Newborn screening program (NSP) is a public health approach aiming at diagnosing important health problems at an early stage with the goal of protecting societies from morbidity and mortality (World Health Organization takes Wilson and Jungner criteria into consideration for a disease to be screened; as SCID is suitable, the first pilot newborn screening was initiated in the state of Wisconsin in the USA in 2008 [8]. After participation of Massachusetts and California in 2009, screening efforts showed progress and in 2010 SCID was included in the national NSP in the USA [9]. Currently, TREC level is measured by real time PCR on DNA samples obtained from dried blood on Guthrie cards, this technique is available in screening programs in 50 states in the USA including Puerto Rico as well as 10 other countries like Israel, Iceland, Norway, Sweden, Germany, New Zealand, Switzerland, Lebanon and Taiwan. In 10 other countries including Turkey, pilot programs or regional screening efforts are reported to be in place [10]. The incidence of SCID reported to be 1 out of 63.500 live births in France [11] In the USA, screening program data was published for 23 states reporting the incidence of SCID as 1:58.000 [2]. As a result of national screenings in Israel and Norway, SCID incidences were the same and were reported as 1:22.500 [4, 12]. Saudi Arabia reported the incidence as 1:2.906 [5]. In this first prospective regional pilot study performed in Turkey from October 2018 to October 2020. SCID was screened with TREC method; in 20.253 babies born in Konya and Ankara provinces, SCID incidence was identified as 1:10.000. This incidence figure coincides with the previously reported retrospective SCID frequency of 1:10.000 identified in Konya [6]. The incidence of SCID in our country is higher than that of the USA, Europe and Israel and lower than that of Saudi Arabia. This is in correlation with the frequency of consanguineous marriages in the societies and the autosomal recessive inheritance of the mutations in 18 of the 20 genes responsible for SCID. In our country, the consanguinity rate for parents is reported around 23.2% [13] while it is 60% in Saudi Arabia. Consanguineous marriages are common in Israel as well. In Norway, the population is 5.4 million; we see common sociocultural determinants among people who have immigrated to this country from Turkey with high consanguinity rates of the couples seeming to be the most important factor responsible for high incidence of SCID.
In Turkey, the incidence of consanguineous marriages increases as we travel to Eastern and Southeastern parts of the country. Based on the Turkish Statistics Agency data in 2016, the incidence is 46.2% in these regions [13]. The SCID incidence of 1/10.000 identified in this regional pilot study does not reflect the reality of our country. There is no doubt that with a NSP to be carried out at a national scale, SCID incidence will be much higher. Although we are aware of this reality, we only screened the babies born in Konya and Ankara since we could carry out the detailed evaluations in order to identify and treat suspected SCID cases rapidly. After screening 20.253 newborns, two babies- one in Ankara and one in Konya- had TREC copy numbers of 0 in the first and duplicate samples. These babies could be reached during the second and third weeks of their lives through the Public Health General Directorate. The first baby was found as he was being treated at the neonatal intensive care unit at a private hospital in Konya with the diagnoses of pneumonia and sepsis. The second baby was identified at the hospital where she had been admitted with a skin infection. CBCs of both babies showed lymphopenia and hypogammaglobinemia, in their flow cytometric evaluations T - B- NK + phenotype was found correlating with SCID. Genetically, the first baby had ADA deficiency, and the second had a RAG1 defect. The mean age for the diagnosis of SCID is 5 months in Turkey [14], both babies were diagnosed much earlier during 4–6 weeks of life. They were followed up under isolation and prophylactic treatments. We aimed at providing the best treatment under the best circumstances without experiencing severe morbidities and tried to optimize the circumstances of the families. The first patient underwent GT at one year of age, the second patient received HSCT from a full matched twin sibling both having access to curative treatments. In our country, routine BCG vaccine is administered at two months, morbidities that could result from this vaccine could be avoided by establishing both of the diagnoses before this vaccination. Following successful curative treatments, both patients are being followed up; they are at six months together with a favorable immune reconstitution. The state of California has the largest SCID series to date 3.252.156 babies were screened from 2010 to 2017 and 50 were diagnosed with SCID. 49 babies could be followed up and 46 (94%) of these survived by means of allogeneic HSCT, GT or enzyme replacement [15].
Although TREC is mainly a screening test for the early diagnosis of SCID, prenatal steroid applications, instances causing secondary lymphopenia and conditions resulting in lymphocyte extravasations (hydrops, chylothorax) can lead to false positive results [15]. Furthermore, TREC screening can identify many Combined Immune Deficiencies (CID) and T cell development defects other than SCID characterized by low numbers of naive T cells. In a study by Kwan et al. screening results of more than 3 million babies were evaluated in the USA; 52 babies were diagnosed with SCID while 411 babies had conditions accompanied by T lymphopenia. 136 babies (33%) in this group had 22q1.1 deletion syndrome Trisomy 18, 21 had Ataxia-Telangiectasia or syndromic T cell defects like CHARGE; 117 babies (28%) had congenital heart disease, vascular leaks or gastrointestinal system losses and 12 cases had idiopathic T lymphopenia [2].
