaimer attached, stating:
These tests are screening tests; no screening test is 100% reliable.
Such a disclaimer is particularly relevant to CF because of the potential for falsely high or low IRTs from other non-CF causes (see section 4.2 above).

Newborn blood spot failsafe solution (NBSFS)

The NBSFS is an IT system that identifies babies born in England who have missed NBS screening. It is in use by all maternity units across England. The system also records repeat requests and screening outcomes to support failsafe processes.

The NBSFS user guide and NBSFS operational level agreements provide more information about how to use the failsafe system.

Communicating results

CF not suspected

‘CF not suspected’ results should be communicated to the parents/carers by CHRDs/CHISs by 6 weeks of age. If a ‘CF not suspected’ result is generated following collection and analysis of a repeat sample for a very high first IRT (> 99.9th centile) then the family should be informed promptly (ideally within 24 hours of the laboratory obtaining the second IRT result) as they are likely to have raised anxiety levels due to a repeat sample being needed.

Probable CF carrier, low likelihood of CF

Again, the family will likely have raised anxiety levels because a repeat blood spot sample was needed. The sample taker should have explained the reason for the repeat and when to expect the result. The family should be informed by a healthcare professional of the ‘probable CF carrier’ result promptly, ideally within 24 hours of the laboratory obtaining the second IRT result. Where possible, the screening laboratory should have a screening nurse specialist contact the family health visitor (or preferably a designated health visitor if there is an appointed lead health visitor for giving screening results to families) by phone. They should also send confirmation of the result to the family in writing.

The laboratory should also inform the baby’s GP of the result in writing. A ‘cystic fibrosis: probable carrier’ template letter is available. The communication with the GP should include a link to the ‘cystic fibrosis carrier’ leaflet as well as information about which healthcare professional has been informed of the final result and will be communicating it to the parents/carers.

The designated health visitor or alternative healthcare professional (who must be trained for the purpose) will visit the family to inform them of the result and give them the ‘cystic fibrosis carrier’ leaflet. The healthcare professional that gives the carrier result to the family should complete and return the ‘cystic fibrosis screening: carrier of CF gene’ follow-up form to the newborn screening laboratory within 24 hours of the visit.

Each laboratory should have an agreed arrangement for the communication and follow-up of ‘probable CF carrier’ results. This should be in line with the ‘managing positive results from cystic fibrosis screening’ guidelines.

The screening protocol is designed to pick up as few carrier babies as possible and therefore should not be regarded as a reliable means of detecting CF carrier babies.

CF suspected

‘CF suspected’ results require follow-up/clinical referral. The laboratory should refer babies with positive screening results for CF the same or next working day following confirmation of a positive result. The screening laboratory should report the screen positive result by phone and in writing to the regional CF centre. For information on the regional CF centres, please refer to the Cystic Fibrosis Trust website.

‘Cystic fibrosis: presumed positive’ template letters are available for the screening laboratory to notify the designated clinician and the baby’s GP of the ‘CF suspected’ result. The regional CF centre will usually be responsible for liaising with a more local CF clinic (as required by local arrangements). The referral notification should include a link to the ‘managing positive results from cystic fibrosis screening’ guidelines. This initiates the clinical referral of screen positive cases.

An appropriately trained healthcare professional should communicate the ‘CF suspected’ result as soon as possible to the parents/carers. The ‘cystic fibrosis suspected’ leaflet can support healthcare professionals to have this conversation. Early contact enables them to start their baby’s treatment as soon as this can be arranged.

It is imperative that the GP understands that the result has been communicated directly to the regional CF Centre (who will contact the parents/carers), and that the letter they (the GP) have received is for information only.

Each screening laboratory should have an agreed arrangement for the follow-up and referral of all ‘CF suspected’ cases, which should be in line with the ‘managing positive results from cystic fibrosis screening’ guidelines.

Question 8

Programme monitoring and data collection

Accepted Answer

Programme standards

The NBS screening programme standards are used to assess the NBS screening process. The standards are a set of measures that have to be met to make sure screening is safe and effective. All health care professionals involved in the NBS screening pathway have a part to play in meeting these standards.

Laboratories should submit standards data to the NBS screening programme on an annual basis. The annual data collection template is shared with the screening laboratories via email each year, with instructions for completion and submission. Data submissions must be accurate, timely and complete. This enables performance monitoring and programme evaluation.

Monitoring clinical outcomes

Request clinical information from clinical referral centres on each presumptive positive case. Complete the presumptive positive follow-up form for each case and submit the form to the NBS screening programme. It is the responsibility of the designated clinician at the clinical referral centres to ensure that these forms are completed and returned to their respective laboratory directors. The regional CF centre should also be informed about diagnostic outcome to facilitate regional and national audit.

