- Email: [email protected]
Assessing the improvements in the newborn screening strategy for cystic fibrosis in the Balearic Islands
Clinical Biochemistry 48 (2015) 419–424
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Assessing the improvements in the newborn screening strategy for cystic fibrosis in the Balearic Islands Josep Miquel Bauça a,⁎, Daniel Morell-Garcia a, Magdalena Vila a, Gerardo Pérez a, Damián Heine-Suñer b, Joan Figuerola c a b c
Servei d'Anàlisis Clíniques, Hospital Universitari Son Espases, Palma de Mallorca, Spain Servei de Genètica, Hospital Universitari Son Espases, Palma de Mallorca, Spain Servei de Pediatria, Hospital Universitari Son Espases, Palma de Mallorca, Spain
a r t i c l e
i n f o
Article history: Received 30 November 2014 Received in revised form 15 January 2015 Accepted 3 February 2015 Available online 11 February 2015 Keywords: Newborn screening False-positive rate Strategy Sweat test
a b s t r a c t Objectives: Newborn screening strategies for cystic fibrosis (CF) are run worldwide, and aim at the early detection of the disorder to significantly improve the quality of life. Elevated levels of immunoreactive trypsinogen (IRT) represent a high likelihood for the screened child to be affected with CF. However, the specificity of IRT is low. The objective of this study was to assess the screening program in the Balearic Islands during the past 14 years. Design & methods: We evaluated all results of the screening program after 14 years, by considering all changes in the protocol and assessing the number of positive samples, the mutations detected, the number of sweat tests performed, the incidence of CF and the presence of false-negative cases. Results: Despite a great variability among the different Balearic Islands, the global incidence of CF was 1:6059 for the 14 years assessed. The incidence in the smaller islands is about 5 times higher than in Majorca (1:2376 versus 1:10,613). After different changes in the protocol, an IRT cut-off value of 60 ng/mL was established. The two most common mutations are ΔF508 and G542X, in accordance with other geographical regions. Conclusions: The changes in the protocol helped reduce the number of sweat tests performed without any increase in the false-negative rate. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Newborn screening is an integrated program that aims at the systematic identification of inborn errors of metabolism, endocrine disorders and other diseases in the first days after birth. With a high sensitivity, it looks for potentially harmful disorders that are not apparent or evident, and its basic principle relies on the fact that early detection and treatment will lead to improved outcomes by reducing morbidity and mortality. Such screening programs started in the United States in 1960s and expanded worldwide, currently representing an important public health initiative in many countries. Nevertheless, the range of screened pathologies may vary according to ethical, epidemiological, economic and political issues. In the Balearic Islands, an archipelago with an overall population of 1,113,506 (Fig. 1), the newborn screening program started in 1979, including only two pathologies: phenylketonuria (PKU) and congenital hypothyroidism (CH). It was not until 1999 when cystic ⁎ Corresponding author at: Ctra. Valldemossa 79, module J+1, 07010 Palma de Mallorca, Spain. Fax: + 34 871 909706. E-mail address: [email protected] (J.M. Bauça).
fibrosis (CF) was also introduced. Since then, many updates and changes have occurred. CF is the most common genetic disorder in Europe and North America, having greater incidence in the Caucasian ethnicity [1,2]. With an autosomal recessive inheritance, it is reported to have an incidence of 1 in 2000–2500 live births. The main clinical features of CF are abnormally high sodium and chloride concentrations in sweat, insufficient digestive function, exocrine pancreatic insufficiency, and an increase of viscous secretions from mucous glands which lead to obstructive lung disease and infections. According to a study performed in Wisconsin, newborn screening for CF enables an immediate and accurate diagnosis before the onset of clinical symptoms [3,4]. Current CF screening approaches are based on the quantification of immunoreactive trypsinogen (IRT) on dried blood spots. Elevated levels of IRT represent a high likelihood for the screened child to be affected with CF. However, the specificity of IRT is low. Alternative screening methods are continuously suggested to improve the efficacy of the programs [5], for instance, by adding determinations to the protocol, such as the pancreatitis-associated protein (PAP) [6,7]. Since ethical aspects regarding benefits and risks are still under debate, we decided to perform an integral evaluation of 14 years of CF screening program, taking into consideration all the changes in the
http://dx.doi.org/10.1016/j.clinbiochem.2015.02.001 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
420
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
Majorca Population: 873,414 Screened: 76,125 CF incidence: 1:10,613
Minorca Population: 94,875 Screened: 7,321 CF incidence: 1:2,376
Ibiza – Formentera Population: 145,217 Screened: 11,448 CF incidence: 1:2,781 Fig. 1. Regional variations in the incidence of screened CF patients, from 2005 to 2013. (source for population: 2011 census).
protocol, assessing the benefits and flaws of each period, and checking all biochemical testing results, genetic analyses and outcomes. Material and methods Study population The study population encompasses all the newborn in the Balearic Islands between January 2000 and October 2013. A total of 151,471 newborns were screened for CF. Although most of the inhabitants are of Caucasian origin, our population is relatively heterogeneous. This evaluation was performed according to the guidelines of the Investigation Ethics Board of the Balearic Islands. Sample collection Dried blood samples (on Whatman-903 filter paper) were obtained at all maternity units; both in public and private settings. Samples were drawn after 48 h of life before hospital discharge, and after two complete doses of either maternal or formula milk. Family data and informed consent were required for possible further genetic testing. Both demographical data registration and biochemical testing (IRT and sweat chloride) were performed in the Biochemistry Laboratory of Hospital Son Espases. Sample quality is controlled visually upon reception. Further samples were required in cases of premature birth (b 32 weeks of gestation), low weight at birth (b 1500 g), blood transfusions, less than 2 complete milk ingestions or in cases of biochemical abnormal results. Screening strategies Different protocols were followed during the period, mainly due to improvements in the cut-off value for IRT. Performance of a sweat test (ST) and sequencing of the CFTR gene were carried out as confirmation (Fig. 2). The assessment of protocols was based on the number of positive samples obtained, the amount of samples with mutations detected, the number of ST performed, the incidence of CF and the presence of false-negative results. Between September 1999 and October 2005, the protocol was based on two IRT quantifications, followed by mutation study and ST altogether (IRT/IRT/DNA + ST). From November 2005 until June 2006, one IRT determination was removed, and the cut-off value was followed at 60 ng/mL. From July 2006 to June 2008, IRT/DNA/ST,
without a variation in the cut-off value. From July 2008 to January 2013, the protocol was the same, but the cut-off value was increased up to 70 ng/mL, by accepting the vendor kit recommendations and by performing a correlation study (data not shown). After February 2013, the cut-off value was lowered again down to 60 ng/mL (see Table 1), due to method reevaluation.
Current algorithm IRT concentrations below 60 or 70 ng/mL were classified as negative for CF, and no additional testing was required. Results above the cut-off value were confirmed in the same sample before further analyses (DNA or ST). In case this ‘apparently positive’ result was confirmed, a filter paper aliquot was sent to the Clinical Genetics Laboratory together with the demographic data of the child.
