*Article* **Long-Term Outcomes of Adult Patients with Homocystinuria before and after Newborn Screening**

#### **Kenji Yamada 1,\*, Kazunori Yokoyama 2, Kikumaro Aoki 2, Takeshi Taketani <sup>1</sup> and Seiji Yamaguchi <sup>1</sup>**


Received: 1 July 2020; Accepted: 29 July 2020; Published: 30 July 2020

**Abstract:** Background: Homocystinuria (HCU) is a rare inherited metabolic disease. In Japan, newborn screening (NBS) for HCU (cystathionine β-synthase deficiency) was initiated in 1977. We compared the outcomes between patients detected by NBS (NBS group) and clinically detected patients (non-NBS group). Methods: We administered questionnaires about clinical symptoms and social conditions to 16 attending physicians of 19 adult HCU patients treated with methionine-free formula. Results: Eighteen patients (nine patients each in the NBS and non-NBS groups) participated. The frequency of patients with ocular, vascular, central nervous system, and skeletal symptoms in the NBS group was lower than that in the non-NBS group. Intellectual disability was observed in one and eight patients in the NBS and non-NBS groups, respectively. Concerning their social conditions, all patients in the NBS group were employed or still attending school, while only two patients in the non-NBS group were employed. Three of the four patients who discontinued treatment presented some symptoms, even in the NBS group. Conclusion: The social and intellectual outcomes of adult Japanese patients with HCU detected by NBS were favorable. However, even in the patients in the NBS group, some symptoms might not be preventable without continuous treatment.

**Keywords:** homocystinuria; cystathionine β-synthase deficiency; newborn screening; long-term outcome; social outcome; vitamin B6; methionine

#### **1. Introduction**

Homocystinuria (HCU) is a rare inherited metabolic disease characterized by the accumulation of homocysteine (Hcy) and its metabolites in the blood and urine [1,2]. HCU is classically categorized into three types depending on the specific enzymes involved in the metabolism of sulfur-containing amino acids that are deficient. The three types are as follows: (1) cystathionine β-synthase (CBS) deficiency (OMIM, 236200); (2) defect in cobalamin metabolism; (3) methylenetetrahydrofolate reductase deficiency. In CBS deficiency, the conversion of Hcy to cystathionine is impaired, and CBS deficiency is known as classic homocystinuria or homocystinuria type I [3]. The major clinical manifestations of CBS deficiency include the dislocation of the optic lenses, osteoporosis, "marfanoid" habitus, learning difficulties, and a predisposition to thromboembolism [4]. Because CBS deficiency is clinically heterogeneous and exhibits a wide range of outcomes, some patients have severe clinical phenotypes from childhood, while other patients may be asymptomatic until adulthood. Furthermore, CBS deficiency is classified into two phenotypes depending on vitamin B6 responsiveness. The B6-responsive type generally results in a milder phenotype [5].

Newborn screening (NBS) for HCU is performed in several countries and regions, including Western Europe, Australia, the United States of America, and Japan [6]. The prevalence varies widely by ethnicity and has been previously reported to range from 1:1800 to 1:1,000,000, with an overall estimated prevalence of 1:344,000 [7–11]. However, the true frequency is unknown and is thought to be higher than the prevalence detected by both NBS and clinical identification [9,12].

In Japan, NBS for HCU was initiated in 1977 and involves measuring methionine (Met) levels as the diagnostic marker. This strategy detects only CBS deficiency. The detection incidence of HCU (CBS deficiency) in Japan has been reported to be 1:800,000 to 1,000,000 births [13]. Early detection by NBS enables therapeutic intervention from early infancy and the maintenance of lower levels of blood Hcy, which may prevent the development of complications and consequently improve clinical outcomes in terms of both mortality and morbidity [14,15]. Although we reported the outcomes of inborn errors of metabolism in Japanese patients, including HCU detected by NBS, fifteen years ago [16], to date, the long-term outcomes of HCU are unknown. Here, we investigated the long-term outcomes in adult HCU patients, including clinically detected cases.

#### **2. Materials and Methods**

We sent questionnaires to 16 attending physicians of 19 adult patients with HCU who were older than 20 years as of October 2017 and who continued to be treated with a Met-free amino acid formula supplied by the Secretariat of Special Formula, Aiiku Maternal and Child Health Center. Patients with HCU types 2 and 3 were excluded from this study.

The questionnaires included questions regarding age, sex, clinical form, physical growth, prior NBS, onset age, first symptoms, metabolic data (i.e., blood methionine and Hcy levels), enzyme activity, genotype, treatments, symptoms (i.e., ocular involvement, vascular symptoms, central nervous system (CNS) disorders, and skeletal malformations), degree of intellectual disability, intermittent use of treatments, educational status, working status, marital status, and additional details. In this study, "marfanoid" included only skeletal symptoms, such as excessive height and/or arachnodactyly.

The subjects were divided into two groups, based on whether they underwent NBS as follows: (1) the NBS group was defined as those who underwent NBS after the initiation of NBS in Japan in 1977; (2) the non-NBS group included those who were born before 1977 or who were born in or after 1977 but did not undergo NBS. The results of the non-NBS group were compared with those of the NBS group; no statistical analysis was performed because of the small sample size in this study.

