3q29 Microduplication and Microdeletion

Class of recurrent copy number variants (CNVs) of band 3q29, generated by non-allelic homologous recombination between segmental duplications: the recurrent 3q29 duplication of about 1.6 Mb (main entity), the reciprocal 3q29 deletion (comparative entity) and atypical, overlapping or nested 3q29 duplications; susceptibility loci with incomplete penetrance and variable expressivity for neurodevelopmental disorders.

Literature and regulatory information updated: 26 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Definition and scope

Band 3q29, at the distal end of the long arm of chromosome 3, contains an interval of about 1.6 Mb of unique sequence flanked by blocks of segmental duplications, also called low copy repeats (LCRs). Misalignment of these blocks in meiosis, followed by unequal crossing-over, produces gametes with loss or gain of the intervening segment, resulting in two reciprocal copy number variants (CNVs): the recurrent 3q29 deletion and the recurrent 3q29 duplication [4, 11]. The region is recorded in the ClinGen dosage sensitivity curation as a single entry, ISCA-37443, with coordinates GRCh38 chr3:196,029,183–197,617,791 and GRCh37 chr3:195,756,054–197,344,662; the same region receives a haploinsufficiency score of 3 for the deletion and a triplosensitivity score of 1 for the duplication [7]. The main entity of this article is the 3q29 microduplication syndrome (OMIM 611936; ORPHA:251038), defined as the recurrent 1.6 Mb duplication, GRCh38 chr3:195,998,000–197,623,000, which contains 21 protein-coding genes, three antisense transcripts, three long non-coding RNAs and one microRNA [2]. The 3q29 microdeletion syndrome (OMIM 609425; ORPHA:65286) is treated as a comparative entity, because its much more extensive clinical characterisation serves as a reference for interpreting the duplication and because the only GeneReviews chapter for the region is devoted to the deletion [4]. Atypical 3q29 duplications, overlapping, nested or non-overlapping with the canonical interval, form the third group, which is of interest mainly for laboratory interpretation [5, 6].

Orphanet classifies the duplication as a malformation syndrome and describes it, in the record of 2 July 2026, as a chromosomal anomaly of "unclear clinical significance" [1]; the systematic description of the 3q29 Registry, published in 2020, presents it as a genomic syndrome with reproducible neurodevelopmental phenotypes [2]. The two readings correspond to two levels of evidence: the duplication is consistently associated with developmental problems in clinically ascertained carriers, but case-control studies have not demonstrated statistically significant enrichment and some carriers are healthy [7, 25]. For this reason, the recurrent 3q29 duplication is treated here as a susceptibility CNV with incomplete penetrance and variable expressivity, without declaring it benign or fully pathogenic.

1.2. History of discovery

The deletion was the first to be described. The first cryptic subtelomeric 3q29 deletion was reported by Rossi and colleagues in 2001 and, in 2005, Willatt and colleagues described the recurrent interstitial deletion in six individuals [11]. In these six patients, the deletion was about 1.5 Mb, with boundaries in the same or adjacent BAC clones, spanned 22 genes, including PAK2 and DLG1, autosomal homologues of the X-linked ID genes PAK3 and DLG3, and was flanked by two almost identical LCRs, which from the outset pointed to NAHR as the mechanism; the phenotype was variable despite the almost identical size, with mild to moderate ID and mild facial dysmorphism (long, narrow face, short philtrum, high nasal bridge) [16]. Many CMA platforms predating 2005 did not cover the region, and at least 14 individuals with the deletion were identified with the first bacterial artificial chromosome (BAC) arrays [4].

The reciprocal duplication was published in 2008 by two groups. Lisi and colleagues, at Johns Hopkins (Baltimore), identified by high-resolution karyotype, array comparative genomic hybridisation (aCGH) and single nucleotide polymorphism (SNP) arrays an interstitial duplication of about 1.61 Mb at the distal end of 3q29 in five members of a three-generation family, with mild to moderate ID and microcephaly, and proposed it as a "new syndrome" and the genomic counterpart of the microdeletion [26]. According to OMIM entry 611936, which is based only on this family, the duplication was between 1.61 and 1.76 Mb, with the proximal breakpoint between SNPs rs11926771 and rs6797622 and the distal one between rs11922324 and rs6583248 (NCBI35 coordinates of about 197.1–198.9 Mb), between two almost identical LCRs [14]. In the ClinGen reassessment, the five carriers were the brother and sister, the mother (proband), a maternal aunt and the maternal grandfather; all except the grandfather had mild to moderate ID, obesity, dysmorphic features and microcephaly, and the grandfather had dysmorphic features and had attended special education; another eight CNVs of unknown significance segregated in the family, and the 3q29 duplication was the only significant imbalance segregating with the phenotype [7]. Ballif and colleagues, in the same year, pointed out that all affected members of this family also had a deletion of about 169 kb at 4q13.1, absent in the unaffected members, and therefore hypothesised that the 3q29 duplication was not, by itself, causal [9].

The study by Ballif and colleagues (Signature Genomic laboratory, Spokane, USA) screened 14,698 consecutive individuals with developmental disorders between March 2004 and September 2007, with a BAC array with almost continuous coverage of the 5.7 Mb of 3q29, and refined the breakpoints on an oligonucleotide array with about 13,000 probes in the region. It identified 14 deletions and 19 3q29 duplications; of the 17 refined duplications, there were 12 different sizes, from 200 kb to 2.4 Mb, and only five had both breakpoints in the LCRs that mediate the common deletion, that is, they were true reciprocal duplications [9]. The authors concluded that the clinical significance of the reciprocal duplication was still uncertain and that reciprocal duplications tend to produce milder phenotypes than deletions, which could explain their lower detection frequency [9]. Goobie and colleagues described in the same period four families from Toronto, Edinburgh and Leiden with duplications of 1.9 to 2.4 Mb and a smallest region of overlap of about 1.6 Mb, highlighted developmental delay and recurrent ophthalmological findings and proposed the first management and anticipatory care guidance [27].

In the following years, case reports broadened the spectrum (epilepsy and cerebral palsy with severe ID, myelomeningocele, familial metabolic phenotype, late onset with cortical atrophy) [19, 28–30]. In 2020 the 3q29 Registry published the first systematic description of 31 duplication carriers, compared with 117 deletion carriers and 64 controls [2]; Coyan and Dyer described 11 new cases from nine families, including three prenatal diagnoses [31]; in 2024 a French multicentre cohort brought together 46 patients from 31 families [6]; and in 2025 a review from Tomsk pooled 64 carriers with known coordinates [5]. In parallel, the 3q29 deletion was established as one of the single genetic risk factors of largest effect for schizophrenia [20, 32, 33], which prompted the 3q29 Project research programme, first at Emory University and then at Rutgers University [4, 22].

1.3. Nomenclature

The cytogenomic nomenclature of the International System for Human Cytogenomic Nomenclature (ISCN) for the interval, in the form used by GeneReviews for the deletion, is seq[GRCh37] del(3)(q29) chr3:195 756 054–197 344 665 [4]; the end coordinate differs by 3 bp from that recorded by ClinGen for the ISCA-37443 region in GRCh37 (197 344 662) [7]; the duplication is written, in the microarray report, as arr[GRCh38] 3q29(start_end)x3, with the laboratory's coordinates. The sizes cited vary between 1.6 and 1.76 Mb because they depend on probe density, on whether the adjacent segmental duplications are included and on the genome assembly [4, 13]. Coordinates should always be accompanied by the assembly (GRCh37 or GRCh38) and should not be converted manually.

2. Molecular Genetics and Pathophysiology

2.1. Locus architecture and mechanism of formation

The breakpoints of the recurrent deletion and duplication lie in two blocks of segmental duplications (SDs) with identity above 98%, and non-allelic homologous recombination (NAHR) between paralogous copies of these blocks generates the two reciprocal rearrangements [4, 11]. Because the process is mediated by repeated sequences, the unique sequence lost or gained is the same in unrelated individuals, although the reported size is larger when the laboratory includes the adjacent SDs and varies with array design [4, 8]. The general principle also applies here: deletion and duplication are expected reciprocal products of the same event, but fewer duplications have been clinically delineated because trisomy of a segment is usually better tolerated than monosomy [34].

Yilmaz and colleagues (Genome Medicine, 2023) studied the locus by optical mapping in 161 unaffected individuals from the 1000 Genomes and CIAPM collections, in 16 probands with the deletion and their parents and in two probands with the duplication, and by PacBio HiFi long-read sequencing in 44 unaffected individuals and one trio [11]. They identified 34 structural haplotypes (19 new), out of a known total of 37, with ancestry-dependent distribution: haplotype H8 in 11% of Africans, H5 in 41% of East Asians, H6 in 19% of Europeans and H1 in 26% of South Asians. In 16 of 18 probands (89%), the breakpoints were located in paralogous copies of a 20 kb segment within the SDs, and in one deletion the breakpoint was refined to 374 bp [11]. The authors defined five rearrangement classes: class I deletion, of 1.69 Mb (GRCh38 195.94–197.64 Mb; eight probands); class II deletion, of 1.63 Mb (195.99–197.64 Mb; seven probands); class III deletion, non-recurrent, of about 1.2 Mb, with 19 of the 21 genes and a proximal break within TFRC, with no difference in severity; class IV duplication, of 1.6 Mb (GRCh38 chr3:196–197.6 Mb), reciprocal to the class II deletion and generated by NAHR between paralogous 5 kb segments in blocks SDB and SDC; and class V duplication, of 942 kb (chr3:196.25–197.2 Mb), non-recurrent, with breaks in unique sequence of identity below 80%, compatible with non-homologous end joining or a replicative mechanism [11]. Fifteen of the 16 deletions were de novo; the remaining one had been inherited from a father with schizophrenia, generalised anxiety disorder and ADHD. There was no association between haplotypes or breakpoint classes and phenotype, and none of the known inversions of the region (289 kb and 2 Mb) was present in the parental chromosomes of origin, so inversions do not appear to predispose to the rearrangement and searching for them by genome sequencing has no utility in counselling [11].

2.2. Gene content

The following table brings together, in centromeric to telomeric order, the coding genes of the ISCA-37443 interval, with the Ensembl coordinates (GRCh38), loss-of-function constraint in gnomAD (Genome Aggregation Database) v4.1.1 (LOEUF, upper bound of the confidence interval of the observed/expected ratio of loss-of-function variants; low values indicate greater intolerance), the probability of haploinsufficiency (pHaplo) of Collins and colleagues, brain expression (BrainSpan), the homozygous null mouse phenotype and the level of evidence assigned by Herriges and Purcell as a possible "driver" gene of the deletion phenotype [10, 35].

The count of 21 genes by Pollak and by Yilmaz and that of 22 genes by Herriges differ because of annotation issues (read-through transcripts and antisense genes, such as NCBP2AS2 and TM4SF19-DYNLT2B) [2, 10]. In the deletion, no single gene explains the phenotype: pathogenic variants in a single gene of the region do not cause the 3q29 deletion syndrome [4]. Herriges and Purcell classified six genes as probable drivers (Tier 1: TFRC, UBXN7, FBXO45, PAK2, NCBP2, DLG1) and proposed that compound haploinsufficiency of UBXN7, FBXO45, RNF168 and SENP5 compromises the ubiquitination and SUMOylation pathway [10]. ClinGen considers that UBXN7, DLG1, SENP5, FBXO45 and TFRC meet criteria for predicted haploinsufficiency [7].

For the duplication, the most relevant point is that, in the machine-learning genomic analysis of Collins and colleagues cited by Herriges and Purcell, only PAK2 and UBXN7 are predicted to be triplosensitive, which is consistent with the lower mean severity of the duplication [10]. The ClinGen curation of the PAK2 gene assigns it haploinsufficiency and triplosensitivity scores of 1 ("little evidence") [38].

2.3. Function of candidate genes and animal models

DLG1 (discs large MAGUK scaffold protein 1, also called SAP97; OMIM 601014) encodes a scaffolding protein of the MAGUK family that participates in the trafficking of glutamatergic AMPA and NMDA receptors to the neuronal membrane [4, 39]. PAK2 (p21-activated kinase 2; OMIM 605022) encodes a kinase that mediates neuronal migration, neurite outgrowth, dendritic spine morphogenesis and neuronal differentiation [4, 38]. Pak2 haploinsufficiency in the mouse compromises the synaptic cytoskeleton and produces autism-like behaviour [40]. BDH1 encodes 3-hydroxybutyrate dehydrogenase 1, a mitochondrial enzyme for the interconversion of acetoacetate and (R)-3-hydroxybutyrate [41].

A mouse model of the 3q29 deletion obtained by CRISPR editing shows learning, memory and social behaviour deficits and growth deficit; mice with isolated loss of Dlg1 or Pak2 do not recapitulate the full phenotype [4, 40, 42]. In Drosophila, silencing of NCBP2 aggravates the effects of other genes of the region, and UBXN7 emerges as a central node in the cortical transcriptome [10]. The sources consulted do not describe an animal model of the recurrent duplication.