Follow-up information is also required on all cases reported as CF carriers. For each case, send the CF carrier follow-up form to the relevant health professional (specialist health visitor/counsellor) for completion. The anonymised data from the form should comprise part of the annual screening standards data submission by the screening laboratory to the NBS screening programme.

Follow-up information is also required on all cases with an inconclusive test result. The newborn screening laboratory should complete the CF inconclusive test follow-up form and send it to the baby’s midwife or health visitor (also send a copy to the baby’s GP).

Babies diagnosed with CF not identified through newborn screening

If a screening laboratory director, clinical team or molecular laboratory director is made aware of a CF case (born after 01 April 2007) that has not been detected via the CF screening programme, it is very important that information on the case is reported to the NBS screening programme.

The clinical team should gather the details in conjunction with the screening laboratory director using the ‘CF diagnosis not identified through screening’ form. The laboratory director should collate and anonymise the data, and send it as soon as possible to NBS screening programme at phe.screeninghelpdesk@nhs.net. All form fields shaded in grey should be removed prior to submitting the form.

There should be no patient identifiable data on the ‘CF diagnosis not identified through screening’ form.

Data on these cases are collated on an annual basis as an important part of the audit and evaluation of the CF screening programme.

Screening safety incidents

All safety concerns and incidents must be reported and managed in accordance with the ‘managing safety incidents in NHS screening programmes’ guidance. This guidance details the accountabilities for reporting, investigating and managing NHS screening programme safety incidents. It covers the management of screening programme:

safety concerns
safety incidents
serious incidents

Question 9

Background to screening for CF

Accepted Answer

Screening for CF was formally introduced as a national newborn screening programme in England in April 2001. Screening was already taking place in Wales, Northern Ireland and some parts of England. Screening started in Scotland in February 2003 and became universal across the UK in October 2007.

The programme seeks to balance the interests of families where a child is identified as having CF versus the interests of the large majority of families where children are unaffected. The programme was also designed to minimise the number of CF carriers unavoidably detected.

Scientific background to the screening protocol

Newborn screening for CF is founded on the work of Crossley and others (1979), who showed that IRT in blood is significantly increased in affected newborns. Screening programmes based on IRT measurement from a dried bloodspot sample were introduced in East Anglia in 1980 and subsequently elsewhere in the UK. A raised IRT level is sensitive for identifying CF but has poor positive predictive value (PPV) and therefore CF programmes require a second ‘tier’ test to improve PPV. In early programmes, this was traditionally IRT in a second blood sample collected at 2 to 4 weeks of age. A 2-stage IRT procedure has the disadvantage of requiring a relatively high number of second samples, which increase the anxiety generated by screening, and the presumptive diagnosis is made relatively late.

Once the CFTR gene was identified in 1989, the option of genetic analysis became possible to replace the second-tier IRT assay. Initially, in 1990, only the most common CFTR variant, ΔF508 (now known as c.1521_1523delCTT p.Phe508del) was used, but more recently panels with a larger number of CFTR gene variants have been used globally.

Over 2000 variants have been identified associated with CF and around 400 occur at sufficient frequency to determine their characterisation (see Farrell and others, 2017; Sosnay and others, 2017).

For the UK programme, the initial panel used for babies with a raised IRT measurement contains the 4 commonest CFTR variants in the UK population. If one variant is recognised, an expanded panel is used. In cases where only one variant is recognised using both the initial and expanded panel, a second IRT measurement is undertaken at day 21. If this is lower than a pre-defined cut-off the baby is classified as a probable carrier. Babies with a raised second IRT are referred for a clinical assessment and sweat testing. The limited initial panel and repeated IRT measurement at day 21 are unique aspects of the UK programme.

Because of the limited initial panel, a safety net is employed in the UK protocol. If the first IRT is extremely high (>99.9th centile), but there is no variant recognised on the limited 4-variant panel, then a second dried blood spot sample is collected at day 21. Babies with a raised second IRT are referred for clinical assessment and sweat testing. This aspect of the UK protocol recognises babies with CF who have unusual CFTR variants (disproportionately babies from a non-European background), but has a poor positive predictive value compared to other parts of the protocol (around 10 to 20 babies are referred for sweat testing for each positive diagnosis).

Heeley, M.E., Field, A.A., Whitaker, J. & Heeley, A.F. (1999) An update of cystic fibrosis screening in East Anglia 1990-1997 with previous ten years included for comparison. Proceedings of the International conference, Dépistage néonatal de la mucoviscidose, 10-11 September, 1998, Université de Caen. Caen, France. ↩