This study was approved by the Institutional Review Board of Shimane University in 28 September 2017 (#20170726-3).

#### **3. Results**

We received answers from 18 of 19 adult patients with HCU (response rate of approximately 95%). Ten males and eight females were included. Although 10 patients underwent NBS, one of these patients failed to be diagnosed with HCU based on NBS, and the diagnosis was made after the onset of symptoms. She was included in the non-NBS group. Eventually, the numbers of patients in the NBS and non-NBS groups were the same (9 per group).

#### *3.1. Profiles of the Patients: Biochemical Findings and Treatments*

The patient profiles are summarized in Table 1. The male/female ratio was 7/2 in the NBS group and 3/6 in the non-NBS group. The median ages were 25.8 years (21.3–36.7) and 44.3 years (32.2–59.2) in the NBS and non-NBS groups, respectively. Regarding vitamin B6 responsiveness, six and five patients in the NBS and non-NBS groups were non-B6 responsive, respectively. Elevated Met was found in a diagnostic test in 12 patients (NBS, eight patients; non-NBS, four patients) with available data. The median Met levels were 1264 μM (456 to 2433) and 565 μM (366 to 3903) in the NBS and non-NBS groups, respectively. Among the nine patients (NBS, five patients; non-NBS, four patients) with available data, the median blood Hcy levels were 71.7 μM (9.1 to 286) and 63.6 μM (22.2 to 292) in the NBS and non-NBS groups, respectively.


**Table 1.** Overview of the participating Japanese patients with HCU.

Met, methionine; Hcy, homocysteine; CBS, cystathionine β-synthase.

In three of the seven patients who underwent genetic testing, five types of variants—namely, p.H65R, p.G116R, p.G259S, p.M382R, and p.F531Gfs\*9—in *CBS*, were identified. Only p.G116R was found in two patients. p.M382R was a novel mutation judged as "causing the disease" based on the Mutation Taster (http://www.mutationtaster.org/) results.

In the NBS group, all nine patients were treated with betaine and dietary therapy with Met-free formula, seven patients followed a protein-restricted diet, and six patients underwent antiplatelet therapy with aspirin. Additionally, three patients were treated with vitamin supplementation or anticoagulant therapy as other treatments. Meanwhile, in the non-NBS group, all nine patients received dietary therapy using a Met-free formula, and four patients followed a protein-restricted diet. Betaine and aspirin were administered to three patients. In the non-NBS group, one patient was treated for hypertension and hyperlipidemia, and another patient received antiplatelet therapy other than aspirin.

In total, four and two patients temporarily discontinued the treatment in the NBS and non-NBS groups, respectively. The reasons included economic problems, insufficient instructions by the physicians, and self-judgment mainly due to fewer subjective symptoms experienced during adulthood, difficulty in visiting distant hospitals, decreased motivation, the need for excessively strict control, and/or the cost of betaine. The discontinuation periods ranged from a few years to approximately ten years.

#### *3.2. Clinical Symptoms*

The major clinical symptoms in this survey are shown in Table 2. The NBS group had fewer complications than the non-NBS group. In particular, four of the nine patients in the NBS group were asymptomatic, and all patients were in their 20s.


**Table 2.** Comparison of the major clinical symptoms in adult patients.

#: marfanoid involves excessive height and/or arachnodactyly. \*: other symptoms included diabetes, arteriosclerosis obliterans, hyperlipidemia, and pneumonia in one patient each. The cumulative total number of patients was counted.

Concerning optic involvement, retinal detachment was noted in one of the nine patients in the NBS group, who was aged 35 years, while among the nine patients in the non-NBS group, ectopia lentis, myopia, and glaucoma were observed in five, one and two patients, respectively.

Regarding vascular system complications, one patient in the NBS group had a thromboembolism in the pulmonary vessels at the age of 26 years, and another patient had cerebrovascular thromboembolic events at the age of 31 years. These two patients had a history of treatment interruption. In the non-NBS group, four patients developed thromboembolism in the period between the age of 6 years and their fourth decade of life, and one of them experienced complications of both cerebrovascular and pulmonary vascular obstructions.

Regarding CNS symptoms, psychiatric disability was noted in two of nine patients in the NBS group. One of these two patients with psychiatric disability also had intellectual disability. In contrast, all nine patients in the non-NBS group had some type of CNS symptoms, such as intellectual disability, epilepsy, psychiatric disability, and/or dystonia. Intellectual disability, including that of a mild degree, was observed in seven patients in the non-NBS group.

Regarding complications of the skeletal system, marfanoid was observed in three patients in the NBS group in their 30s. One patient also had osteoporosis. Meanwhile, all patients in the non-NBS group, except for one patient, presented some skeletal symptoms. Marfanoid, osteoporosis, scoliosis, and pectus excavatum were noted in five, two, five, and two patients, respectively.

Other complications, such as diabetes, arteriosclerosis obliterans, hyperlipidemia, and pneumonia, were noted in one patient each in the non-NBS group.

