ROHHAD Syndrome

Rapid-onset obesity with hypothalamic dysfunction, hypoventilation and autonomic dysregulation, with or without neural crest tumour (ROHHAD/ROHHAD-NET)

Literature and regulatory information updated: 1 November 2026.

0. Initial Reference Table

1. Introduction

1.1. Definition

ROHHAD syndrome (rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation) is an ultra-rare paediatric disorder defined by clinical criteria: rapid and extreme weight gain, after the first year and a half of life, in a previously healthy child of normal build, followed by signs of hypothalamic dysfunction, central alveolar hypoventilation and dysregulation of the autonomic nervous system [3, 15]. Orphanet classifies it as a rare syndromic endocrine disease, characterised by hyperphagia and obesity with onset in childhood, alveolar hypoventilation, dysautonomia, hypothalamic dysfunction and neurobehavioural disturbances, which may be accompanied by central hypothyroidism, other endocrine abnormalities, electrolyte imbalances and respiratory failure [1]. The suffix NET (neural tumor, neuroendocrine tumor or neural crest tumor, depending on the authors) designates the subgroup with a neuroblastic tumour, that is, ganglioneuroma (GN), ganglioneuroblastoma (GNB) or, more rarely, neuroblastoma (NB) [4, 12, 16]. There is no validated molecular test or biomarker, and the diagnosis rests on the clinical picture and on the exclusion of other disorders, in particular congenital central hypoventilation syndrome (CCHS) by study of the PHOX2B gene [18, 22].

1.2. Historical overview

The ATS (American Thoracic Society) and the 2023 collaborative review place the first description in 1965, in an article by Fishman, Samson and Sperling on "primary alveolar hypoventilation syndrome" [17, 22, 28]. Fishman's patient was a boy aged 2.7 years who gained 10.4 kg in 9 months, had repeated apnoeas, especially after taking codeine syrup, hypercapnia with a maximum PaCO2 of 70 mmHg and diabetes insipidus, and required ventilation throughout sleep; the authors called the condition "Ondine's curse", a designation that the Chicago group advises against, because it is a literary misconception and suggests that the patient "forgets" to breathe [17]. In 2000, Katz, McGrath and Marcus, in Baltimore, brought together at least ten reports of late-onset idiopathic central hypoventilation with signs of hypothalamic dysfunction (hyperphagia, hypersomnolence, thermal dysregulation, emotional lability and endocrinopathies), described a case treated with intermittent nasal positive pressure ventilation and proposed that this "late-onset central hypoventilation syndrome with hypothalamic dysfunction" was distinct from CCHS, because of the age at presentation and the hypothalamic dysfunction, absent in CCHS [2]. The OMIM entry for CCHS records this work as the description of a heterogeneous group of patients corresponding to ROHHAD [13].

In 2007, Ize-Ludlow and colleagues analysed 23 children referred to Rush University Medical Center, in Chicago, with complete records and samples in 15 and respiratory physiology studies in 9, and gave the entity its current name, chosen to alert the clinician to the usual sequence of appearance of the signs [3, 22]. Obesity began in the first 10 years, at a median of 3 years, followed by hypothalamic dysfunction and autonomic symptoms (median 3.6 years), and alveolar hypoventilation appeared later (median 6.2 years) [3]. Sequencing of PHOX2B, NTRK2 and BDNF was negative, which showed that the entity was distinct from CCHS [3]. In 2008, Bougnères and colleagues, in Paris, described the endocrine manifestations of six patients admitted for apparently common early-onset obesity and proposed the designation ROHHADNET, to include the neural tumours [4].

From 2015 onwards, exome studies of trios, of discordant monozygotic twins and of tumours, and later genome studies, did not identify a causal gene [8–10, 29]. Between 2022 and 2026, data accumulated in favour of an autoimmune syndrome, partly paraneoplastic, with the discovery of antibodies directed against the ZSCAN1 protein (zinc finger and SCAN domain-containing protein 1), first in children with a neuroblastic tumour and then in patients without a tumour and in an adult woman [23, 30, 31].

1.3. Nomenclature and entities within scope

The ATS stresses that ROHHAD must be distinguished from the late-onset form of CCHS (LO-CCHS, later-onset CCHS), which is a disorder of the PHOX2B gene [22]. The GeneReviews text on CCHS treats ROHHAD as "a rare disorder distinct from LO-CCHS", with a phenotype that "unfolds" with age [15]. The 2022 review in the Handbook of Clinical Neurology concludes that the two entities, considered for decades to be part of the same spectrum, are different [32]. In addition to the classic form, the literature describes forms without rapid-onset obesity, with later onset and in which dysautonomia precedes hypothalamic dysfunction [33], an adult-onset form, at 57 years, with anti-ZSCAN1 antibodies [34, 35], a similar picture after SARS-CoV-2 infection [36] and an intracranial ganglion cell tumour, which led to the proposal to rename the tumour component as neural crest tumour (ROHHAD-NeCT) [37]. Congenital central hypoventilation syndrome, neuroblastoma predisposition due to PHOX2B and ALK and the monogenic and syndromic obesities are considered here only as differential diagnoses. The 2023 collaborative review by centres with experience of the disease, coordinated by the Lurie Children's Hospital group in Chicago, considers that the separation between ROHHAD with and without tumour (ROHHADNET) showed no differences in the core phenotype or in evolution with age, and prefers the more inclusive name ROHHAD, which does not distinguish children with a tumour [17].

2. Molecular Genetics and Pathophysiology

2.1. Absence of a causal gene

ROHHAD has no identified gene, and the AACR (American Association for Cancer Research) lists it in its table of neuroblastoma predisposition syndromes as an entity with no known genes and no germline predisposition [12]. Genetic research began with candidate genes for the development of the hypothalamus and the autonomic nervous system and was later extended to the exome, the genome and structural variants, always with negative results (Table 2.1).

Table 2.1. Genetic studies in ROHHAD (CGH, comparative genomic hybridisation; FISH, fluorescence in situ hybridisation; NGS, next-generation sequencing; MS-MLPA, methylation-specific multiplex ligation-dependent probe amplification; MAF, minor allele frequency; CNV, copy-number variation; SNV, single nucleotide variant; AD, autosomal dominant; AR, autosomal recessive)

Barclay and colleagues concluded that clinical exome sequencing is unlikely to serve as a diagnostic test in true ROHHAD [8]. The case of monozygotic twins, with concordant growth and development up to 8 years and in whom only one developed the disease between 8 and 12 years, argues against the monogenic hypothesis and led the authors to propose autoimmune or epigenetic mechanisms or a combination of genetic predisposition with an acquired trigger [29]. The exome of the two twins showed no coding differences, which keeps the epigenetic hypothesis open [8, 18]. The absence of a monogenic cause in the genome of five patients does not exclude a genetic aetiology, but, according to the authors, strengthens the autoimmune hypothesis [10].

The 2023 collaborative review lists genetic mechanisms that these studies do not exclude. A post-zygotic somatic variant limited to tissues derived from the neural crest or to the central nervous system would be compatible with the neuroblastic tumours and with the two pairs of discordant monozygotic twins described, and would escape analyses performed on blood or saliva; exome comparison between tumour and blood in four patients and between the discordant twins found no candidate genes, which, if mosaicism exists, would confine it to the central nervous system [8, 17]. Epigenetic dysregulation of genes with a normal sequence is the other hypothesis [17]. In favour of a disturbance of neural crest development, the authors cite a facial phenotype similar to that of CCHS, with a shorter and broader face and a downward inflection of the lateral third of the vermilion of the upper lip, a structure of neural crest origin [17].

2.2. Variants found in patients with a clinical diagnosis of ROHHAD

Some patients with a clinical diagnosis of ROHHAD turned out to have another genetic disorder. In a child with early-onset morbid obesity, hypoventilation and autonomic and behavioural abnormalities, trio exome sequencing identified the de novo variant RAI1 NM_030665.4:c.3265C>T, NP_109590.3:p.(Arg1089Ter) (described as p.R1089X), a cause of Smith-Magenis syndrome, and the authors noted that the picture was not typical of either syndrome [42, 43]. In a 21-month-old girl from the Liège cohort who met the criteria for ROHHAD, but had hypotonia, abnormal eye movements and severe psychomotor delay, exome sequencing identified a homozygous truncating P4HTM variant, NM_177939.3:c.1217_1224delinsCACTGTGACA, NP_808808.1:p.(Lys406ThrfsTer3), the cause of HIDEA syndrome (hypotonia, intellectual disability, eye abnormalities, hypoventilation and dysautonomia), of autosomal recessive inheritance, described in subsection 6.6 [24]. In a Greek boy with a ROHHAD picture and syndromic facial features, an 80-gene panel of the leptin-melanocortin pathway showed PLXNA3 NM_017514.5:c.3640C>T, NP_059984.3:p.(Arg1214Trp), of uncertain significance, and PCSK1 NM_000439.5:c.661A>G, NP_000430.3:p.(Asn221Asp), classified as an obesity risk factor, both heterozygous, without any causal relationship being established [43, 44]. These cases justify exome sequencing in the assessment of suspected ROHHAD, not to confirm it, but to exclude disorders with a known cause [20].

2.3. Autoimmune and paraneoplastic hypothesis

Several findings suggest an immune-mediated process. Oligoclonal bands of intrathecal synthesis have been described in the cerebrospinal fluid (CSF) of two children with ROHHAD [45]. A 3-year-old girl had anti-pituitary and anti-hypothalamus autoantibodies in serum and CSF by indirect immunofluorescence, improved partially with plasmapheresis, methylprednisolone, anti-CD20 and azathioprine, but died of cardiorespiratory arrest [41]. The only autopsy included in the Harvengt review, of a 5-year-old girl, showed hypothalamic encephalitis with a mild chronic perivascular and parenchymal inflammatory infiltrate of CD3 T lymphocytes, without vascular necrosis [20]. A 15-year-old boy with febrile encephalopathy, hypoventilation, dysautonomia and abnormalities of temperature and of water-salt balance had pleocytosis, raised neopterin and oligoclonal bands in the CSF, focal inflammation of the periaqueductal grey matter and the hypothalamus on early magnetic resonance imaging (MRI) and improved with corticosteroids, intravenous immunoglobulin (IVIG) and mycophenolate [46]. In 11 patients from Genoa, compared with 11 children with simple obesity matched for age and sex, T lymphocytes were increased, with CD4 predominance and fewer activated CD8 cells, there were more regulatory B lymphocytes and type 1 regulatory T cells, and interleukin 8 (IL-8) and monocyte chemoattractant protein 1 were raised, with higher IL-8 in the subgroup with a tumour [47]. A reproducible improvement with immunosuppression, described in some patients, strengthens this hypothesis [25, 48].

The 2023 collaborative review brings together the published neuropathological data. There are autopsies of ten patients, with onset at 2–6 years and death from respiratory failure, with or without infection, 5 months to 10 years later; eight had an asymptomatic neuroblastic tumour, and MRI performed during life had been normal in six [17]. Four had no evident central nervous system lesion; five had marked encephalitis of the hypothalamus and brainstem and a sixth minimal brainstem inflammation, with perivascular cuffs of CD20 B lymphocytes and CD4 T lymphocytes, parenchymal infiltrates of CD8 T lymphocytes with granzyme B and variable neuronal loss and gliosis [17]. The inflammation affected the arcuate, ventromedial, periventricular and supraoptic nuclei of the hypothalamus and, in the brainstem, the periaqueductal grey matter, the oculomotor complex, the nucleus of the solitary tract, the dorsal motor nucleus of the vagus, the locus coeruleus, the raphe nuclei and the region of the retrotrapezoid nucleus; patients with extensive hypothalamic inflammation had more severe behavioural changes than usual in ROHHAD [17]. Patients with inflammation died closer to onset (5 months to 2 years; mean 10 months) than patients without lesions (1.5 to 9.5 years; mean 5.6 years), and the absence of neuronal loss in several cases and the dissociation between inflammation and tumour lead the authors to consider a combination of a trigger with a predisposition [17]. In the CSF, only 7 of 13 patients assessed had any inflammatory finding: in the Mandel-Brehm series, CSF-restricted oligoclonal bands in 1 of 8 and matched serum and CSF bands in another 3 (unpublished data cited in the review), a raised IgG index in 1 of 6 and raised neopterin in 3 of 6 [17, 30]. In a study of nine patients, five had HLA-DQ complex alleles associated with autoimmune predisposition [17].

In 2022, Mandel-Brehm and colleagues screened the immunoglobulin G (IgG) of nine children with ROHHAD by phage display of the human proteome followed by sequencing (PhIP-Seq) [30]. Among about 20,000 proteins, only ZSCAN1 met the enrichment criteria, and the antibodies were confirmed by two orthogonal assays, a radioligand binding assay (RLBA) and a cell-based assay in 293T cells (CBA), in 7 of 9 patients and in 0 of 125 controls, including 25 children with opsoclonus-myoclonus syndrome (OMAS, opsoclonus-myoclonus-ataxia syndrome) with and without a tumour [30]. The seven positive patients had a neuroblastic tumour and shared an 11-amino-acid epitope in the C-terminal region of the protein, where the C2H2 zinc fingers are located; ZSCAN1 was detected in the tumour by immunohistochemistry and in the human hypothalamus by mass spectrometry [30]. The authors concluded that tumour-associated ROHHAD meets the criteria for a paraneoplastic neurological syndrome, possibly triggered by an immune response to the peripheral tumour, and noted that ZSCAN1 is an intracellular antigen with no rodent orthologue, which explains the failure of classical methods based on rodent tissue [30]. In paraneoplastic syndromes with intracellular antigens, such as those associated with anti-Yo and anti-Hu antibodies, the antibodies are not considered directly pathogenic, and the response to resection and immunosuppression is less consistent than in syndromes with surface antigens [30].

The antibodies were later found in patients without a tumour. In Barcelona, in an in-house cell-based assay, 5 of 6 patients with paired serum and CSF were positive, four only in the CSF, whereas the 7 patients with only serum available and the 199 controls were negative [23]. In Japan, an ELISA developed by the Hiroshima group was positive in 12 of 14 patients without a tumour (85.7%), with higher titres in those with a worse prognosis, and ZSCAN1 was coexpressed in the mouse subfornical organ, a structure with which the serum of these patients showed reactivity [31]. The same group had described, in a cohort of 22 patients with adipsic hypernatraemia without a structural lesion, reactivity to the subfornical organ in 16, higher prolactin in these patients and six subsequent diagnoses of ROHHAD [49]. In the Japanese national survey published in 2026, 70% of 30 patients had antibodies, the anti-ZSCAN1 titre was higher when there was a neural crest tumour and, in one patient, it fell after tumour resection or immunosuppression [50]. A 2026 preprint (not peer reviewed) describes, in a child with ROHHAD, IgG deposited in brain tissue and expanded CSF clones of B lymphocytes and CD4 T lymphocytes directed against the same ZSCAN1 epitopes [51]. The ZSCAN1 gene has its own OMIM entry (*620026) [52]. The sensitivity and specificity of these antibodies for diagnosis have not yet been established [18, 30].

The trigger of autoimmunity in patients without a tumour is unknown. The Japanese group described diagnoses after influenza or chickenpox, and a ROHHAD-like picture has been reported after COVID-19 pneumonia [31, 36]. Pneumonia or intercurrent viral infection also precede many respiratory decompensations [20, 22]. ROHHAD may include aetiologically diverse subgroups, and autoimmunity may not be universal [25].

2.4. Pathophysiology of the manifestations

The hypothalamus integrates the control of satiety, water balance, temperature, sleep and the pituitary axes, which explains the clustering of hyperphagia, sodium abnormalities, hyperprolactinaemia and central hormone deficiencies [18, 53]. Reactivity directed at the subfornical organ, a structure connected to the hypothalamus and pituitary that senses blood sodium and takes part in the thirst centre, is related to adipsic hypernatraemia and to higher prolactin [31, 49]. Central hypoventilation results from autonomic control of breathing with an abnormal response to hypoxia and hypercapnia [20]. In a study of 25 awake physiological recordings from seven patients, the response of tidal volume and inspiratory drive to hypoxia with hypercapnia was reduced and the patients had no behavioural perception of the compromise, but most ventilatory, cardiovascular and cerebrovascular measures were similar to those of controls, which suggests a mild chemosensory deficit that does not, by itself, explain the severity of the hypoventilation [40, 54]. MRI was normal in 21 of 23 patients in the Harvengt review, and the neuroanatomical findings described (periaqueductal and hypothalamic inflammation, hypointensities in the pons and midbrain, cerebral atrophy) are inconsistent between patients [18, 20]. Hypothalamic inflammation was associated with more severe behavioural manifestations [18]. Secondary narcolepsy with absent hypocretin-1 in the CSF, described in one patient, links the picture to the hypocretin system, whose gene HCRT (OMIM *602358) rarely causes type 1 narcolepsy [55, 56], although no variants of HCRT, HCRTR1 or HCRTR2 have been found [39].

3. Clinical Spectrum and Manifestations by System

3.1. Reference cohorts

All clinical data come from case reports, single-centre series and two systematic reviews with individual data, with no population-based studies; series from referral centres and reviews of the published literature tend to over-represent severe cases [18, 25]. Table 3.1 summarises the cohorts used in this article.

