Clinical and Molecular Characteristics of Patients with Craniofacial Malformations Due to Syndromic and Non-Syndromic Etiologies

1. Introduction

By the fourth week of embryonic development, the pharyngeal arches, clefts, and pouches give rise to key craniofacial structures, including the face, palate, and skull [1]. Postnatally, facial growth follows somatic growth, with variations in timing and vectors influenced by biological sex and population characteristics [2,3]. Craniofacial diversity primarily results from subtle differences in relative shape, size, and spatial arrangement of facial features [2].

Craniofacial malformations are common in early childhood, with prevalence inversely related to age. Craniosynostosis affects approximately 1 in 2000-2500 newborns. Previous studies indicate that nearly 15% of full-term newborns may present craniofacial anomalies, a proportion that rises to 38% in preterm infants [5]. Despite this high prevalence, few studies have comprehensively characterized the phenotypic spectrum of these malformations.

A retrospective Spanish study (ENSERIo) reported a median diagnostic delay of five years for patients with rare diseases, with 20% experiencing delays of 10 years or longer [6]. During this period, patients may receive inadequate or inappropriate treatment, adversely affecting their health and quality of life. In regions where molecular testing is limited, diagnostic delays are compounded, creating significant gaps in management.

In South America, the ECLAMC - an ongoing case-control study established in 1967 - monitors congenital abnormalities and collects data from Argentina, Brazil, Bolivia, Chile, Colombia, Peru, and Venezuela. Four Colombian institutions, including Fundacion Valle del Lili, contribute to the registry [4]. This study aims to describe the clinical characteristics of patients with craniofacial malformation, identify associated genetic loci, and assess delays from clinical to molecular diagnosis.

2. Materials and Methods

2.1. Study Design and Population

This retrospective chart review included patients enrolled in the ECLAMC cohort at Fundacion Valle del Lili between 2011 and 2022. The institution serves as a referral center for congenital malformations in southwest Colombia. Patients were identified using ICD-11 codes for congenital malformations of the eye, ear, face, neck, cleft lip, and cleft palate. All patients with documented craniofacial malformation were included.

2.2. Data Collection

Medical records were reviewed to extract demographic data, clinical and radiological findings, and associated anomalies affecting the central nervous system, musculoskeletal, cardiovascular, digestive, and pulmonary systems. Data were collected using BD Clinic (Cali, Colombia) and de-identified for analysis. Ten percent of records were randomly audited for quality assurance.

2.3. Sample Size

Sample size was estimated using OpenEpi (Emory University, United States) for a descriptive study, assuming a finite population of 1776, an expected prevalence of 50% (+/-5%), a 95% confidence interval, and a design effect of 1.

2.4. Statistical Analysis

Categorical variables were summarized using frequencies and percentages. Association was tested using the chi-square test, while continuous variables were analyzed using the t-test or Mann-Whitney U test, as appropriate. A descriptive subgroup analysis of patients with syndromic and non-syndromic craniofacial malformations was performed based on phenotype and genotype. The timing between birth, clinical diagnosis, and molecular diagnosis was graphed in Kaplan-Meier cumulative incidence curves. The Kaplan-Meier method was used first between birth and clinical diagnosis, and subsequently between clinical and molecular diagnosis. Censoring was applied according to the last follow-up visit. Statistical analyses were performed using STATA version 14 (StataCorp LLC, Texas, USA), with significance set at p < 0.05.

2.5. Ethical Statement

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee in Biomedical Research of Fundacion Valle del Lili, Cali, Colombia (protocol code [XXX]; date of approval: [DD/MM/YYYY]). Patient consent was waived given the retrospective de-identified design.

3. Results

At study initiation, 1776 patients were screened according to predefined criteria. Of 330 medical records reviewed, 145 met inclusion criteria. Among these, 85 patients (58.6%) were male and 60 (41.4%) were female, with a median age of 13 years (IQR 6.6-18.7) (Table 1). Most patients resided in urban areas of southwest Colombia. Departments reporting the highest number of cases included Genetics, Neurology, Pediatrics, and Neonatology.

The most frequent craniofacial manifestations involved ocular and periocular structures, observed in 56 patients (38.6%). Findings included downward-sloping palpebral fissures, telecanthus, ocular hyper- or hypotelorism, ptosis, proptosis/exophthalmos, strabismus, and refractive disorders. Neurological features were also common: 75 patients (51.7%) had language disorders and/or intellectual disability, 38 (26.2%) had developmental complications, and 32 (22.1%) had movement disorders such as spasticity, palsy, hypotonia, or arthrogryposis (Table 2).

Of the 145 patients, 115 (79.3%) had an initial clinical diagnosis. Non-syndromic presentations accounted for 73 cases (50.3%), most commonly non-syndromic cleft lip and/or palate (46 patients; 31.7%), isolated non-syndromic craniosynostosis (13; 8.9%), and benign familial macrocephaly (10; 6.9%). Syndromic diagnoses were suspected in 42 patients (29.0%), while 30 (20.7%) lacked both clinical and suspected diagnoses (Table 2).