#### *3.3. Life Outcomes: Intelligence, Education, Employment, and Other Outcomes*

The outcomes in this survey are shown in Table 3. Physical development was within the normal range in six and three patients in the NBS and non-NBS groups, respectively. In total, two and four patients had excessive height in the NBS and non-NBS groups, respectively. Other physical findings, such as short stature and obesity, were noted in one and two patients in the NBS and non-NBS groups, respectively.



\* Other includes obesity or a short stature.

Regarding intelligence, eight patients showed normal intelligence, and one patient had mild intellectual disability among the nine patients in the NBS group. In contrast, only one of the nine patients in the non-NBS group demonstrated normal development, while three patients showed borderline, and one each had moderate and severe intellectual disability. An unknown degree of intellectual disability (may be moderate or severe) was noted in the other three patients in the non-NBS group.

Regarding education status, the final educational attainment was university, vocational school, high school, and school for the handicapped in two, four, one, and one patients, respectively, in the NBS group. In contrast, in the non-NBS group, the numbers of patients were two, zero, two, and three, respectively.

Regarding other outcomes, including employment and marital status, eight of nine patients in the NBS group were employed, and the other patient was still attending school. In the non-NBS group, two patients were employed, and three patients each were unemployed or living in a facility for the handicapped. Two patients in the NBS group were married, while no patient was married in the non-NBS group. Although all nine patients in the NBS group were alive, three patients in the non-NBS group had already died. The causes of death in these three patients were thalamic

hemorrhage, pneumonia, and unknown. No further information regarding the three deceased patients was available.

In the other free description, all attending physicians stated the need for life-long treatment. Additionally, some physicians described the need to support medical expenses, including the treatment cost of betaine, and the necessity for consultation with internal medicine and psychiatry specialists.

#### **4. Discussion**

Our study revealed that NBS substantially contributed to the improvement in the long-term outcomes of Japanese patients with HCU but that the symptoms might progress even in patients detected by NBS. All patients in the NBS group worked or attended school, and all had normal mental development, except for one patient. NBS for HCU was previously reported to be an effective and recommended program [17]. The early start of treatment to maintain low levels of Hcy improved the long-term outcomes [18], and our study supports the finding that outcomes in Japanese patients are similar to those in previous reports.

Our results also indicate that lifelong continuous treatment is important to the achievement of improved long-term outcomes. Three of the four patients who discontinued their treatments presented some symptoms, even in the NBS group, while three of the five patients continuing treatments were asymptomatic. However, because information regarding the Hcy levels during the treatment period, including the discontinuation period, was not available in our study, the relationship between the outcomes and treatment interruption could not be fully elucidated.

Additionally, our results suggest that, even if HCU is detected by NBS and is continuously treated, the condition is likely to progress in some cases. In our study, the condition of the patients in their 30s seemed to be more severe than that of the patients in their 20s, even within the same NBS group. This finding might indicate that the management was improved or that their symptoms progressed. Our results also explored the responsiveness to treatments. Because marfanoid and psychiatric disability were observed in some patients in the NBS group, these symptoms might not be completely prevented even by early detection and intervention. However, scoliosis and pectus excavatum might be responsive to early intervention because these symptoms were present only in the non-NBS group. Ectopia lentis and intellectual disability also seemed to be responsive to early treatment.

Regarding vitamin B6 responsiveness, two of 11 patients could be considered B6-responders in our study, which is similar to a previous report revealing that 15% of Japanese HCU patients were vitamin B6 responders (based on a report in a domestic Japanese journal). The Japanese prevalence of B6 responders was higher than that in Ireland, where 1 in 25 patients was a B6 responders [18], while a report showed that 231 of 629 patients (36.7%) were vitamin B6 responders [19]. However, because not all Japanese patients, such as pyridoxine-responsive patients treated without Met-free formula, were included in our study, our results for vitamin B6 responsiveness might not correctly reflect the true prevalence in Japan.

Concerning the genetic background, it has been suggested that p.G116R might be common among Japanese patients, although genetic testing information was available for only three patients. In the European population, it has been reported that the p.I278T mutation is common and is associated with B6 responsiveness [9]. Additionally, the p.T191M, p.G307S, and p.R336C mutations are relatively common in some populations [8,20,21]. However, these mutations were not observed in the Japanese patients in our study.

Although 1 in 10 patients born between 1977 and 1997 was a false-negative case in our study, our results could not provide information about the sensitivity and specificity of screening tests using Met levels because of the small sample size. It is well known that the sensitivity and specificity of screening tests using Met levels alone are insufficient [22]. The selection of appropriate markers and setting of accurate cut-off values are future challenges for the NBS for HCU in the Japanese population.

There may be some limitations of our study. For example, patients treated with Met-free formula who could be traced by our institution were recruited for our study. Patients treated with only betaine and vitamin B6, patients who died before adulthood, and patients who interrupted treatment with Met-free formula during this survey, were not included. Information regarding the three dead patients was insufficient. Because approximately 30,000,000 babies were screened between 1977 and 1997 in Japan (1,200,000 to 1,800,000 births per year), there should, theoretically, be approximately 40 adult patients with HCU, but only 10 patients could be enrolled in our study. Therefore, our results do not correctly reflect the long-term outcomes of all Japanese patients with HCU. Furthermore, because HCU is a progressive disease, the differences in outcomes between the two groups may be associated with, not only NBS, but also age. We did not collect information to estimate the severity of disease except for symptoms; therefore, we could not compare disease severity between the two groups. Nevertheless, we believe that the severity is similar between the two groups because the Met and/or Hcy levels in the NBS group at diagnosis were similar to those in the non-NBS group. Therefore, we could not definitively conclude that the positive outcomes were all due to NBS. However, because there are few reports in which the long-term outcomes of patients with HCU detected by NBS were compared with those of clinically detected patients on the same scale and because the long-term outcomes of Japanese patients are unknown, our results are important for the investigation of the effect of NBS on HCU.