2.4. Penetrance, expressivity and two-hit models

The penetrance of the duplication is incomplete. Goh and colleagues (European Journal of Human Genetics, 2026) proposed defining penetrance as the probability of manifesting the phenotype due to the variant, excluding the baseline population risk, and applied a Bayesian formula to 83 recurrent CNVs, with pooled cases from eight studies (mostly children with ID or psychomotor developmental delay, PMDD, from high-income countries), gnomAD v4 controls (n = 269,885) and a paediatric ID prevalence of 1.1% [18]. For the 3q29 duplication (18 of 32,587 cases versus 4 of 269,885 controls), ID penetrance was 67% (95% CI 44–91) by the previous method and 29% (13–67) by the new formula; for the deletion (20 of 33,010 versus 6 of 269,885), 59% (39–84) and 22% (11–52) [18]. Both estimates carry the authors' note of complexity in the coordinates or data, with a recommendation of clinical caution, and refer only to ID: ASD, psychiatric disorders and seizures are not included [18]. The authors also point to the 3q29 region as an example in which differences in ancestry or geography between case and control cohorts may contribute to the apparent excess in cases [18]. The 29% estimate coexists with the ClinGen triplosensitivity score of 1, maintained in the 2024 assessment [7].

Expressivity is variable within families. In the Vitale series, two children had the metabolic phenotype of their father, who did not have the duplication, while the duplication was maternal and the mother only had excess weight [5, 30]. In several families, the duplication was inherited from apparently healthy mothers and sisters [5, 13], and a Spanish familial case with moderate ID, craniofacial dysmorphism and musculoskeletal anomalies illustrated variable expressivity among carrier relatives [43]. The authors of the recent series interpret the effect of the duplication, which does not always manifest, as dependent on additional factors: a second genetic event (CNV or nucleotide variant), the polygenic background and, possibly, sex [5, 13, 31]. Bauleo and colleagues invoked a female protective effect to explain unaffected female carriers, but the 3q29 Registry found no sex-dependent susceptibility threshold [2, 13]. In the Kashevarova review, other genetic findings were recorded in 16 families, including 15q11.2 deletion, 17q12 duplication, 8p23.1 duplication, intragenic CNTN5 duplication and unbalanced translocations in six families [5]. In the French cohort, possibly implicated additional genetic findings were identified in 11 patients [6].

The deletion also has unknown penetrance, probably high but below 100%: deletions inherited from parents considered unaffected show, when the parent is assessed with a comprehensive protocol, psychiatric disorders and executive function deficits not previously diagnosed [4, 44].

2.5. Atypical duplications and the critical region hypothesis

Atypical duplications are more frequent than atypical deletions: in the Kashevarova review, 24 of 64 duplication carriers with known coordinates (38%) had atypical duplications, compared with 7 of 64 deletions (11%; p < 0.001) [5]. Tassano and colleagues proposed as the smallest region of overlap a 446.8 kb duplication containing only DLG1 and BDH1, identified in a patient with ID, epilepsy, microcephaly and gastrointestinal problems; the same study described a patient with the common duplication and periventricular grey matter heterotopia nodules [12, 13]. Bauleo and colleagues described an Italian family with a 432.8 kb duplication (GRCh37 196,892,569–197,324,567; DLG1 and BDH1) inherited from the unaffected mother, and concluded that the evidence for a critical region of about 500 kb is questionable, because all the carriers compared had other possible explanations and none had undergone exome or whole-genome sequencing [13]. Kashevarova and colleagues defined two minimal overlapping regions (MORs), MOR1 (arr[GRCh37] 3q29(195751853_196120090), with TFRC, ZDHHC19, SLC51A, PCYT1A, TCTEX1D2, TM4SF19 and UBXN7) and MOR2 (arr[GRCh37] 3q29(197121593_197310451), only BDH1), but considered that whole-gene duplications are usually benign or likely benign and that BDH1 is not predicted to be triplosensitive [5]. In the French cohort, the study of distal duplications smaller than 1 Mb (which almost always contain BDH1 and sometimes DLG1) did not support any critical region: in the families with known inheritance, all were inherited, three from an asymptomatic parent; duplications in this region appear in population databases (4 in 13,759 exomes and 7 in 21,694 gnomAD samples), and the French network of cytogenetics laboratories classifies them mostly as benign or likely benign, whereas classifications in DECIPHER diverge [6].

2.6. Neighbouring genes outside the interval and allelic diseases

Some genes of band 3q29 associated with monogenic diseases lie outside the recurrent interval and are only affected by larger or atypical CNVs. OPA1 (GRCh38 193,593,102–193,697,947) lies about 2.4 Mb proximal to the boundary and RPL35A (197,949,957–197,956,610), associated with Diamond-Blackfan anaemia 5, about 330 kb distal; TNK2, MUC4 and MUC20 are proximal and fall outside ISCA-37443, although some sources include TNK2 in a "canonical region" of 1.6–1.76 Mb, owing to differences in breakpoints and arrays [5, 13, 35]. An atypical distal duplication of 232 kb that included RPL35A was described in a family from Tomsk, inherited from an apparently healthy mother and grandmother [5].

Several genes of the interval cause monogenic diseases through mechanisms other than dosage alteration, and their presence does not imply that the duplication or deletion produces those phenotypes. Knobloch syndrome type 2 (OMIM 618458) results from heterozygous missense variants, almost always de novo, in the kinase domain of PAK2 (p.Glu435Lys, recurrent; p.Thr406Met, p.Asp425Asn and p.Asp372Val, with functional data; p.Gln279Pro, in a fetus), with loss of kinase activity; it manifests as vitreoretinal degeneration with retinal detachment, which may be congenital and bilateral, high myopia, occipital encephalocele or scalp lesions and PMDD or ID with autistic behaviour, and bilateral fetal pleural effusion has been described [37, 38, 45, 46]. These variants are reported in the source only in protein notation, without a transcript. Knobloch syndrome type 1 is a different entity, autosomal recessive, due to biallelic COL18A1 variants [38].

Combined immunodeficiency due to TFRC deficiency (immunodeficiency 46; OMIM 616740; ORPHA:476113) is autosomal recessive and results, in more than 20 patients, mostly of Arab origin (which suggests a founder effect), from the homozygous variant NM_001128148.3(TFRC):c.58T>C p.(Tyr20His), NC_000003.12:g.196075339A>G (historical designation Y20H), which compromises internalisation of transferrin receptor 1; one patient with c.64C>T p.(Arg22Trp), NC_000003.12:g.196075333G>A has been described [36, 47–49]. Haematopoietic stem cell transplantation (HSCT) was curative: eight transplanted patients had resolution, whereas six non-transplanted patients died [50]. Both variants lie within the ISCA-37443 interval [49]. In a 3q29 deletion carrier, a pathogenic variant in trans in a recessive gene of the region, such as TFRC, SLC51A, PCYT1A, DYNLT2B or RNF168, would unmask the corresponding recessive phenotype; this is a general principle of deletion interpretation [34], and the sources consulted do not describe any such case in the 3q29 region. The duplication does not have this unmasking effect.

2.7. Genotype-phenotype correlations

In the recurrent deletion, no genotype-phenotype correlations are known, and the breakpoint class does not alter severity [4, 11]. In the duplication, the Kashevarova review compared all carriers with carriers of recurrent or overlapping duplications larger than 1 Mb and found no significant differences in the frequency of features (for example, PMDD or ID in 61% of all duplications and in 68% of recurrent ones or those larger than 1 Mb; p = 0.549) [5]. The French cohort concluded that recurrent and overlapping duplications are associated with mild neurodevelopmental disorder and that a severe or syndromic presentation should prompt a search for another genetic cause [6]. Larger duplications, of 2.08 and 2.4 Mb, were associated in the Unique association guide with structural ocular anomalies (microphthalmia, iris coloboma, corneal opacity, cataract) not described in the typical duplication [51].

3. Clinical Spectrum and Manifestations by Organ System

3.1. Data sources and ascertainment biases

Clinical knowledge of the 3q29 duplication comes from four types of source, with different biases. Laboratory series and case reports (Lisi, Ballif, Goobie, Tassano, Coyan and Dyer, Bauleo, Kashevarova) describe carriers identified because they had developmental disorders and tend to represent the severe end of the distribution [2, 9]. The 3q29 Registry collects data by online questionnaire completed by parents, with standardised instruments, in a voluntary and almost exclusively white sample, and the families most motivated to participate may have more affected children [2]. The French cohort of Massier brings together carriers identified by CMA in 14 genetics centres, through the network of the Association of French-Speaking Cytogeneticists (ACLF), with clinical data collected by questionnaire to geneticists and a systematic search for second genetic events [6]. The Kashevarova review pools 64 carriers with known coordinates, including affected and healthy relatives, with different denominators for each feature [5]. The figures from these sources are not interchangeable and do not measure the frequency of features in all carriers in the population.

The French multicentre study (December 2018 to October 2022) included 46 carriers from 31 families, including three fetuses, after excluding 18 of the 64 recruited for lack of data; cohort A (33 individuals from 22 families) comprises the recurrent duplication of about 1.6 Mb (23 individuals from 14 families) and overlapping duplications of 1.76–3.03 Mb (10 individuals from 8 families, six due to unbalanced translocation), and cohort B (13 individuals from 9 families) the distal duplications smaller than 1 Mb (129–857 kb) [6]. In cohort A, with known inheritance in 14 families, the duplication was de novo in 3, maternal in 7, paternal in 2 and derived from a paternal balanced translocation in 2; one third of individuals (11/33) had another genetic finding, including pathogenic deletions due to unbalanced translocation (11q24.1q25 of Jacobsen syndrome, 2q37 and 12p13.33p13.32), the 15q11.2 BP1-BP2 deletion, deletions of uncertain significance (17p13.3, 1q44) and VUS in TRRAP and PHF6 [6]. In the complete cohort A, PMDD was reported in 67% (18/27), learning difficulties with special education in 64% (16/25), ID in 20% (5/25) and neuropsychiatric disorders in 43% (9/21), with ASD in two; there was heart disease in four patients and one fetus (with hypoplastic left heart), neonatal hypotonia in 4, unilateral hearing loss in 3 (two with inner ear malformation), obesity in 3 and seizures in 3 [6]. In the subgroup without another known genetic finding (22 individuals from 13 families; negative exome or panel in 5 families and no additional study in 8), PMDD was 69% (13/19), learning difficulties 56% (9/16), ID 19% (3/16), always mild, and neuropsychiatric disorders 29% (4/14); two carriers were asymptomatic, the patients did not share facial dysmorphism and the phenotype was homogeneous within families, as in one family in which the father and four children had the same mild disorder [6]. In the 28 published cases with an apparently isolated duplication reviewed by the authors, severity was more variable (PMDD 13/17, mild to severe ID 12/21), and there were 4 asymptomatic carriers out of 24 [6]. The authors conclude that the recurrent or overlapping duplication is associated with mild neurodevelopmental disorder, with variable expressivity and incomplete penetrance, and that a severe or syndromic presentation justifies a search for another genetic cause [6].

3.2. Prenatal and neonatal period and first year of life

In the 3q29 Registry (31 duplication carriers, 48.4% male, aged 0.3 to 52.2 years, mean 10.0 years; data frozen in April 2019), mean birth weight was 2948 g, with a mean gestational age of 38.1 weeks, compared with 3447 g and 39.2 weeks in the 64 controls (difference in gestational age, p = 0.04); after adjustment for gestational age, sex and race, the deficit was 336 g (p = 0.005), and 376 g in white participants (p = 0.001) [2]. The weight deficit reported in the deletion is 394 g [2, 52]. Postnatal hospital stay was 9.8 ± 14.5 days, compared with 3.8 ± 6.4 days in controls (p = 0.037) [2].

Significant problems in the first year of life were reported in 80.6% (25/31) of carriers, compared with 39.1% of controls: feeding difficulties in 54.8% (17), poor weight gain in 41.9% (13), hypotonia in 38.7% (12) and respiratory distress in 29.0% (9) [2]. In the Kashevarova review, 15 of 56 carriers (27%) and 14 of 31 carriers of recurrent duplications or those larger than 1 Mb (45%) were born small for gestational age [5].

3.3. Development and cognition

Developmental milestones were reached on average 10 to 25 months later than in controls (p < 0.005), but most carriers eventually reached each milestone; mean delay was 10 months in the social-emotional domain, 15 in gross motor, 16 in fine motor and 25 in communication [2]. In Kaplan-Meier analysis, the social smile appeared 3.79 months later (24 carriers versus 63 controls), the first word 13.4 months later (23 versus 59), independent walking 4.56 months later (26 versus 62) and the ability to hold and drink from an open cup 23.29 months later (23 versus 53) [2].

At least one diagnosed learning problem was reported in 71.0% (22/31) of carriers, compared with 4.7% of controls: expressive language delay in 54.8%, global PMDD in 41.9% and receptive language delay in 29.0%; among the 21 carriers over 5 years of age, 33.3% had a mathematics learning difficulty and 33.3% a reading difficulty [2]. PMDD or ID were reported in 45.2% (14/31) [2]. The ID described in the literature is mostly mild to moderate; one case of severe ID with epilepsy and cerebral palsy has been described [2, 28]. In the founding family, all affected members of assessable age had mild to moderate ID [9, 26]. In the Kashevarova review, PMDD or ID were present in 35 of 57 carriers (61%) [5]. The French cohort described a high rate of learning difficulties and a low proportion of ID in apparently isolated duplications [6].

