A 41-month-old female with no perinatal issues showed severe developmental delay, dyskinetic movements, and spasticity, and had been followed under the diagnosis of dyskinetic CP. Her development corresponded to an equivalent age of 8 months in cognition, 5 months in receptive language, 6 months in expressive language, 8 months in fine motor skills, and 7 months in gross motor skills. However, no definitive abnormality was observed in the brain MRI. Due to the discrepancy between clinical symptoms and radiological findings, a clinical exome sequencing (CES) test was performed, revealing a de novo heterozygous nonsense variant, c.1543C>T(p.Arg515Ter), in the CTNNB1, which has been previously reported as a pathogenic variant. She was diagnosed with neurodevelopmental disorder with spastic diplegia and visual defects (OMIM 615075).

A 27-month-old male with no perinatal issues showed gross motor delays and spasticity, with the left side more severely affected. His developmental milestones matched those of a 27-month-old in cognition and fine motor skills, and a 13-month-old in gross motor skills. A brain MRI showed periventricular nodular heterotopia (Fig. 1). Chromosomal microarray (CMA) and trio whole genome sequencing (WGS) tests were performed, which identified a de novo 1.4 Mb copy number loss in Xp22.33 (1,411,421-2,859,126). Based on previous research linking the Xp22.33 deletion to periventricular nodular heterotopia and CP [5,6], this deletion is believed to be a causative CNV in the patient described.

A 12-year-old female, born at 35 weeks gestation weighing 2,170 g, was diagnosed with CP due to persistent delays in gross motor development, dyskinetic movements, and spasticity since infancy. There was no evidence of phenotypic progression. The patient had received multiple botulinum toxin injections for spasticity and was able to walk independently at 48 months of age. As no defined abnormalities were observed in the brain and spine MRI, trio WGS was performed and identified a novel de novo heterozygous nonsense variant, c.310G>T(p.Glu104Ter) in SOX2. She was diagnosed with SOX2 disorder.

A recent whole exome sequencing (WES) trio analysis conducted on a cohort of 250 individuals diagnosed with CP identified 8 genes that may contribute to the disorder’s etiology. The genes identified include CTNNB1 (observed in Case 1), TUBA1A, RHOB, ATL1, DHX32, SPAST, FBXO31, and ALK, indicating a diverse genetic basis underlying the pathogenesis of CP [7]. A recent meta-analysis found that the overall diagnostic yield of CMA in patients with CP was 5%, with 21 CNVs identified, as in Case 2. This also suggests that diverse genetic mutations are involved in CP [8]. Furthermore, recent advances and widespread application of WES and WGS technologies have significantly increased the identification of new genes associated with movement and muscle tone/posture abnormalities in children, as illustrated by the identification of SOX2 in Case 3. Meanwhile, determining whether genes identified through these studies should be classified as “causative” poses a significant challenge. This happens because most studies defined de novo variants observed in CP patients as pathogenic variants and suggested a potential cause of CP through review of previous studies and molecular pathways [9-11]. Morevoer, these genes frequently demonstrate associations with other neurodevelopmental disorders, such as intellectual disabilities and autism spectrum disorders, which often can present various degrees of tone abnormalities and movement problems [12]. Although the International Cerebral Palsy Genomics Consortium suggested in 2019 that the identification of genomic causes should not require reclassification of CP diagnoses [13], establishing a gene as the etiology of CP requires well-designed prospective studies, including comprehensive phenotyping of patients from a large CP cohort, an unambiguous clinical definition of CP, and rigorous genotyping of patients, their family members, and normal controls. Furthermore, confirming newly discovered disease-causing genes requires robust functional studies in vitro and in vivo to verify their pathogenic role. More research on CP-causing genes and CNVs is needed, and previous studies should be reviewed more rigorously. For instance, the recent systematic review on genetic tests for CP reported that the diagnostic yield of exome sequencing ranges from 15% to 34% (95% confidence interval) and identified 22 potential common causative genes in CP patients [8]. Among them, 3 genes (SPAST, ATL1, KIF1A) are associated with hereditary spastic paraplegia (HSP), and 2 others (SLC2A1, ATP1A3) are linked to child-onset dystonia. Notably, WDR45 (associated with Neurodegeneration with Brain Iron Accumulation 5) and MECP2 (linked to Rett syndrome) are also mentioned. Clinically differentiating CP from early-onset HSP, dystonia, or leukodystrophy can be challenging due to overlapping symptoms and difficulty in perceiving progression. Furthermore, WDR45 and MECP2 conditions, which typically show progressive neuromotor symptoms post-infancy and are distinct from CP. Therefore, previous studies, including this review, raise a question: It may be uncertain whether the genes identified through these studies are monogenic, involved in differential diagnosis, or coexisting.