#### **5. Conclusions**

The long-term, particularly social and intellectual, outcomes of Japanese adult patients with HCU detected by NBS were favorable compared with those of patients with clinically detected HCU. However, even in the patients in the NBS group, some symptoms might not be preventable, and long-term outcomes may worsen if treatment is interrupted.

**Author Contributions:** Conceptualization, K.Y. (Kenji Yamada), K.A. and S.Y.; Methodology, K.Y. (Kenji Yamada) and S.Y.; Software, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Validation, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Formal Analysis, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Investigation, K.Y. (Kenji Yamada) and S.Y.; Data Curation, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Writing—Original Draft Preparation, K.Y. (Kenji Yamada); Writing—Review and Editing, K.Y. (Kazunori Yokoyama), K.A., T.T. and S.Y.; Visualization, K.Y. (Kenji Yamada); Supervision, K.A., T.T. and S.Y.; Project Administration, K.A., T.T. and S.Y.; Funding Acquisition, K.Y. (Kenji Yamada). All authors have read and agreed to the published version of the manuscript.

**Funding:** This report was partially supported by AMED (grant number JP20ek0109482) and JSPS KAKENHI (grant number 19K08300). The authors confirm independence from the sponsors; the content of the article was not influenced by the sponsors.

**Acknowledgments:** We thank O. Sakamoto at Tohoku University, M. Ishige and E. Ogawa at Nihon University, K. Shimura at Tokyo Metropolitan Children's Medical Center, A. Tsuchiya at Tsuchiya Clinic, S. Soneda at St. Marianna University School of Medicine, N. Shimozawa at Gifu University, Y. Maruo at Shiga University of Medical Science Hospital, A. Nishiyama at Kobe University, E. Naito at Tokushima University, T. Miyake at Ehime University, Y. Watanabe at Kurume University, Y. Indo at Kumamoto University, K. Shiomi at Miyazaki University, M. Gotanda at Gotanda Medical Clinic, Y. Maruyama at Imakiire General Hospital, and T. Inoue at Fukiage Clinic for providing patient information and cooperating during the study.

**Conflicts of Interest:** The authors declare that there are no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Long-Term Neurological Outcomes of Adult Patients with Phenylketonuria before and after Newborn Screening in Japan**

**Kenji Yamada 1,\*, Seiji Yamaguchi 1, Kazunori Yokoyama 2, Kikumaro Aoki <sup>2</sup> and Takeshi Taketani <sup>1</sup>**


**Abstract:** Japanese newborn screening (NBS) for phenylketonuria (PKU) was initiated in 1977. We surveyed the neurological outcomes of Japanese adult patients with PKU to investigate the long-term effects and of and issues with NBS. Eighty-five patients with PKU aged over 19 years who continued to be treated with a phenylalanine-free amino acid formula were investigated by administering questionnaires regarding clinical characteristics, such as mental ability, education status, and therapeutic condition. Of the 85 subjects, 68 patients were detected by NBS (NBS group), while the other 17 were clinically diagnosed before the initiation of NBS (pre-NBS group). Further, 10 of the 68 NBS patients presented intellectual and/or psychiatric disabilities, 5 of whom had a history of treatment discontinuation; in contrast, 12 of the 17 pre-NBS patients presented with neuropsychiatric symptoms. Regarding social outcomes, almost all patients in the NBS group could live an independent life, while over half of the patients in the pre-NBS group were not employed or lived in nursing-care facilities. Neurological outcomes are obviously improved by NBS in Japan. However, some patients, even those detected by NBS, developed neuropsychiatric symptoms due to treatment disruption. Lifelong and strict management is essential to maintain good neurological and social prognoses for patients with PKU.

**Keywords:** phenylketonuria; newborn screening; long-term outcome; adult patients; Japanese; intellectual disability; psychiatric disability; treatment discontinuation

#### **1. Introduction**

Phenylketonuria (PKU, OMIM No. 261600), an autosomal recessive inherited disease of amino acid metabolism, is a major disease identified by newborn screening (NBS) [1]. Classic PKU is caused by phenylalanine hydroxylase (PAH, OMIM No. 612349) deficiency and is associated with severe intellectual disability, convulsions, hypopigmentation, behavioral abnormalities, and dementia. Because PKU is detected as an elevation in phenylalanine (Phe) levels in NBS, mild hyperphenylalaninemia (HPA), tetrahydrobiopterin (BH4) deficiency, including BH4 metabolic defects, and BH4-responsive PKU and classic PKU are all detectable. Mild HPA is also caused by a defect in PAH but is clinically mild and may require less strict dietary restrictions. BH4-responsive PKU and BH4 deficiency show BH4 responsiveness. The global prevalence of PKU has been reported to be approximately 1 in 4000 to 15,000 births [1–3], but the incidence of NBS in Japan has been reported to be 1 in 50,000 to 70,000 births [4,5].