3.4. Neuropsychiatric disorders and social behaviour

In the 3q29 Registry, diagnosed ASD was reported in 38.7% (12/31) of carriers, an estimated risk about 26 times higher than that of the population; anxiety disorder in 32.3% (10), compared with a population lifetime prevalence of 5.7% (p = 1.05 × 10⁻⁹); depression in 16.1% (5); conduct disorder, oppositional defiant disorder and panic attacks in 6.5% (2) each [2]. At least one psychiatric diagnosis, including ASD, was reported in 58.1% of duplication carriers and 48.7% of deletion carriers (p = 0.500) [2]. The overlap between PMDD, ASD, seizures and anxiety was only partial, so the risks are not concentrated in a severe subgroup [2]. In the Kashevarova review, ASD was present in 10 of 39 carriers (26%) and in 3 of 18 carriers of recurrent duplications or those larger than 1 Mb (17%), and ADHD in 2 of 56 (4%) [5]. The duplication was not enriched in the large ASD cohorts analysed by Sanders and colleagues, which may reflect a lack of statistical power or false negatives in the detection of duplications by exome [2].

The Social Responsiveness Scale (SRS) was administered to 15 duplication carriers and 56 controls. The mean total score (T) was 79.1 in the duplication and 45.9 in controls (p = 1.09 × 10⁻¹⁵), 87.2 in carriers with ASD and 73.7 in carriers without ASD (both p < 10⁻⁸ versus controls); social motivation was less affected than in idiopathic ASD (71.1 versus 78.4; p = 0.040), but more affected than in the deletion (71.1 versus 63.5; p = 0.043), which suggests an intermediate profile [2].

Source: Pollak and colleagues, Table 2; all comparisons with controls had p ≤ 0.0003, in regression adjusted for age, race and sex [2]. SD, standard deviation.

On the Child/Adult Behavior Checklist (CBCL/ABCL), administered to 15 carriers, the withdrawn subscale had a mean of 66.20 ± 12.76, compared with 52.37 ± 5.85 in controls (p = 6.33 × 10⁻⁸), in the borderline range, and more than half of carriers, with or without ASD, were in the borderline or clinical range; on the subscales oriented to the Diagnostic and Statistical Manual of Mental Disorders (DSM), the means were 63.0 for ADHD, 64.4 for anxiety and 65.8 for depression in the duplication, 61.7, 61.4 and 63.1 in the deletion and 51.3, 53.2 and 52.3 in controls (duplication and deletion versus controls, p < 0.0005; duplication versus deletion, p > 0.282) [2].

Psychotic disorders are not established as part of the duplication phenotype. In the large studies of CNVs in schizophrenia in European and East Asian populations, only the 3q29 deletion reached significance; the duplication is not among the associated loci [33, 53]. There are isolated reports: a monozygotic twin with bipolar disorder and a recurrent 1.69 Mb 3q29 duplication, identified by genome sequencing in a study of pairs discordant for psychosis, in which the CNV was private to the affected twin [5, 54]; and a Portuguese adolescent with moderate ID and episodes of dysregulation with psychotic symptoms, described in section 10 [55]. A disruptive behavioural profile was described in a patient with the reciprocal duplication from a Santiago de Compostela series [56].

3.5. Neurological manifestations

The frequency of seizures varies with the source. The 2011 Unique guide did not record seizures in carriers of the typical duplication, but did so in 3 of 7 members with larger duplications [51]; the 3q29 Registry reported seizures in 25.8% (8/31), compared with 1.2% in the general population (p < 2.2 × 10⁻¹⁶), without information on seizure type [2]; and the Kashevarova review found epilepsy in 5 of 59 carriers (8%) and in 4 of 34 carriers of recurrent duplications or those larger than 1 Mb (12%) [5]. The difference reflects different methods (self-report of the presence of seizures versus diagnosis of epilepsy in published reports) and small samples. Epilepsy with severe ID and cerebral palsy, focal occipital epilepsy, febrile losses of consciousness and language regression have been described [5, 13, 28].

Structural brain anomalies were described in 3 of 19 carriers with imaging (16%) [5], including periventricular nodular heterotopia [12], myelomeningocele with midline cranial defects in a newborn [29] and progressive cortical atrophy in a 34-year-old woman with late onset [19]. Microcephaly was the most frequent feature in the founding family (4 of 5) and was present in 5 of 8 carriers in the Ballif series, but in only 6 of 45 (13%) in the 2025 review; macrocephaly was described in 3 of 53 (6%) [5, 9]. Hypotonia was reported in 38.7% of infants in the registry [2].

3.6. Growth, weight and metabolism

The weight phenotype of the duplication reverses with age: carriers are born with lower weight, but obesity has been described from childhood to adulthood, which suggests accelerated weight gain at an age not yet determined, because the registry does not collect current weight and height [2]. Obesity was present in 4 of 8 carriers in the Ballif series and in 12 of 57 (21%) in the Kashevarova review (10 of 35, 29%, in recurrent duplications or those larger than 1 Mb) [5, 9]. In the Vitale family, from southern Italy, all three carriers were obese, two had diabetes, two hypertension and two dyslipidaemia [5, 30]. In the 2025 review, diabetes and hypertension were recorded in 3 of 61 carriers (5%) each, and dyslipidaemia in 2 of 61 (3%) [5]. In the deletion, by contrast, the weight deficit persists until adolescence and is reproduced in both mouse models [2, 42]. This mirror pattern is one of the arguments for a dosage relationship with metabolic phenotypes [2, 5].

3.7. Craniofacial dysmorphism

The dysmorphism of the duplication is mild and inconsistent. The OMIM clinical synopsis, based exclusively on the founding family and not updated since 2011, includes obesity, microcephaly, round or long face, short and downslanting palpebral fissures, large eyes, short and bulbous nose, broad nasal bridge, excessive palmar creases, flat foot, low posterior hairline, cognitive delay and mild to moderate ID [14, 26]. It does not include the manifestations described after 2008 (feeding problems, seizures, ASD, anxiety), which makes it an incomplete description of the current spectrum [2, 14]. In the Ballif series, the round face and bulbous nose of the Lisi family were considered possibly familial [9]. Palpebral fissure anomalies were recorded in 6 of 61 carriers (10%) and nasal anomalies in 8 of 58 (14%) in the 2025 review [5]. Craniosynostosis was described in one of eight carriers of the reciprocal duplication [9].

3.8. Other systems

The frequencies for the remaining systems differ greatly between the registry and the case review, and the following table presents them side by side.

Sources: Ballif and colleagues, Table 2 [9]; Pollak and colleagues [2]; Kashevarova and colleagues, Table 4 [5]. A dash indicates that the source does not present the datum.

Ophthalmological findings were highlighted by Goobie and colleagues as recurrent in their four families, which suggested dosage sensitivity for ocular development [27]; in the Unique guide, structural eye anomalies appeared only in duplications larger than the typical one [51]. The heart defects described in the typical duplication were a VSD and a bicuspid aortic valve; atrial septal defect, atrioventricular septal defect and tetralogy of Fallot appeared in larger duplications [51]. A coarctation of the aorta was treated by balloon angioplasty in a boy with an atypical BDH1 duplication [5]. Café-au-lait spots, haemangiomas, melanoma (1/61), chronic pain and recurrent mucosal candidiasis have been described [5, 19]. Recurrent otitis media may be associated with transient conductive hearing loss in a critical period for language [51].

3.9. Adulthood and late onset

Data on adults are scarce. The founding family included adult carriers with mild ID and obesity, and the carrier grandfather had only dysmorphic features and a history of special education [7]. Streata and colleagues described a 34-year-old Romanian woman with a 1.65 Mb duplication, normal development until about 10 years, followed by attention deficit, hyperkinesia, gait and coordination disturbances, mild ID, speech disorder, dysphagia, chronic pain, progressive cortical atrophy and recurrent mucosal candidiasis; at 34 years she was non-ambulatory [7, 19]. Many transmitting parents have a mild phenotype similar to that of their children [6]. The sources consulted do not describe the life expectancy or mortality of duplication carriers.

3.10. 3q29 deletion: comparative clinical spectrum

The GeneReviews entry for the deletion is based on a review of all cases published up to 2015, on 44 individuals from the 3q29 Deletion Registry and on in-depth phenotyping of 32 individuals by the Emory 3q29 Project [4, 15, 23, 52].

Source: GeneReviews, Table 2 [4].

Developmental milestones are delayed by 6 to 12 months; first words appear on average at 23 months, there is receptive language delay in 33%, verbal apraxia in 19% and dysphasia or aphasia in 10%, and only one of 44 individuals was completely non-verbal; hypotonia in the first year affects 34% and independent walking usually appears at 20 months, without pronounced ataxia despite the cerebellar anomalies [4, 23, 52]. About 34% have mild to moderate ID and fewer than 5% severe ID; 66% have an intelligence quotient (IQ) in the normal range, but with neurodevelopmental or psychiatric disability not captured by IQ; the mean absolute difference between verbal and non-verbal IQ is 14 points, and from 6 years of age 61% have a verbal subtest higher than the non-verbal one [4]. Executive function is impaired in almost half and graphomotor skills in most, with difficulties in writing and literacy [4].

The deletion is associated with ASD in 38%, anxiety in 40%, ADHD in 63%, bipolar disorder in 5% and schizophrenia in at least 20%; about half of carriers with ASD also have an anxiety disorder, and even without a diagnosis of autism there is social disability [4, 57]. Psychosis may begin much earlier than in the general population (mean age 20 to 25 years): psychosis in a 5-year-old boy and a 10-year-old girl and attenuated psychosis syndrome in an 8-year-old boy have been described [4, 44, 58, 59]. In 2026, an adolescent in Spain with neurodevelopmental disorders was described in whom early-onset psychosis was the manifestation that led to the diagnosis of a deletion of about 1.6 Mb, without marked dysmorphic features [60]. Seizures affect 13%, are generally mild and respond to standard treatment [4]. The case reports gathered in OMIM illustrate the variability: in the six patients of Willatt, autism, ataxic gait, chest deformity and long, tapering fingers appeared in at least two, and microcephaly, cleft lip and palate, horseshoe kidney, hypospadias, ligamentous laxity, recurrent otitis media and abnormal pigmentation in one each; in the two mother-daughter pairs of Digilio, with a 1.5 Mb deletion, all had PMDD with mild to moderate ID or learning difficulties, speech delay, low birth weight, microcephaly, high nasal bridge and short philtrum; the infant of Li, with a paternal 1.3–1.4 Mb deletion, had primary pulmonary hypertension, patent ductus arteriosus, subvalvular aortic stenosis and gastro-oesophageal reflux, with normal development at 6 months, and the father had had patent ductus arteriosus, pulmonary stenosis and mild developmental delay, with normal cognition as an adult; and the two patients of Quintero-Rivera (a 10-year-old girl and a 15-year-old Hispanic boy, with de novo deletions of 1.6 and 2.1 Mb) had autism, ID, aggression, anxiety and hyperactivity, and one of them bipolar disorder with psychosis, in a family with bipolar disorder in the father and paternal grandfather [16, 59, 61, 62]. The OMIM clinical synopsis of the deletion records low birth weight, growth delay, long narrow face, short philtrum, large, low-set and posteriorly rotated ears, high nasal bridge, thin upper lip, pectus excavatum or carinatum, long tapering fingers, toe clinodactyly, mild to moderate ID, ataxic gait, autism, psychosis, anxiety, hyperactivity and aggression [16].

In other systems, the deletion is associated with pectus excavatum (25%), pectus carinatum (9%), scoliosis (6%), lower (72%) and upper (47%) limb anomalies; with feeding difficulties (60%), failure to thrive (44%), gastro-oesophageal reflux (50%), chronic constipation (41%), dysphagia (12%), hiatus hernia (5%) and chronic diarrhoea (5%); with strabismus (28%, of which about 30% require surgery); with caries (24%), weak enamel (19%) and enamel hypoplasia (10%); with seasonal allergies (16%), food allergies (13%), IgA deficiency (3%) and IgG deficiency (3%); and with patent ductus arteriosus (12%), VSD (5%) and pulmonary valve stenosis (5%) [4]. The dysmorphism is subtle but not random: prominent forehead and broad nose in almost half, prominent nasal tip in about 35% and thin upper lip vermilion in about 25%; facial analysis by Face2Gene distinguished carriers from controls in 87.3% of cases (p = 0.006) [4, 63].

3.11. Direct comparison between duplication and deletion

Two published comparisons give different results, and the difference is methodological. In the 3q29 Registry, with the same questionnaire applied to both CNVs, frequencies were similar in almost all categories, except heart disease [2]. In the Kashevarova review, which compared published duplications with deletion data from different samples and methods, the deletion had more language delay and ocular, dental, cardiac and musculoskeletal anomalies, and the duplication more ID or PMDD [5].

Sources: Pollak and colleagues, Table 4 and text [2]; Kashevarova and colleagues, Tables 4 and 5 [5]; CNV studies in schizophrenia [20, 33, 53]. Table 5 of Kashevarova prints 39% for excess weight, but 10 of 35 corresponds to 29%, the value in Table 4 of the same source.

4. Clinical Recognition and Red Flags

4.1. When to suspect

The 3q29 duplication does not have a clinical picture recognisable on examination, and the diagnosis is almost always a finding of a microarray requested for PMDD, ID, ASD or multiple congenital anomalies [4, 21]. Clinical suspicion directed at this CNV is not realistic; what matters is requesting CMA for the established indications and knowing how to interpret the result. By age, the contexts in which the duplication has been found are as follows: in the prenatal period, as an incidental finding or in fetuses with another explanatory alteration [5, 31]; in the infant, feeding difficulties, poor weight gain, hypotonia, neonatal respiratory distress and low birth weight [2]; in the young child, expressive language delay, global PMDD, seizures and ASD [2]; at school age, reading and mathematics learning difficulties, anxiety and social disability [2]; and in the adolescent or adult, mild ID with obesity, a familial metabolic phenotype or, rarely, late-onset regression [19, 26, 30].