Sep 1999
IRT(<120 ng/ml) → IRT(< 60 ng/ml) → DNA + ST Nov 2005
IRT (<60 ng/ml) → DNA + ST Jun 2006
IRT (<60 ng/ml) → DNA → ST Jun 2008
IRT (<70 ng/ml) → DNA → ST Jan 2013
IRT (<60 ng/ml) → DNA → ST
Fig. 2. Timeline with the different changes in the protocol.
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
421
Table 1 Results of newborn CF screening in the 4 study periods, according to changes in the algorithm. Period
2000 to Oct 2005 Nov 2005 to Jun 2006 Jul 2006 to Jun 2008 and Feb to Oct 2013 July 2008 to January 2013 Total
Number screened IRT d2-5 positive (% of screened population) Two mutations (n) One mutation (n) No mutations (n) Referred for ST (% of screened population) confirmed CF (n) Incidence ST positive (% of total ST) ST negative Number of false-negative (% of CFTR mutations)
58,656 234 (0.39) 9 13 212 234 (0.39) 11 1:5,332 9 (3.85) 225 2 (9.09)
6,715 72 (1.07) 1 8 63 72 (1.07) 1 1:6715 1 (1.38) 71 0
If one or more mutations were detected during the genetic testing, a sweat test was performed. If the sweat test was normal, the newborn was classified as ‘CF carrier’ and the family was offered extensive genetic family counseling (Fig. 3). IRT levels have been shown to decrease with age [8], even in the presence of pathology. Therefore, samples received at the laboratory after 40 days of birth were not processed. Exceptions were cases of adopted children or flaws in the newborn screening program itself, as IRT concentrations decrease after the first weeks of life, genetic testing was directly carried out. In special cases, such as preterm deliveries or low birth weight cases, a new sample is drawn after 15 days of life; and if the newborn is admitted at the Intensive Care Unit, the sample is drawn after hospital discharge. Biochemical IRT quantification For the quantification of immunoreactive trypsinogen (IRT), we used a solid phase, two-site fluoroimmunometric assay (Wallac Victor2D, Perkin-Elmer). Cut-off values were 60 or 70 ng/mL, depending on ongoing strategy.
32,588 337 (1.03) 4 29 304 33 (0.10) 4 1:8147 3 (9.09) 30 0
53,512 532 (0.99) 9 34 488 43 (0.08) 9 1:5,946 11 (25.6) 33 0
151,471 1,175 (0.78) 23 84 1067 382 (0.25) 25 1:6059 24 (6.3) 359 2 (1.87)
Internal Quality Control samples with different IRT concentrations were analyzed once every day, prior to the analysis of the patients' samples themselves. External Quality Control was accomplished according to two independent programs: CDC Centers for Disease Control and Prevention Proficiency Testing (Atlanta, GE), three times a year, and AECNE (Asociación Española de Cribado Neonatal) monthly. Genetic study Mutation analysis of the CFTR gene was performed in the same dried blood sample for the screening of 33 prevalent mutations in our region. The mutation analysis of the CFTR gene was performed at the Clinical Genetics Laboratory, a different department at the same hospital, by single nucleotide polymorphism genotyping based on an oligonucleotide ligation assay (OLA) (CFv3 kit, Abbott Diagnostics) and subsequent sequencing. The screening was carried out for the following mutations: ΔF508, G542X, N1303K, W1282X, G551D, 1717-1GNA, R553X, I507del, 711+1GNT, 3905insT, R560T, 1898+1G-NA, I148T, 621+1G-NT, 3849+10kbCNT, 2183AA-NG, 394delTT, 2789+5GNA, R1162X, 3659delC, R117H, R334W, R347P, G85E, 1078delT, A455E, 2184delA, S549N, S549R, V520F, 3876delA, 3120+1GNA, R347H, using the Celera
Fig. 3. Current CF newborn screening algorithm.
422
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
Table 2 Results of sweat testing according to CFTR mutation analysis in patients with elevated serum IRT. Number of mutations
Number of referred for ST
ST +
ST borderline
ST −
Two mutations (n = 24) One mutation (n = 84) No mutations (n = 1067)
24 84 275
21 2 1
2 8 0
1 74 274
Kit (Abbott Diagnostics). According to the manufacturer, it is able to detect 76% of mutations in our population. Sweat test A sweat test was performed at the Pediatrics Unit in our hospital on all newborns with at least one mutation. Sweat samples were taken after pilocarpine stimulation on the forearm. Chloride was measured by conductivity in a macroduct (cut-off value b80 mmol/L) or nanoduct (cut-off value b60 mmol/L) (Wescor analyzers). The confirmation for high conductivity values was performed by indirect potentiometry (Architect c16000, Abbott Diagnostics) at the Biochemistry Laboratory. Samples with chloride concentrations below 35 mmol/L were classified as ‘negative’, above 60 mmol/L as ‘positive’, whereas concentrations between 36–60 mmol/L were classified as ‘borderline’ [9,10]. Borderline results were advised to be repeated after 2 weeks. Statistical analysis ANOVA test was used for the comparison of IRT values among the three groups (healthy, CF carriers and CF). All statistical analyses were performed using SPSS 15.1 (SPSS Inc., Chicago, IL, USA), and statistical significance was considered when p b 0.05. Kolmogorov–Smirnov test was used to assess sample normality, and Grubbs' test for robustness and outlier removal. Results In the analyzed period, between January 2000 and October 2013, a total of 151,471 newborns were screened, 1,175 of which had IRT values above the established cut-off (0.78%). In all these cases, the CFTR gene was studied. For data analysis, four periods have been described, according to the different changes in the screening algorithm (Table 1). The first period represents the start of the screening program. In this strategy, a value of 120 ng/mL for IRT was used as cut-off for samples obtained 2–5 days after birth, whereas 60 ng/mL was the cut-off for late samples (25–40 days of life). Due to a change in the analytical methodology in 2005, a single cut-off for any sample was established at 60 ng/mL, which led to an increase in the number of positive results up to 1.07% of the screened population. During the first and second periods, both CFTR genetic testing and sweat testing were performed on all patients with positive IRT values. Since only 10% of such cases showed mutations, and b4% yielded a pathological sweat test, this protocol was considered to add unnecessary costs, and was therefore changed (Table 1). In the third period, sweat test was performed only on patients with at least one mutation in the CFTR gene. In this way, it was possible to reduce by 90% the number of sweat tests performed. In the fourth period, since the number of positive results was still high (1%), the cut-off value for IRT was changed to 70 ng/mL. Still, the rate of positive results was not altered (0.99%). Interestingly, the incidence for CF during this period was significantly higher as with two previous strategies (1:5,946 versus 1:8147), and no false-negative case occurred. As a result, the cut-off value was lowered again down to 60 ng/mL. Up to 107 patients showed CFTR gene mutations. Of these, 23 had 2 mutations in the CFTR gene, and 84 had only one mutation. All of them