NBS for PKU first spread from the mid to late 1960s in North America and the United Kingdom, while nationwide NBS for PKU was initiated in 1977 in Japan. Almost all cases of PKU are detected by NBS and treated immediately after birth. The treatment is mainly a diet with protein restriction and Phe-free formula and supplementation with BH4 in some cases. With this therapy, the life and neurological prognoses of patients with PKU

**Citation:** Yamada, K.; Yamaguchi, S.; Yokoyama, K.; Aoki, K.; Taketani, T. Long-Term Neurological Outcomes of Adult Patients with Phenylketonuria before and after Newborn Screening in Japan. *Int. J. Neonatal Screen.* **2021**, *7*, 21. https://doi.org/10.3390/ ijns7020021

Academic Editor: Toshihiro Tajima

Received: 3 March 2021 Accepted: 12 April 2021 Published: 14 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

have been reported to improve, at least in childhood, resulting in a significantly decreased societal economic burden [6,7]. Therefore, PKU is considered one of the best targets for NBS in terms of a cost–benefit analysis.

However, even in patients who are initially treated and achieve good metabolic control, reversible (and sometimes irreversible) neuropsychiatric symptoms can develop if this control is lost in later childhood or adulthood [8]. In addition, over time, some patients may manifest subtle intellectual and neuropsychiatric issues even with strict adherence to a low-Phe diet [9–12]. In fact, the long-term outcome of PKU is not necessarily favorable.

In this study, we expanded upon our previous studies [5,13,14] and investigated the social and neurological outcomes of adult patients aged over 20 years who were diagnosed with PKU before and after NBS in Japan using a questionnaire administered to attending doctors.

#### **2. Materials and Methods**

We sent questionnaires to the 33 attending physicians of 85 adult patients with PKU aged over 20 years as of October 2016 (born before 1996) who had continuously required a special formula (including Phe-free comprehensive amino acid powder and low-Phe peptide powder) supplied by the Secretariat of Special Formula, Aiiku Maternal and Child Health Center for dietary therapy.

Attending physicians answered the questionnaires based on clinical records. The questionnaire included questions on sex, age, clinical form of PKU, history of treatment interruption and the reason, follow-up department, physical development, neuropsychiatric disability, educational status, work status, and marital status and allowed for an open-ended description of the patient.

Regarding the question on neuropsychiatric disability, attending physicians were able to answer "normal intelligence", "borderline intellectual disability", "intellectual disability", and/or "psychiatric disability". When "intellectual disability" was selected, patients with an IQ (intelligence quotient) less than 50 or patients living in nursing-care facilities were characterized as having a "moderate-severe intellectual disability", while those with an IQ of 50–70 or who are able to fend for themselves and patients with borderline intellectual disability were characterized as having a "borderline-mild intellectual disability". Additionally, the details of psychiatric disability were judged based on the open-ended description. Regarding physical development, short stature was defined as a height <160 cm for males and <148 cm for females. Obesity and leanness were defined as BMI >25 and <18.5 kg/m2, respectively.

The subjects were divided into the following two groups: (1) the NBS group was defined as patients who underwent NBS, and (2) the pre-NBS group consisted of patients who were born before 1977 (when NBS was initiated) or were born after 1977 but did not receive NBS. Each of the groups was subdivided into age groups of approximately 5 years. The results of the NBS and pre-NBS groups were compared without a statistical analysis because of the small-scale nature of this study.

This study was approved by the Institutional Review Board of Shimane University on 13 October 2016 (#20160915-2).

#### **3. Results**

We received answers from the attending physicians of all 85 adult patients with PKU (response rate of 100%). All participants were alive. Because only one patient had not undergone NBS despite being born in 1979, she was enrolled in the pre-NBS group (and classified into the 39–44-year-old subgroup). Eventually, 68 and 17 patients were placed into the NBS and pre-NBS groups, respectively. As shown in Table 1, the NBS group consisted of 34 males and 33 females, while 7 males and 10 females constituted the pre-NBS group. The median age (range) of the NBS and pre-NBS groups was 28.5 years (20.5 to 38.2) and 43.9 years (37.7 to 50.8), respectively. Regarding the clinical form of PKU, almost all patients had classic PKU, but six patients with mild HPA and two patients with BH4-

responsive PKU were included in the NBS group. In this study, almost all participants (66 of the 68 patients in the NBS group and 16 of the 17 patients in the pre-NBS group) were managed by pediatricians even after reaching adulthood.


**Table 1.** Overview of the participating patients with phenylketonuria (PKU).

HPA, hyperphenylalaninemia; BH4, tetrahydrobiopterin.