Table 3.1. Main series and reviews

3.2. Chronology of the disease

The acronym is intended to recall the usual order of appearance, but there is no obligatory sequence, and the phenotype is completed over months or years [15, 17, 20]. In the Harvengt review, the initial event was always rapid weight gain, at a median age of 3.1 years (interquartile range, IQR, 2.8–4; n=43), with hyperphagia from a median of 3 years (IQR 2–3.6; n=23); hypothalamic dysfunction appeared a median of less than 1 year (0.9 years) after the obesity, dysautonomia at a median of 4.95 years, central hypoventilation at a median of 5.3 years (IQR 4–8.5; n=28), or 2.2 years (0–3.75) after the onset of obesity, and tumours at a median of 4.75 years (IQR 4–8.45) [20]. Central hypoventilation was diagnosed within the first 5 years after the obesity in 83% of patients [20]. Autonomic features appear between 3 and 11 years and the endocrinopathies cluster between 4 and 8 [18]. The sentinel obesity appears mostly at 3–4 years, but has been described up to 8–9, and in the Padua series it began between 1.8 and 8.6 years, with a median diagnostic delay of 1.6 years (0.3–3.7) [53]. Rapid obesity is not always the first sign: in the Chinese series, two patients had a neurogenic tumour and cognitive and behavioural changes before the weight gain [59], and in the Padua series one child began with changes in thirst and diuresis and excessive sweating [53].

3.3. Obesity and hyperphagia

Weight gain occurs over 3 to 12 months in a previously healthy child of normal build [15]. The sources quantify it in different ways: 20 to 40 pounds (about 9 to 18 kg) in 4 to 6 months according to the ATS in 2010, 20 to 30 pounds (about 9 to 14 kg) according to the Chicago group in 2022, 6 to 10 kg in 6 to 12 months according to the 2024 review and 10 to 15 kg in 6 to 12 months according to a Spanish group [18, 22, 34, 62]. The absence of regular weight monitoring makes it difficult to date the onset, and family photographs help to reconstruct the curve [18]. In the child described by Harvengt, weight went from 15 kg (0 standard deviations, SD) at 3 years and 6 months to 22 kg (+2 SD) three months later and 26 kg at 4 years, with intense hunger and night-time eating in the first three months [20]. In the Lee review, rapid obesity was present in 83.3% of patients with individual data and in 65% of patients from series [16]. The weight curve differs from that of Prader-Willi syndrome (PWS), in which the first year is marked by hypotonia and feeding difficulties and weight gain is less abrupt [40]. The subsequent trajectory is variable: in the Padua series, body mass index (BMI) exceeded the 95th centile by mid-childhood in all patients except one with dysphagia, and at last follow-up the median was 23.9 kg/m2 (median z 2.7) [53].

3.4. Hypothalamic and endocrine dysfunction

GeneReviews lists altered water balance, hyperprolactinaemia, hypothyroidism, altered onset of puberty, growth hormone (GH) deficiency and corticotrophin (ACTH) insufficiency [15]. In the Lee review, hypothalamic dysfunction was present in 83.3% of patients with individual data: GH deficiency in 25.3%, diabetes insipidus in 18.8%, polyuria and polydipsia in 16.7%, central precocious puberty in 14.6% and hypogonadotrophic hypogonadism and premature thelarche in 4.2% each [16]. In patients with laboratory data, there was dysnatraemia in 30 of 31 (hypernatraemia in 25, hyponatraemia in 5), hyperprolactinaemia in 27 of 28, hypothyroidism in 18 of 30 and low insulin-like growth factor 1 (IGF-1) in 12 of 16, with sodium levels described up to 198 mmol/L [16]. Hyperprolactinaemia ranged from 44.7 to 380 ng/mL in the Harvengt review, was a constant early feature in three patients from Bristol and was raised in all Japanese patients at the first visit, although it may be transient and, in the Padua series, mild and present in 3 of 5 [18, 20, 50, 53, 63]. In the Chicago group, hyperprolactinaemia is early and almost constant, the GH response to stimulation testing is considered reduced below 10 ng/mL, and puberty may be precocious, with premature adrenarche, or delayed [17].

The sodium and water abnormalities include hypernatraemia, hyponatraemia, syndrome of inappropriate antidiuretic hormone secretion (SIADH) and arginine vasopressin deficiency (AVP-D, formerly called central diabetes insipidus); eight patients in the Harvengt review began in intensive care with these abnormalities and five with polyuria and polydipsia, some treated with desmopressin [18, 20]. Bougnères described increased sodium without diabetes insipidus in all six patients, a Cushing-like profile in one, glucocorticoid deficiency with normal ACTH in two, complete GH deficiency with low IGF-1 in four, hypogonadotrophic hypogonadism in four and TSH and T4 abnormalities in five of five [4]. In Padua, central hypothyroidism was the most frequent finding (5 of 6), followed by central adrenal insufficiency (4 of 6, two partial and two secondary to corticosteroid therapy), GH deficiency (3 of 4 tested) and AVP-D (3 of 6) [53]. In the Harvengt review, all hypothalamic abnormalities appeared before 10 years, with the exception of delayed puberty, and each patient had a median of two signs [20]. Cushing syndrome may be the presentation, as in a child with seizures due to hypertensive encephalopathy [23].

3.5. Breathing: obstructive apnoea and central hypoventilation

Central alveolar hypoventilation appears first during sleep and later also during wakefulness, often preceded by obstructive sleep apnoea (OSA) [15, 40]. In two Canadian series, OSA was the initial finding in 4 of 6 children, and at follow-up all had nocturnal hypoventilation and required non-invasive ventilation (NIV); nocturnal hypoventilation appeared at a median of 7.2 years, compared with 3.5 years for obesity [57]. In Toronto, all eight children had OSA and two nocturnal hypoventilation on baseline polysomnography; the obstructive apnoea-hypopnoea index did not differ from controls matched for BMI, age and sex (median 11.1 versus 14.4 events/h), but the desaturation index was higher (37.9 versus 14.7 events/h), and while awake and at rest some children had desaturations with central pauses, with no desaturation on the 6-minute walk test [58]. Treatment of OSA does not prevent progression to central hypoventilation [40]. Half of the Canadian children had an irregular breathing pattern while awake [57]. Patients with ROHHAD appear more susceptible to hypoxaemia than children with CCHS, in whom hypercapnia predominates [18].

In the Lee review, 13 of 13 patients with blood gas analysis had hypoxaemia and 14 of 15 hypercapnia at diagnosis, and 43.8% required respiratory support at that time [16]. In the Harvengt review, 11 patients had acute respiratory failure with intubation, eight of them at the start of the diagnostic process, at a median of 5 years; 12 underwent tracheostomy at a median of 4.8 years, five for acute failure and seven to optimise chronic ventilation; NIV started at a median of 6.25 years; there were episodes of cyanosis in seven and pneumonia as a trigger in three [20]. The ATS describes cardiorespiratory arrest after intercurrent viral infection in almost half of the children, before OSA and hypoventilation had been identified [22]. Recurrent hypoxaemia may cause cor pulmonale and right ventricular hypertrophy, and pulmonary hypertension due to chronic hypoventilation has been described [62, 64].

3.6. Autonomic dysregulation

Dysautonomia is heterogeneous and often subtle [18]. Table 3.2 reproduces the data from the Harvengt review, with information on 35 of 43 patients.

Table 3.2. Autonomic manifestations in the Harvengt 2020 review [20]

Strabismus and altered sweating were the first autonomic signs after weight gain, and the signs were independent of each other [20]. The ATS adds low body temperature, cold extremities, severe bradycardia, decreased pain perception and abnormal pupillary responses [22]. A tendency to hypothermia, with a preference for light clothing regardless of ambient temperature, profuse sweating and icy hands and feet distinguish ROHHAD from PWS [40]. Intestinal dysmotility may manifest as constipation or chronic diarrhoea, with rectal prolapse in two patients [18, 20]. Constipation is frequent, but has not so far been associated with Hirschsprung disease, and the pain threshold is often raised [17]. In the Argentinian series, strabismus was the most common autonomic sign (6 of 10), and in the Padua series there was divergent strabismus or convergence insufficiency in 5 of 6, with decreased pain and excessive sweating being the most prominent autonomic features [53, 60]. When actively sought, dysautonomia is much more frequent than in published reports: in the Chinese series of 16 patients, cardiovascular manifestations were recorded in 81.2%, gastrointestinal manifestations and strabismus in 62.5%, sleep disturbances in 50% and an abnormal pain threshold in 25%, compared with 12.8%, 8.5%, 12.8%, 6.4% and 10.6% in cases published after 2019 [59]. Bradycardia may be unpredictable, even during periods of apparent stability, and prolonged sinus pauses led to pacemaker implantation [20, 29, 53].

3.7. Neuroblastic tumours

GeneReviews and the ATS report tumours in about 40% of children, the AACR estimates an overall risk of at least 40–50%, the Lee review found 52.1% in patients with individual data and 42% in those from series, and the Harvengt review 56% [12, 15, 16, 20, 22]. Single-centre series range from 0 of 6 in Padua, 1 of 5 in New Delhi and 5 of 10 in Argentina to 8 of 9 in the Mandel-Brehm cohort, and the range of 30 to 100% between series was highlighted by the latter group [30, 53, 60, 61]. The tumours are mostly mature. In the Harvengt review there were 18 GN, 2 GNB, 2 NB, one of which was aggressive and metastatic, and one hamartoma with neural elements; the location was adrenal in 9, thoracic in 5, and cervical paravertebral, abdominal paravertebral and retropancreatic in 1 each [20]. In the Lee review there were 15 GN (60%), 9 GNB (36%) and 1 hamartoma, mostly abdominal but also mediastinal, some bulky (posterior mediastinal mass of 10 × 10 cm; paravertebral mass of 8 × 3.5 cm with hydroureteronephrosis) [16]. The ATS frequently associates them with scoliosis [22].

Half of the tumours were diagnosed within 0.75 years after the onset of weight gain and 70% in the following two years, but there was one aggressive tumour 11 years after onset [20]. In ROHHAD, tumours were almost always found on screening, without symptoms, unlike NB in the general population; urinary catecholamines have never been described as abnormal, compared with 75% positivity in NB in general, and only one patient had metastases at diagnosis and only one MYCN amplification [20]. Metastatic NBs have been described, as well as cases in which ROHHAD arose as a paraneoplastic syndrome in an already diagnosed NB, with rapid obesity years after the tumour [65, 66]. In one patient, the tumour was associated with opsoclonus-myoclonus-ataxia syndrome, in another with hypertension and Cushing syndrome, in another with paraneoplastic cerebellitis, and there was a first intracranial ganglion cell tumour and a cerebellar ganglion cell tumour with a BRAF mutation [10, 16, 37, 67]. A case of hepatocellular carcinoma at 26 years arose on hepatic fibrosis due to steatosis, and is not a neural crest tumour [68].

3.8. Behaviour, cognition, sleep and neurological manifestations

Behavioural problems range from mild to severe, especially when ventilation is inadequate [15]. In the Lee review, 60.4% of patients had behavioural disturbances, mainly irritability and aggressiveness (20.8%), and 33.3% neurological abnormalities, including seizures in 14.6% [16]. In the Harvengt review, behavioural problems appeared at a median of 3.8 years, with aggressiveness in seven, hyperactivity in two and irritability in three; there were hallucinations in three (median 10 years), major anxiety in four, psychosis in two and blunted affect in four; aggressiveness and mood changes appeared, at the median, 1.5 years before the diagnosis and treatment of hypoventilation [20]. Ize-Ludlow described depression, bipolar disorder, emotional lability, obsessive-compulsive disorder, oppositional defiant disorder, Tourette syndrome and hallucinations [16]. In Padua, mood changes, emotional dysregulation, agitation, aggression towards others and progressive cognitive deterioration affected 5 of 6 patients, and half had difficulty adhering to treatment, including intolerance of life-support devices [53]. A dramatic change in behaviour may be the presenting feature [18]. Night terrors superimposed on delirium have been described [69].

Neurocognitive functioning ranges from above to below average, with particular risk in working memory, processing speed, perceptual reasoning and visuographic skills, and early childhood appears to be the period of greatest vulnerability, possibly because of repeated episodes of hypoxia [70]. Most patients have a normal intelligence quotient, and the Chicago authors attribute delay, when present, to hypoxaemia due to undiagnosed or poorly treated hypoventilation [40]. Developmental regression was described in four patients in the Harvengt review, at a median of 2.8 years [20]. Hypersomnolence is frequent, and there was narcolepsy confirmed by low hypocretin in two patients, one of them with cataplexy, visual hallucinations and sleep-onset REM periods [20, 55]. In the Padua series, excessive daytime sleepiness improved with continuous nocturnal ventilation, which suggests fragmented and unrefreshing sleep [53]. Status epilepticus, febrile encephalopathy and encephalitis-like pictures have been described [23, 46, 71]. With optimised care, the children followed in Chicago had scores equal to or above the mean of controls on the Wechsler intelligence scales, the Beery test of visual-motor integration and the Woodcock-Johnson battery, and the authors use annual formal neurocognitive assessment to confirm that management is adequate [17, 70].

3.9. Metabolism and liver

Insulin resistance, glucose intolerance, type 2 diabetes, hypertriglyceridaemia, raised low-density lipoprotein (LDL) cholesterol and metabolic dysfunction-associated steatotic liver disease (MASLD) appear at very early ages, MASLD from 4 years [18]. In the Harvengt review, there was hypercholesterolaemia in two patients, one with triglycerides of 1062 mg/dL, metabolic syndrome in two and diabetes at 14 years in one [20]. A 27-year-old woman, diagnosed at 15, developed insulin resistance, steatosis, hepatic fibrosis and splenomegaly before 14 years and a 1.6 cm hepatocellular carcinoma at 26 years [68]. In Padua, 4 of 6 patients had steatosis with normal enzymes and 2 of 4 insulin resistance [53]. It is not known whether these abnormalities are intrinsic to the syndrome or a consequence of obesity [18].

3.10. Other manifestations

Recurrent fever, papular rash, enuresis, scoliosis, rhabdomyolysis, megaloblastic anaemia, thrombocytopenia, acanthosis, Raynaud phenomenon, coeliac disease and renal failure have been described, each in a few patients [16, 60]. In the Argentinian series there were type 1 and type 2 diabetes, seizures, hypotension and hypertension and primary hypothyroidism [60]. The Orphanet phenotypic annotations include, as very frequent, dysmorphic features (macrocephaly, hypertelorism, anteverted nares) and sensorineural hearing loss that do not appear in the clinical descriptions of the systematic reviews; the absence of dysmorphism is, on the contrary, a usual finding in the reports [1, 20].

3.11. Atypical forms

ROHHAD without rapid-onset obesity, described in four French children, began later, with dysautonomia followed by hypothalamic dysfunction, and was diagnosed through the hypoventilation; all had OSA and one died at 2 years and 10 months [33]. The first adult-onset case, at 57 years, met the 2007 criteria and had anti-ZSCAN1 antibodies [34, 35]; the Barcelona series, with partly shared authors, includes a 58-year-old woman with previous autoimmune hepatitis, published earlier without antibody testing, who appears to correspond to the same case and who died of hypoventilation [23]. Cases of very early onset, with obesity before 2 years and hypoventilation, encephalitis and tumour within less than four months, illustrate the possible severity [71].

3.12. Natural history and treated disease

The older series and the reviews mix patients diagnosed after cardiorespiratory arrest and patients followed from onset, and do not allow natural history to be separated rigorously from treated disease [16, 20]. Reports of a favourable course concern patients recognised early and adequately ventilated: the affected twin had neurocognitive development equal to that of her sister, a child followed in Chicago had improvement ("refolding") of many aspects of the phenotype over ten years, and the authors of GeneReviews observed recovery of spontaneous breathing while awake with rigorous management [15, 29, 62]. Aggressiveness improved with home NIV and regular sleep schedules in the Harvengt patient [20].

4. Clinical Recognition and Warning Signs

4.1. When to suspect

Suspicion arises in a child over 18 months of age, previously healthy and of normal weight, who gains weight abruptly, with hyperphagia, without another evident cause [3, 20]. Bougnères recommended considering ROHHAD in all cases of isolated, rapid and early-onset obesity, because the syndrome mimics common endocrine disorders and genetic obesities and can cause cardiorespiratory arrest [4]. Suspicion is greater when the obesity is accompanied by hyperprolactinaemia without a structural pituitary lesion, abnormalities of sodium or thirst, central hypothyroidism, GH deficiency, precocious or delayed puberty, thermal instability, strabismus, abnormal sweating, decreased pain or behavioural change [18, 20, 63]. Severe obesity of recent onset and short history is also one of the indications in the Oxford Desk Reference for brain MRI, to exclude a hypothalamic tumour or trauma [72]. A normal first year of life makes PWS unlikely, since hypotonia and neonatal feeding difficulties are universal in it [40].

The oncologist should consider ROHHAD when a child with a neuroblastic tumour develops rapid obesity, even without hypoventilation or endocrinopathy [66]. The respiratory physician should consider it in the presence of unexplained chronic hypoventilation in an obese child, especially if it persists after pneumonia or treatment of OSA [22, 40, 73]. The endocrinologist should consider it in the presence of a hypothalamic syndrome without an identified cause, which warrants a sleep study [33], and the neurologist in the presence of an encephalopathy with hypoventilation, dysautonomia and abnormalities of water-salt balance [46, 71]. Adipsic hypernatraemia of autoimmune cause preceded the diagnosis of ROHHAD in six Japanese patients [49].

4.2. Warning and emergency signs

Table 4.1. Warning signs

Severe decompensations often occur after a trivial respiratory infection, pneumonia or sedation, and acute respiratory failure was the main cause of emergency admission in the Padua series [20, 22, 53]. Respiratory arrest due to central hypoventilation is the main cause of death, and all patients in the Lee review with blood gas analysis were hypoxaemic at the time of diagnosis [16, 18].