The 3q29 deletion should be considered, in addition to the general indications for CMA, in early-onset psychosis with neurodevelopmental comorbidity, even without dysmorphic features [60, 64], and in a child with failure to thrive, feeding disorder, ADHD and anxiety [4].

4.2. Red flags and urgent signs

In carriers already diagnosed, the signs requiring prompt assessment are a first seizure, developmental or language regression, severe poor weight gain or dysphagia with risk of aspiration and, in the deletion, the emergence of prodromal or psychotic symptoms, especially after starting a psychostimulant [2, 4, 23]. A severe or syndromic presentation in a duplication carrier is, by itself, a diagnostic red flag: it requires a search for another genetic cause before attributing the picture to the CNV [6]. Suicidal ideation and severe aggression towards others, described in the Portuguese case, require urgent child and adolescent psychiatric assessment [55].

4.3. Newborn screening

There is no newborn screening for the 3q29 duplication or deletion. In the Norwegian MoBa study, which genotyped 12,252 trios of newborns and parents, no 3q29 duplication was observed and one deletion was identified in a newborn [65].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnostic definition

There are no clinical diagnostic criteria. The 3q29 duplication syndrome is defined by the presence of a heterozygous duplication (three copies) spanning the recurrent ISCA-37443 interval [2, 7], and the deletion by heterozygous loss of the same interval [4]. Significantly larger or smaller duplications may have different clinical significance and should be interpreted according to their gene content [4, 5].

5.2. Tests, in order

CMA, by oligonucleotide (aCGH), SNP or combined arrays, is the first-line test in children with unexplained PMDD or ID, ASD or multiple congenital anomalies, with a diagnostic yield of 15–20%, compared with about 3% for G-banded karyotype; it has been clinically available since 2004 [8, 21]. Karyotype is reserved for evident chromosomal syndromes, a family history of chromosomal rearrangement or multiple miscarriages [21]. High-resolution karyotype detects deletions of 3–5 Mb and duplications larger than about 5 Mb, whereas most CMAs detect CNVs from 100 kb [8]; the 1.6 Mb 3q29 CNV is not visible on conventional karyotype [4].

PCR, polymerase chain reaction; MLPA, multiplex ligation-dependent probe amplification.

5.3. Limits of the tests

CMA does not detect nucleotide variants, insertions-deletions smaller than about 1000 bp, repeat expansions, balanced rearrangements or methylation defects [8]. It does not determine the genomic location of a duplicated sequence: most duplications are in tandem, but a minority are inverted duplications or insertions elsewhere, and only FISH or genome sequencing clarify location and orientation [8]. This limitation is relevant in the 3q29 region, because the duplication has been described in association with a 6p25 deletion due to a paternal unbalanced translocation, with an 8.5 Mb 3q28q29 duplication due to a complex three-chromosome translocation in the paternal grandmother and, in the French cohort, with unbalanced translocations in seven of the ten carriers of duplications smaller than 1 Mb with additional analysis [5, 66, 67]. SNP arrays also detect uniparental isodisomy and are more sensitive than aCGH for mosaicism, although they do not indicate its percentage well [34]; an attenuated signal on CMA may suggest mosaicism, as in a case of an atypical 1.27 Mb 3q29 deletion in which mosaicism was not confirmed by an independent method [68]. Analysis of leucocyte DNA does not detect exclusively germline mosaicism [4]. Microarray coordinates are relative, depend on probe resolution and position, and the calculated size may differ from the expected one because of the SDs at the breakpoints [5, 8]. Exome sequencing and some research approaches have a high false-negative rate for duplications [2]. A negative exome or panel does not exclude the CNV, and neither does a CMA that did not cover the 3q29 region (platforms predating 2005) [4].

5.4. Classification and reporting

The 2020 technical standards of the American College of Medical Genetics and Genomics (ACMG) and ClinGen classify constitutional CNVs into five classes by semiquantitative scoring: pathogenic (≥0.99 points), likely pathogenic (0.90–0.98), of uncertain significance, VUS (−0.89 to 0.89), likely benign (−0.90 to −0.98) and benign (≤−0.99) [69, 70]. CNVs associated with consistent phenotypes in several publications, "with well-documented penetrance and expressivity, even if reduced and/or variable", may be pathogenic, and the classification of the variant is made independently of its contribution to the diagnosis of the specific patient; for a CNV of uncertain significance, the report should list the validated genes of the interval [69]. A CNV inherited from an unaffected parent is, as a rule, likely benign, unless it causes a known disease with reduced penetrance, and duplication of a gene whose mechanism is haploinsufficiency is a criterion for uncertain significance [8].

Because the ISCA-37443 region has a triplosensitivity score of 1 and is not an established triplosensitive region, the recurrent duplication does not receive automatic points for overlap with a dosage-sensitive region, which explains the heterogeneity of classification between laboratories [7, 69]. In ClinVar, in 2019–2020, of the 19 submissions for the 1.6 Mb duplication, 13 (68%) were pathogenic, five of uncertain significance and one conflicting [2]. The deletion, with haploinsufficiency 3, is pathogenic [7]. Nested duplications of DLG1 and BDH1 or of BDH1 alone have been classified as VUS, and a 486 kb proximal duplication with TFRC to UBXN7, inherited from a healthy mother, was classified as likely pathogenic by the authors [5, 13].

The report of a 3q29 duplication should state the genome assembly and coordinates, the minimum and maximum size, the copy number, the genes included, the ACMG/ClinGen classification with the criteria used, reference to the ClinGen curation and to incomplete penetrance, and the recommendation of parental testing and genetic counselling [2, 69]. A 3q29 duplication should never be written in Human Genome Variation Society (HGVS) nucleotide notation; the nomenclature is ISCN, for example arr[GRCh37] 3q29(195633970_196120090)x3 mat, in the form used by Kashevarova for a maternal proximal duplication [5].

5.5. Practical algorithm

The Miller consensus algorithm, reproduced in the Oxford Desk Reference, starts with CMA and, in the presence of a CNV of uncertain significance or incomplete penetrance, turns to DECIPHER (gene content, phenotype of other patients, typical inheritance) and to parental testing. Parental results distinguish a de novo CNV, a CNV inherited from an affected parent, a CNV inherited from an unaffected parent (reduced penetrance) and a parent carrying a balanced rearrangement (increased recurrence risk); the algorithm cautions about incomplete penetrance, unmasking of recessive alleles, imprinting, amplification of a parental duplication and complex inheritance [34]. Applied to the 3q29 duplication, the algorithm can be summarised as follows: (1) CMA confirms a copy gain at 3q29; (2) comparison of the coordinates with ISCA-37443 to distinguish a recurrent, overlapping, nested or non-overlapping duplication; (3) testing of both parents by CMA and, if an insertion or another imbalance is suspected, FISH or karyotype; (4) clinical assessment of the carrier parent; (5) if the phenotype is severe, syndromic or discordant with the duplication spectrum, exome or genome and review of other CNVs; (6) genetic counselling with explicit incomplete penetrance [2, 6, 8, 34].

5.6. Prenatal diagnosis

Prenatal CMA may detect the duplication or deletion in a pregnancy without known risk, but prenatal results do not reliably predict the phenotype [4]. Coyan and Dyer described three prenatal diagnoses of the duplication: two fetuses had another explanatory alteration (pathogenic EFTUD2 variant with mandibulofacial dysostosis and microcephaly; trisomy 21 in a recurrent pregnancy loss) and the third had anencephaly, a phenotype considered incompatible with the reduced penetrance of the duplication [5, 31]. Some laboratories do not routinely report, on prenatal CMA, CNVs with substantial incomplete penetrance, because the outcome is uncertain and the CNV is unlikely to explain the ultrasound findings, and there is no clear consensus on cascade or prenatal testing for these CNVs [34]. Goh and colleagues argue that more accurate penetrance estimates may change the decision to report a CNV prenatally; the 3q29 duplication, with an ID penetrance of 29%, is not in the group of CNVs with penetrance close to 0% [18]. A familial 8.5 Mb 3q28q29 duplication, due to a complex translocation, was identified in a fetus with slightly increased nuchal translucency who was born at term with normal development; it is a rearrangement different from the recurrent duplication [67].

6. Differential Diagnosis

6.1. Principles

The differential diagnosis of the 3q29 duplication is, in practice, the aetiological differential diagnosis of PMDD, mild ID, ASD and epilepsy, because all the manifestations described in the duplication and deletion occur in other genomic disorders [4]. There are three distinct tasks: distinguishing the recurrent duplication from other rearrangements of band 3q29 (reciprocal deletion, atypical duplications, larger duplications due to unbalanced translocation); comparing it with other recurrent neurosusceptibility CNVs, which have different penetrances and their own surveillance implications; and, in a carrier with a severe or syndromic presentation, searching for a coexisting second genetic cause, which may explain the phenotype better than the duplication [5, 6].

6.2. Comparative table

AD, autosomal dominant; AR, autosomal recessive.

6.3. Main differentials in prose

The most important distinction is between the duplication and the reciprocal deletion, which share coordinates, genes, mechanism and much of the neurodevelopmental spectrum, but differ in the risk of schizophrenia, the direction of the weight phenotype, the frequency of heart disease and the proportion of de novo cases [2, 4, 20]. The distinction is made only by copy number on CMA, and the laboratory must confirm the direction (gain or loss), because the implications for psychiatric surveillance are different.

The second distinction is between the recurrent duplication and atypical duplications. Nested duplications or those not overlapping the canonical interval should not automatically receive the designation "3q29 duplication syndrome", because the evidence that they are causal is even more limited than that for the recurrent duplication [5, 6, 13]. The third is the presence of another imbalance: when the 3q29 duplication appears on a CMA with another terminal gain or loss, or when it is smaller than 1 Mb, an unbalanced translocation should be sought in the proband and a balanced rearrangement in the parents [5, 34, 66].

Among the other neurosusceptibility CNVs, the 3q29 duplication has an ID penetrance (29%) higher than that of several much more frequent CNVs, such as the 15q11.2 deletion (1.3%), the 16p13.11 duplication (1.1%) and the 22q11.2 duplication (4.0%), and close to that of the 15q13.3 deletion (22%) and the 3q29 deletion (22%), but with wide confidence intervals and the authors' caution about data complexity [18]. Finally, in a carrier with severe ID, refractory epilepsy, major malformations or marked dysmorphism, the coexistence of a monogenic disease or another CNV is frequent in published series, and the French cohort recommends additional genetic analysis in these cases [5, 6, 31].

7. Treatment and Risk Management Strategies

7.1. General principles

There is no specific treatment, approved or in trials, for the 3q29 duplication or deletion; the only study registered on ClinicalTrials.gov is observational [4, 22]. Management is symptomatic, directed at each manifestation, and relies on three sources of different nature. For the deletion, there are clinical care recommendations from the Emory 3q29 Project, based on in-depth phenotyping of 32 individuals, incorporated into GeneReviews [4, 23]. For the duplication, there are the preliminary management and anticipatory care guidance of Goobie and colleagues (2008), known through the Unique association guide, and the authors' recommendations of the 3q29 Registry (2020), which are not formal guidelines [2, 27, 51]. Applying recommendations designed for the deletion to the duplication is an extrapolation, justified by the overlap of the neurodevelopmental spectrum, but limited by the differences in psychiatric, cardiac and weight risk between the two CNVs [2, 5]. No drug is indicated for the 3q29 duplication or deletion: all doses given are those of the indications approved in the official labels for the corresponding comorbidity (ADHD, anxiety, irritability, psychosis, epilepsy, reflux, constipation), and their use in this population is a declared extrapolation.

7.2. Non-pharmacological interventions by manifestation

7.3. Pharmacotherapy of comorbidities

7.3.1. ADHD

In the deletion, because of the risk of psychosis, the Emory 3q29 Project recommends using stimulants for ADHD only with caution and with explicit monitoring for emerging psychotic symptoms, and suggests considering drugs less likely to precipitate frank psychosis, such as bupropion or atomoxetine, as alternatives to amfetamines or methylphenidate [4, 23]. In the duplication, no risk of schizophrenia has been demonstrated [33, 53], ADHD was recorded in 4% of published cases and ADHD scores on the CBCL were similar to those of the deletion [2, 5]; applying this precaution to the duplication is an extrapolation, prudent especially when there is a personal or family history of psychosis.

The US label of methylphenidate (Ritalin®) states that stimulants may cause new psychotic or manic symptoms, with a frequency of about 0.1% with stimulant and 0% with placebo in the pooled analysis of short trials, and that it may exacerbate pre-existing psychosis [77]. The Portuguese summary of product characteristics (SmPC) of methylphenidate (Rubifen®) goes further: it contraindicates it in patients with a diagnosis or history of psychotic symptoms, schizophrenia, mania, severe depression, suicidal tendencies or poorly controlled severe bipolar disorder, and in haemodynamically significant congenital heart disease, and requires psychiatric monitoring at each dose adjustment, then at least at 6-month intervals and at every visit [78]. The cardiac contraindication is relevant in the deletion, with congenital heart disease in 25% [4].