Genetic or molecular testing was performed in 24 patients (15.9%), yielding confirmed diagnoses in 19 cases. Thirty-eight molecular studies were performed, including 14 individual exomes (36.8%), 10 array CGH tests (26.3%), five G-band karyotypes (13.1%), five trio-exomes (13.1%), one MLPA (2.6%), one RASopathies panel (2.6%), and one AMER1 gene sequencing (2.6%). Identified syndromes included Cowden, Muenke, CHARGE, Joubert, and Pfeiffer syndromes, among others (Table 3).

Figure 1 shows a heat map of altered genes and their expression across brain tissues, demonstrating that the first 12 genes exhibited localized expression while others showed broader expression patterns. Kaplan-Meier analysis (Figure 2) indicated that 80% of patients received a clinical diagnosis by age 10 years. In contrast, only 23% received molecular confirmation within 103 months after their clinical diagnosis, with a median delay of 70 months.

4. Discussion

Most existing studies on craniofacial malformations in this population are epidemiological or focus on clinical prognosis, rather than genotypic characterization. We found no prior studies documenting the genotypic features of patients with craniofacial malformations in our population beyond well-recognized syndromes, such as craniosynostosis, cleft lip/palate, or midline malformations. In our cohort, genetic testing was performed only in patients with a strong clinical suspicion of an underlying syndrome. Notably, none of the patients with isolated craniofacial malformations underwent genetic testing to investigate a molecular explanation. Furthermore, some patients remained undiagnosed despite both clinical and genetic evaluation, while others were found to carry genetic variants that did not fully explain their phenotypic presentation.

The most common abnormalities identified on physical examination in this study were ocular and eyelid malformations, followed by cleft lip and palate. These findings contrast with a previous study conducted in Colombia, in which ear malformations were the most frequently reported anomalies [7]. Our study not only differs in identifying the eyes as the most affected anatomical region, but also highlights that ocular and eyelid malformations were more than twice as frequent as ear malformations (38.6% vs. 15.2%, respectively).

Cleft lip and palate are relatively common congenital anomalies that can present as isolated, non-syndromic conditions or as part of a Mendelian syndrome [8]. Consistent with a previous characterization study conducted in Colombia, where cleft lip/palate represented the most frequent cause of facial fissures and most cases were non-syndromic [8,9], we found that among the 72 patients with non-syndromic craniofacial anomalies, 46 had non-syndromic cleft lip/palate.

A significant diagnostic delay was observed in this cohort, both clinical and at the molecular level. As shown in Figure 2, many patients waited several years before receiving a definitive clinical diagnosis, and approximately 80% received a diagnosis within the first 10 years of life. Only 24 patients (15.8%) underwent molecular/genetic testing, and not all received a definitive diagnosis. Among those tested, several known craniofacial syndromes were identified, including Cowden syndrome, Muenke syndrome, CHARGE syndrome, and Joubert syndrome (Table 3). Cases of syndromic craniosynostosis, such as Pfeiffer syndrome caused by FGFR mutations, were also found [10,11]. For example, one female patient was diagnosed by NGS with a heterozygous FGFR2 variant (10q26.3 c.1087+1218G>C) previously reported in the Orphanet database.

We identified variants in the following genes: PTEN, OTC, EFTUD2, ACVR1, AMER1, FGFR3, RPS6KA3, CHD7, KATNIP, PTPN11, EFNB1, STK11, COCH, FGFR2, MITF, L1CAM, ANKS6, and AHDC1. In 13% of cases, molecular testing did not yield a definitive diagnosis, but variants of uncertain significance (VUS) were detected in several genes, including TP63, STAMBP, OCLN, and FRMPD4. One patient with a suspected syndromic diagnosis and reported VUS in these genes later died. A multidisciplinary board concluded that the TP63 variant was the most likely cause, and recommended segregation studies in the patients parents. Unfortunately, after the patients death, the family did not pursue further genetic counseling.

We identified novel variants associated with craniofacial malformations, as well as previously reported variants relevant to genotype-phenotype correlations. Importantly, we observed a longer delay in molecular diagnosis than reported in other countries, likely reflecting limited access to specialized, high-complexity centers in the region. Only 23% of patients with a clinical diagnosis had molecular confirmation (IQR 15% to 34%). Among those tested, the median diagnostic delay was 70 months, leading to suboptimal management, delayed interventions, and reduced rehabilitative potential.

5. Limitations

This study is limited by potential underreporting due to its retrospective design, as well as selection bias arising from the non-systematic review of clinical records. Future research should focus on systematic data collection to reduce bias and provide reliable clinical and molecular datasets. Such efforts will improve the accuracy of extrapolating findings to the general population, and ultimately support earlier diagnosis and management of patients with craniofacial malformations.

6. Conclusionss

Clinical and molecular characterization distinguishes syndromic from non-syndromic craniofacial malformations and informs comprehensive care. Significant diagnostic delays – particularly at the molecular level – highlight the urgent need for early genetic testing to guide treatment and provide timely counseling for patients and their families. Expanding access to molecular diagnostics in Latin America is critical to reducing these gaps.