4.3. Newborn screening

There is no newborn screening, because the disease begins after a period of normal development and has no known marker [18]. The lack of regular weight measurements in childhood makes it difficult to date the onset of weight gain, and family photographs help to reconstruct it [18].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

The criteria used in the two systematic reviews are those of Ize-Ludlow 2007: (1) onset of rapid and extreme weight gain after 1.5 years (typically between 2 and 7) in a previously non-obese and apparently normal child; (2) evidence of hypothalamic dysfunction; (3) central alveolar hypoventilation; (4) signs of autonomic dysregulation [16, 20]. In the formulation summarised by the 2024 review, hypothalamic dysfunction is defined by at least one of: rapid-onset obesity, hyperprolactinaemia, central hypothyroidism, altered water balance, failed GH stimulation test, corticotrophin deficiency or delayed or precocious puberty, and alveolar hypoventilation must appear after 2 years [18]. Harvengt accepted as eligible patients with criteria 1 and 3 and one sign of criterion 2, and also included young children with a neuroendocrine tumour and obesity before the appearance of criteria 2 and 3, because they might still be evolving [20]. As the criteria do not appear simultaneously and hypoventilation is late, a child may have ROHHAD before meeting all of them; the case treated by Paz-Priel had no hypoventilation and the authors warn that including hypoventilation in the definition introduces a bias towards severe cases [20, 25]. For dysautonomia, a Dutch group proposed a complementary score of 0 to 10 to be added to the 2023 criteria for hypothalamic syndrome, which do not consider autonomic signs; the four patients tested had a median of 9 [74]. The 2023 collaborative review takes the opposite position to Harvengt regarding incomplete cases: it accepts a spectrum of severity, but requires all the components of the acronym for a definitive diagnosis and asks that studies separate patients with the complete phenotype from those without central hypoventilation or another component [17].

According to the ATS, the diagnosis of ROHHAD is based on the clinical presentation, the associated features and documented exclusion of other diagnoses, including CCHS by PHOX2B testing demonstrating the absence of polyalanine repeat expansions (PARMs) and non-polyalanine repeat variants (NPARMs) [22].

5.2. Respiratory assessment

Polysomnography (PSG), or a cardiorespiratory sleep study, with continuous measurement of transcutaneous or end-tidal CO2, is the test that confirms hypoventilation [18, 20]. The European CCHS guideline defines sleep hypoventilation in children as PCO2 above 50 mmHg (6.7 kPa) for more than 25% of total sleep time, with normal values of PaCO2 of 35–45 mmHg, SaO2 of at least 95% and end-tidal CO2 of 30–40 mmHg [75]. The diagnostic algorithm for alveolar hypoventilation in that guideline combines PSG or polygraphy, MRI, electromyography and nerve conduction studies to exclude neuromuscular causes, and includes a PHOX2B branch and a ROHHAD branch [75]. Breathing should also be assessed while awake, at rest and, when possible, on exertion, because many patients have deficits of respiratory control while awake [18, 58]. A normal initial PSG, or one showing only OSA, does not exclude the disease, and Harvengt recommends repeating PSG annually for at least 5 years in cases of high suspicion [20, 57]. Full blood count, bicarbonate, chest radiograph, electrocardiogram and echocardiogram look for signs of chronic hypoventilation, such as polycythaemia, compensated respiratory acidosis and pulmonary hypertension, in line with the ATS guidance for later-presenting hypoventilation [22].

In the Chicago group's protocol, each assessment is performed in a hospital laboratory, with a daytime study while awake and a night-time study during sleep and continuous recording of SpO2, end-tidal CO2 (ETCO2), ECG, thoracic and abdominal respiratory inductance plethysmography, blood pressure, temperature and regional cerebral oxygenation by near-infrared spectroscopy (NIRS); the awake recording reproduces activities of daily living (playing, reading, concentration tasks, watching a video, eating, walking, running) and includes spontaneous breathing trials while awake and asleep [17]. Ventilatory challenges with four gas mixtures are added (100% O2; 95% O2 with 5% CO2; 14% O2 with 7% CO2; 5 to 7 breaths of 100% N2), which test central and peripheral chemoreceptors separately; in children with ROHHAD, ETCO2 rose with all mixtures and saturation fell markedly with the hypoxic mixtures, without perception of dyspnoea [17, 54]. Conventional PSG does not study wakefulness or daily activities, and hypoventilation while awake may go unnoticed until exertional cyanosis appears [17].

5.3. Endocrine, metabolic and neurological assessment

Endocrine assessment looks for hyperprolactinaemia, abnormalities of sodium and osmolality (with investigation of AVP-D and SIADH), central hypothyroidism (free T4 and TSH), GH deficiency (IGF-1 and stimulation tests), ACTH deficiency (for example, a tetracosactide test) and pubertal abnormalities [16, 18]. The Oxford Desk Reference includes, in the genetic assessment of severe obesity, free T4 and TSH, 24-hour urinary cortisol when there are signs of Cushing syndrome, gonadotrophins, GH, fasting glucose and insulin, a glucose tolerance test, serum leptin, calcium and PTH, and brain MRI if the picture is of recent onset [72]. Brain MRI is recommended at any age to exclude structural abnormalities, and is normal in the great majority of patients [18, 20]. Metabolic assessment includes blood glucose, insulin, lipids and liver function, because these abnormalities appear early [18, 20]. Autonomic assessment includes an electrocardiogram (ECG), 24-hour Holter monitoring, serial recordings of core and peripheral temperature and, when possible, sweat tests [18]. A baseline neurocognitive assessment is recommended as a component of routine care [70]. The Chicago protocol includes, at first suspicion, MRI of the brain and brainstem with a hypothalamic-pituitary protocol, with and without contrast and without sedation, a complete metabolic panel, urine osmolality, prolactin, free T4, TSH, IGF-1, IGF-binding protein 3 (IGFBP-3) and cortisol, full blood count with reticulocytes, lipid profile and leptin, echocardiogram, Holter monitoring preferably for 72 hours and neurocognitive assessment; in the presence of precocious or delayed puberty, it adds dehydroepiandrosterone sulfate (DHEA-S), luteinising hormone (LH), follicle-stimulating hormone (FSH), testosterone and a gonadotrophin-releasing hormone (GnRH) test [17]. Objective autonomic characterisation uses pupillometry, circadian variation of core and peripheral temperatures, the quantitative sudomotor axon reflex test (QSWEAT) and, when possible, the thermoregulatory sweat test [17].

5.4. Screening for tumours

Screening for a neuroblastic tumour is part of the initial assessment, with imaging of the chest and abdomen [16, 18]. The AACR recommends abdominal ultrasound, chest radiograph and urinary vanillylmandelic acid (VMA) and homovanillic acid (HVA) normalised to creatinine, and MRI when ultrasound is technically inadequate or another test is suspicious; in ROHHAD, body habitus may limit ultrasound and MRI should be considered as the primary or an alternating modality [12]. Harvengt observed that urinary catecholamines were never positive in ROHHAD tumours, so considers this test falsely reassuring, and proposes MRI of the chest and abdomen alternating with 123I-metaiodobenzylguanidine (MIBG) scintigraphy [20]. The 2024 review suggests whole-body MRI, chest radiograph and/or adrenal ultrasound, urinary catecholamines and plasma metanephrines at diagnosis [18]. Positron emission tomography (PET) with 18F-FDG or with DOTA-NOC has been used in individual cases, and an intracranial tumour justified the proposal to include the skull in screening [27, 37, 71]. The tumour is confirmed by pathology after resection, with classification as GN, GNB or NB and staging [18, 71]. The Chicago group proposes a chest radiograph (for the sympathetic chain) and abdominal ultrasound (for the adrenal glands), without sedation, at first suspicion and every 3 or 4 months up to 7 years, and does not recommend routine MRI, because of the risk of sedation and of cardiorespiratory arrest in these children and because of cost, although it acknowledges that MRI is more sensitive for small tumours; after 7 years, without a tumour, it proposes annual radiograph, ultrasound and urinary VMA/HVA, and warns that these metabolites do not exclude a tumour when used alone, because low-grade tumours often do not secrete them [17].

5.5. Genetic testing

There is no genetic test for ROHHAD, and the purpose of testing is to exclude other diagnoses [12, 18]. PHOX2B testing should include fragment analysis for polyalanine expansions and sequencing for NPARMs, because large expansions may not be detected by gene panels [22, 76]. Methylation of the 15q11–q13 region (MS-MLPA) excludes PWS [18, 40]. Harvengt recommends exome sequencing to exclude differential diagnoses such as Smith-Magenis syndrome [20], and the Liège group suggests interrogating P4HTM when there is abnormal neurological development or eye abnormalities [24]. P4HTM has only been included in intellectual disability panels since 2019, and an old panel or a non-reanalysed exome does not exclude HIDEA syndrome [14]. The Oxford Desk Reference reports a yield of about 10% in genetic investigations of severe obesity referred to the genetics clinic, with chromosomal microarray analysis, panel, exome or genome [72]. A negative PHOX2B result in a patient with rapid obesity and hypothalamic dysfunction is compatible with ROHHAD, but a negative exome does not confirm ROHHAD; clinical exome sequencing does not serve as a diagnostic test [8]. When a variant is found, the report should state the gene, the MANE (Matched Annotation from NCBI and EMBL-EBI) reference transcript with version, HGVS (Human Genome Variation Society) notation, zygosity and class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP), and a variant of uncertain significance, such as the PLXNA3 variant described in one patient, does not explain the phenotype or change the clinical diagnosis [43, 44].

5.6. Anti-ZSCAN1 antibodies

Anti-ZSCAN1 antibodies are available only in research laboratories, and the existing data have important limitations: small series (9, 13, 14 and 30 patients), different methods (PhIP-Seq with RLBA and CBA, CBA alone and an in-house ELISA) and few controls in the Japanese study (20) [23, 30, 31, 50]. In the cell-based assay, sensitivity was much higher in CSF than in serum (6 of 6 versus 2 of 6 in the Mandel-Brehm series; four of five positives only in the CSF in Barcelona), and the authors recommend testing the CSF [23, 30]. The Japanese ELISA detects the antibodies in serum and provides a quantitative titre [31, 50]. A negative test does not exclude ROHHAD, and a high titre does not by itself predict the presence of a tumour [23, 30]. Sensitivity and specificity for clinical use have not yet been established [18]. The CSF may also show oligoclonal bands, mild pleocytosis or raised neopterin [30, 46, 71].

5.7. Practical algorithm

Table 5.1. Assessment of suspected ROHHAD

5.8. Prenatal and preimplantation diagnosis

As there is no gene or germline marker, there is no prenatal or preimplantation diagnosis for ROHHAD; these are possible only when another disorder with a known molecular cause is identified in the proband [12, 72].

6. Differential Diagnosis

6.1. Comparative table

Table 6.1. Differential diagnosis of ROHHAD

6.2. Congenital and later-onset central hypoventilation

CCHS is the most important differential diagnosis, because it shares hypoventilation, dysautonomia and neuroblastic tumours, and the ATS and the European guideline require its exclusion by PHOX2B testing [22, 75]. LO-CCHS, with 20/24 and 20/25 genotypes or rare NPARMs, manifests after the neonatal period, mainly after anaesthesia, sedation, severe lung infection or treatment of OSA, and careful review of the history often reveals signs from birth [22]. It is distinguished from ROHHAD by the absence of rapid-onset obesity and hypothalamic dysfunction and by the presence of a PHOX2B variant [2, 22]. In CCHS, children are more prone to hypercapnia and in ROHHAD to hypoxaemia [18]. Neuroblastic tumours occur in about 5–6% of patients with CCHS, mainly with NPARMs or long PARMs [12, 16].

6.3. Prader-Willi syndrome

PWS has, like ROHHAD, childhood obesity, hypoventilation and autonomic dysfunction, but the apparently shared features differ: weight gain is less abrupt, hypoventilation is not universal and rarely requires ventilation, there is no inability to maintain water balance or hyperprolactinaemia, thermoregulation is unstable with high fevers rather than hypothermia, and there are no tumours [40]. The most useful information is the first year of life, always abnormal in PWS, with hypotonia, feeding difficulties, lethargy and motor delay, and normal in ROHHAD [40]. In PWS, 75% of patients have a 15q11–q13 deletion, 24% maternal uniparental disomy and 1% an imprinting defect, and SNRPN methylation detects 99% [72].

The OMIM entry for PWS reinforces the value of the neonatal period. In a British study comparing clinical with laboratory diagnosis, five signs (hypotonia at birth, weak cry or inactivity, weak suck, feeding difficulties and hypogonadism) were present in all patients with a positive genetic diagnosis, and the absence of any of them predicted a negative result [78]. The revised criteria for requesting DNA testing depend on age: up to 2 years, hypotonia with weak suck; from 2 to 6 years, global developmental delay is added; from 6 to 12 years, excessive eating with central obesity; in adolescence and adulthood, cognitive impairment, central obesity, hypothalamic hypogonadism and typical behaviours, such as temper tantrums and obsessive-compulsive traits [78]. A child with ROHHAD, whose development and feeding in the first years were normal, does not meet these criteria [40, 78]. Some features overlap and do not discriminate: a raised pain threshold, central adrenal insufficiency (15 of 25 randomly selected patients with PWS in a metyrapone test, in a Dutch study) and sleep apnoeas, which in prepubertal children with PWS had a median apnoea-hypopnoea index of 5.1/h, mainly due to central apnoeas [78]. Regarding temperature, the entry describes recurrent hyperthermia in infancy and hypothermia on exposure to cold, but two surveys concluded that clinically significant thermoregulatory abnormalities are neither frequent in nor specific to PWS [78]. In the mortality series of the North American patient association (486 deaths between 1973 and 2015, mean age 29.5 years), respiratory failure was the most frequent cause (31%) [78]. The frequency of PWS was estimated at about 1 in 25,000 in one review, 1 in 16,062 in a North Dakota survey and, in the Anglia and Oxford region (United Kingdom), 1 in 45,000, with a lower limit of prevalence of 1 in 52,000 and of birth incidence of 1 in 29,000, which makes it much more frequent than ROHHAD [78]. Methylation may be normal in the rare patients with a balanced translocation within or downstream of SNRPN, and a PWS-like phenotype with normal methylation should prompt a search for maternal uniparental disomy of chromosome 14 and for 6q16 deletions or rearrangements disrupting SIM1, which the entry includes as a chromosome 6-associated PWS-like syndrome [78, 84].

6.4. Monogenic obesities and other syndromes

Monogenic obesities of the leptin-melanocortin pathway generally begin earlier, in the first year of life or after weaning, and have no central hypoventilation or dysautonomia, with distinctive features such as red hair and ACTH deficiency in POMC deficiency or neonatal enteropathy and raised proinsulin in PCSK1 deficiency [72]. The Oxford Desk Reference recommends, in the presence of severe obesity of recent onset, brain MRI to exclude a hypothalamic tumour [72]. Smith-Magenis syndrome and HIDEA syndrome have been found in patients who met the criteria for ROHHAD, and the presence of dysmorphism, of developmental delay preceding the obesity or of eye abnormalities should prompt extended genetic investigation; HIDEA syndrome is described in detail in subsection 6.6 [24, 42]. The coexistence of variants in leptin-melanocortin pathway genes with a ROHHAD picture has been described in one patient, without proof of causality [44].

6.5. Autoimmune syndromes and acquired hypothalamic lesions

ROHHAD shares with OMAS the association with neuroblastic tumours and persistence after complete tumour resection, which suggests an autoimmune process initiated by the tumour and maintained in its absence [25]. Autoimmune adipsic hypernatraemia may precede ROHHAD [49]. Autoimmune encephalitides, such as anti-NMDAR encephalitis, share altered consciousness, autonomic instability and hypoventilation, and testing for anti-NMDAR, CASPR2, LGI1, AMPA, GABA-B and DPPX antibodies was negative in the case described in section 10 [71]. Acquired hypothalamic obesity due to craniopharyngioma or another structural lesion has a lesion visible on MRI [72].

6.6. HIDEA syndrome (P4HTM)

6.6.1. Definition, history and nomenclature

HIDEA syndrome is an autosomal recessive neurodevelopmental disorder caused by biallelic loss-of-function variants of P4HTM, whose acronym brings together hypotonia, hypoventilation, intellectual disability, dysautonomia, epilepsy and eye abnormalities [6, 11]. In 2014, Kaasinen and colleagues described six patients with profound intellectual disability, hypotonia, strabismus, difficulty with visual fixation, planovalgus feet, mild elbow contractures, interphalangeal hypermobility and coarse facial features of gradual appearance, all living in the same village in northern Finland and descended from a common seventeenth-century ancestor; homozygosity mapping identified a locus at 3p22.1–3p21.1 (11.5 Mb, LOD 11), and the genome of one patient showed homozygous variants in three genes in that region (TKT, P4HTM and USP4), without it being possible to decide which was causal [5]. The authors then proposed the acronym HIDEA in the sense of "hypotonia, intellectual disability and eye abnormalities" [6]. In 2019, Rahikkala and colleagues identified, through international collaboration and exome sequencing, patients from American, Turkish and Syrian families with biallelic variants only in P4HTM, demonstrated the insolubility of the mutant proteins and added hypoventilation, central and obstructive sleep apnoea and dysautonomia to the phenotype, redefining the acronym [6]. The entity has OMIM number #618493 (gene *614584, 3p21.31) and Orphanet code ORPHA:436141, classified as a malformation syndrome with a prevalence below 1/1,000,000; Orphanet withdrew as obsolete a second code with the full descriptive name (ORPHA:656273) [7]. The syndrome falls within the scope of this article because patients with HIDEA may meet the clinical criteria for ROHHAD, and the distinction can only be made by genetic testing and by observing neurological evolution [24].