Approximately 30% of patients (21 of the 68 patients in the NBS group and 5 of the 17 patients in the pre-NBS group) had ever discontinued treatment. Although the description of treatment discontinuation was unequal due to the open-ended questionnaire, concrete information was obtained for only eight patients in the NBS group. Among them, the median age at treatment discontinuation was 18 years (range, 11 to 20 years), and the median duration was 11 years (range, a few years to 23 years). In the pre-NBS group, the details of treatment discontinuation were provided by one patient, and she was a 42-year-old female who had several intermittent histories of treatment discontinuation for a total of a few decades beginning at 3 years of age. As shown in Table 2, the most common reasons for discontinuation before and after NBS were financial problems and selfjudgment, including a decrease in motivation and busy school or work schedules, followed by the unpleasant taste of the special formula and erroneous recommendations from the attending physicians. On the other hand, the most common reason for restarting dietary therapy is the prevention of maternal PKU, followed by improvement of the medical subsidy system and the appearance of psychiatric or behavioral abnormalities.

**Table 2.** Reasons for discontinuing or restarting dietary treatment.


Figure 1 shows the comparison of intellectual outcomes before and after NBS in different age groups. Normal intelligence was observed in 60 of the 68 patients (88%) in the NBS group; specifically, all but two patients less than 35 years old had normal intelligence. Meanwhile, even in the NBS group, 6 of 13 patients over 35 years of age exhibited a certain degree of intellectual disability, including 5 patients with borderline intellectual disability. In contrast to the results of the NBS group, only 6 of the 17 patients (35%) in the pre-NBS group showed normal intelligence. Additionally, the degree of intellectual disability in the pre-NBS group was more severe than that in the NBS group.

**Figure 1.** Degree of intellectual disability in patients stratified by age before and after newborn screening.\*, A 37-year-old female was included in the "39–44-year-old subgroup" of the pre-newborn screening (NBS) group because she did not undergo NBS.

Table 3 shows a comparison of other clinical symptoms and social outcomes. Regarding the psychiatric status, transient psychiatric disabilities were observed in three patients in the NBS group, all of whom had histories of treatment interruption, during their intermittent treatment. Another patient presented with a psychiatric disability. Four patients presented with an abnormal psychiatric status in the NBS group (6%). On the other hand, 6 of 17 patients (35%) in the pre-NBS group had psychiatric disabilities. When the number of patients with intellectual and/or psychiatric disabilities was counted collectively as neuropsychiatric diseases, the total was 10 patients in the NBS group. More specifically, six patients were diagnosed with only an intellectual disability, two patients were diagnosed with intellectual and psychiatric disabilities, and two patients were diagnosed with only a transient psychiatric impairment during treatment interruption. Meanwhile, 12 patients in the pre-NBS group were diagnosed with neuropsychiatric diseases (six patients with only intellectual disability, five with intellectual and psychiatric disabilities, and one with only psychiatric disability).


**Table 3.** Comparison of clinical and social parameters.

Short stature was defined as a height <160 cm for males and <148 cm for females. Obesity and leanness were defined as BMI >25 and <18.5, respectively. \*, A 37-year-old female is included in the group of "39–44 years old" because she did not undergo NBS. #, Graduation from "high school" includes five patients who are still in college.

> The relationship between neuropsychiatric symptoms and treatment interruption is described below. In the NBS group, 5 of 21 patients with treatment interruption had a neuropsychiatric disease, while 5 of 47 patients who were continuing treatment had a neuropsychiatric disease. A 35-year-old female patient who had the earliest and longest treatment discontinuation period (for approximately 23 years beginning at an age of 11 years) in the NBS group presented with a psychiatric disability and "moderate-severe intellectual disability" that was the worst intellectual level in the NBS group. Although 5 of 10 patients in the NBS group with a neuropsychiatric disability had a history of treatment discontinuation, the details, such as how long and when treatment discontinued, and Phe levels, were not accurately obtained. On the other hand, in the pre-NBS group, four of five patients with treatment interruption had a neuropsychiatric disease, while 8 of 12 patients who were continuing treatment had a neuropsychiatric disease. From another perspective, 4 of 12 patients with a neuropsychiatric disability had a history of treatment discontinuation. A 42-year-old female in the pre-NBS group who discontinued treatment at 3 years old based on the recommendation of an attending physician, as mentioned above,

was unable to be employed, despite her normal intelligence, due to a psychiatric disability induced by intermittent treatment discontinuation for a few decades.

The rate of normal physical size in the NBS group was higher than that in the pre-NBS group, but no obvious differences in physical measurements, including short stature, obesity, or leanness, were observed between the two groups.

Information on education status was obtained for 57 of the 68 patients in the NBS group and 12 of the 17 patients in the pre-NBS group. Fifty-five of 57 (96%) patients in the NBS group had an education level of high school or higher, while 7/12 (58%) patients in the pre-NBS group graduated from high school or higher educational institutions.

Regarding employment status, all but two patients (97%) in the NBS group were employed or attended universities. One of the two unemployed patients was a married female with normal intelligence and was managed to prevent maternal PKU. She was likely a full-time housewife, although this information was not provided. Meanwhile, seven patients (41%) in the pre-NBS group were employed. Five patients each in the pre-NBS group were unemployed or were living in a facility for individuals with a disability. Four of five unemployed patients had intellectual and/or psychiatric disabilities. The other unemployed patient had normal intelligence but also had a tracheostomy due to other respiratory diseases. In summary, all but one of the patients in the NBS group was able to live an independent life, while over half of the patients in the pre-NBS group were not employed or lived in facilities for individuals with disabilities.