In our study, we identified 10 babies with low TREC copy numbers. Thus, total recall rate for examination by a pediatric immunologist and immune workup via flow cytometry was 0.049%. This rate was close to 0.04% reported in 72.411 samples by the French study DEPISTREC that totally 190,517 newborns screened prospectively [11]. Among our 10 babies, two newborns were identified to have 0 copies for TREC. As has been elaborated in detail above, both of them were diagnosed as SCID within 4–6 weeks before experiencing significant morbidities and they had access to curative treatments. In the literature, copy numbers of 0–3 TREC/µl are reported to have high positive determinant characteristics for T cell lymphopenia (for SCID or else) [16]. In two of the other eight babies, TREC copy numbers were identified as 35 and 36/DBS. Both of these babies were severely premature and had low BW; they had died during the early newborn period, one of them had phenylketonuria as well. Therefore, physical examination and flow cytometry analyses were not performed for these two babies. For the remaining six babies, TREC copy numbers changed from 37 to 45 copies/DBS. All six of them were invited to the project centers and all were examined. Their histories, physical examinations, CBC results and PBLS analyzed by flow cytometry were normal. After eliminating all possible primary or secondary causes, these babies were accepted to be healthy. In our study, we did not identify any babies with CID other than SCID, syndromic CID, or secondary or idiopathic lymphopenia. We think that his might be related to having a small and limited sample size.
TREC is a screening test, not a diagnostic test. The cases have to be characterized by confirmation tests. Certain physiological reasons, especially age and prematurity can have an effect on TREC copy numbers [15, 17]. In line with the information in the literature, 3.800 (19%) were preterm (GW ≤ 37 weeks); these babies had significantly low TREC copy numbers compared to term babies (p < 0.001). When preterm newborns were further categorized as late (32-≤37 weeks) (n: 3.718); mild (28-≤31 weeks) (n: 68) and severe preterm (≤ 27 weeks) (n: 14); corresponding median TREC copy numbers were 255, 200 and 193 respectively. As can be seen here, TREC copy numbers decreased as GW decreased. 9.8% of an increase has been defined for TREC per GW [18]. In this study, TREC copy numbers were significantly lower in mild preterm babies compared to late preterm ones (p < 0,001). Median TREC copy numbers of severe preterm newborns were lower than those of late and mild preterm ones despite not reaching statistical significance (p < 0.05). We think this can be explained by having only 14 severe preterm babies.
In our study, in a sample size of 20.223, the BW is 500 − 5.410 gr with a median BW of 3.250 gr. 18.206 (90%) newborns were normal, 1.180 (5.8%) had low birth and 837 (4.1%) were born as big babies. TREC copy numbers of newborns with low BW were lower than those with normal BW (p < 0.001). Babies with BWs of ≤ 1.500 gr and 1.501-2.500 gr had significant differences in TREC copy numbers (p < 0.001). When newborns BWs of ≤ 1.500 gr were compared with normal weight and big babies of ≥ 4001 gr, the difference between TREC copy counts was statistically significant respectively (p = 0.016) (p = 0.023). Mode of delivery (normal vaginal vs C-section) and singleton or multiple births did not have any effect on TREC copy numbers.
One of the positive control groups in our study was 100 healthy adult blood bank donors; in this group TREC copy numbers decreased as age increased, it decreased significantly at 53–57 years of age reaching 0 at 58–64 years and remained at 0 afterwards. Thymus glands shrinks together with increasing age and it goes through a physiological involution during late adolescent period. Functionally, thymopoiesis lost with age is a physiological process, too and in this limited number of healthy adults, we see a decrease in TREC copy numbers with increasing age that is independent of β-Actin.
Furthermore, qualitative and quantitative differences in blood samples and sampling, the method used and calibration also effects TREC measurements. In most of the studies found in the literature, the DNA was isolated from a blood sample obtained from a single disc of 3.2 mm. In our study, TREC was measured from DNA samples isolated from dried blood on Guthrie cards by having 3 discs of 3.2 mm with a puncher by using real time PCR based on our own method. From each Guthrie card, 3 discs of 3.2 mm that seemed to be completely covered with blood were collected and isolated DNA was used for measurements requiring repetitions. For each measurement, 5 microliter of a sample was used and a copy number cut-off value was identified as copies per DBS. Previously, retrospective TREC measurements were performed to identify a cut-off value; TREC numbers of > 46 copy/DBS were regarded as normal, ≤ 46 copy/DBS as low. In the literature, each disc is regarded to contain 3 microliters (µl) of blood as a routine [12]. From this point of view, TREC copy number of 46 copies/DBS defined as a cut-off would correspond to 15 copies/µl. In the initial studies that were conducted nearly 10 years ago with homemade methods, cut-off values in measured copy numbers were different. For example, in the first results reported from California, the values were 36 copies/DBS, as commercial kits were developed, the measurement time got shorter and the methodology and cut-off values decreased. In different country studies by using EnLite (Perkin Elmer) or kits, TREC cut-off values of 15 copies/µl or even 5 copies/ µl have been reported [4, 5, 17].
Cost calculations are reported as 4.25 dollars in the USA, 4.21 Euro in France; worldwide 4–5 dollars per test is reported in general [10, 11]. In our study, the cost of TREC measurement and SCID screening was calculated as 5.1 dollars/test which refers about to 45 TL (Turkish Lira). During the natural course of the disease, as the diagnosis of SCID is delayed, increases in the number and severity of infections and ensuing hospital and intensive care admissions result in significant increases in treatment costs. In a previously performed cost analysis in our country, for patients with a delayed diagnosis, a health expenditure of at least 1.2 million TL was reported [19]. NBSP for SCID has been implemented as a cost-effective program both economically sound and is a scientific need at the same time.
In conclusion, this study is the first prospective study in Turkey that screens SCID by measuring TREC copy numbers in DBS. By means of this study, methods to be used in NBSP for SCID have been developed; action plans for having access to patients, confirmation plans and treatments have been devised in the presence of suspicious samples and when duplicate measurements were needed. For early diagnosed SCID patients, genetic diagnosis and curative treatment were performed with success. We think that the SCID incidence of 1:10.000 identified in Ankara and Konya provinces with the help of this study is possibly much higher across our country. Compared to many countries in which SCID is included in NBSP, the incidence is much higher in Turkey as has been identified objectively. All the data and the results obtained from our study demonstrate the need to include SCID into our national NBSP expeditiously.