6.6.2. Gene, protein and mechanism

P4HTM encodes transmembrane prolyl 4-hydroxylase (P4H-TM), an enzyme of the endoplasmic reticulum membrane with a short N-terminal cytoplasmic region, a transmembrane helix and, in the lumen, a calcium-binding EF-hand domain and a prolyl 4-hydroxylase catalytic domain; the active site has two histidine residues and one aspartate residue which, with the co-substrate 2-oxoglutarate, coordinate Fe2+ [11]. The protein is expressed mainly in the brain and eye; in vitro it hydroxylates hypoxia-inducible factor 1α (HIF1α) but not type I collagen, and its inactivation stabilises HIFα in cortical neurons under normoxia, although the location of the catalytic domain in the reticulum lumen suggests that other substrates exist [6]. P4H-TM also takes part in calcium signalling in astrocytes and in gliotransmission, a mechanism that has been invoked for the intellectual disability [11, 24]. The gene has two nine-exon transcripts: the MANE (Matched Annotation from NCBI and EMBL-EBI) Select transcript NM_177939.3 encodes the 502-amino-acid isoform, which is predominant in human fibroblasts and myoblasts, and NM_177938.2 encodes the 563-amino-acid isoform, which served as the reference in the initial publications, and the short transcript excludes parts of exons 6 and 7, so that the c. numbering of variants located downstream of those parts differs between the two [6, 11]. All pathogenic variants described give rise to premature stop codons, an intragenic deletion or substitutions affecting the active site or the overall stability of the protein, and three of them, expressed in insect cells, gave insoluble products; the mechanism is loss of function [6, 11]. P4h-tm-null mice have retinal abnormalities with visual impairment and renal abnormalities, and show hyperactivity, reduced fear and anxiety responses and increased social interaction [6, 88]. In May 2026 a p4htm-null zebrafish model was published with constitutive HIF activation, erythrocytosis, cardiac remodelling and hepatic steatosis, in which pericardial oedema and renal abnormalities reversed with HIF-α inhibition; this is a preclinical finding, with no clinical application [89].

Muscle biopsy showed mitochondrial abnormalities in several patients: low activity of complex III or of complexes I–IV, enlarged and abnormally shaped mitochondria on electron microscopy and cytochrome c oxidase (COX)-negative fibres at the autopsy of a 61-year-old woman, who also had cerebellar atrophy with loss of Purkinje cells [6]. In a British infant, complex I activity in muscle was below 30% of the control mean, without ragged red fibres or COX-negative fibres, a value that meets published criteria for mitochondrial disease; with four of five abnormal biopsies in the patients described up to then, the authors proposed that P4H-TM deficiency behaves as a mitochondrial disease, possibly through HIF, RNA polymerase II or the transcription factor ATF4 [14]. A complex I deficiency in a muscle biopsy therefore does not exclude HIDEA syndrome.

6.6.3. Variant spectrum and correlations

Table 6.2 brings together the variants with a complete molecular description in the sources consulted, normalised to the MANE transcript. Two are enriched in Finland and lie on shared founder haplotypes: c.1073G>A, whose effect on RNA is the loss of exon 6 and whose gnomAD allele frequency is 20 times higher in Finland (7.8 × 10^-4) than in other Europeans (4 × 10^-5), with no homozygotes, with a shared haplotype of about 6.9 Mb, and c.1238C>T, with a haplotype of about 7 Mb [6, 11]. c.949del has been described in two Syrian families, which also suggests a founder effect [14]. Severity varies between siblings with the same genotype, and missense and truncating variants can give equally severe phenotypes, with no established genotype-phenotype correlation; the authors attribute the variability to genetic background, environmental and stochastic factors [6, 11].

Table 6.2. P4HTM variants described in HIDEA syndrome (MANE transcript NM_177939.3; protein NP_808808.1)

In 2023, an analysis of exomes from children with severe obesity from consanguineous families in Pakistan identified rare homozygous missense P4HTM variants in 5 of the probands (about 1.5% of the cohort) and in two more probands of Indian and Moroccan origin, all male [19]. These children had obesity with onset between 0.3 and 3 years, with a BMI-for-age standard deviation score between 5.7 and 8.0, hypotonia and developmental delay with cognitive impairment, generally normal leptin, insulin and cortisol, and three of the five Pakistani probands died in the first years of follow-up, two of them with pneumonia [19]. In the UK Biobank, null P4HTM variants, all heterozygous, were associated with a higher BMI [19]. The authors noted that the variants found in severe obesity were all missense, whereas most families with a neurological phenotype without obesity had truncating variants, and raised the hypothesis of allelic heterogeneity; this hypothesis has no functional demonstration, and the 2022 review found obesity in 17 of 29 patients, including patients homozygous for the exon 6 deletion and one patient with a truncating allele [6, 11, 19]. P4HTM is expressed in all regions and cell types of the mouse and human hypothalamus [19].

6.6.4. Clinical manifestations

The 2022 Kraatari-Tiri review, with 30 patients (19 male, aged between 9 months and 59 years at last examination), is the largest series; its figures are from published patients, without a population base, and with variable denominators because of missing data [11]. All had global developmental delay or intellectual disability (30/30), generally severe or profound, and almost all hypotonia (29/30), muscular and central, with a hypotonic face, open mouth, tented upper lip and low nasal bridge in 22 of 23 [6, 11]. Of 25 patients, 15 learned to walk, between 18 months and 4 years, almost always with an abnormal, waddling gait and with valgus or varus knees and flat feet; 11 of 20 were non-verbal at the time of the study, and first words appeared, when they did, between 1 and 4 years [11]. Epilepsy affected 17 of 30, eye abnormalities 22 of 27 (erratic or roving eye movements, cortical blindness, refractive errors, achromic fundi), strabismus 16 of 24, mainly alternating exotropia with vertical deviation, and nystagmus 8 of 26 [11]. In a review of 21 patients, hypopigmented fundi were present in 7 of 16 and optic pallor or atrophy in 2 of 17, with no consistent abnormalities of the electroretinogram or visual evoked potentials [14]. Brain magnetic resonance imaging (MRI) was normal in 13 of 19; the others had cerebral atrophy, white matter abnormalities or both, and Maddirevula described dysmyelination [11, 91].

Respiratory disease is the main cause of morbidity and death. Central sleep apnoea was recorded in 10 of 22 patients and obstructive apnoea in 9 of 21, and 13 of 26 required bilevel positive airway pressure (BiPAP) or other support, at night or during infections [11]. A slow respiratory rate, which does not adapt to metabolic needs, has been described from the first year, and hypoventilation, when formally studied, was of central origin [6, 14]. Recurrent pneumonias affected 12 of 20 patients, with severe respiratory distress and need for ventilatory support, and respiratory infections were the cause of 4 of the 7 recorded deaths, which occurred between 7 months and 61 years [11, 14]. In a 32-week preterm infant, born to consanguineous parents, intermittent bradypnoea, atypical for bronchopulmonary dysplasia, led to exome sequencing, and the child remained on continuous BiPAP (18/8 cmH2O, rate of 30 per minute), nasogastric tube feeding, with cortical blindness and severe global delay [92]. Dysautonomia manifests as impaired thermoregulation, with recurrent hypothermia or hyperthermia and reduced sweating (6 of 22), and as constipation (9 of 26) [11]. There are also parasomnias in 8 of 22 [11]. Endocrine or hypothalamic abnormalities were described in three patients, about 14% of cases published up to 2023: one early adrenarche, one hypothyroidism and one isolated premature pubarche [24]. Obesity (>95th centile) affected 17 of 29, and in the Liège patient rapid weight gain, with absence of satiety, began shortly before 2 years and stabilised at around 4 [11, 24]. No neuroblastic tumours have been described, but some of the patients are still young [24].

Seizures began, in most of the 18 patients with epilepsy reviewed by Wang, between birth and 6 months, with focal, generalised tonic-clonic, tonic or atonic seizures or epileptic spasms; the exceptions were onset at 22 years and a single seizure at 2.5 years [21]. The seven patients with a detailed description became seizure-free or responded well to antiepileptic drugs, with no cases of drug-resistant epilepsy, and the presence or control of seizures was not related to development [21]. In 2025 a Turkish girl was described with early-infantile-onset developmental and epileptic encephalopathy, cortical atrophy and a thin corpus callosum, which broadens the spectrum [90]. Dystonia was reported in one patient with generalised cerebral atrophy [11]. Maddirevula described a child with only hypotonia, developmental delay and abnormal eye movements, and suggests that many mild cases are diagnosed by reverse phenotyping after the exome identifies the variants [91].

6.6.5. Distinguishing HIDEA from ROHHAD

The Liège patient was admitted at 21 months with persistent hypercapnia, after four severe viral respiratory infections between 6 and 18 months (three with invasive ventilation); she had central hypoventilation on polysomnography, rapid weight gain, raised prolactin, severe strabismus and hypotonia, and later hypothermia of 34 °C at 4 years, which met the criteria for ROHHAD [24]. PHOX2B testing, microarray and 15q11–q13 methylation were normal, as was an intellectual disability panel filtered from the exome; only whole-exome analysis, requested at 5 years because of the psychomotor delay (she sat at 2.5 years, walked at 3.5 and said a few words at 6), identified the homozygous P4HTM variant [24]. The authors acknowledge that, in the initial phase, no finding separated the two hypotheses and that the repeated episodes of severe bronchiolitis probably reflected untreated chronic hypoventilation [24]. Table 6.3 summarises the features that guide the distinction.

Table 6.3. HIDEA syndrome and ROHHAD: features compared

The Liège group recommends studying P4HTM, in addition to PHOX2B, whenever a child with a ROHHAD picture has abnormal neurological development or eye abnormalities, and requesting reinterpretation of old exomes, because the gene was not part of intellectual disability panels before 2019 [14, 24].

6.6.6. Diagnosis

The diagnosis is molecular: identification of two pathogenic or likely pathogenic P4HTM variants in trans, with phase confirmed by parental testing [6, 11]. In practice, the syndrome has been found by exome or genome sequencing, often after extensive negative investigations: in the British infant, neurometabolic tests in blood, urine and cerebrospinal fluid, electromyography (suggestive, but not diagnostic, of a myasthenic disorder), and testing for spinal muscular atrophy, Prader-Willi syndrome, myotonic dystrophy, congenital myasthenic syndromes, PHOX2B expansions and mitochondrial DNA were normal, and the initial genome analysis using the mitochondrial disease and intellectual disability panels gave no result, because P4HTM was not yet included in those panels; only reanalysis targeted at rare variants with protein impact identified the two variants [14]. A panel that does not include P4HTM, or an exome analysed before 2019, does not exclude the syndrome [14, 24]. Interpretation should use the MANE transcript NM_177939.3, because variants located downstream of the parts of exons 6 and 7 absent from the short transcript have different numbering on NM_177938.2 and c.1073G>A, predicted as missense, acts through loss of exon 6 in the RNA [6, 11]. There is no clinical enzymatic biomarker; muscle biopsy with respiratory chain studies may show complex I deficiency, but is non-specific and not required for diagnosis [14]. The initial assessment includes polysomnography with CO2 measurement, electroencephalogram, brain MRI, ophthalmological examination with fundoscopy and developmental assessment [6, 11, 14].

6.6.7. Treatment and risk management

There is no targeted treatment, authorised or in trials, for HIDEA syndrome, and care is supportive [11, 14]. The recommendation common to the series is to perform polysomnography in all patients and assess the need for non-invasive ventilation to prevent sudden death, lowering the threshold for ventilatory support during respiratory infections [6, 11]. In the published series, the ventilation used was nocturnal BiPAP or BiPAP during infections and high-flow nasal cannula; an adult patient underwent tracheostomy after pneumonia with invasive ventilation, and the authors of the Singapore case advocate early tracheostomy when ventilatory dependence is continuous [6, 92]. In the Liège child, nocturnal non-invasive ventilation controlled the hypoventilation, with a single acute decompensation at 6 years, and weight stabilised with diet and assisted physical activity [24]. In a British infant, a trial of pyridostigmine and salbutamol, given after electromyography suggestive of a myasthenic disorder, had minimal effect; the patient died at 7 months after failing repeated extubation attempts during a viral respiratory infection [14].

Epileptic seizures were treated with standard antiepileptic drugs, mainly valproate and phenobarbital, and also lamotrigine, vigabatrin, carbamazepine and topiramate, as monotherapy in most [21]. In the Chinese infant with focal clonic seizures, valproate was started at 18.5 mg/kg/day and the child became seizure-free on 26 mg/kg/day [21]. The sources do not define a regimen specific to HIDEA syndrome; choice and dosage follow the general rules for epilepsy in infants and children and the respective summaries of product characteristics. Wang and colleagues proposed that valproate, because it inhibits HIF-1α in cell models, might have a disease-modifying effect, because their patient did not develop hypoventilation, obesity or dysautonomia; this is a hypothesis based on a single case, without a comparison group [21]. Hay and colleagues, by contrast, urge caution with strategies modulating HIF hydroxylases proposed for neuroprotection, given the severity of the disease caused by loss of P4H-TM [14]. The other measures include physiotherapy and motor rehabilitation, optical correction and treatment of strabismus, feeding support, which may require a tube, treatment of constipation and temperature control [11, 14, 92].

6.6.8. Genetic counselling and epidemiology

Inheritance is autosomal recessive: the parents of a patient are, as a rule, asymptomatic heterozygous carriers, each sibling has a 25% chance of being affected and 50% of being a carrier, and the carriers identified in the published families were healthy [6, 14]. In a Turkish report, a heterozygous carrier brother had autism spectrum disorder, without any relationship to the variant being established [93]. Once the two variants have been identified in the proband, carrier testing in relatives and in partners from at-risk populations, prenatal diagnosis and preimplantation genetic testing are possible [14]. The risk is higher in populations with frequent consanguinity (Saudi Arabia, Iran, Pakistan, Syria) and in Finland, where two founder variants account for most cases, which increases the pre-test probability in patients of Finnish origin [11, 19, 91]. Patients of non-consanguineous parents have been described in the United Kingdom, Belgium and the United States, through compound heterozygosity or rare homozygous variants [6, 14, 24]. The population frequency is unknown: Orphanet classifies the prevalence as below 1/1,000,000, the 2022 review brought together 30 patients and, between 2023 and 2025, at least 11 more were described; the sources consulted include no patients from Portugal or Brazil [7, 11, 19, 21, 24, 90, 93]. The relevant European Reference Network is ERN-ITHACA (rare congenital malformations and rare intellectual disability), through which one of the families in the 2022 series was recruited [11].

6.6.9. Prognosis and surveillance

Prognosis depends mainly on respiratory disease: of the 7 deaths known up to 2022, 4 occurred during respiratory infections; six occurred between 7 months and 8 years, and the latest at 61 years, with acute pneumonia; there are patients alive in the fifth and sixth decades [6, 11, 14]. In the Pakistani cohort with severe obesity, 7 deaths at an early age were recorded (3 probands and 4 relatives with the same variants), which the authors relate to hypoventilation following pneumonia [19]. There is no published surveillance protocol; Table 6.4 brings together the recommendations of the series.

Table 6.4. Proposed surveillance in HIDEA syndrome

7. Treatments and Risk Management Strategies

7.1. Principles and organisation of care

There is no curative treatment or authorised drug for ROHHAD, and in 2024 there was no specific management guideline or consensus; the approach is symptomatic, by domain, and multidisciplinary, with endocrinology, respiratory and sleep medicine, dietetics, psychology and psychiatry, cardiology, neurology, radiology and, often, oncology [17, 18]. Harvengt proposes that all providers be supported by a rare disease reference centre, and the 2024 review advocates centralised multidisciplinary clinics [18, 20]. The Padua series integrated paediatric palliative care at a median of 0.8 years from symptom onset, with individualised care plans, regular meetings, telemedicine and home support, and with 3 to 50 admissions per patient [53]. The available recommendations come from reviews and series (Harvengt 2020, Hawton 2024), from the AACR for tumour surveillance and, by extrapolation, from the CCHS guidelines of the ATS (2010) and the European consortium (2020) for ventilation; immunotherapy doses come from case reports and, by analogy, from the international consensus on OMAS [12, 18, 20, 22, 75, 87]. The 2023 collaborative review describes the organisation of the reference centres: a team coordinated by the paediatrician, with endocrinology, respiratory medicine, sleep and autonomic medicine, psychiatry and psychology, nutrition, cardiology, neurology, oncology and gastroenterology, and reassessments at intervals that follow each child's pace (Table 11.1) [17].

7.2. Ventilatory support

Early recognition and correction of hypoventilation is the measure with the greatest prognostic impact, because respiratory failure and cardiorespiratory arrest are the main causes of death and recurrent hypoxaemia impairs neurodevelopment [16, 18, 70]. All patients in the Harvengt review started with nocturnal NIV; tracheostomy was required in 12 of 43, for acute failure or to optimise chronic ventilation for up to 24 hours a day [20]. The ATS describes that many children can be supported with mask ventilation at night only and that a subgroup requires 24-hour ventilation via tracheostomy, and the 2024 review reports tracheostomy in up to 50% [18, 22]. In the Padua series, five of six patients eventually underwent tracheostomy, at a median of 0.6 years from the first symptom; one did not tolerate NIV and tracheostomy was advised against because of severe behavioural dysregulation and the risk of accidental decannulation [53]. Invasive ventilation was delivered in pressure-controlled or pressure-support modes with a backup rate, with inspiratory pressures of 10–28 cmH2O, positive end-expiratory pressure of 3–5 cmH2O and tidal volumes of 5–9 mL/kg, adjusted to blood gases and sleep [53]. Average volume-assured pressure support (AVAPS) improved adherence to NIV and pulmonary hypertension in an adolescent girl [64]. Diaphragm pacing via the phrenic nerve, implanted thoracoscopically, frees the patient from the ventilator for 12 to 15 hours a day and has been used in children with ROHHAD [18, 94]. The 2023 collaborative review is more reserved: in the only patient with ROHHAD described with phrenic pacers, it was very difficult to find settings that avoided hypoxaemia and hypercapnia, because adipose tissue separates the external antenna from the implanted receiver and the diaphragm has to displace the abdominal mass with each cycle, and pacing should never be used at night in a decannulated patient, because of the risk of severe OSA [17].