underwent a further sweat test. Diagnosis was confirmed for 23 of them (8.7% had one mutation). All children with one mutation and a negative result for sweat test were classified as ‘healthy carriers’, and familial genetic counseling was advised. A total of 275 children without any mutation underwent sweat test, finding only 1 CF positive case among them (Table 2). Diagnosis of CF was established by a pediatrician, according to the Cystic Fibrosis Foundation guidelines [11]. The classical form of CF was diagnosed in 24 cases. The atypical presentation, when defined as the presence of 2 mutations and a borderline value for the sweat test, was found in 2 cases, whereas when defined as a borderline sweat test, independently of the presence of mutations, was found in 8 cases. These altogether 10 atypical cases were classified as ‘suspicion of CF’, and subjected to clinical follow-up (Table 2). Deletion of Phe508 (ΔF508) was present in 83.3% of the alleles of patients with CF. Homozygosity frequency was 25%. The second most common mutation was G542X, in 10.7% of the screened population alleles. Although the manufacturer claims that the kit is able to detect 76% of mutations in our population, our results show that detection rate was 87%. Table 3 summarizes the most frequent mutations (N1%). Only one frequent mutation in our community (L206W) is not included in the kit's detection panel. Moreover, some mutations not included in the kit were also detected (712-1GNT, R75Q and V317A), with an allele frequency of 0.76%. During the whole period, there were two false-negative cases (IRT below cut-off value), which showed a meconium ileus and a borderline value for the sweat test, as well as a mutation not included in the kit [12,13]. Taking all together, a statistical analysis was performed to evaluate the possibility of IRT values being able to predict the results of the mutation analysis. Kolmogorov–Smirnov test assured a normal distribution of the different populations. Outliers were removed and percentiles were calculated according to Grubbs' robustness test. Mean IRT values were compared depending on the detected mutations. For IRT values close to cut-off (60–70 ng/mL), differences were not statistically significant for the prediction of mutations, since 91% of patients did not show any genetic alteration (Table 4). IRT values higher than 140 ng/mL indicated a possibility of 90% for the patient to have 2 mutations and suffer from CF (p = 0.0001). Global incidence of CF was 1:6,059 for the 14 years assessed. If we only take into account results from the last change in protocol, and cut-off values 60–70 ng/mL since 2005, the accumulated incidence is 1:6,610 (CI 95%: 1:2879–1:7554). Significant differences were found across the Balearic archipelago, being the incidence in the smaller islands about 5 times higher than in Majorca (1:2,376 versus 1:10,613) (Fig. 1).
Table 3 Percentage common mutations (Celera kit 33, Abbott Diagnostics). Mutation
Screened CF patients (n = 131 alleles)
ΔF508 G542X R117H N1303K 2789+5GNA R1162X R553X L206W
58.8% 10.7% 6.9% 4.6% 3.0% 1.5% 1.5% 1.5%
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424 Table 4 Predictive value of the IRT levels (cut-off N60 μg/L) for finding mutations in the CFTR gene analysis. Number of mutations
Mean IRT level (ng/mL)
95th percentile
5th percentile
Two mutations (n = 24) One mutation (n = 84) No mutations (n = 1067)
147.31 86.66 85.70
287.52 139.88 133.90
66.88 b60.0 b60.0
Comparison of IRT mean concentrations
p-value
2 mutations group vs 1 mutation group 2 mutations group vs no mutation group 1 mutation group vs no mutation group
0.006 0.0001 0.96
423
and Castilla-Leon, 1:4800 in Aragon, 1:6244 in Catalonia and 1:6602 in the Balearic Islands [31,32]. Data from the Spanish Association of Newborn Screening (AECNE) approximate the incidence to 1:4807, although great variability among regions is reported. In accordance to this, incidence was 1:8,342 in Madrid in 2011, 1:10,432 in 2012 [33], and in the Basque Country it was 1:3000 [34]. In summary, the four changes in the screening strategy have resulted in a reduction in the total number of patients requiring the performance of a sweat test (from 1.07% to 0.10%), and helped maintaining a stable global incidence for CF during the last 10 years of 1:6,610 (IC 95%: 1:2879–1:7554).
References Discussion The benefits of CF screening and detection at birth are more than evident, both for the newborn themselves and for the families. In our case, we follow the IRT/DNA protocol, but to date, many different strategies for its detection have been described [14–16]. One of the main objectives is to reduce the rate of carrier detection, which has been widely regarded as an ethical problem in bibliography [14,17], and implies the performance of genetic counseling [18]. For us, this fact does not pose any shortcoming, since it helps reduce the incidence of this pathology. However, a reasonable period of time is needed to accurately evaluate such changes. The autosomal recessive inheritance of cystic fibrosis evidences a high number of carriers and affected people among populations with high inbreeding tendencies, due to geographic isolation. In small islands such as the Balearic, the geographic situation would be mostly responsible for the high incidence found in our study. In addition to this relative degree of endogamy, which has been extensively studied, but weak conclusions have been drawn to date, another explanation for the five-fold incidence of CF detected in smaller islands could result from a smaller sample size. Moreover, the influence of genetic variability, a characteristic of small islands, may result from the strong Anglo-Saxon and French domination in Minorca in the 18th century might explain the mutation pattern observed. In fact, demographic studies in the UK describe G542X (1.85%) as the most frequent mutation in CFTR after ΔF508 (74.1%) and G551D (3.37%). This stays in contrast with R117H, which is the third most frequent in Caucasian population (1.25%) [19–22]. Furthermore, population studies in the Mediterranean area and North Africa have found a higher allele frequency for G542X and N1303K in comparison to R117H, being the latter more frequent in Northwestern Europe [23,24]. These data are consistent with findings from a French study [17]. The most frequent mutation of CF worldwide is ΔF508, although great geographic dispersion is present, with a gradient from Northwest to Southeast (Denmark 88%, Italy 50% and Turkey 20%) [25,26]. In the Balearic Islands, ΔF508 is present in 58.8% cases, similar to Galicia (59.3%). In the South of Spain, incidence is 43.5%, and in the Basque Country, it is 87%. The second most common mutation is G542X, with an incidence of 10.7%. This value is similar to that found in Andalusia (11.4%), and significantly higher than the one in Asturias (5.0%), Castilla-Leon (3.8%) or Galicia (4.9%), according to the Cystic Fibrosis Newborn Screening Assessment, issued by the Spanish Ministry of Health in 2013. L206W, which is known to be the sixth most prevalent in Spain (1.6%), is reported to cause pancreatic insufficiency in 78% of the patients [27], and is more prevalent in Hispanic population (6.2%) and African Americans (1.2%) [28–30]. In Europe, the prevalence of CF varies between countries, from 1:1350 in Ireland to 1:25,000 in Finland, with a median birth prevalence of 1:3500 [25]. In Spain, thanks to the progressive implementation of newborn screening programs in the different provinces, a lower incidence is being detected. In 2009, CF incidence was 1:4439 in Galicia