Regarding the marital status, 19 of 50 patients in the NBS group were married, while 2 of 15 patients in the pre-NBS group were married.

When given the opportunity to freely describe the patient, many attending physicians stated the difficulty in continuing a Phe-restricted diet, the necessity for lifelong treatment, the need to support medical expenses, and/or the necessity for consultation with internal medicine and psychiatric specialists.

#### **4. Discussion**

Our study examined the physical, neurological and social status of adult patients who were diagnosed with PKU before and after NBS in Japan and revealed that NBS obviously contributed to the improvement of long-term outcomes; our results were similar to those of previous reports [1,15–17]. While almost all patients in the NBS group exhibited normal mental development and were able to live an independent life, over half of the patients in the pre-NBS group had neurological and psychiatric problems and were more severely disabled. However, even in the NBS group, 10 patients had intellectual and/or psychiatric disabilities. Because five of them had histories of treatment interruption, continuous and lifelong treatment is essential for good neurological outcomes.

Meanwhile, five patients in the NBS group whose disease remained well-controlled from birth presented with intellectual and/or psychiatric disabilities despite having no history of treatment interruption. In particular, the neurological outcomes of patients in their late 30s (born from 1977 to 1981) were not as good as those in the younger age group. Four of five patients with neuropsychiatric diseases despite continuing the treatment were in their late 30s. This difference is due to less strict target levels of Phe at the beginning stage of NBS [3,18]. In fact, the target levels of plasma Phe concentration were 4–8 mg/dL in babyhood and 4–12 mg/dL in childhood in Japanese guidelines at that time. In addition, because the recommended Phe levels after the age of 6 years were not indicated until the early 1990s, some Japanese physicians considered that dietary therapy could be relaxed and the dietary restriction could be discontinued after patients reached school age. Because the restriction of Phe was not sufficient even before 20 years, a patient aged in the early 20s had a borderline intellectual disability, despite having no history of poor Phe management. As well as this patient, patients who were detected by NBS, even those with proper management, showed poor neurological outcomes in previous studies [16,19]. Hence, in the present guidelines, the restriction of Phe has become stricter. Namely, not only continuous but also more restricted diet therapy is necessary for good neurological outcomes.

Normal intelligence was observed in some patients in the pre-NBS group despite treatment after onset for unexplained reasons. Additionally, some untreated individuals with classic PKU have normal intelligence despite the elevated plasma Phe concentration [1]. However, the reasons are still unknown. In our study, the symptoms at the onset, what triggered their diagnosis, and when and what type of treatment they had received were not recorded, while our results revealed that their clinical forms were all classic, their age ranged from 39 to 46 years, and one patient had a treatment interruption for 7 years, indicating that the factors constituting a good neurological prognosis are unknown.

Our study also revealed the reasons why patients with PKU discontinued and restarted treatment. Although the unpleasant taste of the special formula is well known [20], treatment interruption was more frequently caused by economic problems in our study. In fact, because the medical public support system for adult patients with PKU has improved since 2015 in Japan, some patients were able to restart treatment. Therefore, lifelong administrative-economic support is necessary to continue treatment. Furthermore, treatment was often neglected due to the self-judgment of being free of symptoms and personal circumstances, such as busy school life and working during adulthood or after childhood. Although many patients detected by Japanese NBS are strictly controlled in childhood by physicians, nutritionists, medical staff, and family, they are released from strict management after adulthood due to independence from those supporters. Therefore, adult patients tend to neglect regular visits and strict dietary therapy due to poor subjective symptoms, and neurological symptoms progress over a chronic course. On the other hand, the primary reason for restarting treatment was pregnancy and the prevention of maternal PKU. Therefore, we strongly suggest that appropriate and continuous patient education is the most effective method to prevent loss to follow-up.

In our study, psychiatric disabilities were likely to progress in the later period of life. One of the reasons may be a difference in the quality of disease management between childhood and adulthood, as mentioned above. In fact, some patients present with psychiatric disabilities after adulthood due to inadequate self-management or treatment intermittence despite having normal intelligence and receiving higher education in childhood. Additionally, it was previously reported that neuropsychiatric symptoms are more prevalent in older adults with PKU [21]. Namely, PKU is considered a progressive disease even after adulthood. Higher blood levels of Phe inhibit myelination in early childhood and functionally impair myelin in late childhood or adulthood, even after normal myelination [22]. Thus, continuous treatment is important for favorable neuropsychiatric outcomes.

Regarding physical development, no obvious differences were detected between the NBS and pre-NBS groups. Nevertheless, optimal growth outcomes were not attained in a previous study, even with advances in dietary treatments [23]. Although our study did not indicate major issues related to growth, physical development should be continuously evaluated for each individual with PKU.