The Chicago group recommends, from diagnosis, home pulse oximetry and capnography, continuous during sleep and with hourly checks while awake, respiratory assessments every 2 or 3 months during the first 3 years after diagnosis or suspicion and every 6 to 12 months once the phenotype is complete, cardiopulmonary resuscitation training for parents and, ideally, 24-hour home nursing, with targets of SpO2 ≥95% and ETCO2 of 35–45 mmHg [17]. Many patients start with nocturnal bilevel positive airway pressure (BPAP) and progress to 24-hour ventilation, often invasive via tracheostomy, depending on the severity of the central deficit, the difficulty in obtaining adequate volumes with the mass of the chest wall and leaks at high pressures; a cannula with a tight-to-the-shaft (TTS) cuff allows it to be deflated while awake, for speech, and inflated during sleep, to reduce leaks [17]. The authors advise against daytime mask ventilation in young children or those who do not keep it on, because of the risk of cardiorespiratory arrest, and propose tracheostomy for them; in a subgroup with improvement (refolding) of spontaneous breathing while awake, some patients begin to feel dyspnoea and may extend ventilator-free periods or move from tracheostomy to mask [17].

Targets and monitoring derive from CCHS and apply by extrapolation. The ATS considers an end-tidal CO2 of 30–50 mmHg (ideally 35–40) and an SpO2 of at least 95% reasonable, with alarms usually at SpO2 ≤85% and end-tidal CO2 ≥55 mmHg, continuous oximetry and capnography throughout sleep and periodic checks while awake, and trained carers, because these patients show no signs of respiratory distress in the presence of hypoxaemia or hypercapnia; thoracic impedance apnoea monitors do not detect hypoventilation [22]. The European guideline recommends, during assisted ventilation, SpO2 ≥95% and PCO2 ideally of 35–45 mmHg, and considers mild hyperventilation during sleep not validated [75]. The 2024 review states that early nocturnal ventilation may improve daytime ventilation, and daytime sleepiness improved with continuous nocturnal ventilation in the Padua series [18, 53]. Better ventilation was associated with improved behaviour and stabilisation of BMI in the Harvengt patient [20]. The European CCHS guideline warns that these patients do not increase ventilation in response to hypoxia and should not receive oxygen alone without assisted ventilation, and that, on a commercial flight, a ventilator should be available in the cabin if SpO2 falls below 90% [75].

7.3. Hypothalamic and endocrine dysfunction

Hyperprolactinaemia may require treatment with cabergoline, a dopamine D2 agonist [18]. GH deficiency, confirmed by stimulation tests, is treated with replacement; there is a theoretical risk of worsening airway obstruction, extrapolated from PWS with sleep apnoea [18]. In PWS itself, the OMIM entry records that six months of GH did not worsen sleep-disordered breathing in prepubertal children, with the recommendation to monitor breathing during upper respiratory infections, and that GH increased resting ventilation, occlusion pressure and the response to CO2 in 9 children aged 7 to 14 years; these data have not been studied in ROHHAD [78]. In the Padua series, two of three patients treated with GH had significant adverse effects: oedema and a 1.5 kg gain in one month with 0.026 mg/kg/day, repeated with a lower dose, which led to permanent discontinuation, and fluid retention, lethargy, hypotension and hyponatraemia during a viral illness with 0.07 mg/kg/day, which required discontinuation and resumption at 0.02 mg/kg/day; the authors suggest susceptibility to instability of water balance and cautious, individualised use [53]. AVP-D is treated with desmopressin and monitoring of fluid intake, and transient SIADH requires sodium monitoring [18]. In Padua, two of three patients with AVP-D required long-term desmopressin (maximum 60 µg/day alternating with 45 µg/day), and severe hypernatraemia during an infection was refractory to desmopressin [53]. Adipsic hypernatraemia requires a prescribed fluid intake to be defined, because the patient does not feel thirst [18, 49]. Central hypothyroidism is treated with levothyroxine, with good control in the Padua series [53]. ACTH deficiency, investigated by a tetracosactide test, is treated with hydrocortisone replacement (up to 13.6 mg/m2/day in Padua) and stress doses; adrenal insufficiency may be partial or secondary to corticosteroid therapy [18, 53]. Central precocious puberty and hypogonadotrophic hypogonadism require pubertal staging and hormone treatment; in one girl, sex steroid replacement induced menarche [18, 53]. Regular assessment of the hypothalamic-pituitary axis is mandatory, with no agreed frequency [18]. The Chicago group adds that joint follow-up by paediatric endocrinology, a dietitian and the family can slow weight gain and sometimes reverse the BMI trajectory, and that electrolyte abnormalities are controlled by regulating salt and water intake and with desmopressin when there is AVP-D; it proposes an endocrine panel every 2 or 3 months in the first 3 years (Table 11.1) [17].

7.4. Dysautonomia and cardiac abnormalities

Sinus pauses and severe bradycardia may require a pacemaker: in the Harvengt patient, a 21-day Holter recording showed a 15-second pause concomitant with syncope, and a pacemaker with a ventricular lead and a minimum rate of 60/min, implanted at 5 years and 4 months, eliminated the syncopes [20]. The affected twin described by Patwari required a pacemaker for profound bradycardia [29]. The 2024 review recommends an electrocardiogram and 24-hour Holter monitoring, and the European CCHS guideline uses annual 48–72-hour Holter monitoring, ambulatory blood pressure monitoring and echocardiography [18, 75]. Thermal dysregulation is managed with simple measures, such as adjusting clothing and drinking warm fluids, and with paracetamol for hyperthermia, with variable response [18, 53]. Decreased pain requires safety precautions, intestinal dysmotility may require gastroenterology, and strabismus and refractive errors are treated by ophthalmology [18, 53]. The Chicago group describes reduced heart rate variability while awake and asleep and exaggerated bradycardia during sleep, which it considers not associated with sudden asystole, although the odd patient has received a pacemaker for extreme nocturnal bradycardia; it recommends ECG and 72-hour Holter monitoring, with an annual ECG, and in its follow-up table repeats the echocardiogram and Holter every 6 months [17]. The 15-second pause with syncope in the Harvengt patient shows that significant sinus pauses do occur [20]. Hypothermia and hyperthermia are both possible, so clothing should follow ambient temperature, and constipation is treated with diet and, if necessary, prokinetic drugs under the guidance of paediatric gastroenterology [17].

7.5. Neuroblastic tumours

Identified tumours are resected, with good long-term outcomes for mature tumours, but with no evidence that resection alters the progression of the other features of the syndrome [18, 25]. After resection, staging defines whether additional oncological treatment is required, and metastatic NBs require intensive treatment according to oncology protocols [18, 65, 66]. In the case described in section 10, resection of a GNB with a lymph node metastasis followed by IVIG was associated with improvement of the encephalopathy [71]. Surgery requires specific anaesthetic preparation (section 7.12) [18, 95]. The Chicago group recommends resection even of mature tumours, to avoid compression of neighbouring structures and possible dissemination; there are no documented recurrences, and resection does not alter the unfolding of the phenotype [17].

7.6. Weight, hyperphagia and metabolism

Weight control requires an experienced dietitian and paediatric endocrinologist, with the aim of losing weight or maintaining weight while the child grows, and there is little evidence on effective treatments for hyperphagia [18, 96]. A systematic review of dietary and lifestyle interventions in paediatric hypothalamic obesity brought together 12 studies with 118 patients, only one with ROHHAD-NET and 66 with PWS: hypocaloric diets (8–10 kcal/cm/day) and the balanced restrictive diet (30% fat, 45% carbohydrate and 25% protein) stabilised or reduced weight without impairing growth, but the studies had a high risk of bias, with no control group, and the effect of lifestyle interventions was lost in the long term [96]. In the Harvengt patient, a strict diet controlled by the parents and endurance physical activity stabilised BMI [20]. GLP-1 receptor agonists have been used off-label: semaglutide with phentermine in a child, with weight loss and less hyperphagia; liraglutide in an adult woman, with a large reduction in insulin requirements and limited weight loss; and semaglutide 2.4 mg/week in a 24-year-old woman, with a loss of 8.5% in 12 months [18, 68, 97, 98]. Roux-en-Y gastric bypass in a patient from Toronto reduced BMI by 9.5 kg/m2 in 48 months, allowed insulin to be stopped and improved glycated haemoglobin, lipids and liver enzymes [99]. Dysglycaemia can be characterised by continuous glucose monitoring and treated with metformin, insulin or GLP-1 agonists, with variable response and reports of hypoglycaemia [18]. MASLD has no proven treatment other than weight loss [18]. Dyslipidaemia has been treated with statins from 5 years of age, with limited efficacy; in a 13-year-old girl there was rhabdomyolysis and acute kidney injury two weeks after switching from lovastatin to rosuvastatin [18, 100].

Setmelanotide (Imcivree®). Setmelanotide is a melanocortin 4 receptor (MC4R) agonist of the leptin-melanocortin pathway. In a German boy with ROHHAD, ventilator-dependent, with MASLD and aggressive outbursts, off-label setmelanotide produced a 28% weight loss (from 97 to 70 kg), regression of steatosis, reduced ventilatory support and improved behaviour that allowed antipsychotics to be reduced; discontinuation at 18 months, because insurance cover was refused, led to a 10% gain in 3 months (from 70 to 77 kg) [26]. In March 2026, the FDA (Food and Drug Administration) and the European Commission extended the indication of setmelanotide to acquired hypothalamic obesity from 4 years of age, based on a phase 3 trial in patients with hypothalamic obesity due to injury or dysfunction of the hypothalamus, with no reference to ROHHAD [85, 101]. The European Medicines Agency (EMA) summary defines acquired hypothalamic obesity as that occurring when the hypothalamus stops working properly, after an injury or because of an underlying disease; the inclusion of ROHHAD in this definition is not established in the sources consulted, and the FDA label excludes other obesities associated with genetic syndromes [85, 101]. In the absence of a dose studied in ROHHAD, the only published dose is that of the label for acquired hypothalamic obesity, whose application would be an off-label extrapolation (Table 7.1).

Table 7.1. Setmelanotide in acquired hypothalamic obesity: label doses (extrapolation to ROHHAD)

In the EU phase 3 trial (RM-493-040; 81 patients on setmelanotide and 39 on placebo), the mean change in BMI at 52 weeks was −16.46% (95% CI −19.56 to −13.37) versus +3.29% (95% CI 1.11 to 5.47), with an adjusted mean difference of −19.83% (95% CI −24.55 to −15.10; p<0.0001), and 79.5% versus 10.4% of patients reduced BMI by at least 5% [85]. The FDA label, with 142 patients analysed, gives a mean difference of −18.40% in the percentage change in BMI [101]. The warnings on sodium, hydration and adrenal insufficiency are particularly relevant in ROHHAD, in which AVP-D, adipsic hypernatraemia and ACTH deficiency are frequent [16, 85].

7.7. Behaviour, psychiatry and sleep

Behavioural changes require psychology and psychiatry and prior adjustment of ventilation, because aggressiveness and mood changes often precede the diagnosis of hypoventilation [18, 20]. Antipsychotics, antidepressants and methylphenidate have been used, and benzodiazepines with caution for anxiety, always weighing the risk of respiratory depression and weight gain [18]. In the Padua series, first-line atypical antipsychotics (olanzapine up to 10 mg/day, risperidone up to 2.5 mg/day, aripiprazole up to 10 mg/day), haloperidol (2.5 mg/day) and combinations with promazine, clotiapine, chlorpromazine and benzodiazepines had limited efficacy, did not prevent behavioural deterioration or attempts to remove life-support devices and caused extrapyramidal symptoms with risperidone, hallucinations with methylphenidate and hyperthermia with clotiapine and chlorpromazine [53]. Sleep disturbances were treated with melatonin (2–5 mg/day), 5-hydroxytryptophan, hydroxyzine (0.5–1 mg/kg/day) and mirtazapine (15–30 mg/day), with minimal or transient benefit [53]. A child's night terrors improved with anticipatory awakening and haloperidol, and delirium with quetiapine and clonidine [69]. High-dose cyclophosphamide was associated with improved behaviour and neuropsychological function in two children [48]. Secondary narcolepsy requires its own symptomatic treatment [55]. Regular neurocognitive assessment guides educational support [70].

7.8. Immunotherapy

Immunotherapy is based on the autoimmune and paraneoplastic hypothesis and is documented only in reports and small series, with variable response and no controlled trial [18, 20]. In the Lee review, 7 of 48 patients received IVIG, 7 corticosteroids, 6 cyclophosphamide and 5 rituximab, and in the Harvengt review 3 received IVIG, 6 cyclophosphamide, 5 rituximab and 3 corticosteroids, with insufficient data to assess the effect [16, 20]. Table 7.2 brings together the regimens published in ROHHAD and the doses from the international consensus on OMAS, another paraneoplastic syndrome of neuroblastic tumours, cited by analogy; all these uses are off-label.

Table 7.2. Immunotherapy: regimens described in ROHHAD and reference doses in OMAS (off-label use; paediatric population; no maximum dose defined in the sources other than the values given)

The authors of the Paz-Priel report interpret the repeated improvement after each start of immunosuppression and the relapse as the effect wore off as criteria of temporality, consistency and dose-response relationship, but acknowledge that spontaneous improvement cannot be excluded and that ROHHAD may include subgroups in which autoimmunity is not relevant [25]. One review suggests starting with low-dose cyclophosphamide (750 mg/m2 every 28 days) and, if there is no response, moving to the high dose (50 mg/kg/day for 4 days), and the Barcelona group advocates a trial of immunotherapy when symptoms are severe, progressive or refractory [18, 23]. In paraneoplastic syndromes with an intracellular antigen, the response to immunosuppression and tumour resection is less consistent, and may be better with earlier and more intensive regimens [30]. The anti-ZSCAN1 titre fell after resection or immunosuppression in a Japanese patient, but did not change after acute treatment in another, and its role as a response marker is still a proposal [50, 71]. The Johns Hopkins trial of high-dose cyclophosphamide was withdrawn because of insufficient recruitment [104].

The 2023 collaborative review summarises the course of treated patients. One child with onset after a viral infection improved in behaviour with IVIG, worsened on stopping it and stabilised after 9 months of treatment, with resolution of sodium instability and bradycardia, but retained hypersomnolence, food obsession and behavioural changes; another, treated with cyclophosphamide, IVIG and dexamethasone, reduced ventilatory support without improvement in the other features, and added rituximab gave only transient improvement in appetite and weight [17]. In the Spanish case, IVIG 2 g/kg per month for 8 months stabilised sodium and temperature with no effect on behaviour, low-dose cyclophosphamide (750 mg/m2 per month for 4 months) gave transient improvements after each dose, and the high dose improved function and behaviour, including insensitivity to pain, 9 months later, before sudden death off the ventilator [17, 102]. In the Mandel-Brehm series, 6 of 7 patients treated with immunotherapy had some improvement, with different regimens [17, 30]. Data on the durability of responses are lacking, and no report has correlated immunotherapy with systematic physiological assessments or biomarkers [17].

7.9. Regulatory status

Table 7.3. Regulatory status of drugs with described use in ROHHAD

The situation in Brazil (registration with ANVISA) was not consulted for this article.

7.10. Trials and studies

Table 7.4. Trials and studies relevant to ROHHAD

There are no randomised trials in ROHHAD, and the only registered interventional trial was withdrawn [104]. The setmelanotide trial included patients with acquired hypothalamic obesity of various causes and does not show whether the effect is reproduced in ROHHAD, in which hypothalamic dysfunction is of a different nature; the effect reported in a boy with ROHHAD (28% weight loss) is not comparable with the mean of a trial [26, 85]. The OMAS trial refers to another disease, and its transposition to ROHHAD is an analogy [87]. The immunotherapy reports are subject to publication bias, short follow-up (maximum 18 months in the Harvengt review) and the impossibility of separating the effect of the drug from tumour resection, ventilation and spontaneous evolution [20, 25].

7.11. Developments in the last 12 months

Between November 2025 and October 2026, the main regulatory event was the extension of setmelanotide to acquired hypothalamic obesity by the FDA (19 March 2026) and the European Commission (20 March 2026), an authorisation that is not specific to ROHHAD [85, 101]. In terms of evidence, a case report of off-label setmelanotide (June 2026) [26], a Japanese national observational study relating the anti-ZSCAN1 titre to the presence of a tumour and to the response to treatment (July 2026) [50], two retrospective series, from Argentina (10 patients) and Padua (6 patients) [53, 60], the first intracranial ganglion cell tumour [37] and a review on paediatric central apnoea and hypoventilation [106] were published. A preprint not yet peer reviewed describes the convergence of CSF B and T lymphocyte clones on ZSCAN1 epitopes in a child [51]. No new interventional trials, CHMP opinions or authorisations specific to ROHHAD were identified [104].

In the genetic differential diagnosis, a p4htm-null zebrafish model was published in May 2026, with constitutive HIF activation reversed by HIF-α inhibition, a preclinical finding on HIDEA syndrome (subsection 6.6.2) [89].