[1] Paracchini V, Seia M, Raimondi S, Costantino L, Capasso P, Porcaro L, et al. Cystic fibrosis newborn screening: distribution of blood immunoreactive trypsinogen concentrations in hypertrypsinemic neonates. JIMD Rep 2012;4:17–23. [2] Kosorok MR, Wei WH, Farrell PM. The incidence of cystic fibrosis. Stat Med 1996; 15(5):449–62. [3] Sanders DB, Lai HJ, Rock MJ, Farrell PM. Comparing age of cystic fibrosis diagnosis and treatment initiation after newborn screening with two common strategies. J Cyst Fibros 2012;11(2):150–3. [4] Maclean JE, Solomon M, Corey M, Selvadurai H. Cystic fibrosis newborn screening does not delay the identification of cystic fibrosis in children with negative results. J Cyst Fibros 2011;10:333–7. [5] Lindau-Shepard BA, Pass KA. Newborn screening for cystic fibrosis by use of a multiplex immunoassay. Clin Chem 2010;56:445–50. [6] Sarles J, Berthézène P, Le Louarn C, Somma C, Perini JM, Catheline M, et al. Combining immunoreactive trypsinogen and pancreatitis-associated protein assays, a method of newborn screening for cystic fibrosis that avoids DNA analysis. J Pediatr 2005; 147(3):302–5. [7] Krulišová V, Balaščaková M, Skalická V, Piskáčková T, Holubová A, Paděrová J, et al. Prospective and parallel assessments of cystic fibrosis newborn screening protocols in the Czech Republic: IRT/DNA/IRT versus IRT/PAP and IRT/PAP/DNA. Eur J Pediatr 2012;171(8):1223–9. [8] Kirby LT, Applegarth DA, Davidson AG, Wong LT, Hardwick DF. Use of a dried blood spot in immunoreactive-trypsin assay for detection of cystic fibrosis in infants. Clin Chem 1981;27:678–80. [9] Gibson LE, Cooke RE. A test for concentration of electrolytes in sweat in cystic fibrosis of the pancreas utilizing pilocarpine by iontophoresis. Pediatrics 1959; 23(3):545–9. [10] van der Merwe DE, Ubbink JB, Delport R, Becker P, Dhatt GS, Vermaak WJ. Biological variation in sweat sodium chloride conductivity. Ann Clin Biochem 2002;39:39–43. [11] Rosenstein BJ, Cutting GR. The diagnosis of cystic fibrosis: a consensus statement. Cystic Fibrosis Foundation Consensus Panel. J Pediatr 1998;132(4):589–95. [12] Nelson MR, Adamski CR, Tluczek A. Clinical practices for intermediate sweat tests following abnormal cystic fibrosis newborn screens. J Cyst Fibros 2011;10:460–5. [13] Kirk JM. A continuing role for sweat testing in an era of newborn screening for cystic fibrosis. Clin Biochem 2011;44:487–8. [14] Southern KW. Determining the optimal newborn screening protocol for cystic fibrosis. Thorax 2012;67(4):281–2. [15] Vernooij-van Langen AM, Loeber JG, Elvers B, Triepels RH, Gille JJ, Van der Ploeg CP, et al. Novel strategies in newborn screening for cystic fibrosis: a prospective controlled study. Thorax 2012;67(4):289–95. [16] Hale JE, Parad RB, Dorkin HL, Gerstle R, Lapey A, O'Sullivan BP, et al. Cystic fibrosis newborn screening: using experience to optimize the screening algorithm. J Inherit Metab Dis 2010;33(Suppl. 2):S255–61. [17] Munck A, Dhondt JL, Sahler C, Roussey M. Implementation of the French nationwide cystic fibrosis newborn screening program. J Pediatr 2008;153(2):228–33. [18] Baker MW, Groose M, Hoffman G, Rock M, Levy H, Farrell PM. Optimal DNA tier for the IRT/DNA algorithm determined by CFTR mutation results over 14 years of newborn screening. J Cyst Fibros 2011;10(4):278–81. [19] McCormick J, Green MW, Mehta G, Culross F, Mehta A. Demographics of the UK cystic fibrosis population: implications for neonatal screening. Eur J Hum Genet 2002; 10:583–90. [20] Schwarz MJ, Malone GM, Haworth A, Cheadle JP, Meredith AL, Gardner A, et al. Cystic fibrosis mutation analysis: report from 22 UK regional genetics laboratories. Hum Mutat 1995;6:326–33. [21] Bobadilla JL, Macek Jr M, Fine JP, Farrell PM. Cystic fibrosis: a worldwide analysis of CFTR mutations—correlation with incidence data and application to screening. Hum Mutat 2002;10:575–606. [22] Juan-Fita MJ, Egea-Mellado JM, González-Gallego I, Moya-Quiles MR, FernándezSánchez A, et al. Expanded newborn screening in the Region of Murcia. Threeyear experience. Med Clin (Barc) 2012;13:566–71. [23] Estivill X, Bancells C, Ramos C. Geographic distribution and regional origin of 272 cystic fibrosis mutations in European populations. Hum Mutat 1997;10(2): 135–54. [24] Sobczyńska-Tomaszewska A, Ołtarzewski M, Czerska K, Wertheim-Tysarowska K, Sands D, Walkowiak J, et al. Newborn screening for cystic fibrosis: Polish' 4 years experience with CFTR sequencing strategy. Eur J Hum Genet 2012;21(4):391–6.
424
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
[25] Castellani C, Macek Jr M, Cassiman JJ, Duff A, Massie J, ten Kate LP, et al. Benchmark for cystic fibrosis carrier screening: a European consensus document. J Cyst Fibros 2010;9:165–78. [26] Southern KW, Munck A, Pollitt R, Travert G, Zanolla L, Dankert-Roelse J, et al. A survey of newborn screening for cystic fibrosis in Europe. J Cyst Fibros 2007;1:57–65. [27] The Clinical and Functional TRanslation of CFTR (CFTR2). available at http://cftr2.org. [Last accessed: January 2015]. [28] Sugarman EA, Rohlfs EM, Silverman LM, Allitto BA. CFTR mutation distribution among U.S. Hispanic and African American individuals: evaluation in cystic fibrosis patient and carrier screening populations. Genet Med 2004;6(5):392–9. [29] Rozen R, Ferreira-Rajabi L, Robb L, Colman N. L206W mutation of the cystic fibrosis gene, relatively frequent in French Canadians, is associated with atypical presentations of cystic fibrosis. Am J Med Genet 1995;57(3):437–9.
[30] Claustres M, Desgeorges M, Moine P, Morral N, Estivill X. CFTR haplotypic variability for normal and mutant genes in cystic fibrosis families from southern France. Hum Genet 1996;98(3):336–44. [31] Andalusian Attending Guidelines for Cystic Fibrosis. Plan de Atención a Personas Afectadas por Enfermedades Raras en Andalucía; May 2011. [32] Gartner S, Cobos N. Neonatal screening for cystic fibrosis. An Pediatr (Barc) 2009; 71(6):481–2. [33] Health Service in Madrid. Service for the prevention of illnesses. General Direction for Primary Care; 2012. [34] Newborn screening for cystic fibrosis in the Basque Country. Advising council for congenital errors of metabolism of the CAPV. Basque-Navarra Bull Pediatr 2010; 42:53–60.