Bupropion, however, has an important limitation in this population: it is not approved for ADHD, its paediatric safety and efficacy have not been established, the risk of seizures is dose-dependent and the drug is contraindicated in patients with a seizure disorder, which is pertinent in a CNV with seizures in 13% (deletion) to 25.8% (duplication) of carriers [2, 4, 79]. The sources consulted do not provide a paediatric bupropion dose for ADHD, and its use in children is off-label and without a labelled dose. Atomoxetine has a boxed warning for suicidal thoughts and behaviours in paediatric patients and requires screening for bipolar disorder before initiation [80].

7.3.2. Anxiety, irritability and psychotic symptoms

For anxiety, CBT is the first-line intervention recommended in GeneReviews, with medication when needed [4]. Escitalopram has a US Food and Drug Administration (FDA) indication for generalised anxiety disorder from 7 years, and sertraline has a paediatric FDA indication only for obsessive-compulsive disorder; the use of sertraline in other paediatric anxiety disorders is off-label in the USA [81, 82]. For irritability and aggression associated with ASD, risperidone and aripiprazole have FDA indications in children, whereas in the European Union (EU) aripiprazole has no recommended posology in this indication and risperidone is authorised in Portugal for short-term symptomatic treatment (up to 6 weeks) of persistent aggression in conduct disorder in children over 5 years and adolescents with below-average intellectual functioning [83–86]. For early-onset psychosis in the deletion, the schizophrenia indications in adolescents apply, which also diverge between jurisdictions (table 7.4). A retrospective review of 1196 records from a mental health clinic for people with developmental disorders identified 24 carriers of schizophrenia risk variants, 11 of whom were treated with antipsychotics, which were reasonably effective and tolerated; none had a 3q29 CNV [87]. The sources consulted do not describe antipsychotic response data in series of duplication carriers; the Portuguese case in section 10 is the only detailed report [55].

7.3.3. Epilepsy

There is no preferred antiepileptic in the deletion or the duplication [4]. Levetiracetam is used as an example of a broad-spectrum drug with labelled paediatric posology; its label describes non-psychotic behavioural changes in 38% of paediatric patients aged 4 to 16 years (19% with placebo) and psychotic symptoms in 17% of patients aged 1 month to under 4 years (5% with placebo), 2% of those aged 4 to 16 years (2% with placebo) and 1% of adults (0.2% with placebo), effects to be monitored in a population with neuropsychiatric risk [88]. The choice of antiepileptic follows the seizure type and the electroclinical syndrome, not a rule specific to the CNV.

7.3.4. Reflux and constipation

Omeprazole and macrogol are examples of drugs with paediatric posology by weight or age in the labels consulted [89, 90]. Over-the-counter macrogol 3350 in the USA (MiraLAX®) is labelled only from 17 years, so paediatric use is off-label in that country, whereas the British SmPC of Movicol® Paediatric authorises treatment of chronic constipation from 1 year (2 years for the chocolate-flavoured version) and of faecal impaction from 5 years [90, 91]. The joint guideline of the European and North American Societies for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN/NASPGHAN) places polyethylene glycol as first line in paediatric functional constipation [76].

7.3.5. Dosing table

MAOIs, monoamine oxidase inhibitors. The doses are those of the labels consulted for the indication stated and have not been studied in carriers of 3q29 CNVs.

7.4. Comparative regulatory status

No medicine has a marketing authorisation for the 3q29 duplication or deletion, in the USA or the European Union, and there are no orphan medicinal product designations for these CNVs in the sources consulted. The relevant regulatory differences concern the paediatric indications of symptomatic drugs.

7.5. Clinical studies and therapeutic evidence

There are no interventional trials registered on ClinicalTrials.gov for the 3q29 duplication or deletion; this search did not cover the EU clinical trials register or ISRCTN, and a negative result in one registry does not prove the absence of studies in other registries [22]. The efficacy of symptomatic drugs in this population is inferred from trials in general populations with the same comorbidity, without direct comparison and without subgroup data for carriers. The frequency of psychiatric effects of stimulants in a group with a high baseline risk of psychosis, such as the deletion, may differ from that observed in short-term trials in patients without this risk, and the absence of cases in small samples is not equivalent to zero risk.

7.6. Developments in the last twelve months

The literature search for the period between October 2025 and October 2026 did not identify any new cohort study dedicated to the 3q29 duplication, and the sources consulted do not record interventional trials or regulatory processes relating to these CNVs [22]. The relevant publications were of five types. In human genetic studies, Goh and colleagues recalculated the ID penetrance (duplication 29%, deletion 22%), published in October 2025 [18], and the first study of rare CNVs in schizophrenia in East Asian populations (20,903 cases and 23,258 controls) confirmed the 3q29 deletion as a risk locus with an odds ratio of 31.2 (p = 5.74 × 10⁻⁵; carrier frequency of 0.00118 in cases and 0.00015 in controls), with no significant difference from the European odds ratios, and did not associate the duplication [53]. In a gene-by-gene evidence review, Herriges and Purcell proposed six driver genes of the deletion and predicted that only PAK2 and UBXN7 are triplosensitive [10]. In case reports, early-onset psychosis as a sentinel manifestation of the deletion in a Spanish adolescent [60], an atypical 1.27 Mb deletion with suspected mosaicism in a patient with ASD and deficiencies of several micronutrients [68] and a Chinese child with ADH5/ALDH2 deficiency and a concomitant 3q29 duplication, in whom the duplication was considered responsible mainly for the neurological delay [75], were described. In methods, a remote phenotyping protocol was validated in the deletion [94]. In guideline syntheses, a scoping review of 17 European clinical guidelines published since 2010 on genetic testing in neuropsychiatric disorders showed wide variation: seven ASD guidelines recommend testing mainly if there is ID or dysmorphism, and the Royal College of Psychiatrists (2023) is the only one that recommends looking for rare CNVs in schizophrenia with co-occurring features, citing the 22q11.2, 3q29 and 16p11.2 deletions as clear risk factors [64].

7.7. Agents to avoid, precautions and special situations

In the deletion, psychostimulants should be used with caution and monitoring for psychosis; in Portugal, methylphenidate is contraindicated in patients with a history of psychotic symptoms [23, 78]. Bupropion is contraindicated in patients with a seizure disorder, which requires confirming the absence of epilepsy before considering it as an alternative to stimulants [79]. In the duplication, the tendency to excess weight in late childhood and adulthood, with diabetes and dyslipidaemia in some families, calls for metabolic surveillance with atypical antipsychotics, which cause weight gain [5, 86]. Levetiracetam is associated with behavioural and psychotic effects that may be confused with the neuropsychiatric course of the CNV [88].

The sources consulted do not describe pregnancy in female carriers of the duplication or deletion, nor specific anaesthetic or perioperative complications; the precautions of each drug described in the pregnancy and lactation sections of the respective labels apply, as do those of the comorbidities present (heart disease, reflux, hypotonia). Transition to adult care includes transfer from child and adolescent psychiatry to adult psychiatry, continued surveillance for psychotic symptoms in the deletion, learning support and vocational guidance, education about sexual development and information about the probability of transmitting the CNV [4, 51].

7.8. Guideline recommendations and divergences

8. Reproductive Options and Family Counselling

8.1. Inheritance pattern and proportion of inherited cases

Both CNVs are transmitted in an autosomal dominant manner: each child of a carrier has a 50% probability of inheriting the CNV [4, 51]. The difference between them lies in their origin. The deletion is typically de novo: in the review by Cox and Butler, the deletion was inherited in 2 of 28 families with parental testing (7%), or in 10 of 32 deletions with known inheritance in the count cited by ClinGen, and in the Yilmaz study 15 of 16 deletions were de novo [4, 7, 11]. The duplication is mostly inherited: in the Ballif series, the three reciprocal duplications with parental samples were all inherited and only 2 of 10 3q29 duplications of any size were de novo; in the Coyan and Dyer series, 1 of 9 families with the recurrent duplication had a de novo duplication and the remaining eight had unknown inheritance; in the French cohort, some were de novo, but most had been inherited from a parent with a similar mild phenotype [6, 7, 9]. The exact proportion of inherited duplications is unknown, partly because many families do not undergo parental testing because of cost or refusal [2, 55]. In the families with atypical duplications described by Kashevarova and by Bauleo, the inherited duplications were of maternal origin, but the number of families does not allow a parent-of-origin bias to be assessed [5, 13].

8.2. Recurrence risks by scenario

8.3. Cascade testing and assessment of relatives

The 3q29 Registry recommends that parental testing and genetic counselling be the standard of care for all families of a duplication carrier, because the carrier status of a parent is critical for counselling and for reproductive choices [2]. Parents should be tested with a method that detects the proband's CNV and, when they do not have it, investigation for a balanced rearrangement should be considered [4]. Clinical assessment of the carrier parent is necessary to interpret causality: an "unaffected" parent may have undiagnosed phenotypes, as shown by the comprehensive assessment of a parent transmitting the deletion, with schizoaffective disorder, ADHD, panic, social anxiety and executive function deficits not previously diagnosed, with a mean IQ of 94 [34, 44]. When a parent has the deletion, the proband's older and younger siblings should be tested to identify those who would benefit from developmental surveillance and neuropsychiatric monitoring [4]. In the duplication, the same principle applies to developmental surveillance, but there is no consensus on cascade testing in CNVs with incomplete penetrance [34].

A phenotypic discrepancy between a parent and a child carrying the same CNV may result from incomplete penetrance, complex inheritance with other susceptibility loci, unmasking of a recessive allele, amplification of a duplication or extension of a deletion [34]. In the Vitale family, the two children had the metabolic phenotype of the non-carrier father, and the duplication was maternal [5].

8.4. Prenatal diagnosis and preimplantation genetic testing

Prenatal diagnosis by chorionic villus sampling or amniocentesis and preimplantation genetic testing (PGT) are technically possible when a parent has the CNV or after a child with a de novo CNV [4, 34]. Prenatal results do not predict the phenotype, and views on the use of prenatal testing differ between professionals and families; the decision is personal [4]. When the CNV is de novo, testing may be offered for reassurance; when a parent is a carrier without a phenotype, whether the CNV is the cause of the proband's condition should be weighed before offering prenatal diagnosis [34]. For the duplication, with an ID penetrance estimated at 29% and a wide confidence interval, and with frequent healthy carriers, the information to be conveyed is that of an increased risk, without the possibility of predicting severity [6, 18]. The ideal time to discuss the risk and options is before pregnancy, and counselling should be offered to young adult carriers or those at risk [4].

8.5. Counselling of minors and people with ID

Testing of asymptomatic minor siblings is justified when the result changes developmental or neuropsychiatric surveillance [4]. Goobie's guidance includes, in adolescence, education about sexual development and the probability of transmitting the duplication [51]. In counselling people with learning difficulties, sufficient time must be set aside and appropriate consent obtained for any intervention [34].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency in the population and what each figure measures

The prevalence of the deletion is about 1:30,000–1:40,000, based on Iceland and the UK Biobank; in other populations it is unknown [4]. For the duplication, population estimates range from about 1:75,000 to 1:8000, and clinical cohorts show frequencies of 1:1969 to 1:5000 [2, 5]. The excess in cases is consistent, but has not reached statistical significance in any single case-control study [17, 25]. In the complete UK Biobank (about 420,000 participants of white British or Irish ancestry), the 3q29 deletion was too rare for analysis of cognitive and functional outcomes [98]. These figures measure frequencies in samples with different designs and should not be presented as national prevalence. At least 50 carriers of the recurrent duplication had been published by 2021, and 35 cases were counted in the 2020 Portuguese report [4, 55].

9.2. Ancestry, geography and founder populations

The published duplication series come mainly from the USA (Baltimore, Spokane, Cincinnati and the 3q29 Registry, from Emory and Rutgers), Canada, the United Kingdom, the Netherlands, France, Italy, Spain, Romania, Russia, Turkey and China [5, 6, 9, 12, 13, 19, 26–28, 31, 74, 75]. The 3q29 Registry is overwhelmingly white (87.1% of duplication carriers; no Black participants), which limits generalisation [2]. The sources consulted do not describe founder populations for the duplication, which is consistent with its recurrent origin through NAHR; the haplotype structure of the locus does, however, differ between populations, without a demonstrated association with formation of the rearrangement [11]. For the deletion, the risk of schizophrenia has been confirmed in European and East Asian populations, with no significant difference in odds ratios [53], and Mulle's original study included a sample of Ashkenazi Jews (245 cases and 490 controls) [32]. The founder variant TFRC c.58T>C, in patients mostly of Arab origin, belongs to the recessive immunodeficiency and not to the CNV [47].

In Portugal, the only publication found in the sources consulted is the Coimbra case described in section 10 [55]. The sources consulted do not include data from Brazil.

9.3. Special populations

Data in adults and older people are scarce and based on transmitting parents and isolated reports [7, 19]. Institutionalised children and families with severe psychosocial adversity, as in the Portuguese case, accumulate factors that confound attribution of the phenotype to the CNV [55]. Monozygotic twins discordant for psychosis showed a 3q29 duplication private to the affected twin, which illustrates the possibility of post-zygotic events [54].