7.12. Anaesthesia, procedures and agents to avoid

Anaesthesia requires preparation: in 14 children with CCHS or ROHHAD undergoing thoracoscopic implantation of phrenic pacing, there were bradyarrhythmias in 43% (one required rescue cardiac pacing), hypothermia below 35.5 °C in 36% despite warming, hypercapnia above 50 mmHg during lung isolation in 57%, desaturation below 90% in 29% and a need for supplemental oxygen postoperatively in 64%, with no mortality [94]. In a child with ROHHAD-NET, excision of a paravertebral ganglioneuroma was complicated by hypotension, prolonged surgery, prolonged postoperative ventilation and initial weaning failure, with discharge on home BiPAP [95]. Short-acting anaesthetics with minimal respiratory effect, anxiolytics, preoperative optimisation and postoperative monitoring of electrolytes, seizures, hormone deficiencies (with corticosteroid cover if there is ACTH deficiency), temperature and behaviour are recommended [18, 53]. The ATS points out that anaesthetics, central nervous system depressants, severe lung infections and treatment of OSA may unmask latent central hypoventilation [22]. For the same reason, sedatives, opioids, benzodiazepines and sedating antipsychotics require respiratory monitoring [18, 53]. The European CCHS guideline warns that these patients should not receive oxygen alone without assisted ventilation, a warning that applies to ROHHAD by extrapolation [75]. Statins require muscle monitoring, after the rhabdomyolysis described with rosuvastatin [100]. GH should be used with monitoring of water balance and the airway [18, 53].

7.13. Pregnancy, fertility and transition

The sources consulted describe no pregnancies in women with ROHHAD. Hypogonadotrophic hypogonadism is frequent, and cyclophosphamide carries a risk of infertility [4, 25]. In CCHS, ventilatory needs increase over the course of pregnancy [75]. Few adults have been described: two young adults with follow-up up to 27 years in the Harvengt review, a 30-year-old woman in clinical remission in the Mandel-Brehm cohort, a woman with onset at 57 years and anti-ZSCAN1 antibodies, described in Spain and apparently also included in the Barcelona series (there aged 58, by partly shared authors), and a 27-year-old woman with hepatocellular carcinoma [20, 23, 30, 34, 35, 68]. Transition to adult care should maintain ventilation, hormone replacement, metabolic and hepatic surveillance and tumour screening, which Harvengt proposes spacing to every 2 years in adults [20, 68].

7.14. Recommendations and divergences between sources

Table 7.5. Available recommendations and divergences

The main divergence concerns tumour screening: the AACR adapts the hereditary NB predisposition protocol to ROHHAD, with urinary catecholamines and an end point at 10 years, whereas Harvengt, on the basis of ROHHAD tumours, considers catecholamines useless and advocates continued imaging, including MIBG scintigraphy, because there was an aggressive tumour 11 years after onset [12, 20]. The AACR bases the end point at 10 years on data from NB in general (98% diagnosed by that age), which are not specific to ROHHAD [12]. The proposal to include the skull in imaging arose after these recommendations [37]. The Chicago protocol departs from the other proposals by avoiding routine MRI, because of the risk of sedation, and by extending annual radiograph, ultrasound and catecholamines beyond 7 years with no defined end point; it also diverges from Harvengt on the case definition, by requiring the complete phenotype [12, 17, 20].

8. Reproductive Options and Family Counselling

8.1. Recurrence risk

ROHHAD has no identified gene or known germline predisposition, and all cases described are sporadic [1, 12]. The sources consulted report no affected siblings or parent-to-child transmission, and provide no estimate of recurrence risk; the absence of described familial cases does not equate to zero risk [16, 20]. Trio exome sequencing showed only non-recurrent de novo variants, and the pair of discordant monozygotic twins, with no coding differences between them, indicates that shared genetic constitution is not sufficient to cause the disease [8, 29]. The autoimmune and paraneoplastic hypothesis, supported by anti-ZSCAN1 antibodies, also does not imply Mendelian transmission [18, 30]. In counselling, these data make it possible to explain to parents that no hereditary mechanism is known, without assigning a numerical risk value that the sources do not provide.

8.2. When another cause is identified

Genetic investigation of suspected ROHHAD may identify another disorder with its own recurrence risk: CCHS due to PHOX2B, autosomal dominant, with 5–10% of expansions inherited from a mosaic parent [22]; Smith-Magenis syndrome due to a de novo RAI1 variant [42]; HIDEA syndrome, autosomal recessive, with a 25% risk for each sibling, founder variants in Finland and carrier testing, prenatal diagnosis and preimplantation testing possible once the two variants have been identified (subsection 6.6.8) [11, 14, 24]; PWS, with a risk that depends on the mechanism: close to zero with a de novo deletion, estimated at less than 1 in 1000 to 0.4% when the karyotype is normal and there is only one affected child, up to 50% when there is biparental inheritance of an imprinting centre defect, and variable with the type of rearrangement in familial chromosomal rearrangements [72, 78]; and the recessive (LEP, LEPR, POMC, PCSK1) or dominant (MC4R, SIM1) monogenic obesities [72]. For this reason, the diagnosis of ROHHAD should only be communicated as definitive after these entities have been excluded [20, 22].

8.3. Prenatal and preimplantation diagnosis and testing of relatives

There is no prenatal diagnosis, preimplantation genetic testing or cascade testing for ROHHAD; these are possible only when another disorder is confirmed in the laboratory in the proband [12, 72]. There is no indication to test asymptomatic siblings, but informing parents about respiratory and tumour warning signs is part of counselling [12, 18].

8.4. Fertility and reproduction of the patient

Hypogonadotrophic hypogonadism, present in 4 of 6 of Bougnères' patients and in the two assessed in the Padua series, and exposure to cyclophosphamide may compromise fertility [4, 25, 53]. The sources consulted do not describe reproduction in patients with ROHHAD.

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency

Population prevalence is unknown and there are no incidence studies [18]. Orphanet classifies the prevalence below 1/1,000,000 and records 96 published cases or families, based on the 2018 systematic review and expert opinion [1]. That review brought together 158 patients in 46 studies up to July 2018, and the Harvengt review analysed 43 patients with chronology and left out about 70 because of possible duplication or imprecise description [16, 20]. The ATS counted fewer than 55 children in 2010, the Spanish group about 120 cases in 2024 and the 2023 and 2024 reviews fewer than 200 or about 200 [17, 18, 22, 34]. These counts measure published cases, not prevalence, and the lack of a diagnostic test, presentation to different specialties and incomplete forms suggest underdiagnosis [18]. The Japanese national survey, the only attempt at a national census, brought together 30 patients with antibodies tested, but the abstract does not state the reference population [50].

9.2. Sex, age and ethnicity

There is a female predominance: 65 girls and 35 boys (1.9:1) among the 100 patients of known sex in the Lee review, 2.3:1 among patients with individual data and 29 girls and 8 boys in the Harvengt review [16, 20]. The authors of the review note that this predominance contrasts with the male predominance of acquired OSA [16]. There is no clear geographical distribution [18]. Up to 2018 no adults had been described, and adult-onset cases subsequently emerged [16, 23, 34].

9.3. Regional data

The published series come from the USA (Chicago, Boston, Baltimore, Michigan, Atlanta, Durham), Canada (Toronto and Calgary), Europe (France, Belgium, Italy, Spain, the Netherlands, the United Kingdom, Germany, Greece, Bulgaria and Russia), Asia (Japan, China, Korea, India, Singapore, Thailand and Turkey) and Argentina [2–4, 8, 10, 16, 20, 23, 26, 27, 36, 42, 44, 46, 50, 53, 57, 59–61, 63, 68, 69, 73, 100, 107]. The Padua series included white patients, a Bengali child and a North African child, all with equal access to the Italian national health service [53]. The Barcelona series had the participation of neurologists from São Paulo and Salvador da Bahia, but the article does not identify the origin of each patient [23]. The sources consulted include no data from Portugal and no Brazilian series, and describe no founder populations, which is consistent with the absence of a known hereditary cause [18].

9.4. Special populations

In obese children, ultrasound may be technically limited, and the AACR recommends considering MRI as the primary or an alternating modality [12]. Preterm infants and children with a pre-existing neurodevelopmental disorder may also have ROHHAD: in the Padua series, four of six had previous neurodevelopmental disorders and two had been born preterm [53]. Patients with highly dysregulated behaviour may tolerate neither NIV nor tracheostomy, which requires shared decisions on the proportionality of care [53].

9.5. European Reference Networks and patient associations

Endo-ERN (European Reference Network for rare endocrine conditions, https://endo-ern.eu) was used to discuss a child with ROHHAD through the Clinical Patient Management System (CPMS), which led to a whole-body PET scan for tumour screening [27]. The GeneReviews text on CCHS lists as patient associations the ROHHAD Association, in the United Kingdom, ROHHAD Association Belgium and ROHHAD Fight, Inc., in the USA (www.rohhadfight.org), as well as the Children's Neuroblastoma Cancer Foundation [15]. The international ROHHAD registry, based at the Center for Autonomic Medicine in Pediatrics (CAMP) of Lurie Children's Hospital, in Chicago, recruits patients with a suspected or confirmed diagnosis (NCT03135730) [15, 104]. ROHHAD Fight and the ROHHAD Association have funded studies by the Chicago group [62].

10. Illustrative Clinical Case

Tocan and colleagues described, in 2024, a Japanese girl with very early-onset ROHHAD-NET and high-titre anti-ZSCAN1 antibodies [71]. She was born at 39 weeks and 2 days, weighing 3894 g (+2.88 SD) and measuring 50.0 cm, without asphyxia or a family history of hereditary disease; suspected congenital hypothyroidism on newborn screening was excluded [71]. Physical and neurological development was normal up to 18 months, when she weighed 11.1 kg (BMI +0.75 SD); she was a shy child, who cried with pain and was a selective eater [71]. Shortly before 2 years she stopped being selective, began to eat increasing quantities of all foods and gained weight rapidly, as shown by the growth curve (Figure 1: https://pmc.ncbi.nlm.nih.gov/articles/PMC10931978/figure/ijms-25-02820-f001/) [71]. Over the following two months strabismus appeared, she stopped crying in uncomfortable situations, she began to snore and her parents noticed shallow breathing with pauses [71].

At 2 years and 2 months she had a flu-like illness and, three days later, was found unresponsive, with a Glasgow score of 3 and a PCO2 of 96 mmHg, which fell to 43 mmHg after intubation; tests for influenza and COVID-19 were negative and the electroencephalogram showed high-amplitude slow waves [71]. Over the following days she developed hypernatraemia of 162 mmol/L, and abdominal computed tomography showed a lobulated 30 mm left adrenal mass with an adjacent enlarged lymph node [71]. On arrival at the referral centre she weighed 18.5 kg (+3.93 SD), measured 91.0 cm and had a BMI of 22.34 kg/m2 (+4.52 SD); tests showed sodium of 151 mmol/L, prolactin of 26.7 ng/mL (reference 1.2–14.4), IGF-1 of 11 ng/mL (32–213), free T4 of 0.69 ng/dL with normal TSH, cortisol of 4.1 µg/dL with ACTH of 9.8 pg/mL, mildly raised neuron-specific enolase (22.8 ng/mL) and normal urinary VMA, HVA and catecholamines; the CSF had 3 cells/µL, sodium of 160 mmol/L and oligoclonal bands [71]. MRI on day 1 showed diffusion hyperintensity in the deep white matter, and on day 15 mild global cerebral atrophy and new hyperintensities in the caudate nuclei; 18F-FDG PET combined with computed tomography (PET-CT) showed uptake in the adrenal mass (maximum standardised uptake value, SUVmax, 4.69) and in the lymph node (Figure 2: https://pmc.ncbi.nlm.nih.gov/articles/PMC10931978/figure/ijms-25-02820-f002/) [71].

With the clinical diagnosis of ROHHAD-NET, based on rapid obesity, central hypoventilation, hypernatraemia and hyperprolactinaemia, strabismus and the adrenal tumour, the mass was resected on day 18: it was an intermixed ganglioneuroblastoma with a metastasis in an adjacent lymph node (Figure 3: https://pmc.ncbi.nlm.nih.gov/articles/PMC10931978/figure/ijms-25-02820-f003/) [71]. After daily IVIG between days 23 and 27, consciousness improved and the hyperintensities regressed on MRI on day 37 [71]. Anti-Nax, anti-myelin oligodendrocyte glycoprotein (MOG), anti-NMDA, anti-CASPR2, anti-LGI1, anti-AMPA, anti-GABA-B and anti-DPPX antibodies were negative, and the serum anti-ZSCAN1 titre was 348 before and 350 after acute treatment (reference below 40) [71]. She was extubated on day 22 to NIV, but maintained a PaCO2 of 70–90 mmHg during sleep, and underwent tracheostomy on day 32, with PaCO2 falling to 50–60 mmHg [71]. At 2 years and 8 months she walked and talked during the day, remained on pressure-support ventilation during sleep, with a venous PCO2 of 48 mmHg, sodium of about 136 mmol/L and unchanged low IGF-1 and hyperprolactinaemia [71].

Compared with the medians of the Harvengt review, this child had obesity at 1.9 years (median 3.1), behavioural change at 2 years (3.8) and hypoventilation and tumour at 2.2 years (5.3 and 4.75), with progression from obesity to symptomatic hypoventilation in less than four months [20, 71]. The case illustrates respiratory decompensation triggered by a trivial infection, the usefulness of imaging and histology, normal catecholamines in a tumour with a lymph node metastasis and the debated role of immunotherapy [20, 71]. The authors did not study PHOX2B, considering CCHS unlikely in view of the previously normal development, the rapid obesity and the absence of neonatal hypoventilation; the ATS guidelines, however, require documented exclusion of CCHS by PHOX2B testing for the diagnosis of ROHHAD [22, 71]. Serial measurement of anti-ZSCAN1 is proposed by the authors as a possible marker of severity, a use that has not been validated [18, 71].

11. Prognosis and Long-Term Follow-up

11.1. Mortality and causes of death

ROHHAD is a potentially fatal disease [18, 20]. In the Harvengt review, 6 of 32 patients with known outcome died, at a median of 4.6 years in the three with recorded age, from sudden cardiorespiratory arrest (2), tube obstruction during ventilation (1) and septic shock with multiorgan failure (1); 26 were alive at the end of follow-up, at a median of 8 years, 70% under 12 years of age [20]. In the Lee review there were 4 deaths among 48 patients with individual data (3 sudden cardiac arrests and 1 multiorgan failure after sepsis) and 12 among 110 patients from series (9 sudden cardiac arrests), and cardiac arrest, probably preceded by respiratory arrest, was the main cause of death [16]. The ATS reports cardiorespiratory arrest in almost half of the children, the 2024 review a risk of 50–60%, citing Patwari and Wolfe (2014), and the Ize-Ludlow series recorded cardiac arrest in 9 of 15 [16, 18, 22]. Recent single-centre series show high mortality: 3 of 10 in Argentina, 4 of 5 in New Delhi (mean age 5 years; one sudden cardiorespiratory arrest and three respiratory infections) and 4 of 6 in Padua (median 9.9 years, mainly from cardiorespiratory arrest) [53, 60, 61]. Bougnères described two deaths at 8.5 and 12 years, and the literature before 2020 placed death at around 10 years [4, 20]. In the Mandel-Brehm cohort, two of nine patients died suddenly at 8 and 13 years, and one woman was alive at 30 years [30]. These estimates come from small series and published cases and may be subject to publication bias; there are no life expectancy data [18].

11.2. Factors associated with prognosis

Late diagnosis, which delays ventilation, is the most frequently cited factor: all patients with blood gas analysis in the Lee review were hypoxaemic at diagnosis, and early death may reflect more severe initial presentations or late diagnosis [16, 20]. With early recognition and rigorous management, normal neurocognitive development, improvement of many features over ten years and recovery of spontaneous breathing while awake have been described [15, 29, 62]. Poor adherence to life-support devices, because of behavioural dysregulation, increased the risk of death in Padua [53]. Higher anti-ZSCAN1 titres were associated with a worse prognosis in the 2024 Japanese study, a finding not yet confirmed [31]. Mature tumours have a good prognosis after resection, but there have been metastatic NBs and a late aggressive tumour [18, 20, 65]. In adults, liver disease may progress to hepatocellular carcinoma [68]. At Lurie Children's Hospital, some children had gradual improvement of several components after years of disease, always with early identification and rigorous conservative management of ventilation and of the endocrine and autonomic abnormalities, and with tumour resection; with optimised support, they take part in age-appropriate activities with mild to moderate exertion [17].

11.3. Surveillance

There is no validated surveillance protocol for ROHHAD; Table 11.1 brings together the specific proposals (AACR 2024, Harvengt 2020, Hawton 2024 and the Chicago group's protocol described in the 2023 collaborative review) and, by declared extrapolation, the intervals of the European CCHS guideline. In the Chicago protocol, most assessments are repeated every 2 or 3 months during the first 3 years and spaced out once the phenotype is complete [17].

Table 11.1. Surveillance in ROHHAD

The AACR does not recommend NB screening after 10 years, but Harvengt advocates continuing tumour screening, because there was an aggressive tumour 11 years after onset, and proposes imaging every 2 years in adults [12, 20]. The Chicago group also sets no end point: after 7 years, without a tumour, it maintains annual chest radiograph, abdominal ultrasound and urinary VMA/HVA [17]. Families should know the signs of NB (abdominal mass or pain, opsoclonus-myoclonus or ataxia, subcutaneous nodules, watery diarrhoea, periorbital ecchymosis) and the signs of respiratory decompensation [12, 22].

