In the United States, newborn screening is an important state-based public health program that began over 40 years ago with the development of a screening test for phenylketonuria using newborn bloodspots dried onto a filter paper card [56, 57]. Many factors could influence a decision to include a given condition in a newborn screening program, such as the severity of the condition, the availability of effective treatment, the age of onset, and the complexity, availability or cost of the test . Fragile X screening has captured increasing attention lately for both potential benefits and concerns that affect the development of a screening program. Fragile X screening was not recommended for newborn screening in the American College of Medical Genetics report of 2006  primarily because of the lack of an accurate screening test and the absence of data on benefits at that time. In the past few years the advent of clinical trials of targeted treatments for FXS and indications of positive outcomes in early phase studies [60–64] have been exciting developments that promote the need for newborn screening for FXS. Some of the targeted treatments and additional interventions are being studied in children in the toddler period and these interventions will likely enhance the developmental/behavioral interventions for young children . In addition, the development of a new PCR-based screening approach utilized here has further stimulated the discussion around newborn screening in fragile X.
Accurate estimates of frequency of FMR1 mutations in the general population are needed to better estimate fragile X allele frequencies for all racial and ethnic groups and to determine the ramifications of any population screening program in terms of numbers of identified cases. The increasing number of disorders attributed to the premutation has also encouraged better epidemiology data. Indeed, great interest has been focused on premutation carrier detection, since premutation alleles have been found to be associated with FXPOI [13, 14, 66] and FXTAS [67–69] and sometimes with neurodevelopmental disorders, such as ASDs and ADHD [5, 9, 70], which can respond to treatments .
Here, we report allele frequency distributions found in a pilot newborn screening study from three sites in the US, using a novel PCR-based approach to demonstrate the feasibility of screening for FMR1 mutations in a large sample size and with samples collected on blood spot cards. This is the largest newborn sample size screened in the US for both males and females and for the detection of expanded alleles throughout the normal to full mutation range. We found that the most common alleles were those containing 29 and 30 CGG repeats, regardless of ethnicity, in agreement with previous reports. The screening identified 170 newborns carrying a gray zone allele (45 to 54 CGG repeats) with a prevalence of 1:66 in females and 1:112 in males. Some studies [52, 53] have advocated for expanding the gray zone to 40 to 54 CGG repeats because there is an elevation in the FMR1 mRNA expression levels in this range and there may be evidence of risk of clinical involvement, including an increased rate of primary ovarian insufficiency (POI) compared to the general population [18, 19]. In addition, an increased prevalence of gray zone alleles has also been recently reported in subjects with parkinsonism [52, 72] and several cases of FXTAS have been reported in gray zone [20, 73]. Thus, we also report the prevalence in this expanded gray zone range as 1:32 in males and 1:18 in females based on the total number of newborns screened. Our findings regarding the prevalence of the premutation alleles (1:209 in females and 1:430 in males) are within the range of what was previously reported in females , but in males we observed a prevalence almost two-fold higher than that in the Canadian study (1:813) , lower than in the Spanish population  but in line with a recent population-based screening study of older adults in Wisconsin, US (1:468 in males) . It is interesting to note that from our study the female to male prevalence rate for the premutation is 2.05, in agreement with the predicted ratio described by Hagerman . Although the size of the premutation alleles varied between 55 and 130 CGG repeats in females and between 56 and 125 CGG repeats in males, it is interesting to note that 70% of the premutation alleles contained <70 CGG repeats, in agreement with a recent report . This may be of relevance for estimating the frequency of FMR1 related disorders in the general population since individuals with >70 repeats are more likely to have premutation disorders . If we consider that the prevalence of a premutation allele in males is approximately 1:400 and if FXTAS is affecting approximately 40% of the premutation male carriers, then we would expect that 1.6 males out of 2,000 in the general population would develop the neurodegenerative syndrome. As was described in a recent study , FXTAS is far less likely in patients with <70 repeats. Thus, despite rare reports of FXTAS in the gray zone  and in the low end of the premutation range, it is likely the frequency of FXTAS in the general population is lower than 1.6/2,000. However, mild neurological problems, such as neuropathy or balance problems associated with the premutation, are likely to be close to this prevalence and more common than in those with a definitive diagnosis of FXTAS.
Only one male newborn, out of the total 7,312 males screened, was found to have a full mutation at the UCDMC site. A large screening of newborns (n = 36,154) reported a prevalence of 1:5,161 in males ; however, our sample size is too small to be confident of a prevalence estimate for the full mutation. Indeed, one would need in excess of 70,000 samples to estimate a prevalence of 1:5,000 and 95% CI within a 50% margin of error.
Although the CGG size distribution did not show a difference between the two genders and among different ethnic groups, differences were detected in the prevalence of expanded alleles. Specifically, the prevalence of gray zone alleles was higher in White males compared to Black and Hispanic males. Differences in the prevalence between the different ethnic groups were also observed for the premutation alleles; however, they did not reach statistical significance likely due to the small number. It is important to consider the potential difference in prevalence of premutation alleles in different populations as this could explain both the differences in premutation prevalence and the incidence of FXS among different studies.
Identifying and reporting babies with a premutation is somewhat controversial, with important arguments on both sides of the equation. One argument in favor of disclosure is the potential benefit for extended family members, in terms of genetic and reproductive counseling. Some of these family members may be suffering from clinical problems related to the premutation or full mutation segregating in the family, and can benefit from knowledge of their condition to help direct treatment . Identification of babies with the premutation can also lead to early intervention or treatment when needed with appropriate follow-up . Although premutation babies are far less likely to show developmental problems than full mutation babies, some are at risk for learning problems, ASD, or seizures, and early intervention will be important to implement if developmental problems emerge in follow-up [5, 9, 70, 71].
On the negative side of identifying FMR1 premutation carriers at the time of birth is that the family is told of possible future problems related to the premutation that may or may not develop, including FXTAS, and this may cause excessive worries for the family, especially since the certainty of problems will be unknown. Many families may not want to know about carrier status, and a robust consent process is needed to assure that families understand the kind of information that could be learned from FX screening. The high rate of carrier detection makes clear the burden that screening would place on genetic counseling.
The identification of a newborn with the premutation or the full mutation can create the need for cascade testing throughout the family. Some family members will be interested in knowing if they are carriers, especially if they have medical problems that may relate to premutation involvement. These types of problems include depression, anxiety [12, 78, 79], autoimmune problems, such as fibromyalgia or hypothyroidism [8, 11], hypertension , sleep apnea , neuropathy, FXPOI and FXTAS. In our study, the largest family so far identified through cascade testing after the newborn was identified as a carrier had 16 additional carriers identified, including a great grandmother with probable FXTAS , several great aunts with neurological problems, others with emotional difficulties and female carriers with significant needs for reproductive counseling. Although it is unclear whether all of these problems are a direct result of the premutation alone, it is clear that there is a need to test extended family members in relation to premutation and full mutation disorders. However, the time and energy of the counseling and health care professionals for cascade testing of identified families may be a limiting factor on how many individuals in one family tree can be identified.