9.4. Reference networks and patient associations

The relevant European reference network is ERN-ITHACA, dedicated to rare malformation syndromes, ID and other neurodevelopmental disorders; the sources consulted did not identify a guideline from this network specific to 3q29 CNVs [24]. Patient resources include Unique (Rare Chromosome Disorder Support Group, Oxted, United Kingdom; www.rarechromo.org), with a guide on 3q29 duplications, Chromosome Disorder Outreach (chromodisorder.org), the 3q29 Foundation (www.3q29.org) and the 3q29 Project and Registry of Rutgers University (https://3q29.com), which recruits carriers of the deletion and duplication for research [4, 22, 51]. DECIPHER (https://decipher.sanger.ac.uk) allows comparison of gene content and phenotypes of other patients [34].

10. Illustrative Clinical Case

10.1. Portuguese adolescent with 3q29 microduplication and psychiatric presentation

In 2020, Reis and colleagues, from the Child and Adolescent Psychiatry and Paediatric Neurology departments of the Centro Hospitalar e Universitário de Coimbra, described a 14-year-old girl with a 3q29 microduplication and an unusual psychiatric presentation [55]. The patient was placed in residential care at 9 years of age because of parental neglect; the parents were non-consanguineous, the father had a substance use disorder and a brother had macrocephaly, hirsutism and cognitive impairment. At 9 years she presented with oppositional behaviour, aggression towards others, learning difficulties with grade retention, nystagmus, simple motor tics and caries. Ophthalmology identified bilateral, symmetrical, congenital horizontal and torsional nystagmus, with preserved visual acuity. In paediatric neurology, length was at the 10th–25th centile and weight and head circumference at the 50th centile; there was a long face, short nose with a bulbous tip, thick eyelashes, upslanting palpebral fissures and a haemangioma-like lesion on the right thigh, persistent motor tic disorder, general incoordination and dysdiadochokinesia. Brain MRI showed a 9.5 mm pineal cyst, with no other changes, stable one year later (Figure 1 of the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC7138485/figure/FIG1/). Full blood count, biochemistry and metabolic, autoimmune, thyroid and copper studies were normal, and two awake and sleep video-electroencephalograms (EEG) were normal [55].

aCGH revealed a 3q29 microduplication, whose coordinates the article does not give; the brother with dysmorphic features had the same alteration, and only the mother agreed to be tested, with no alteration in this chromosome, so paternal origin is presumed but was not demonstrated [55]. At 10 years, the Wechsler Intelligence Scale for Children (WISC-III) gave a full-scale IQ of 48, compatible with moderate intellectual developmental disorder, and school adaptations, speech therapy and psychomotor therapy were instituted. Caregivers reported persecutory ideation. At 11 years, after court-ordered contact with her parents, she had a major depressive episode with irritability, agitation, anhedonia, secondary daytime enuresis and suicidal ideation, treated with sertraline 50 mg/day and psychotherapy, with complete remission in 6 months [55].

In the following two years, episodes of emotional dysregulation appeared, up to three per week, with an altered but alert state of consciousness, drooling, loss of bladder control, self-harm and aggression towards others and uncontrollable crying for 1 to 2 hours, always in the presence of someone and without tongue biting; the EEG showed no paroxysmal activity. At 12 years, risperidone 1 mg/day was started, with a moderate response (one episode per week), and aripiprazole 15 mg/day was then added (one episode per month); in the face of renewed recurrence, the medication was changed to paliperidone 6 mg/day and valproate semisodium 1000 mg/day, with behavioural techniques. During the crises there was disorganised speech, with derailment, incoherence and perseveration, grossly disorganised behaviour and possible command auditory hallucinations, with a good response to chlorpromazine 25 mg intramuscularly or orally [55]. The authors discussed, according to DSM-5, brief psychotic disorder (excluded by duration of less than one day), bipolar II disorder (no hypomania), post-traumatic stress disorder, disruptive mood dysregulation disorder and attenuated psychosis syndrome (a criterion proposed for research), without arriving at a classifiable diagnosis. Non-pharmacological interventions included emotional support, behavioural techniques for caregivers and teachers, suspension of parental visits, a small class with a special education teacher and a day hospital with weekly group therapy; at the last assessment, the patient had fewer than one episode per week and was attending school on paliperidone and valproate [55].

The case shows the difficulty of attributing a severe psychiatric picture to the duplication in a context of severe psychosocial adversity, which the authors acknowledge as a confounding factor, as is the absence of structured interviews [55]. The succession of combined antipsychotics and the introduction of valproate in an adolescent are individual choices in the report and do not constitute a recommendation for duplication carriers; there are no pharmacotherapy guidelines for this CNV, and the Portuguese indication for risperidone in aggression in conduct disorder in children with ID is limited to short-term treatment [86]. Sertraline was used in a paediatric indication that is off-label in the USA [82]. The episodic psychotic symptoms do not, by themselves, demonstrate an association between the duplication and psychosis, which the large CNV studies in schizophrenia did not find [33, 53].

10.2. Italian family with a nested duplication inherited from an unaffected mother

Bauleo and colleagues (University of Calabria) described a 4-year-old boy with ASD and moderate ID, a developmental quotient of 76 at 41 months on the Brunet-Lézine scale, severe language regression at 13 months after acute gastroenteritis, walking at 16 months, stereotypies and sleep disturbance, with a normal EEG and no dysmorphic features; his 2-year-old brother had a mixed developmental disorder with stereotyped behaviour, a developmental quotient of 79, oral hypotonia, recurrent otitis and hypoxic-ischaemic distress in the first week of life [13]. Both had a 432.8 kb 3q29 duplication (GRCh37 196,892,569–197,324,567) with only DLG1 and BDH1, inherited from the unaffected mother, also present in an unaffected 10-year-old sister; the proband also had a paternal 397.8 kb intragenic duplication at 11q22.1, with nine exons of CNTN5. Fragile X testing and the ASD (182 genes) and neurodevelopmental (95 genes) panels were negative, and both CNVs were classified as VUS (pedigree in Figure 1: https://pmc.ncbi.nlm.nih.gov/articles/PMC10094080/figure/mgg32130-fig-0001/) [13]. The authors concluded that the evidence for a minimal critical region with DLG1 and BDH1 is questionable, because all the carriers compared had other possible explanations and none had undergone exome or genome sequencing [13].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

In the duplication, most carriers reach developmental milestones, with a mean delay of 10 to 25 months, and the ID described is mostly mild; learning problems (71.0%), ASD (38.7%), anxiety (32.3%) and seizures (25.8%) in the registry are the main sources of disability [2]. Some carriers have no apparent phenotype, and ID penetrance is 29% (95% CI 13–67) by the new Goh formula [6, 18]. In adulthood, excess weight, diabetes, hypertension and dyslipidaemia have been described in some families [5, 30], and one case of late-onset neurological deterioration with loss of ambulation has been published [19]. The sources consulted do not describe the life expectancy or mortality of carriers, and the absence of these data does not mean that they are normal.

In the deletion, neurodevelopmental and psychiatric conditions account for most of the disability [4]. The risk of schizophrenia is the dominant prognostic factor in adolescence and adulthood: odds ratio of 41.1 (95% CI 5.6–1953.6) in the meta-analysis of 25,314 cases and 62,432 controls, possibly the largest single known risk factor, above the 22q11.2 deletion [20]. The frequency of psychosis or schizophrenia of at least 20% comes from a small sample of young people still at risk, so the lifetime risk is probably higher and the figure should not be read as a lifetime estimate [4]. Seizures are generally mild and responsive to treatment [4].

11.2. Evaluation at diagnosis

11.3. Surveillance

For the deletion, the GeneReviews summary indicates an annual ophthalmological examination, whereas the surveillance table of the same chapter leaves the frequency to the ophthalmologist; the annual examination is adopted, unless the ophthalmologist advises otherwise [4]. For the duplication there is no evidence-based surveillance protocol; the table combines Goobie's 2008 guidance, the registry authors' recommendations and, by declared extrapolation, the deletion recommendations for the systems in which the duplication also has manifestations [2, 4, 27]. The monitoring intervals with methylphenidate (blood pressure, pulse, weight, height and psychiatric status at least every 6 months) are those of the Portuguese SmPC and match section 7 [78].

12. Summary and Concluding Remarks

The recurrent 3q29 duplication is the gain of one copy of an interval of about 1.6 Mb (ISCA-37443; GRCh38 chr3:196,029,183–197,617,791) with 21 to 22 genes, generated by non-allelic homologous recombination between segmental duplications, reciprocal to the 3q29 deletion [7, 11]. Its status is that of a susceptibility CNV: in clinically ascertained carriers it is associated with feeding problems and hypotonia in infancy, language delay, learning difficulties, mostly mild ID, ASD, anxiety, seizures and a tendency to excess weight, but most duplications are inherited from parents with a mild or absent phenotype, the enrichment in cases was not statistically significant in case-control studies, the ClinGen triplosensitivity score is 1 and the ID penetrance was recalculated at 29%, with a wide confidence interval [2, 6, 7, 18, 25]. The reciprocal deletion, typically de novo, is a high-penetrance syndrome, with ADHD, anxiety, ASD, failure to thrive, heart disease, posterior fossa anomalies and a risk of schizophrenia about 40 times higher than that of the population [4, 20].

Diagnosis is by CMA, which does not locate the duplication; a duplication smaller than 1 Mb, associated with another imbalance or in a patient with a severe presentation requires a search for a chromosomal rearrangement or a second genetic cause [5, 6, 8]. The report should state the assembly, coordinates, genes, ACMG/ClinGen classification and incomplete penetrance, and parental testing, with clinical assessment of the carrier, is part of the diagnosis [2, 69]. There is no specific treatment; management combines early intervention, developmental therapies, ASD assessment, seizure surveillance, feeding and gastroenterological support, dental and ophthalmological care and, in adolescence, metabolic surveillance [2, 4, 51]. The caution with psychostimulants, designed for the deletion, applies to the duplication only by extrapolation; in Portugal, the methylphenidate SmPC contraindicates it with a history of psychotic symptoms and with haemodynamically significant congenital heart disease, and bupropion, suggested as an alternative, is contraindicated in epilepsy [23, 78, 79]. Genetic counselling should convey the 50% transmission risk per pregnancy from a carrier and the impossibility of predicting the phenotype, especially in prenatal diagnosis [4, 34]. The main gaps are the absence of population cohorts of the duplication with direct assessment, of data in adults, of ancestral diversity in the cohorts and of any therapeutic trial; the observational study NCT02447861, recruiting until 2028, is the main ongoing source of longitudinal data [2, 22].