Our results also indicated that the transition to an adult internal medicine department is not easy at this time in Japan. In fact, this issue has been observed in not only Japan but also other countries [24] and for not only PKU patients but also patients with other rare diseases [25]. The explanation for this finding is that internal medicine physicians for adults have little knowledge of PKU or are unable to appropriately treat this disease. In addition and very importantly, patients with PKU realize this limitation and generally prefer to be treated by pediatricians who are experts in PKU care and treatment. Furthermore, patients feel comfortable visiting a place where they know the care providers rather than a new and strange facility. Therefore, the transition to adult department is a major issue for all adult patients with PKU. Furthermore, the issue of the transition and when to make transition have not been described yet in Japanese guidelines, although preparations for the transition from pediatrics to adult internal medicine should begin at approximately 12 years of age based on European guidelines [18,26]. Namely, transitions are likely to be delayed in Japan compared with Europe.

Finally, our study may have some limitations. Only patients who were treated with Phe-free formula and could be traced after adulthood were recruited for our study. According to the annual report of Japanese NBS, approximately 400 adult patients were estimated to be diagnosed with PKU during the period from 1977 to 1996. However, in the present study, only 68 patients were enrolled in the NBS group. Our study did not include a considerable number of adult patients with PKU who no longer received treatment at the hospital, discontinued treatment with Phe-free formula, were treated with only BH4, and died before adulthood. Furthermore, because the participants in this study had continuously or intermittently ordered the special formula, they were willing to be treated even after reaching adulthood. Therefore, our results might be shifted to better outcomes of PKU than the actual situation. Nevertheless, our results are useful for many Asian physicians because the long-term outcomes of East Asian patients with PKU have not been well understood to date. Improvements in the patient registration, follow-up and medical support systems for adult and pediatric patients with PKU will be essential for achieving lifelong favorable outcomes.

#### **5. Conclusions**

The long-term outcomes of adult patients with PKU detected by NBS were much more favorable than those of patients in the pre-NBS group in Japan, as previously reported in other countries. However, some patients, even those who underwent early detection using NBS, with histories of treatment intermittence suffered from neuropsychiatric symptoms. Lifelong and strict management is essential to maintain a good prognosis for patients with PKU.

**Author Contributions:** Conceptualization, K.Y. (Kenji Yamada), K.A. and S.Y.; Methodology, K.Y. (Kenji Yamada) and S.Y.; Software, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Validation, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Formal Analysis, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Investigation, K.Y. (Kenji Yamada) and S.Y.; Data Cu-ration, K.Y. (Kenji Yamada) and K.Y. (Kazunori Yokoyama); Writing–Original Draft Preparation, K.Y. (Kenji Yamada); Writing–Review and Editing, K.Y. (Kazunori Yokoyama), K.A., T.T. and S.Y.; Visualization, K.Y. (Kenji Yamada); Supervision, K.A., T.T. and S.Y.; Project Administration, K.A., T.T. and S.Y.; Funding Acquisition, K.Y. (Kenji Yamada) and S.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This report was partially supported by AMED (grant numbers JP16ek0109050, JP19ek0109276, and JP20ek0109482) and JSPS KAKENHI (grant numbers 19K08300 and 19K08347). The authors confirm that the sponsors had no role in the study; the content of the article was not influenced by the sponsors.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Shimane University on 13 October 2016 (protocol code #20160915-2 and date of approval).

**Informed Consent Statement:** Patient consent was waived because this study is retrospective based on already-existing data using questionnaire for attending doctor and because it is difficult to obtain informed consent of some patients due to not visiting the hospital within the research period. However, this study concept and implementation are widely announced on our university website

**Acknowledgments:** We thank Y. Okano (Okano Kodomo Clinic, Izumi), T. Takahashi (Takahashi Clinic, Kobe), M. Takahashi (Isesaki Municipal Hospital, Isesaki), T. Fukao (Gifu University, Gifu), Y. Watanabe (Kurume University, Kurume), M. Araki (Kochi University, Kochi), M. Furujo (Okayama Medical Center, Okayama), J. Toyama (Nishiniigata Chuo Hospital, Niigata), H. Motizuki (Saitama Children's Medical Center, Saitama), C. Numakura (Yamagata University, Yamagata), K. Kosugiyama (Teine Keijinkai Hospital, Sapporo), A. Noguchi (Akita University, Akita), H. Awano and M. Yagi (Kobe University, Kobe), K. Shiota (St. Luke's International Hospital, Tokyo), A. Matsunaga and T. Fushimi (Chiba Children's Hospital, Chiba), K. Hamaguchi (Kuramochi Hospital, Tochigi), T. Yorifuji, (Osaka City General Hospital, Osaka), D. Tokuhara, H. Shintaku, K. Hamasaki, and M. Saito (Osaka City University, Osaka), M. Oshio (Kyushu Hospital, Kitakyushu), Y. Hasegawa (Tokyo Metropolitan Children's Medical Center, Tokyo), O. Sakamoto (Tohoku University, Sendai), T. Ito and Y. Nakajima (Fujita Health University, Toyoake), T. Asano (Nippon Medical School Chibahokusoh Hospital, Inzai), M. Ishige and E. Ogawa (Tokyo Health Service Association, Tokyo), K. Yoshimura and R. Tsurusawa (Fukuoka University, Fukuoka), M. Owada (Tokyo Health Service Association, Tokyo), M. Shiroo (Kyusyu Hospital Kitakyusyu), M. Inoue (Misakae No Sono, Isohaya), and T. Kakiba (Matsue Red Cross Hospital, Matsue) for participating in and assisting with our study.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**


#### **References**