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Assessing the improvements in the newborn screening strategy for cystic fibrosis in the Balearic Islands Josep Miquel Bauça a,⁎, Daniel Morell-Garcia a, Magdalena Vila a, Gerardo Pérez a, Damián Heine-Suñer b, Joan Figuerola c a b c
Servei d'Anàlisis Clíniques, Hospital Universitari Son Espases, Palma de Mallorca, Spain Servei de Genètica, Hospital Universitari Son Espases, Palma de Mallorca, Spain Servei de Pediatria, Hospital Universitari Son Espases, Palma de Mallorca, Spain
a r t i c l e
i n f o
Article history: Received 30 November 2014 Received in revised form 15 January 2015 Accepted 3 February 2015 Available online 11 February 2015 Keywords: Newborn screening False-positive rate Strategy Sweat test
a b s t r a c t Objectives: Newborn screening strategies for cystic fibrosis (CF) are run worldwide, and aim at the early detection of the disorder to significantly improve the quality of life. Elevated levels of immunoreactive trypsinogen (IRT) represent a high likelihood for the screened child to be affected with CF. However, the specificity of IRT is low. The objective of this study was to assess the screening program in the Balearic Islands during the past 14 years. Design & methods: We evaluated all results of the screening program after 14 years, by considering all changes in the protocol and assessing the number of positive samples, the mutations detected, the number of sweat tests performed, the incidence of CF and the presence of false-negative cases. Results: Despite a great variability among the different Balearic Islands, the global incidence of CF was 1:6059 for the 14 years assessed. The incidence in the smaller islands is about 5 times higher than in Majorca (1:2376 versus 1:10,613). After different changes in the protocol, an IRT cut-off value of 60 ng/mL was established. The two most common mutations are ΔF508 and G542X, in accordance with other geographical regions. Conclusions: The changes in the protocol helped reduce the number of sweat tests performed without any increase in the false-negative rate. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Newborn screening is an integrated program that aims at the systematic identification of inborn errors of metabolism, endocrine disorders and other diseases in the first days after birth. With a high sensitivity, it looks for potentially harmful disorders that are not apparent or evident, and its basic principle relies on the fact that early detection and treatment will lead to improved outcomes by reducing morbidity and mortality. Such screening programs started in the United States in 1960s and expanded worldwide, currently representing an important public health initiative in many countries. Nevertheless, the range of screened pathologies may vary according to ethical, epidemiological, economic and political issues. In the Balearic Islands, an archipelago with an overall population of 1,113,506 (Fig. 1), the newborn screening program started in 1979, including only two pathologies: phenylketonuria (PKU) and congenital hypothyroidism (CH). It was not until 1999 when cystic ⁎ Corresponding author at: Ctra. Valldemossa 79, module J+1, 07010 Palma de Mallorca, Spain. Fax: + 34 871 909706. E-mail address: [email protected] (J.M. Bauça).
fibrosis (CF) was also introduced. Since then, many updates and changes have occurred. CF is the most common genetic disorder in Europe and North America, having greater incidence in the Caucasian ethnicity [1,2]. With an autosomal recessive inheritance, it is reported to have an incidence of 1 in 2000–2500 live births. The main clinical features of CF are abnormally high sodium and chloride concentrations in sweat, insufficient digestive function, exocrine pancreatic insufficiency, and an increase of viscous secretions from mucous glands which lead to obstructive lung disease and infections. According to a study performed in Wisconsin, newborn screening for CF enables an immediate and accurate diagnosis before the onset of clinical symptoms [3,4]. Current CF screening approaches are based on the quantification of immunoreactive trypsinogen (IRT) on dried blood spots. Elevated levels of IRT represent a high likelihood for the screened child to be affected with CF. However, the specificity of IRT is low. Alternative screening methods are continuously suggested to improve the efficacy of the programs [5], for instance, by adding determinations to the protocol, such as the pancreatitis-associated protein (PAP) [6,7]. Since ethical aspects regarding benefits and risks are still under debate, we decided to perform an integral evaluation of 14 years of CF screening program, taking into consideration all the changes in the
http://dx.doi.org/10.1016/j.clinbiochem.2015.02.001 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
420
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
Majorca Population: 873,414 Screened: 76,125 CF incidence: 1:10,613
Minorca Population: 94,875 Screened: 7,321 CF incidence: 1:2,376
Ibiza – Formentera Population: 145,217 Screened: 11,448 CF incidence: 1:2,781 Fig. 1. Regional variations in the incidence of screened CF patients, from 2005 to 2013. (source for population: 2011 census).
protocol, assessing the benefits and flaws of each period, and checking all biochemical testing results, genetic analyses and outcomes. Material and methods Study population The study population encompasses all the newborn in the Balearic Islands between January 2000 and October 2013. A total of 151,471 newborns were screened for CF. Although most of the inhabitants are of Caucasian origin, our population is relatively heterogeneous. This evaluation was performed according to the guidelines of the Investigation Ethics Board of the Balearic Islands. Sample collection Dried blood samples (on Whatman-903 filter paper) were obtained at all maternity units; both in public and private settings. Samples were drawn after 48 h of life before hospital discharge, and after two complete doses of either maternal or formula milk. Family data and informed consent were required for possible further genetic testing. Both demographical data registration and biochemical testing (IRT and sweat chloride) were performed in the Biochemistry Laboratory of Hospital Son Espases. Sample quality is controlled visually upon reception. Further samples were required in cases of premature birth (b 32 weeks of gestation), low weight at birth (b 1500 g), blood transfusions, less than 2 complete milk ingestions or in cases of biochemical abnormal results. Screening strategies Different protocols were followed during the period, mainly due to improvements in the cut-off value for IRT. Performance of a sweat test (ST) and sequencing of the CFTR gene were carried out as confirmation (Fig. 2). The assessment of protocols was based on the number of positive samples obtained, the amount of samples with mutations detected, the number of ST performed, the incidence of CF and the presence of false-negative results. Between September 1999 and October 2005, the protocol was based on two IRT quantifications, followed by mutation study and ST altogether (IRT/IRT/DNA + ST). From November 2005 until June 2006, one IRT determination was removed, and the cut-off value was followed at 60 ng/mL. From July 2006 to June 2008, IRT/DNA/ST,
without a variation in the cut-off value. From July 2008 to January 2013, the protocol was the same, but the cut-off value was increased up to 70 ng/mL, by accepting the vendor kit recommendations and by performing a correlation study (data not shown). After February 2013, the cut-off value was lowered again down to 60 ng/mL (see Table 1), due to method reevaluation.
Current algorithm IRT concentrations below 60 or 70 ng/mL were classified as negative for CF, and no additional testing was required. Results above the cut-off value were confirmed in the same sample before further analyses (DNA or ST). In case this ‘apparently positive’ result was confirmed, a filter paper aliquot was sent to the Clinical Genetics Laboratory together with the demographic data of the child.