13. References

1. Orphanet. 3q29 microduplication syndrome (ORPHA:251038) [Internet]. Paris: INSERM; 2026 [cited 2026 Oct 26]. Available from: https://www.orpha.net/en/disease/detail/251038
2. Pollak RM, Zinsmeister MC, Murphy MM, Zwick ME, Mulle JG. New phenotypes associated with 3q29 duplication syndrome: Results from the 3q29 registry. Am J Med Genet A. 2020;182(5):1152-1166. doi:10.1002/ajmg.a.61540. PMID: 32154651. PMC7785085.
3. Orphanet. 3q29 microdeletion syndrome (ORPHA:65286) [Internet]. Paris: INSERM; 2026 [cited 2026 Oct 26]. Available from: https://www.orpha.net/en/disease/detail/65286
4. Mulle JG, Gambello MJ, Sanchez Russo R, Murphy MM, Burrell TL, Klaiman C, et al. 3q29 Recurrent Deletion. 2016 Sep 22 [updated 2021 Jul 1]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK385289/ PMID: 27656750.
5. Kashevarova AA, Lopatkina ME, Vasilyeva OY, Fedotov DA, Lobanov AD, Fonova EA, et al. Delineation of the Genetic Architecture and Clinical Polymorphism of 3q29 Duplication Syndrome: A Review of the Literature and a Report of Two Novel Patients With Single-Gene BDH1 Duplications. Mol Genet Genomic Med. 2025;13(1):e70047. doi:10.1002/mgg3.70047. PMID: 39739615. PMC11686339.
6. Massier M, Doco-Fenzy M, Egloff M, Le Guillou X, Le Guyader G, Redon S, et al. 3q29 duplications: A cohort of 46 patients and a literature review. Am J Med Genet A. 2024;194(7):e63531. doi:10.1002/ajmg.a.63531. PMID: 38421086.
7. ClinGen Dosage Sensitivity Curation. 3q29 recurrent region (includes DLG1) (ISCA-37443) [Internet]. ClinGen; last evaluated 2024 Feb 18 [cited 2026 Oct 26]. Available from: https://search.clinicalgenome.org/kb/gene-dosage/region/ISCA-37443
8. Wallace SE, Bean LJH. Educational Materials — Genetic Testing: Current Approaches. 2017 Mar 14 [revised 2025 Nov 20]. In: Adam MP, Bick S, Mirzaa GM, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK279899/
9. Ballif BC, Theisen A, Coppinger J, Gowans GC, Hersh JH, Madan-Khetarpal S, et al. Expanding the clinical phenotype of the 3q29 microdeletion syndrome and characterization of the reciprocal microduplication. Mol Cytogenet. 2008;1:8. doi:10.1186/1755-8166-1-8. PMID: 18471269. PMC2408925.
10. Herriges AR, Purcell RH. Driver or passenger? A new assessment of genes in the schizophrenia-associated 3q29 deletion locus for contribution to neurodevelopmental disorders. J Neurodev Disord. 2026;18(1). doi:10.1186/s11689-026-09696-y. PMID: 42015005. PMC13326166.
11. Yilmaz F, Gurusamy U, Mosley TJ, Hallast P, Kim K, Mostovoy Y, et al. High level of complexity and global diversity of the 3q29 locus revealed by optical mapping and long-read sequencing. Genome Med. 2023;15(1):35. doi:10.1186/s13073-023-01184-5. PMID: 37165454. PMC10170684.
12. Tassano E, Uccella S, Giacomini T, Severino M, Siri L, Gherzi M, et al. 3q29 microduplication syndrome: Description of two new cases and delineation of the minimal critical region. Eur J Med Genet. 2018;61(8):428-433. doi:10.1016/j.ejmg.2018.02.011. PMID: 29501613.
13. Bauleo A, Pace V, Montesanto A, De Stefano L, Brando R, Puntorieri D, et al. 3q29 microduplication syndrome: New evidence for the refinement of the critical region. Mol Genet Genomic Med. 2023;11(4):e2130. doi:10.1002/mgg3.2130. PMID: 36691815. PMC10094080.
14. Online Mendelian Inheritance in Man (OMIM). #611936 Chromosome 3q29 duplication syndrome [Internet]. Baltimore (MD): Johns Hopkins University; created 2008 Apr 7, last edited 2011 Oct 28 [cited 2026 Oct 26]. Available from: https://www.omim.org/entry/611936
15. Cox DM, Butler MG. A clinical case report and literature review of the 3q29 microdeletion syndrome. Clin Dysmorphol. 2015;24(3):89-94. doi:10.1097/MCD.0000000000000077. PMID: 25714563. PMC5125389.
16. Online Mendelian Inheritance in Man (OMIM). #609425 Chromosome 3q29 deletion syndrome [Internet]. Baltimore (MD): Johns Hopkins University; created 2005 Jun 17, last edited 2013 Oct 7 [cited 2026 Oct 26]. Available from: https://www.omim.org/entry/609425
17. Kaminsky EB, Kaul V, Paschall J, Church DM, Bunke B, Kunig D, et al. An evidence-based approach to establish the functional and clinical significance of copy number variants in intellectual and developmental disabilities. Genet Med. 2011;13(9):777-84. doi:10.1097/GIM.0b013e31822c79f9. PMID: 21844811. PMC3661946.
18. Goh S, Dudding-Byth T, Pinese M, Kirk EP. Updated penetrance estimates for recurrent copy number variants - an improved definition and formula. Eur J Hum Genet. 2026;34(1):119-127. doi:10.1038/s41431-025-01948-0. PMID: 41094176. PMC12816643.
19. Streata I, Riza AL, Sosoi S, Burada F, Ioana M. Phenotype Heterogeneity in 3q29 Microduplication Syndrome. Curr Health Sci J. 2020;46(2):193-197. doi:10.12865/CHSJ.46.02.14. PMID: 32874693. PMC7445637.
20. Mulle JG. The 3q29 deletion confers >40-fold increase in risk for schizophrenia. Mol Psychiatry. 2015;20(9):1028-9. doi:10.1038/mp.2015.76. PMID: 26055425. PMC4546529.
21. Miller DT, Adam MP, Aradhya S, Biesecker LG, Brothman AR, Carter NP, et al. Consensus statement: chromosomal microarray is a first-tier clinical diagnostic test for individuals with developmental disabilities or congenital anomalies. Am J Hum Genet. 2010;86(5):749-64. doi:10.1016/j.ajhg.2010.04.006. PMID: 20466091. PMC2869000.
22. Rutgers, The State University of New Jersey. The 3q29 Deletion and 3q29 Duplication: Architecture of Behavioral Phenotypes. ClinicalTrials.gov identifier NCT02447861 [Internet]. Last update posted 2026 May 20 [cited 2026 Oct 26]. Available from: https://clinicaltrials.gov/study/NCT02447861
23. Sanchez Russo R, Gambello MJ, Murphy MM, Aberizk K, Black E, Burrell TL, et al. Deep phenotyping in 3q29 deletion syndrome: recommendations for clinical care. Genet Med. 2021;23(5):872-880. doi:10.1038/s41436-020-01053-1. PMID: 33564151. PMC8105170.
24. ERN-ITHACA. European Reference Network on Rare Malformation Syndromes, Intellectual and Other Neurodevelopmental Disorders [Internet] [cited 2026 Oct 26]. Available from: https://ern-ithaca.eu/
25. Coe BP, Witherspoon K, Rosenfeld JA, van Bon BW, Vulto-van Silfhout AT, Bosco P, et al. Refining analyses of copy number variation identifies specific genes associated with developmental delay. Nat Genet. 2014;46(10):1063-71. doi:10.1038/ng.3092. PMID: 25217958. PMC4177294.
26. Lisi EC, Hamosh A, Doheny KF, Squibb E, Jackson B, Galczynski R, et al. 3q29 interstitial microduplication: a new syndrome in a three-generation family. Am J Med Genet A. 2008;146A(5):601-9. doi:10.1002/ajmg.a.32190. PMID: 18241066.
27. Goobie S, Knijnenburg J, Fitzpatrick D, Sharkey FH, Lionel AC, Marshall CR, et al. Molecular and clinical characterization of de novo and familial cases with microduplication 3q29: guidelines for copy number variation case reporting. Cytogenet Genome Res. 2008;123(1-4):65-78. doi:10.1159/000184693. PMID: 19287140. PMC2920184.
28. Fernández-Jaén A, Castellanos Mdel C, Fernández-Perrone AL, Fernández-Mayoralas DM, de la Vega AG, Calleja-Pérez B, et al. Cerebral palsy, epilepsy, and severe intellectual disability in a patient with 3q29 microduplication syndrome. Am J Med Genet A. 2014;164A(8):2043-7. doi:10.1002/ajmg.a.36559. PMID: 24838842.
29. Lawrence MB, Arreola A, Cools M, Elton S, Wood KS. 3q29 Chromosomal duplication in a neonate with associated myelomeningocele and midline cranial defects. Clin Dysmorphol. 2017;26(4):221-223. doi:10.1097/MCD.0000000000000193. PMID: 28763312.
30. Vitale A, Labruna G, Mancini A, Alfieri A, Iaffaldano L, Nardelli C, et al. 3q29 microduplication in a small family with complex metabolic phenotype from Southern Italy. Clin Chem Lab Med. 2018;56(7):e167-e170. doi:10.1515/cclm-2017-1090. PMID: 29306918.
31. Coyan AG, Dyer LM. 3q29 microduplication syndrome: Clinical and molecular description of eleven new cases. Eur J Med Genet. 2020;63(12):104083. doi:10.1016/j.ejmg.2020.104083. PMID: 33039685.
32. Mulle JG, Dodd AF, McGrath JA, Wolyniec PS, Mitchell AA, Shetty AC, et al. Microdeletions of 3q29 confer high risk for schizophrenia. Am J Hum Genet. 2010;87(2):229-36. doi:10.1016/j.ajhg.2010.07.013. PMID: 20691406. PMC2917706.
33. Marshall CR, Howrigan DP, Merico D, Thiruvahindrapuram B, Wu W, Greer DS, et al. Contribution of copy number variants to schizophrenia from a genome-wide study of 41,321 subjects. Nat Genet. 2017;49(1):27-35. doi:10.1038/ng.3725. PMID: 27869829. PMC5737772.
34. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 660–665.
35. Ensembl. Genes in region 3:196029183-197617791 (GRCh38) [Internet]. Hinxton (UK): EMBL-EBI; [cited 2026 Oct 26]. Available from: https://www.ensembl.org/Homo_sapiens/Location/Overview?r=3:196029183-197617791
36. Jabara HH, Boyden SE, Chou J, Ramesh N, Massaad MJ, Benson H, et al. A missense mutation in TFRC, encoding transferrin receptor 1, causes combined immunodeficiency. Nat Genet. 2016;48(1):74-8. doi:10.1038/ng.3465. PMID: 26642240. PMC4696875.
37. Antonarakis SE, Holoubek A, Rapti M, Rademaker J, Meylan J, Iwaszkiewicz J, et al. Dominant monoallelic variant in the PAK2 gene causes Knobloch syndrome type 2. Hum Mol Genet. 2021;31(1):1-9. doi:10.1093/hmg/ddab026. PMID: 33693784.
38. GeneGuidance. Curation of the PAK2 gene and Knobloch syndrome type 2 [user database; export of 26 Sep 2026]. Accessed 2026 Oct 26.
39. GeneGuidance. DLG1 gene record: identifiers (HGNC:2900; OMIM 601014; NM_004087) and function [user database; export of 26 Sep 2026]. Accessed 2026 Oct 26.
40. Wang Y, Zeng C, Li J, Zhou Z, Ju X, Xia S, et al. PAK2 Haploinsufficiency Results in Synaptic Cytoskeleton Impairment and Autism-Related Behavior. Cell Rep. 2018;24(8):2029-2041. doi:10.1016/j.celrep.2018.07.061. PMID: 30134165.
41. GeneGuidance. BDH1 gene record: identifiers (HGNC:1027; OMIM 603063; NM_004051) [user database; export of 26 Sep 2026]. Accessed 2026 Oct 26.
42. Rutkowski TP, Purcell RH, Pollak RM, Grewenow SM, Gafford GM, Malone T, et al. Behavioral changes and growth deficits in a CRISPR engineered mouse model of the schizophrenia-associated 3q29 deletion. Mol Psychiatry. 2021;26(3):772-783. doi:10.1038/s41380-019-0413-5. PMID: 30976085. PMC6788962.
43. Aleixandre Blanquer F, Manchón Trives I, Forniés Arnau MJ, Alcaraz Mas LA, Picó Alfonso N, Galán Sánchez F. [3q29 microduplication syndrome]. An Pediatr (Barc). 2011;75(6):409-12. doi:10.1016/j.anpedi.2011.08.002. PMID: 21982553.
44. Murphy MM, Burrell TL, Cubells JF, Epstein MT, Espana R, Gambello MJ, et al. Comprehensive phenotyping of neuropsychiatric traits in a multiplex 3q29 deletion family: a case report. BMC Psychiatry. 2020;20(1):184. doi:10.1186/s12888-020-02598-w. PMID: 32321479. PMC7179007.
45. Schnur RE, Dvořáček L, Kalsner L, Shapiro FL, Grebeňová D, Yanni D, et al. New kinase-deficient PAK2 variants associated with Knobloch syndrome type 2. Clin Genet. 2024;106(4):518-524. doi:10.1111/cge.14578. PMID: 38894571.
46. Domenach L, Rooryck C, Legendre M, Bouchghoul H, Beneteau C, Margot H. Antenatal phenotype associated with PAK2 pathogenic variants: bilateral pleural effusion as a warning sign. BMC Med Genomics. 2025;18(1):35. doi:10.1186/s12920-025-02096-6. PMID: 39994693. PMC11853806.
47. GeneGuidance. Curation of the TFRC gene and immunodeficiency 46 [user database; export of 26 Sep 2026]. Accessed 2026 Oct 26.
48. Aljohani AH. TFRC Germline Variants and Inborn Error of Immunity: Mechanistic Insights into Iron-Immune Crosstalk. J Clin Immunol. 2026;46(1). doi:10.1007/s10875-026-01999-y. PMID: 41714512. PMC12992366.
49. 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. PMID: 28967166.
50. Whangbo JS, Chou J, Al-Dhekri H, Harris M, Geha RS, Pai SY, et al. Hematopoietic Stem Cell Transplantation Is a Curative Therapy for Transferrin Receptor 1 (TFRC) Deficiency. J Allergy Clin Immunol Pract. 2021;9(2):753-759.e2. doi:10.1016/j.jaip.2020.10.018. PMID: 33096268.
51. Unique – Rare Chromosome Disorder Support Group. 3q29 duplications and microduplications [Internet]. Oxted (UK): Unique; 2011 (version 1) [cited 2026 Oct 26]. Available from: https://rarechromo.org/media/information/Chromosome%20%203/3q29%20Duplications%20and%20Microduplications%20FTNW.pdf
52. Glassford MR, Rosenfeld JA, Freedman AA, Zwick ME, Mulle JG. Novel features of 3q29 deletion syndrome: Results from the 3q29 registry. Am J Med Genet A. 2016;170A(4):999-1006. doi:10.1002/ajmg.a.37537. PMID: 26738761. PMC4849199.