Three recent WGS studies in relatively large cohorts, including 327 children with their parents [14], 150 singletons [15], and 120 families with CP [16], are considered methodologically rigorous. These studies carefully defined the clinical diagnostic criteria and timing of CP, and thoroughly detailed the inclusion and exclusion criteria. The findings showed that 11.3%, 24.7%, and 45.0% of cases in these respective studies contained candidate CP causative variants. However, further research is still needed to conclusively determine the pathogenicity of these variants. This should involve rigorous peer review and expert consensus, long-term longitudinal studies rather than cross-sectional ones, as well as conducting in vitro and in vivo studies to validate the findings.

A 14-year-old female patient, with no significant perinatal complications, was initially diagnosed with CP, showing spasticity and dystonia in both lower extremities. She underwent various treatments, including oral baclofen, botulinum toxin injections, and rehabilitation therapy. Living in a child welfare institution, her diurnal fluctuations in motor symptoms were mistakenly attributed to psychiatric issues. The brain and spine MRI showed no abnormalities. Genetic testing using a CES test identified a heterozygous missense variant, c.607G>A(p.Gly203Arg), in GCH1, which had previously been reported as pathogenic. Subsequently, she was diagnosed with DOPA-responsive dystonia (OMIM 128230). The low-dose L-dopa treatment markedly improved her condition, resulting in a disability-free state.

The primary objective in the differential diagnosis of CP is to prevent the misdiagnosis of conditions that might be amenable to specific therapeutic interventions, potentially alleviating symptoms. This encompasses a range of nongenetic conditions, such as tethered cord syndrome, as well as numerous genetic disorders, including neurometabolic disorders, such as Case 4. Furthermore, an accurate differential diagnosis provides several key advantages: reducing unnecessary diagnostic tests, improving medical management, preventing disease recurrence in future siblings, informing at-risk family members, and alleviating potential feelings of ambivalence or guilt among relatives.

The literature suggests that differential diagnosis of CP often uses terms such as “atypical CP,” “CP mimics,” and “CP masqueraders” [17-20]. These terms share common elements: a lack of documented risk factors or neuroimaging results consistent with a history of brain injury. However, there is no consensus on these terms, as they are defined differently in various sources. Furthermore, as previously noted, the challenges extend beyond unclear differential diagnoses, encompassing ambiguous distinctions of monogenic disorders and co-occurring conditions. Therefore, effective use of these terms requires a clear definition or consensus. “Differential Genetic Disorders of CP” is proposed as a more accurate descriptor, as the main goal is to facilitate differential diagnosis and avoid misdiagnoses. Particularly, recent genetic therapy research has expanded, with some treatments already approved, such as for aromatic L-amino acid decarboxylase deficiency. As these treatable conditions continue to increase, it is crucial to identify clinical red flags that require a differential diagnosis before diagnosing CP.

Genetic testing is recommended for CP diagnosis under certain clinical conditions (Table 1) [18,21].

A 6-year-old male was referred to our clinic for proper diagnosis of developmental delay and ataxia. The patient was the 4th child of nonconsanguineous healthy parents, with a normal family and perinatal history. He started standing on his own at 10 months old and, at 12 months old, began taking 2 or 3 steps but immediately fell down. He visited several hospitals due to ataxia symptoms and underwent brain MRI scans and genetic testing, but was unable to obtain a definitive diagnosis. Although his ataxia symptoms remained similar, his overall gross motor skills gradually improved as he aged. The patient underwent rehabilitation treatment under the diagnosis of CP, ataxic type. Brain MRI at 1, 5, and 7 years showed interval development of mild atrophy in the cerebellar vermis and hemispheres (Fig. 2). The CES and CMA tests revealed a heterozygous pathogenic nonsense variant in ATM, c.742C>T (p.Arg248Ter) inherited from the father and a compound heterozygous CNV, 11q22.3(108151766-108183226)x1, 31,460 bp (deletion of exons 24-40 of ATM) inherited from the mother. The patient was ultimately diagnosed with ataxia-telangiectasia.