Sep 1999
IRT(<120 ng/ml) → IRT(< 60 ng/ml) → DNA + ST Nov 2005
IRT (<60 ng/ml) → DNA + ST Jun 2006
IRT (<60 ng/ml) → DNA → ST Jun 2008
IRT (<70 ng/ml) → DNA → ST Jan 2013
IRT (<60 ng/ml) → DNA → ST
Fig. 2. Timeline with the different changes in the protocol.
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
421
Table 1 Results of newborn CF screening in the 4 study periods, according to changes in the algorithm. Period
2000 to Oct 2005 Nov 2005 to Jun 2006 Jul 2006 to Jun 2008 and Feb to Oct 2013 July 2008 to January 2013 Total
Number screened IRT d2-5 positive (% of screened population) Two mutations (n) One mutation (n) No mutations (n) Referred for ST (% of screened population) confirmed CF (n) Incidence ST positive (% of total ST) ST negative Number of false-negative (% of CFTR mutations)
58,656 234 (0.39) 9 13 212 234 (0.39) 11 1:5,332 9 (3.85) 225 2 (9.09)
6,715 72 (1.07) 1 8 63 72 (1.07) 1 1:6715 1 (1.38) 71 0
If one or more mutations were detected during the genetic testing, a sweat test was performed. If the sweat test was normal, the newborn was classified as ‘CF carrier’ and the family was offered extensive genetic family counseling (Fig. 3). IRT levels have been shown to decrease with age [8], even in the presence of pathology. Therefore, samples received at the laboratory after 40 days of birth were not processed. Exceptions were cases of adopted children or flaws in the newborn screening program itself, as IRT concentrations decrease after the first weeks of life, genetic testing was directly carried out. In special cases, such as preterm deliveries or low birth weight cases, a new sample is drawn after 15 days of life; and if the newborn is admitted at the Intensive Care Unit, the sample is drawn after hospital discharge. Biochemical IRT quantification For the quantification of immunoreactive trypsinogen (IRT), we used a solid phase, two-site fluoroimmunometric assay (Wallac Victor2D, Perkin-Elmer). Cut-off values were 60 or 70 ng/mL, depending on ongoing strategy.
32,588 337 (1.03) 4 29 304 33 (0.10) 4 1:8147 3 (9.09) 30 0
53,512 532 (0.99) 9 34 488 43 (0.08) 9 1:5,946 11 (25.6) 33 0
151,471 1,175 (0.78) 23 84 1067 382 (0.25) 25 1:6059 24 (6.3) 359 2 (1.87)
Internal Quality Control samples with different IRT concentrations were analyzed once every day, prior to the analysis of the patients' samples themselves. External Quality Control was accomplished according to two independent programs: CDC Centers for Disease Control and Prevention Proficiency Testing (Atlanta, GE), three times a year, and AECNE (Asociación Española de Cribado Neonatal) monthly. Genetic study Mutation analysis of the CFTR gene was performed in the same dried blood sample for the screening of 33 prevalent mutations in our region. The mutation analysis of the CFTR gene was performed at the Clinical Genetics Laboratory, a different department at the same hospital, by single nucleotide polymorphism genotyping based on an oligonucleotide ligation assay (OLA) (CFv3 kit, Abbott Diagnostics) and subsequent sequencing. The screening was carried out for the following mutations: ΔF508, G542X, N1303K, W1282X, G551D, 1717-1GNA, R553X, I507del, 711+1GNT, 3905insT, R560T, 1898+1G-NA, I148T, 621+1G-NT, 3849+10kbCNT, 2183AA-NG, 394delTT, 2789+5GNA, R1162X, 3659delC, R117H, R334W, R347P, G85E, 1078delT, A455E, 2184delA, S549N, S549R, V520F, 3876delA, 3120+1GNA, R347H, using the Celera
Fig. 3. Current CF newborn screening algorithm.
422
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
Table 2 Results of sweat testing according to CFTR mutation analysis in patients with elevated serum IRT. Number of mutations
Number of referred for ST
ST +
ST borderline
ST −
Two mutations (n = 24) One mutation (n = 84) No mutations (n = 1067)
24 84 275
21 2 1
2 8 0
1 74 274
Kit (Abbott Diagnostics). According to the manufacturer, it is able to detect 76% of mutations in our population. Sweat test A sweat test was performed at the Pediatrics Unit in our hospital on all newborns with at least one mutation. Sweat samples were taken after pilocarpine stimulation on the forearm. Chloride was measured by conductivity in a macroduct (cut-off value b80 mmol/L) or nanoduct (cut-off value b60 mmol/L) (Wescor analyzers). The confirmation for high conductivity values was performed by indirect potentiometry (Architect c16000, Abbott Diagnostics) at the Biochemistry Laboratory. Samples with chloride concentrations below 35 mmol/L were classified as ‘negative’, above 60 mmol/L as ‘positive’, whereas concentrations between 36–60 mmol/L were classified as ‘borderline’ [9,10]. Borderline results were advised to be repeated after 2 weeks. Statistical analysis ANOVA test was used for the comparison of IRT values among the three groups (healthy, CF carriers and CF). All statistical analyses were performed using SPSS 15.1 (SPSS Inc., Chicago, IL, USA), and statistical significance was considered when p b 0.05. Kolmogorov–Smirnov test was used to assess sample normality, and Grubbs' test for robustness and outlier removal. Results In the analyzed period, between January 2000 and October 2013, a total of 151,471 newborns were screened, 1,175 of which had IRT values above the established cut-off (0.78%). In all these cases, the CFTR gene was studied. For data analysis, four periods have been described, according to the different changes in the screening algorithm (Table 1). The first period represents the start of the screening program. In this strategy, a value of 120 ng/mL for IRT was used as cut-off for samples obtained 2–5 days after birth, whereas 60 ng/mL was the cut-off for late samples (25–40 days of life). Due to a change in the analytical methodology in 2005, a single cut-off for any sample was established at 60 ng/mL, which led to an increase in the number of positive results up to 1.07% of the screened population. During the first and second periods, both CFTR genetic testing and sweat testing were performed on all patients with positive IRT values. Since only 10% of such cases showed mutations, and b4% yielded a pathological sweat test, this protocol was considered to add unnecessary costs, and was therefore changed (Table 1). In the third period, sweat test was performed only on patients with at least one mutation in the CFTR gene. In this way, it was possible to reduce by 90% the number of sweat tests performed. In the fourth period, since the number of positive results was still high (1%), the cut-off value for IRT was changed to 70 ng/mL. Still, the rate of positive results was not altered (0.99%). Interestingly, the incidence for CF during this period was significantly higher as with two previous strategies (1:5,946 versus 1:8147), and no false-negative case occurred. As a result, the cut-off value was lowered again down to 60 ng/mL. Up to 107 patients showed CFTR gene mutations. Of these, 23 had 2 mutations in the CFTR gene, and 84 had only one mutation. All of them