53. Chen Y, Feng Q, Lam M, Yu M, Sun Y, Huai C, et al. Contribution of copy number variants to schizophrenia in East Asian populations. Nat Genet. 2026;58(10):2514-2524. doi:10.1038/s41588-026-02732-6. PMID: 42728386. PMC13649597.
54. Ormond C, Ryan NM, Hedman AM, Cannon TD, Sullivan PF, Gill M, et al. Whole genome sequencing study of identical twins discordant for psychosis. Transl Psychiatry. 2024;14(1):313. doi:10.1038/s41398-024-02982-0. PMID: 39080272. PMC11289105.
55. Reis F, Pereira C, Laureano M, Cartaxo T. An Unusual Psychiatric Presentation of the 3q29 Microduplication Syndrome. Cureus. 2020;12(3):e7203. doi:10.7759/cureus.7203. PMID: 32269882. PMC7138485.
56. Quintela I, Barros-Angueira F, Perez-Gay L, Dacruz D, Castro-Gago M, Carracedo A, et al. [Molecular characterisation and phenotypic description of two patients with reciprocal chromosomal aberrations in the region of the 3q29 microdeletion/microduplication syndromes]. Rev Neurol. 2015;61(6):255-60. PMID: 26350776.
57. Pollak RM, Murphy MM, Epstein MP, Zwick ME, Klaiman C, Saulnier CA, et al. Neuropsychiatric phenotypes and a distinct constellation of ASD features in 3q29 deletion syndrome: results from the 3q29 registry. Mol Autism. 2019;10:30. doi:10.1186/s13229-019-0281-5. PMID: 31346402. PMC6636128.
58. Sagar A, Bishop JR, Tessman DC, Guter S, Martin CL, Cook EH. Co-occurrence of autism, childhood psychosis, and intellectual disability associated with a de novo 3q29 microdeletion. Am J Med Genet A. 2013;161A(4):845-9. doi:10.1002/ajmg.a.35754. PMID: 23443968. PMC3685481.
59. Quintero-Rivera F, Sharifi-Hannauer P, Martinez-Agosto JA. Autistic and psychiatric findings associated with the 3q29 microdeletion syndrome: case report and review. Am J Med Genet A. 2010;152A(10):2459-67. doi:10.1002/ajmg.a.33573. PMID: 20830797.
60. Vidal W, Masotto B, Viñas-Jornet M, Ramos-Quiroga JA, Español-Martín G. Case Report: Early-onset psychosis as a sentinel manifestation of 3q29 deletion syndrome in an adolescent with neurodevelopmental disorders. Front Child Adolesc Psychiatry. 2026;5:1823061. doi:10.3389/frcha.2026.1823061. PMID: 42529018. PMC13415694.
61. Digilio MC, Bernardini L, Mingarelli R, Capolino R, Capalbo A, Giuffrida MG, et al. 3q29 Microdeletion: a mental retardation disorder unassociated with a recognizable phenotype in two mother-daughter pairs. Am J Med Genet A. 2009;149A(8):1777-81. doi:10.1002/ajmg.a.32965. PMID: 19610115.
62. Li F, Lisi EC, Wohler ES, Hamosh A, Batista DA. 3q29 interstitial microdeletion syndrome: an inherited case associated with cardiac defect and normal cognition. Eur J Med Genet. 2009;52(5):349-52. doi:10.1016/j.ejmg.2009.05.001. PMID: 19460468.
63. Mak BC, Sanchez Russo R, Gambello MJ, Fleischer N, Black ED, Leslie E, et al. Craniofacial features of 3q29 deletion syndrome: Application of next-generation phenotyping technology. Am J Med Genet A. 2021;185(7):2094-2101. doi:10.1002/ajmg.a.62227. PMID: 33938623. PMC8250870.
64. Arican I, van der Horst M, Bass N, Zinkstok JR. What do European clinical guidelines say about genetic testing for people with neuropsychiatric disorders? A scoping review. Psychiatr Genet. 2026;36(1):1-12. doi:10.1097/YPG.0000000000000407. PMID: 41408787. PMC12834271.
65. Smajlagić D, Lavrichenko K, Berland S, Helgeland Ø, Knudsen GP, Vaudel M, et al. Population prevalence and inheritance pattern of recurrent CNVs associated with neurodevelopmental disorders in 12,252 newborns and their parents. Eur J Hum Genet. 2021;29(1):205-215. doi:10.1038/s41431-020-00707-7. PMID: 32778765. PMC7852900.
66. McGoey R, Varma A, Lacassie Y. Siblings with phenotypic overlap with Toriello-Carey syndrome and complex cytogenetic imbalances including 3q29 microduplication and 6p25 microdeletion: Review of the literature and additional evidence for genetic heterogeneity. Am J Med Genet A. 2010;152A(12):3068-73. doi:10.1002/ajmg.a.33721. PMID: 21108391.
67. Wen X, Zhu J, Cai L, Tang G, Zeng W, Luo Y, et al. A familial 3q28q29 duplication induced mild intellectual disability: case presentation and literature review. Am J Transl Res. 2022;14(3):1663-1671. PMID: 35422908. PMC8991147.
68. Yatsuga S. A Suspected Mosaic 3q29 Deletion in a Patient With Autism Spectrum Disorder and Multiple Micronutrient Deficiencies: A Case Report. Case Rep Genet. 2026;2026:9353827. doi:10.1155/crig/9353827. PMID: 42923983. PMC13688419.
69. Riggs ER, Andersen EF, Cherry AM, Kantarci S, Kearney H, Patel A, et al. Technical standards for the interpretation and reporting of constitutional copy-number variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics (ACMG) and the Clinical Genome Resource (ClinGen). Genet Med. 2020;22(2):245-257. doi:10.1038/s41436-019-0686-8. PMID: 31690835. PMC7313390.
70. Riggs ER, Andersen EF, Cherry AM, Kantarci S, Kearney H, Patel A, et al. Correction: Technical standards for the interpretation and reporting of constitutional copy-number variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics (ACMG) and the Clinical Genome Resource (ClinGen). Genet Med. 2021;23(11):2230. doi:10.1038/s41436-021-01150-9. PMID: 33731880.
71. Guida V, Sinibaldi L, Pagnoni M, Bernardini L, Loddo S, Margiotti K, et al. A de novo proximal 3q29 chromosome microduplication in a patient with oculo auriculo vertebral spectrum. Am J Med Genet A. 2015;167A(4):797-801. doi:10.1002/ajmg.a.36951. PMID: 25735547.
72. Rosenfeld JA, Coe BP, Eichler EE, Cuckle H, Shaffer LG. Estimates of penetrance for recurrent pathogenic copy-number variations. Genet Med. 2013;15(6):478-81. doi:10.1038/gim.2012.164. PMID: 23258348. PMC3664238.
73. Kessi M, Peng J, Yang L, Duan H, Tang Y, Yin F. A Case With 4 de Novo Copy Number Variations With Clinical Features That Overlap 1q43q44 Microdeletion and 3q29 Microduplication Syndromes. Child Neurol Open. 2018;5:2329048X18798200. doi:10.1177/2329048X18798200. PMID: 30263904. PMC6153526.
74. Geckinli B, Turkyilmaz A, Alavanda C, Sager G, Arslan Ates E, Soylemez MA, et al. Novel, homozygous RAB3GAP1 c.2606 + 1G>A, p.Glu830ValfsTer9 variant and chromosome 3q29 duplication in a Turkish individual with Warburg micro syndrome. Clin Dysmorphol. 2023;32(2):55-61. doi:10.1097/MCD.0000000000000454. PMID: 36876345.
75. Chenzhuo S, Lingling F, Ang W, Hui C, Ruixin W, Bixi Y, et al. A case of ADH5/ALDH2 deficiency combined with 3q29 microduplication syndrome. BMC Pediatr. 2025;25(1):717. doi:10.1186/s12887-025-05948-3. PMID: 41039406. PMC12492562.
76. Tabbers MM, DiLorenzo C, Berger MY, Faure C, Langendam MW, Nurko S, et al. Evaluation and treatment of functional constipation in infants and children: evidence-based recommendations from ESPGHAN and NASPGHAN. J Pediatr Gastroenterol Nutr. 2014;58(2):258-74. doi:10.1097/MPG.0000000000000266. PMID: 24345831.
77. Sandoz Inc. RITALIN (methylphenidate hydrochloride) tablets. Prescribing information. NDA010187. Revised 2025 Feb [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=d6fb2750-cdab-4749-ba0d-7534840a5892
78. Laboratorios Rubió S.A. RUBIFEN 5 mg, 10 mg e 20 mg comprimidos (cloridrato de metilfenidato) — summary of product characteristics [Internet]. Lisbon: INFARMED, Infomed; text revised September 2025 [cited 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=36077&tipo_doc=rcm
79. Bausch Health US. WELLBUTRIN XL (bupropion hydrochloride extended-release tablets). Prescribing information. NDA021515. Revised 2026 Feb [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=a435da9d-f6e8-4ddc-897d-8cd2bf777b21
80. Eli Lilly and Company. STRATTERA (atomoxetine) capsules, for oral use. Prescribing information. NDA021411. Revised 2026 Jun [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=309de576-c318-404a-bc15-660c2b1876fb
81. Allergan. LEXAPRO (escitalopram) tablets. Prescribing information. NDA021323. Revised 2023 Oct [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=13bb8267-1cab-43e5-acae-55a4d957630a
82. Viatris Specialty. ZOLOFT (sertraline hydrochloride). Prescribing information. NDA019839. Revised 2023 Aug [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=fda754f6-d0f3-4dce-a17a-927d64f912f7
83. Janssen Pharmaceuticals. RISPERDAL (risperidone). Prescribing information. NDA020272. Revised 2026 May [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=7e117c7e-02fc-4343-92a1-230061dfc5e0
84. Otsuka America Pharmaceutical. ABILIFY (aripiprazole) tablets. Prescribing information. NDA021436. Revised 2025 Jan [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=c040bd1d-45b7-49f2-93ea-aed7220b30ac
85. European Medicines Agency. Abilify (aripiprazole) — EPAR, Product Information (Annex I, Summary of Product Characteristics) [Internet]. Amsterdam: EMA; first authorisation 2004 Jun 4 [cited 2026 Oct 26]. Available from: https://www.ema.europa.eu/en/documents/product-information/abilify-epar-product-information_en.pdf
86. Janssen Farmacêutica Portugal. RISPERDAL comprimidos revestidos por película e solução oral (risperidona) — summary of product characteristics [Internet]. Lisbon: INFARMED, Infomed [cited 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=7607&tipo_doc=rcm
87. Colijn MA. Antipsychotic Use Among Intellectually Disabled Individuals With Rare Genetic Variants That Confer Risk for Schizophrenia. J Intellect Disabil Res. 2026;70(4):440-445. doi:10.1111/jir.70065. PMID: 41320189. PMC12950631.
88. UCB. KEPPRA (levetiracetam) tablets, oral solution. Prescribing information. NDA021035/NDA021505. Revised 2026 Jan [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=3ca9df05-a506-4ec8-a4fe-320f1219ab21
89. Covis Pharma. PRILOSEC (omeprazole magnesium) for delayed-release oral suspension. Prescribing information. NDA022056. Revised 2024 Mar [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=b6761f84-53ac-4745-a8c8-1e5427d7e179
90. Norgine Limited. Movicol Paediatric Plain 6.9 g powder for oral solution — Summary of Product Characteristics [Internet]. Electronic Medicines Compendium (UK); revised 2025 Mar 11 [cited 2026 Oct 26]. Available from: https://www.medicines.org.uk/emc/product/52/smpc (e Movicol Paediatric Chocolate flavour 6.8 g, https://www.medicines.org.uk/emc/product/4308/smpc)
91. Bayer HealthCare. MiraLAX (polyethylene glycol 3350) powder for oral solution, OTC Drug Facts. ANDA090685. 2025 Dec [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=0a5b2121-928f-8f49-e063-6394a90a8aaf
92. Takeda Pharmaceuticals America. VYVANSE (lisdexamfetamine dimesylate) capsules and chewable tablets. Prescribing information. NDA021977/NDA208510. Revised 2026 Apr [cited 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=704e4378-ca83-445c-8b45-3cfa51c1ecad
93. European Medicines Agency. Keppra (levetiracetam) — EPAR, Product Information (Annex I, Summary of Product Characteristics) [Internet]. Amsterdam: EMA; first authorisation 2000 Sep 29 [cited 2026 Oct 26]. Available from: https://www.ema.europa.eu/en/documents/product-information/keppra-epar-product-information_en.pdf
94. Pollak RM, Sefik E, Klaiman C, Harner MK, Bishop DV, Purcell JR, et al. Development and Validation of a Cross-Dimensional Screening Protocol for Remote Phenotyping Applied to Individuals With 3q29 Deletion Syndrome. J Intellect Disabil Res. 2026;70(7):758-768. doi:10.1111/jir.70116. PMID: 42090187. PMC13238388.
95. Petrin AL, Daack-Hirsch S, L'Heureux J, Murray JC. A case of 3q29 microdeletion syndrome involving oral cleft inherited from a nonaffected mosaic parent: molecular analysis and ethical implications. Cleft Palate Craniofac J. 2011;48(2):222-30. doi:10.1597/09-149. PMID: 20500065. PMC2964377.
96. Stefansson H, Meyer-Lindenberg A, Steinberg S, Magnusdottir B, Morgen K, Arnarsdottir S, et al. CNVs conferring risk of autism or schizophrenia affect cognition in controls. Nature. 2014;505(7483):361-6. doi:10.1038/nature12818. PMID: 24352232.
97. Kendall KM, Rees E, Escott-Price V, Einon M, Thomas R, Hewitt J, et al. Cognitive Performance Among Carriers of Pathogenic Copy Number Variants: Analysis of 152,000 UK Biobank Subjects. Biol Psychiatry. 2017;82(2):103-110. doi:10.1016/j.biopsych.2016.08.014. PMID: 27773354.
98. Kendall KM, Bracher-Smith M, Fitzpatrick H, Lynham A, Rees E, Escott-Price V, et al. Cognitive performance and functional outcomes of carriers of pathogenic copy number variants: analysis of the UK Biobank. Br J Psychiatry. 2019;214(5):297-304. doi:10.1192/bjp.2018.301. PMID: 30767844. PMC6520248.