When investigation identifies HIDEA syndrome, surveillance follows Table 6.4, centred on polysomnography, ventilatory support during infections and neurological and ophthalmological assessment [6, 11].

12. Summary and Concluding Remarks

ROHHAD is an ultra-rare paediatric syndrome, described since 1965 and named since 2007, defined by rapid weight gain after a period of normal development, followed by hypothalamic dysfunction, central hypoventilation and dysautonomia, and accompanied by mature neuroblastic tumours in about 40–56% of patients [3, 16, 20]. Obesity begins at a median of 3.1 years, hypothalamic dysfunction follows within less than a year, and central hypoventilation appears a median of 2.2 years later, within the first 5 years in 83% of patients, often preceded by obstructive apnoea [20, 57]. Hyperprolactinaemia and sodium abnormalities are the most constant laboratory findings [16].

The diagnosis is clinical and requires exclusion of CCHS by PHOX2B testing and of other genetic causes of obesity and hypoventilation, such as PWS, Smith-Magenis syndrome and HIDEA syndrome [20, 22, 24]. No gene has been identified, despite exome, genome and discordant monozygotic twin studies, and anti-ZSCAN1 antibodies, found in the CSF and serum of patients with and without a tumour, support an autoimmune syndrome, partly paraneoplastic, but are not yet a validated test [8, 10, 23, 30, 31].

Treatment is symptomatic and multidisciplinary: adequate ventilation, started in time, is the measure that most influences survival and neurodevelopment, and is accompanied by correction of hormone deficiencies with attention to water balance, tumour resection and management of weight, metabolism and behaviour [16, 18, 70]. Immunotherapy and setmelanotide have only case reports in ROHHAD; setmelanotide was authorised in 2026 for acquired hypothalamic obesity, without ROHHAD being named in the labels [25, 26, 85, 101]. Surveillance of hypoventilation and tumours should be regular, with divergence between the AACR and Harvengt on the modality and duration of tumour screening [12, 20]. Published mortality is high, mainly from cardiorespiratory arrest, but patients recognised early and well ventilated may have a favourable course [20, 29, 53, 62]. The international registry and collaboration between centres and European Reference Networks are the route to obtaining data on natural history and treatment [18, 27, 104].