underwent a further sweat test. Diagnosis was confirmed for 23 of them (8.7% had one mutation). All children with one mutation and a negative result for sweat test were classified as ‘healthy carriers’, and familial genetic counseling was advised. A total of 275 children without any mutation underwent sweat test, finding only 1 CF positive case among them (Table 2). Diagnosis of CF was established by a pediatrician, according to the Cystic Fibrosis Foundation guidelines [11]. The classical form of CF was diagnosed in 24 cases. The atypical presentation, when defined as the presence of 2 mutations and a borderline value for the sweat test, was found in 2 cases, whereas when defined as a borderline sweat test, independently of the presence of mutations, was found in 8 cases. These altogether 10 atypical cases were classified as ‘suspicion of CF’, and subjected to clinical follow-up (Table 2). Deletion of Phe508 (ΔF508) was present in 83.3% of the alleles of patients with CF. Homozygosity frequency was 25%. The second most common mutation was G542X, in 10.7% of the screened population alleles. Although the manufacturer claims that the kit is able to detect 76% of mutations in our population, our results show that detection rate was 87%. Table 3 summarizes the most frequent mutations (N1%). Only one frequent mutation in our community (L206W) is not included in the kit's detection panel. Moreover, some mutations not included in the kit were also detected (712-1GNT, R75Q and V317A), with an allele frequency of 0.76%. During the whole period, there were two false-negative cases (IRT below cut-off value), which showed a meconium ileus and a borderline value for the sweat test, as well as a mutation not included in the kit [12,13]. Taking all together, a statistical analysis was performed to evaluate the possibility of IRT values being able to predict the results of the mutation analysis. Kolmogorov–Smirnov test assured a normal distribution of the different populations. Outliers were removed and percentiles were calculated according to Grubbs' robustness test. Mean IRT values were compared depending on the detected mutations. For IRT values close to cut-off (60–70 ng/mL), differences were not statistically significant for the prediction of mutations, since 91% of patients did not show any genetic alteration (Table 4). IRT values higher than 140 ng/mL indicated a possibility of 90% for the patient to have 2 mutations and suffer from CF (p = 0.0001). Global incidence of CF was 1:6,059 for the 14 years assessed. If we only take into account results from the last change in protocol, and cut-off values 60–70 ng/mL since 2005, the accumulated incidence is 1:6,610 (CI 95%: 1:2879–1:7554). Significant differences were found across the Balearic archipelago, being the incidence in the smaller islands about 5 times higher than in Majorca (1:2,376 versus 1:10,613) (Fig. 1).
Table 3 Percentage common mutations (Celera kit 33, Abbott Diagnostics). Mutation
Screened CF patients (n = 131 alleles)
ΔF508 G542X R117H N1303K 2789+5GNA R1162X R553X L206W
58.8% 10.7% 6.9% 4.6% 3.0% 1.5% 1.5% 1.5%
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424 Table 4 Predictive value of the IRT levels (cut-off N60 μg/L) for finding mutations in the CFTR gene analysis. Number of mutations
Mean IRT level (ng/mL)
95th percentile
5th percentile
Two mutations (n = 24) One mutation (n = 84) No mutations (n = 1067)
147.31 86.66 85.70
287.52 139.88 133.90
66.88 b60.0 b60.0
Comparison of IRT mean concentrations
p-value
2 mutations group vs 1 mutation group 2 mutations group vs no mutation group 1 mutation group vs no mutation group
0.006 0.0001 0.96
423
and Castilla-Leon, 1:4800 in Aragon, 1:6244 in Catalonia and 1:6602 in the Balearic Islands [31,32]. Data from the Spanish Association of Newborn Screening (AECNE) approximate the incidence to 1:4807, although great variability among regions is reported. In accordance to this, incidence was 1:8,342 in Madrid in 2011, 1:10,432 in 2012 [33], and in the Basque Country it was 1:3000 [34]. In summary, the four changes in the screening strategy have resulted in a reduction in the total number of patients requiring the performance of a sweat test (from 1.07% to 0.10%), and helped maintaining a stable global incidence for CF during the last 10 years of 1:6,610 (IC 95%: 1:2879–1:7554).
References Discussion The benefits of CF screening and detection at birth are more than evident, both for the newborn themselves and for the families. In our case, we follow the IRT/DNA protocol, but to date, many different strategies for its detection have been described [14–16]. One of the main objectives is to reduce the rate of carrier detection, which has been widely regarded as an ethical problem in bibliography [14,17], and implies the performance of genetic counseling [18]. For us, this fact does not pose any shortcoming, since it helps reduce the incidence of this pathology. However, a reasonable period of time is needed to accurately evaluate such changes. The autosomal recessive inheritance of cystic fibrosis evidences a high number of carriers and affected people among populations with high inbreeding tendencies, due to geographic isolation. In small islands such as the Balearic, the geographic situation would be mostly responsible for the high incidence found in our study. In addition to this relative degree of endogamy, which has been extensively studied, but weak conclusions have been drawn to date, another explanation for the five-fold incidence of CF detected in smaller islands could result from a smaller sample size. Moreover, the influence of genetic variability, a characteristic of small islands, may result from the strong Anglo-Saxon and French domination in Minorca in the 18th century might explain the mutation pattern observed. In fact, demographic studies in the UK describe G542X (1.85%) as the most frequent mutation in CFTR after ΔF508 (74.1%) and G551D (3.37%). This stays in contrast with R117H, which is the third most frequent in Caucasian population (1.25%) [19–22]. Furthermore, population studies in the Mediterranean area and North Africa have found a higher allele frequency for G542X and N1303K in comparison to R117H, being the latter more frequent in Northwestern Europe [23,24]. These data are consistent with findings from a French study [17]. The most frequent mutation of CF worldwide is ΔF508, although great geographic dispersion is present, with a gradient from Northwest to Southeast (Denmark 88%, Italy 50% and Turkey 20%) [25,26]. In the Balearic Islands, ΔF508 is present in 58.8% cases, similar to Galicia (59.3%). In the South of Spain, incidence is 43.5%, and in the Basque Country, it is 87%. The second most common mutation is G542X, with an incidence of 10.7%. This value is similar to that found in Andalusia (11.4%), and significantly higher than the one in Asturias (5.0%), Castilla-Leon (3.8%) or Galicia (4.9%), according to the Cystic Fibrosis Newborn Screening Assessment, issued by the Spanish Ministry of Health in 2013. L206W, which is known to be the sixth most prevalent in Spain (1.6%), is reported to cause pancreatic insufficiency in 78% of the patients [27], and is more prevalent in Hispanic population (6.2%) and African Americans (1.2%) [28–30]. In Europe, the prevalence of CF varies between countries, from 1:1350 in Ireland to 1:25,000 in Finland, with a median birth prevalence of 1:3500 [25]. In Spain, thanks to the progressive implementation of newborn screening programs in the different provinces, a lower incidence is being detected. In 2009, CF incidence was 1:4439 in Galicia