A 30-month-old male with no perinatal complications showed global developmental delays and spasticity. He achieved independent walking at 16 months. The parents first noticed developmental delays compared to other children when the patient was around 12 months old, but they were unsure exactly when the delays began. Since the brain MRI showed no abnormalities, a CES test was conducted, identifying a de novo heterozygous missense variant, c.38G>A(p.Arg13His), in KIF1A, which has been previously reported as pathogenic. He was diagnosed with HSP 30 (OMIM 615357) and CP. The patient is currently 60 months old and has not experienced any further progression of neurological symptoms.

While HSP primarily affects the lower limbs and progresses over time, CP typically presents as quadriplegia and is non-progressive. However, the distinction is often not clear-cut in clinical practice, especially when CP symptoms predominantly affect the lower limbs or when the onset of HSP is very early. Additionally, distinguishing progressive from non-progressive neuromotor abnormalities in CP is considered crucial for the genetic differential diagnosis of CP but remains challenging. This complexity arises because disease progression can be slow and sometimes only becomes evident later in life. In some cases, patients exhibit very early onset or severe symptoms from birth, making it hard to determine the clinical progression, as in Case 6. Furthermore, progression does not always impact motor function, but can manifest in other symptoms, such as worsening seizures, loss of vision or speech, or cognitive decline. The clinical diversity of each genetic disorder further complicates this issue, as progression varies significantly among individuals, from early onset to late onset. Moreover, specific treatments, such as baclofen pump implantation, selective dorsal rhizotomy, and deep brain stimulation for spasticity or dystonia, can halt the progression of neuromotor symptoms. Furthermore, clinical symptoms may not always match radiological findings or changes, such as in Case 5. This adds another layer of complexity to CP diagnosis [22]. Is it appropriate to diagnose CP while waiting for genetic test results, as in Case 5? In clinical practice, certain cases may be suspected of having a differential genetic disorder mimicking CP or present ambiguous features; however, a definitive genetic diagnosis frequently remains elusive. This scenario raises a complex issue: should a diagnosis of CP be used when the genetic investigations are inconclusive? Consequently, it is advised that clinicians and researchers move beyond traditional nomenclature, using these diagnostic ambiguities as opportunities to refine and redefine categories within neurodevelopmental disorders.

A 4-year-old male, born prematurely at 31 weeks weighing 1,200 g, was diagnosed with CP due to gross motor delays, spasticity, and ataxic gait. He achieved independent walking at 30 months. A brain MRI showed posthemorrhagic ventricular enlargement (Fig. 3). Due to intellectual disability, macrocephaly, and facial dysmorphism, a CES test was performed. It identified a de novo heterozygous frameshift variant, c.5285dup(p.Val1764SerfsTer), in the NSD1. He was diagnosed with co-occurring Sotos syndrome (OMIM 117550) and CP.

With the advancement of genetic testing and its increasing clinical application, cases where two or more genetic diseases occur simultaneously are becoming more common. This dual diagnosis can happen even with similar phenotypes. In some cases, this scenario applies to CP and genetic neurodevelopmental disorders. Generally, the prevalence of CP is 0.2%, while the prevalence of neurodevelopmental disorders such as intellectual disability and autism spectrum disorder ranges from 1.3% to 3.6%, with about 40% of these cases being genetically confirmed [23-25]. This means that roughly 0.001% to 0.003% of children have CP and a genetic neurodevelopmental disorder simultaneously. Therefore, when diagnosing patients with CP, it is important to consider not only the possibility of a monogenic cause and differential diagnosis but also the coexistence of conditions. Moreover, when CP and genetic neurodevelopmental disorders are diagnosed simultaneously, it is hard to determine which of the dual diagnoses has a greater impact on the patient’s physical impairment such as in Case 7.