13. References

1. Orphanet. Rapid-onset childhood obesity-hypothalamic dysfunction-hypoventilation-autonomic dysregulation syndrome (ORPHA:293987). Paris: INSERM; 2026 [accessed 2026-10-31]. Available from: https://www.orpha.net/en/disease/detail/293987
2. Katz ES, McGrath S, Marcus CL. Late-onset central hypoventilation with hypothalamic dysfunction: a distinct clinical syndrome. Pediatr Pulmonol. 2000;29(1):62-8. doi:10.1002/(sici)1099-0496(200001)29:1<62::aid-ppul10>3.0.co;2-m. PMID: 10613788.
3. Ize-Ludlow D, Gray JA, Sperling MA, Berry-Kravis EM, Milunsky JM, Farooqi IS, et al. Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation presenting in childhood. Pediatrics. 2007;120(1):e179-88. doi:10.1542/peds.2006-3324. PMID: 17606542.
4. Bougnères P, Pantalone L, Linglart A, Rothenbühler A, Le Stunff C. Endocrine manifestations of the rapid-onset obesity with hypoventilation, hypothalamic, autonomic dysregulation, and neural tumor syndrome in childhood. J Clin Endocrinol Metab. 2008;93(10):3971-80. doi:10.1210/jc.2008-0238. PMID: 18628522.
5. Kaasinen E, Rahikkala E, Koivunen P, Miettinen S, Wamelink MM, Aavikko M, et al. Clinical characterization, genetic mapping and whole-genome sequence analysis of a novel autosomal recessive intellectual disability syndrome. Eur J Med Genet. 2014;57(10):543-51. doi:10.1016/j.ejmg.2014.07.002. PMID: 25078763.
6. Rahikkala E, Myllykoski M, Hinttala R, Vieira P, Nayebzadeh N, Weiss S, et al. Biallelic loss-of-function P4HTM gene variants cause hypotonia, hypoventilation, intellectual disability, dysautonomia, epilepsy, and eye abnormalities (HIDEA syndrome). Genet Med. 2019;21(10):2355-63. doi:10.1038/s41436-019-0503-4. PMID: 30940925. PMC6774999.
7. Orphanet. HIDEA syndrome (ORPHA:436141). Paris: INSERM; 2026 [accessed 2026-11-01]. Available from: https://www.orpha.net/en/disease/detail/436141
8. Barclay SF, Rand CM, Borch LA, Nguyen L, Gray PA, Gibson WT, et al. Rapid-Onset Obesity with Hypothalamic Dysfunction, Hypoventilation, and Autonomic Dysregulation (ROHHAD): exome sequencing of trios, monozygotic twins and tumours. Orphanet J Rare Dis. 2015;10:103. doi:10.1186/s13023-015-0314-x. PMID: 26302956. PMC4548308.
9. Grossi A, Rusmini M, Cusano R, Massidda M, Santamaria G, Napoli F, et al. Whole genome sequencing in ROHHAD trios proved inconclusive: what's beyond?. Front Genet. 2023;14:1031074. doi:10.3389/fgene.2023.1031074. PMID: 37609037. PMC10440434.
10. van Engelen N, van Santen HM, van Dijk F, Kleisman MM, Merks JHM, Schouten-van Meeteren AYN, et al. Genetic evaluation of five patients with ROHHAD-NET using whole genome sequencing and optical genome mapping. Orphanet J Rare Dis. 2025;20(1):412. doi:10.1186/s13023-025-03938-3. PMID: 40775360. PMC12333276.
11. Kraatari-Tiri M, Soikkonen L, Myllykoski M, Jamshidi Y, Karimiani EG, Komulainen-Ebrahim J, et al. HIDEA syndrome is caused by biallelic, pathogenic, rare or founder P4HTM variants impacting the active site or the overall stability of the P4H-TM protein. Clin Genet. 2022;102(5):444-50. doi:10.1111/cge.14203. PMID: 35908151. PMC9804808.
12. Kamihara J, Diller LR, Foulkes WD, Michaeli O, Nakano Y, Pajtler KW, et al. Neuroblastoma predisposition and surveillance — an update from the 2023 AACR Childhood Cancer Predisposition Workshop. Clin Cancer Res. 2024;30(15):3137-43. doi:10.1158/1078-0432.CCR-24-0237.
13. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 209880 — CENTRAL HYPOVENTILATION SYNDROME, CONGENITAL, 1; CCHS1. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/209880
14. Hay E, Wilson LC, Hoskins B, Samuels M, Munot P, Rahman S. Biallelic P4HTM variants associated with HIDEA syndrome and mitochondrial respiratory chain complex I deficiency. Eur J Hum Genet. 2021;29(10):1536-41. doi:10.1038/s41431-021-00932-8. PMID: 34285383. PMC8484625.
15. Weese-Mayer DE, Rand CM, Khaytin I, Slattery SM, Yap KL, Marazita ML, Berry-Kravis EM. Congenital Central Hypoventilation Syndrome. 2004 Jan 28 [updated 2021 Jan 28]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1427/
16. Lee JM, Shin J, Kim S, Gee HY, Lee JS, Cha DH, et al. Rapid-Onset Obesity with Hypoventilation, Hypothalamic, Autonomic Dysregulation, and Neuroendocrine Tumors (ROHHADNET) Syndrome: A Systematic Review. Biomed Res Int. 2018;2018:1250721. doi:10.1155/2018/1250721. PMID: 30584530. PMC6280256.
17. Khaytin I, Victor AK, Barclay SF, Benson LA, Slattery SM, Rand CM, et al. Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD): a collaborative review of the current understanding. Clin Auton Res. 2023;33(3):251-268. doi:10.1007/s10286-023-00936-y. PMID: 37162653.
18. Hawton K, Giri D, Crowne E, Greenwood R, Hamilton-Shield J. The Enigma That Is ROHHAD Syndrome: Challenges and Future Strategies. Brain Sci. 2024;14(11). doi:10.3390/brainsci14111046. PMID: 39595809. PMC11591771.
19. Saeed S, Ning L, Badreddine A, Mirza MU, Boissel M, Khanam R, et al. Biallelic Mutations in P4HTM Cause Syndromic Obesity. Diabetes. 2023;72(9):1228-34. doi:10.2337/db22-1017. PMID: 37083980. PMC7617486.
20. Harvengt J, Gernay C, Mastouri M, Farhat N, Lebrethon MC, Seghaye MC, et al. ROHHAD(NET) Syndrome: Systematic Review of the Clinical Timeline and Recommendations for Diagnosis and Prognosis. J Clin Endocrinol Metab. 2020;105(7). doi:10.1210/clinem/dgaa247. PMID: 32407531.
21. Wang YJ, Li SX, Hu WG, Zhao LL, Lan M, Chen JL. Clinical characteristics of patients with P4HTM variant-associated epilepsy and therapeutic exploration: a case report and literature review. Front Neurol. 2024;15:1428076. doi:10.3389/fneur.2024.1428076. PMID: 39582684. PMC11581967.
22. Weese-Mayer DE, Berry-Kravis EM, Ceccherini I, Keens TG, Loghmanee DA, Trang H. An official ATS clinical policy statement: Congenital central hypoventilation syndrome: genetic basis, diagnosis, and management. Am J Respir Crit Care Med. 2010;181(6):626-44. doi:10.1164/rccm.200807-1069ST. PMID: 20208042.
23. Serafim AB, Olivé-Cirera G, Ortega-González Á, Kruer MC, Weese-Mayer D, Rand CM, et al. Antibodies Against ZSCAN1 in Pediatric and Adult Patients With Non-Paraneoplastic ROHHAD Syndrome. Neurol Neuroimmunol Neuroinflamm. 2024;11(5):e200276. doi:10.1212/NXI.0000000000200276. PMID: 38917381. PMC11204383.
24. Harvengt J, Lumaka A, Fasquelle C, Caberg JH, Mastouri M, Janssen A, et al. HIDEA syndrome: A new case report highlighting similarities with ROHHAD syndrome. Front Genet. 2023;14:1137767. doi:10.3389/fgene.2023.1137767. PMID: 37035730. PMC10073441.
25. Paz-Priel I, Cooke DW, Chen AR. Cyclophosphamide for rapid-onset obesity, hypothalamic dysfunction, hypoventilation, and autonomic dysregulation syndrome. J Pediatr. 2011;158(2):337-9. doi:10.1016/j.jpeds.2010.07.006. PMID: 20727534. PMC3976575.
26. Grünewald A, Steidel E, Müntjes C, Hühne T, Degenhardt F, Stehling F, et al. Melanocortin-4 Receptor Agonist Treatment of Hypothalamic Obesity in ROHHAD Syndrome. Pediatrics. 2026;157(6). doi:10.1542/peds.2025-074432. PMID: 42114845.
27. Galcheva S, Hachmeriyan M, Krumova D, Georgiev R, Dancheva Z, Begemann M, et al. A girl with ROHHAD syndrome – a rare cause of rapid-onset obesity and hypothalamic dysfunction [abstract P1-475]. Horm Res Paediatr. 2023;97(Suppl):P1-475. Available from: https://abstracts.eurospe.org/hrp/0097/hrp0097p1-475
28. FISHMAN LS, SAMSON JH, SPERLING DR. PRIMARY ALVEOLAR HYPOVENTILATION SYNDROME (ONDINE'S CURSE). Am J Dis Child. 1965;110:155-61. doi:10.1001/archpedi.1965.02090030165011. PMID: 14320765.
29. Patwari PP, Rand CM, Berry-Kravis EM, Ize-Ludlow D, Weese-Mayer DE. Monozygotic twins discordant for ROHHAD phenotype. Pediatrics. 2011;128(3):e711-5. doi:10.1542/peds.2011-0155. PMID: 21807698.
30. Mandel-Brehm C, Benson LA, Tran B, Kung AF, Mann SA, Vazquez SE, et al. ZSCAN1 Autoantibodies Are Associated with Pediatric Paraneoplastic ROHHAD. Ann Neurol. 2022;92(2):279-291. doi:10.1002/ana.26380. PMID: 35466441. PMC9329235.
31. Nakamura-Utsunomiya A, Yamaguchi K, Goshima N. Anti-ZSCAN1 Autoantibodies Are a Feasible Diagnostic Marker for ROHHAD Syndrome Not Associated with a Tumor. Int J Mol Sci. 2024;25(3). doi:10.3390/ijms25031794. PMID: 38339072. PMC10855538.
32. Ceccherini I, Kurek KC, Weese-Mayer DE. Developmental disorders affecting the respiratory system: CCHS and ROHHAD. Handb Clin Neurol. 2022;189:53-91. doi:10.1016/B978-0-323-91532-8.00005-7. PMID: 36031316.
33. Desse B, Tran A, Butori M, Marchal S, Afanetti M, Barthélemy S, et al. ROHHAD syndrome without rapid-onset obesity: A diagnosis challenge. Front Pediatr. 2022;10:910099. doi:10.3389/fped.2022.910099. PMID: 36120648. PMC9471950.
34. Ortega-González Á, Perea-Rozas R, Martínez-García A, Rodríguez-Argente F, Ortega-Moreno Á. ROHHAD syndrome spectrum in an adult: a possible new variant. ERJ Open Res. 2024;10(1). doi:10.1183/23120541.00583-2023. PMID: 38259814. PMC10801756.
35. Ortega-González Á, Perea-Rozas R, Martínez-García A, Rodríguez-Argente F, Serafim AB, Olivé-Cirera G, et al. ROHHAD syndrome in an adult: presence of ZSCAN1 antibodies. ERJ Open Res. 2025;11(3). doi:10.1183/23120541.00895-2024. PMID: 40337335. PMC12053991.
36. Artamonova IN, Petrova NA, Lyubimova NA, Kolbina NY, Bryzzhin AV, Borodin AV, et al. Case Report: COVID-19-Associated ROHHAD-Like Syndrome. Front Pediatr. 2022;10:854367. doi:10.3389/fped.2022.854367. PMID: 35433531. PMC9009510.
37. Doelman-Oldenburger NJ, Schouten-van Meeteren AYN, Kranendonk MEG, Boshuisen K, Bannier MAGE, van Santen HM. Case Report: First description of an intracranial "NET" in ROHHAD syndrome; time to rename to the ROHHAD-NeCT syndrome?. Front Endocrinol (Lausanne). 2026;17:1828990. doi:10.3389/fendo.2026.1828990. PMID: 42290863. PMC13259900.
38. Rand CM, Patwari PP, Rodikova EA, Zhou L, Berry-Kravis EM, Wilson RJ, et al. Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation: analysis of hypothalamic and autonomic candidate genes. Pediatr Res. 2011;70(4):375-8. doi:10.1203/PDR.0b013e318229474d. PMID: 21691246.
39. Barclay SF, Rand CM, Gray PA, Gibson WT, Wilson RJ, Berry-Kravis EM, et al. Absence of mutations in HCRT, HCRTR1 and HCRTR2 in patients with ROHHAD. Respir Physiol Neurobiol. 2016;221:59-63. doi:10.1016/j.resp.2015.11.002. PMID: 26555080.
40. Barclay SF, Rand CM, Nguyen L, Wilson RJA, Wevrick R, Gibson WT, et al. ROHHAD and Prader-Willi syndrome (PWS): clinical and genetic comparison. Orphanet J Rare Dis. 2018;13(1):124. doi:10.1186/s13023-018-0860-0. PMID: 30029683. PMC6053704.
41. Giacomozzi C, Guaraldi F, Cambiaso P, Niceta M, Verrillo E, Tartaglia M, et al. Anti-Hypothalamus and Anti-Pituitary Auto-antibodies in ROHHAD Syndrome: Additional Evidence Supporting an Autoimmune Etiopathogenesis. Horm Res Paediatr. 2019;92(2):124-132. doi:10.1159/000499163. PMID: 31039576.
42. Thaker VV, Esteves KM, Towne MC, Brownstein CA, James PM, Crowley L, et al. Whole exome sequencing identifies RAI1 mutation in a morbidly obese child diagnosed with ROHHAD syndrome. J Clin Endocrinol Metab. 2015;100(5):1723-30. doi:10.1210/jc.2014-4215. PMID: 25781356. PMC4422892.
43. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348. Tool available from: https://variantvalidator.org
44. Kostopoulou E, Efthymiadou A, Giannakopoulos A, Chrysis D. A patient with ROHHAD syndrome and leptin-melanocortin pathway gene variations. Correlation or just a coincidence? Case report. Hormones (Athens). 2026;25(2):685-691. doi:10.1007/s42000-025-00754-z. PMID: 41593286. PMC13582127.
45. Sartori S, Priante E, Pettenazzo A, Marson P, Suppiej A, Benini F, et al. Intrathecal synthesis of oligoclonal bands in rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation syndrome: new evidence supporting immunological pathogenesis. J Child Neurol. 2014;29(3):421-5. doi:10.1177/0883073812469050. PMID: 23292759.
46. Chow C, Fortier MV, Das L, Menon AP, Vasanwala R, Lam JC, et al. Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD) syndrome may have a hypothalamus-periaqueductal gray localization. Pediatr Neurol. 2015;52(5):521-5. doi:10.1016/j.pediatrneurol.2014.11.019. PMID: 25746964.
47. Fava D, Morandi F, Prigione I, Angelelli A, Bocca P, Pistorio A, et al. Blood Lymphocyte Subsets and Proinflammatory Cytokine Profile in ROHHAD(NET) and non-ROHHAD(NET) Obese Individuals. J Endocr Soc. 2023;7(9):bvad103. doi:10.1210/jendso/bvad103. PMID: 37564886. PMC10411042.
48. Jacobson LA, Rane S, McReynolds LJ, Steppan DA, Chen AR, Paz-Priel I. Improved Behavior and Neuropsychological Function in Children With ROHHAD After High-Dose Cyclophosphamide. Pediatrics. 2016;138(1). doi:10.1542/peds.2015-1080. PMID: 27313069. PMC4925071.
49. Nakamura-Utsunomiya A, Goda S, Hayakawa S, Sonoko S, Hoorn EJ, Blanchard A, et al. Identification of clinical factors related to antibody-mediated immune response to the subfornical organ. Clin Endocrinol (Oxf). 2022;97(1):72-80. doi:10.1111/cen.14737. PMID: 35419873.
50. Nakamura-Utsunomiya A, Hasegawa K, Ono T, Kanno J, Shima H, Suzuki Y, et al. Clinical characteristics and anti-ZSCAN1 antibody titer analysis in a nationwide survey of ROHHAD (-NET) syndrome. J Clin Endocrinol Metab. 2026;111(8):2232-2241. doi:10.1210/clinem/dgag109. PMID: 41821305.
51. Bodansky A, Ahyong V, Dayao M, Asaki J, Vanasupa S, Dandekar R, et al. AEGIS reveals epitope- and clone-resolved convergence of CNS B and T cell autoreactivity in ROHHAD [preprint]. bioRxiv. 2026 May 11. doi:10.64898/2026.05.06.722823.
52. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 620026 — ZINC FINGER- AND SCAN DOMAIN-CONTAINING PROTEIN 1; ZSCAN1. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/620026
53. Mercante A, Salerno A, Marinetto A, Santini A, Tascini B, Raffagnato A, et al. ROHHAD syndrome: an interdisciplinary perspective. Front Pediatr. 2026;14:1777286. doi:10.3389/fped.2026.1777286. PMID: 42317651. PMC13273399.
54. Carroll MS, Patwari PP, Kenny AS, Brogadir CD, Stewart TM, Weese-Mayer DE. Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD): Response to ventilatory challenges. Pediatr Pulmonol. 2015;50(12):1336-45. doi:10.1002/ppul.23164. PMID: 25776886.
55. Dhondt K, Verloo P, Verhelst H, Van Coster R, Overeem S. Hypocretin-1 deficiency in a girl with ROHHAD syndrome. Pediatrics. 2013;132(3):e788-92. doi:10.1542/peds.2012-3225. PMID: 23940246.
56. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 602358 — HYPOCRETIN; HCRT. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/602358
57. Reppucci D, Hamilton J, Yeh EA, Katz S, Al-Saleh S, Narang I. ROHHAD syndrome and evolution of sleep disordered breathing. Orphanet J Rare Dis. 2016;11(1):106. doi:10.1186/s13023-016-0484-1. PMID: 27473663. PMC4967322.
58. Selvadurai S, Benzon D, Voutsas G, Hamilton J, Yeh A, Cifra B, et al. Sleep-disordered breathing, respiratory patterns during wakefulness and functional capacity in pediatric patients with rapid-onset obesity with hypothalamic dysfunction, hypoventilation and autonomic dysregulation syndrome. Pediatr Pulmonol. 2021;56(2):479-485. doi:10.1002/ppul.25199. PMID: 33270379.
59. Wang Y, Xu Y, Xie R, Cao B, Ding Y, Guo J, et al. Emphasizing autonomic dysregulation evaluation contributes to the diagnosis of ROHHAD syndrome. Endocr Connect. 2024;13(11). doi:10.1530/EC-24-0189. PMID: 39235844. PMC11562685.
60. Aziz M, Leske V, Guerdile MJ, Armeno M, Ciaccio M, Gil SM. Clinical Description of Ten Pediatric Patients with Clinical Signs of ROHHAD-NET Syndrome and Review of the Literature. Horm Res Paediatr. 2025;:1-11. doi:10.1159/000549050. PMID: 41118323.
61. Priyadarshini S, Verma A, Paul PG, Sharma R, Jain V. ROHHAD NET in Retrospect: Key Lessons Learned from Five Cases. Indian J Pediatr. 2025;92(11):1194-1200. doi:10.1007/s12098-024-05291-y. PMID: 39722101.
62. Khaytin I, Stewart TM, Zelko FA, Kee MAL, Osipoff JN, Slattery SM, et al. Evolution of physiologic and autonomic phenotype in rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation over a decade from age at diagnosis. J Clin Sleep Med. 2022;18(3):937-944. doi:10.5664/jcsm.9740. PMID: 34694990. PMC8883099.
63. Hawton K, Hilliard T, Langton-Hewer SC, Burren C, Crowne EC, Hamilton-Shield JP, et al. Rapid-onset obesity, hypothalamic dysfunction, hypoventilation, and autonomic dysregulation syndrome - neuro-endocrine tumours (ROHHAD-NET): case series and learning points. J Pediatr Endocrinol Metab. 2023;36(4):418-423. doi:10.1515/jpem-2022-0376. PMID: 36696572.
64. Stowe RC, Afolabi-Brown O. Pulmonary hypertension and chronic hypoventilation in ROHHAD syndrome treated with average-volume assured pressure support. Pediatr Investig. 2019;3(4):253-256. doi:10.1002/ped4.12168. PMID: 32851331. PMC7331430.
65. Calvo C, Storey C, Morcrette G, Akl P, Fréneaux P, Pierron G, et al. Metastatic neuroblastoma in a patient with ROHHAD: A new alert regarding the risk of aggressive malignancies in this rare condition. Pediatr Blood Cancer. 2019;66(10):e27906. doi:10.1002/pbc.27906. PMID: 31309712.
66. Zhang S, Zhao W, Jin M, Wang H, Gong C, Liu M, et al. Rapid-onset obesity, hypothalamic dysfunction, hypoventilation, and autonomic dysregulation as a paraneoplastic syndrome in neuroblastoma: case report. Transl Pediatr. 2025;14(9):2393-2400. doi:10.21037/tp-2025-269. PMID: 41141676. PMC12552169.
67. Tiwari S, Arimbrakkunnan M, Yadav T, Vyas V, Rathod KK, Vishwajeet V. Thoracic Ganglioneuroma Presenting with Paraneoplastic Cerebellitis and ROHHAD-NET Syndrome. Ann Indian Acad Neurol. 2022;25(2):307-309. doi:10.4103/aian.aian_372_21. PMID: 35693680. PMC9175397.
68. Jalal Eldin AW, Tombayoglu D, Butz L, Affinati A, Meral R, Ontan MS, et al. Natural history of ROHHAD syndrome: development of severe insulin resistance and fatty liver disease over time. Clin Diabetes Endocrinol. 2019;5:9. doi:10.1186/s40842-019-0082-y. PMID: 31333877. PMC6617654.
69. Donaldson RA, Kothari S, Diaz Stransky A, Rajagopalan A, Weese-Mayer DE. Night Terrors and Delirium in a Child With Presumed Rapid-Onset Obesity With Hypothalamic Dysfunction, Hypoventilation, and Autonomic Dysregulation. Cureus. 2026;18(4):e107609. doi:10.7759/cureus.107609. PMID: 42181399. PMC13198437.
70. Zelko FA, Welbel RZ, Rand CM, Stewart T, Fadl-Alla A, Khaytin I, et al. Neurocognition as a biomarker in the rare autonomic disorders of CCHS and ROHHAD. Clin Auton Res. 2023;33(3):217-230. doi:10.1007/s10286-022-00901-1. PMID: 36289132.
71. Tocan V, Nakamura-Utsunomiya A, Sonoda Y, Matsuoka W, Mizuguchi S, Muto Y, et al. High-Titer Anti-ZSCAN1 Antibodies in a Toddler Clinically Diagnosed with Apparent Rapid-Onset Obesity with Hypothalamic Dysfunction, Hypoventilation, and Autonomic Dysregulation Syndrome. Int J Mol Sci. 2024;25(5). doi:10.3390/ijms25052820. PMID: 38474067. PMC10931978.
72. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 254–256.
73. Butpech T, Thawongit S, Wannapaschaiyong P, Limkul L, Tovichien P, Udomittipong K. A 2.5-year-old obese girl with persistent hypoventilation post-pneumonia: when is it not just the lungs?. Breathe (Sheff). 2026;22(3):260049. doi:10.1183/20734735.0049-2026. PMID: 42482832. PMC13386199.
74. Doelman-Oldenburger NJ, Müller HL, van Santen HM. Novel autonomic dysregulation score in children with hypothalamic syndrome. Endocr Connect. 2025;14(12). doi:10.1530/EC-25-0336. PMID: 41251290. PMC12673360.
75. Trang H, Samuels M, Ceccherini I, Frerick M, Garcia-Teresa MA, Peters J, et al. Guidelines for diagnosis and management of congenital central hypoventilation syndrome. Orphanet J Rare Dis. 2020;15(1):252. doi:10.1186/s13023-020-01460-2. PMID: 32958024. PMC7503443.
76. Kamihara J, Bourdeaut F, Foulkes WD, Molenaar JJ, Mossé YP, Nakagawara A, et al. Retinoblastoma and neuroblastoma predisposition and surveillance. Clin Cancer Res. 2017;23(13):e98-e106. doi:10.1158/1078-0432.CCR-17-0652.
77. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 603851 — PAIRED-LIKE HOMEOBOX 2B; PHOX2B. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/603851
78. OMIM® Online Mendelian Inheritance in Man. #176270 PRADER-WILLI SYNDROME; PWS. Johns Hopkins University; last edited 16/07/2025. https://omim.org/entry/176270. Accessed 29 October 2026.
79. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 614962 — LEPTIN DEFICIENCY OR DYSFUNCTION; LEPD. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/614962
80. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 614963 — LEPTIN RECEPTOR DEFICIENCY; LEPRD. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/614963
81. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 609734 — OBESITY, EARLY-ONSET, WITH ADRENAL INSUFFICIENCY AND RED HAIR; OBAIRH. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/609734
82. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 600955 — PROPROTEIN CONVERTASE 1/3 DEFICIENCY. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/600955
83. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 155541 — MELANOCORTIN 4 RECEPTOR; MC4R. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/155541
84. Online Mendelian Inheritance in Man, OMIM®. MIM Number: 603128 — SIM bHLH TRANSCRIPTION FACTOR 1; SIM1. Baltimore (MD): Johns Hopkins University; 2026 [accessed 2026-10-31]. Available from: https://omim.org/entry/603128
85. European Medicines Agency. Imcivree (setmelanotide): EPAR — product information and procedural steps taken after authorisation. Amsterdam: EMA; 2026 [accessed 2026-10-31]. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/imcivree
86. Online Mendelian Inheritance in Man, OMIM®. mim2gene.txt (MIM 257500, Obesity-hypoventilation syndrome). Baltimore (MD): Johns Hopkins University [accessed 2026-10-31]. Available from: https://omim.org/downloads
87. Rossor T, Yeh EA, Khakoo Y, Angelini P, Hemingway C, Irani SR, et al. Diagnosis and Management of Opsoclonus-Myoclonus-Ataxia Syndrome in Children: An International Perspective. Neurol Neuroimmunol Neuroinflamm. 2022;9(3). doi:10.1212/NXI.0000000000001153. PMID: 35260471. PMC8906188.
88. Leinonen H, Koivisto H, Lipponen HR, Matilainen A, Salo AM, Dimova EY, et al. Null mutation in P4h-tm leads to decreased fear and anxiety and increased social behavior in mice. Neuropharmacology. 2019;153:63-72. doi:10.1016/j.neuropharm.2019.04.023. PMID: 31029587.
89. Liang Y, Dou X, Huang H, Tang Y, Zhang B, Qiu W, et al. p4htm acts as a non-redundant regulator of HIF signaling to govern erythropoiesis and renal homeostasis in zebrafish. Biochem Pharmacol. 2026;250(Pt 2):118035. doi:10.1016/j.bcp.2026.118035. PMID: 42114683.
90. Alomarı O, Bebek O, Turkyilmaz A, Sager SG. Novel compound heterozygous P4HTM variants in a girl with developmental and epileptic encephalopathy: First case report of P4HTM variant-associated epileptic encephalopathy. Seizure. 2025;124:35-8. doi:10.1016/j.seizure.2024.11.009. PMID: 39612909.
91. Maddirevula S, Ben-Omran T, AlMureikhi M, Eyaid W, Arabi H, Alkuraya H, et al. Further delineation of HIDEA syndrome. Am J Med Genet A. 2020;182(12):2999-3006. doi:10.1002/ajmg.a.61885. PMID: 32965080.
92. Lim AM, Tan PL, Visruthan NK, Fong N, Viegelmann GC, Tan YH. HIDEA syndrome: A rare cause of congenital hypoventilation in a premature infant. Pediatr Pulmonol. 2022;57(7):1826-9. doi:10.1002/ppul.25966. PMID: 35546426.
93. Gülcü Üstün NS. A pathogenic P4HTM gene variant in two brothers with autism spectrum disorder. Psychiatr Genet. 2024;34(2):68-9. doi:10.1097/YPG.0000000000000364. PMID: 38441145.
94. Ballard HA, Leavitt OS, Chin AC, Kabre R, Weese-Mayer DE, Hajduk J, et al. Perioperative anesthetic management of children with congenital central hypoventilation syndrome and rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation undergoing thoracoscopic phrenic nerve-diaphragm pacemaker implantation. Paediatr Anaesth. 2018;28(11):963-973. doi:10.1111/pan.13475. PMID: 30251310.
95. Puri S, Yaddanapudi S, Menon P. Peri-operative management of a child with ROHHAD-NET syndrome undergoing neural crest tumour excision. Anaesth Rep. 2022;10(1):e12172. doi:10.1002/anr3.12172. PMID: 35756834. PMC9205752.
96. Van Roessel IMAA, Van Den Brink M, Dekker J, Ruitenburg-van Essen BG, Tissing WJE, van Santen HM. Feasibility, safety, and efficacy of dietary or lifestyle interventions for hypothalamic obesity: A systematic review. Clin Nutr. 2024;43(8):1798-1811. doi:10.1016/j.clnu.2024.05.028. PMID: 38955055.
97. Marpuri I, Ra E, Naguib MN, Vidmar AP. Weight management in youth with rapid-onset obesity with hypothalamic dysregulation, hypoventilation, autonomic dysregulation, and neural crest tumor (ROHHAD-NET): literature search and case report. J Pediatr Endocrinol Metab. 2022;35(4):543-548. doi:10.1515/jpem-2021-0600. PMID: 34954931.
98. Bonnet JB, Ramillon-Cury C, Tournayre S, Sultan A, Avignon A. Management of obesity in an individual with ROHHAD syndrome with semaglutide 2.4 mg/week: a case report. Clin Auton Res. 2025;35(1):153-156. doi:10.1007/s10286-024-01072-x. PMID: 39340692.
99. Granato A, Ryan PM, Wong A, Hamilton JK, Danska JS. Gut Microbiome Alterations Accompany Metabolic Normalization Following Bariatric Surgery in ROHHAD Syndrome. JCEM Case Rep. 2024;2(6):luae091. doi:10.1210/jcemcr/luae091. PMID: 38832003. PMC11146140.
100. Vijayakanthi N, Felner EI, Romero R, Daley TC. Rhabdomyolysis due to rosuvastatin in a patient with ROHHAD syndrome. J Clin Lipidol. 2021;15(6):789-792. doi:10.1016/j.jacl.2021.09.002. PMID: 34600840.
101. Rhythm Pharmaceuticals. IMCIVREE (setmelanotide) injection, for subcutaneous use. Prescribing information. Silver Spring (MD): U.S. Food and Drug Administration; revised 2026 Apr [accessed 2026-10-31]. NDA 213793. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=213793
102. Ibáñez-Micó S, Marcos Oltra AM, de Murcia Lemauviel S, Ruiz Pruneda R, Martínez Ferrández C, Domingo Jiménez R. Rapid-onset obesity with hypothalamic dysregulation, hypoventilation, and autonomic dysregulation (ROHHAD syndrome): A case report and literature review. Neurologia. 2017;32(9):616-622. doi:10.1016/j.nrl.2016.04.008. PMID: 27340018.
103. Hawton KAC, Doffinger R, Ramanan AV, Langton Hewer SC, Evans HJ, Giri D, et al. Rituximab therapy in ROHHAD(NET) syndrome. J Pediatr Endocrinol Metab. 2022;35(8):1102-1106. doi:10.1515/jpem-2022-0085. PMID: 35470643.
104. ClinicalTrials.gov. Search "ROHHAD": NCT02441491 (Treatment of ROHHAD, withdrawn), NCT03135730 (International ROHHAD Registry), NCT02602769 (Whole Transcriptome Profiling and Metabolic Phenotyping in Children With ROHHAD Syndrome). Bethesda (MD): National Library of Medicine; 2026 [accessed 2026-10-31]. Available from: https://clinicaltrials.gov/search?term=ROHHAD
105. INFARMED. Infomed — Imcivree (setmelanotido), procedure EMEA/H/C/5089. Lisbon: INFARMED; 2026 [accessed 2026-10-31]. Available from: https://extranet.infarmed.pt/INFOMED-fo/
106. Narang I, Au CT, Chen ML. Central Sleep Apnea and Hypoventilation Disorders in Children. Sleep Med Clin. 2026;21(3):377-396. doi:10.1016/j.jsmc.2026.05.004. PMID: 42567628.
107. Özcan G, Özsu E, Şiklar Z, Çobanoğlu N. A Rare Cause of Sleep-Disordered Breathing: ROHHAD Syndrome. Front Pediatr. 2020;8:573227. doi:10.3389/fped.2020.573227. PMID: 33330273. PMC7714909.