[1] Paracchini V, Seia M, Raimondi S, Costantino L, Capasso P, Porcaro L, et al. Cystic fibrosis newborn screening: distribution of blood immunoreactive trypsinogen concentrations in hypertrypsinemic neonates. JIMD Rep 2012;4:17–23. [2] Kosorok MR, Wei WH, Farrell PM. The incidence of cystic fibrosis. Stat Med 1996; 15(5):449–62. [3] Sanders DB, Lai HJ, Rock MJ, Farrell PM. Comparing age of cystic fibrosis diagnosis and treatment initiation after newborn screening with two common strategies. J Cyst Fibros 2012;11(2):150–3. [4] Maclean JE, Solomon M, Corey M, Selvadurai H. Cystic fibrosis newborn screening does not delay the identification of cystic fibrosis in children with negative results. J Cyst Fibros 2011;10:333–7. [5] Lindau-Shepard BA, Pass KA. Newborn screening for cystic fibrosis by use of a multiplex immunoassay. Clin Chem 2010;56:445–50. [6] Sarles J, Berthézène P, Le Louarn C, Somma C, Perini JM, Catheline M, et al. Combining immunoreactive trypsinogen and pancreatitis-associated protein assays, a method of newborn screening for cystic fibrosis that avoids DNA analysis. J Pediatr 2005; 147(3):302–5. [7] Krulišová V, Balaščaková M, Skalická V, Piskáčková T, Holubová A, Paděrová J, et al. Prospective and parallel assessments of cystic fibrosis newborn screening protocols in the Czech Republic: IRT/DNA/IRT versus IRT/PAP and IRT/PAP/DNA. Eur J Pediatr 2012;171(8):1223–9. [8] Kirby LT, Applegarth DA, Davidson AG, Wong LT, Hardwick DF. Use of a dried blood spot in immunoreactive-trypsin assay for detection of cystic fibrosis in infants. Clin Chem 1981;27:678–80. [9] Gibson LE, Cooke RE. A test for concentration of electrolytes in sweat in cystic fibrosis of the pancreas utilizing pilocarpine by iontophoresis. Pediatrics 1959; 23(3):545–9. [10] van der Merwe DE, Ubbink JB, Delport R, Becker P, Dhatt GS, Vermaak WJ. Biological variation in sweat sodium chloride conductivity. Ann Clin Biochem 2002;39:39–43. [11] Rosenstein BJ, Cutting GR. The diagnosis of cystic fibrosis: a consensus statement. Cystic Fibrosis Foundation Consensus Panel. J Pediatr 1998;132(4):589–95. [12] Nelson MR, Adamski CR, Tluczek A. Clinical practices for intermediate sweat tests following abnormal cystic fibrosis newborn screens. J Cyst Fibros 2011;10:460–5. [13] Kirk JM. A continuing role for sweat testing in an era of newborn screening for cystic fibrosis. Clin Biochem 2011;44:487–8. [14] Southern KW. Determining the optimal newborn screening protocol for cystic fibrosis. Thorax 2012;67(4):281–2. [15] Vernooij-van Langen AM, Loeber JG, Elvers B, Triepels RH, Gille JJ, Van der Ploeg CP, et al. Novel strategies in newborn screening for cystic fibrosis: a prospective controlled study. Thorax 2012;67(4):289–95. [16] Hale JE, Parad RB, Dorkin HL, Gerstle R, Lapey A, O'Sullivan BP, et al. Cystic fibrosis newborn screening: using experience to optimize the screening algorithm. J Inherit Metab Dis 2010;33(Suppl. 2):S255–61. [17] Munck A, Dhondt JL, Sahler C, Roussey M. Implementation of the French nationwide cystic fibrosis newborn screening program. J Pediatr 2008;153(2):228–33. [18] Baker MW, Groose M, Hoffman G, Rock M, Levy H, Farrell PM. Optimal DNA tier for the IRT/DNA algorithm determined by CFTR mutation results over 14 years of newborn screening. J Cyst Fibros 2011;10(4):278–81. [19] McCormick J, Green MW, Mehta G, Culross F, Mehta A. Demographics of the UK cystic fibrosis population: implications for neonatal screening. Eur J Hum Genet 2002; 10:583–90. [20] Schwarz MJ, Malone GM, Haworth A, Cheadle JP, Meredith AL, Gardner A, et al. Cystic fibrosis mutation analysis: report from 22 UK regional genetics laboratories. Hum Mutat 1995;6:326–33. [21] Bobadilla JL, Macek Jr M, Fine JP, Farrell PM. Cystic fibrosis: a worldwide analysis of CFTR mutations—correlation with incidence data and application to screening. Hum Mutat 2002;10:575–606. [22] Juan-Fita MJ, Egea-Mellado JM, González-Gallego I, Moya-Quiles MR, FernándezSánchez A, et al. Expanded newborn screening in the Region of Murcia. Threeyear experience. Med Clin (Barc) 2012;13:566–71. [23] Estivill X, Bancells C, Ramos C. Geographic distribution and regional origin of 272 cystic fibrosis mutations in European populations. Hum Mutat 1997;10(2): 135–54. [24] Sobczyńska-Tomaszewska A, Ołtarzewski M, Czerska K, Wertheim-Tysarowska K, Sands D, Walkowiak J, et al. Newborn screening for cystic fibrosis: Polish' 4 years experience with CFTR sequencing strategy. Eur J Hum Genet 2012;21(4):391–6.
424
J.M. Bauça et al. / Clinical Biochemistry 48 (2015) 419–424
[25] Castellani C, Macek Jr M, Cassiman JJ, Duff A, Massie J, ten Kate LP, et al. Benchmark for cystic fibrosis carrier screening: a European consensus document. J Cyst Fibros 2010;9:165–78. [26] Southern KW, Munck A, Pollitt R, Travert G, Zanolla L, Dankert-Roelse J, et al. A survey of newborn screening for cystic fibrosis in Europe. J Cyst Fibros 2007;1:57–65. [27] The Clinical and Functional TRanslation of CFTR (CFTR2). available at http://cftr2.org. [Last accessed: January 2015]. [28] Sugarman EA, Rohlfs EM, Silverman LM, Allitto BA. CFTR mutation distribution among U.S. Hispanic and African American individuals: evaluation in cystic fibrosis patient and carrier screening populations. Genet Med 2004;6(5):392–9. [29] Rozen R, Ferreira-Rajabi L, Robb L, Colman N. L206W mutation of the cystic fibrosis gene, relatively frequent in French Canadians, is associated with atypical presentations of cystic fibrosis. Am J Med Genet 1995;57(3):437–9.
[30] Claustres M, Desgeorges M, Moine P, Morral N, Estivill X. CFTR haplotypic variability for normal and mutant genes in cystic fibrosis families from southern France. Hum Genet 1996;98(3):336–44. [31] Andalusian Attending Guidelines for Cystic Fibrosis. Plan de Atención a Personas Afectadas por Enfermedades Raras en Andalucía; May 2011. [32] Gartner S, Cobos N. Neonatal screening for cystic fibrosis. An Pediatr (Barc) 2009; 71(6):481–2. [33] Health Service in Madrid. Service for the prevention of illnesses. General Direction for Primary Care; 2012. [34] Newborn screening for cystic fibrosis in the Basque Country. Advising council for congenital errors of metabolism of the CAPV. Basque-Navarra Bull Pediatr 2010; 42:53–60.