In January 2024, individuals presenting to the Department of Clinical Genetics at Hong Kong Children’s Hospital with suspected pathogenic variants in the IRF6 or GRHL3 genes were assessed. Eight patients with molecularly confirmed Van der Woude syndrome (VWS) from four unrelated families were identified, aged between 17 and 58 years, with a male-to-female ratio of 5:3. All four index cases had an affected parent. Pedigrees of the four families are shown in online supplementary Figure 1.

Cleft palate was observed in 75% (n=6) of individuals, of whom two had bilateral cleft palate and two had submucous cleft palate. Lower lip pits were present in 62.5% (n=5). Cleft lip and/or alveolus was evident in four patients (50%), usually affecting both sides. Bifid uvula was observed in only one individual, while hypodontia was seen in two. One patient had ankyloglossia and two patients (25%) developed vitiligo.

The clinical findings are summarised in the Table. Clinical photos demonstrating oral findings from Family 4 are shown in the Figure.

A different variant was identified in each of the index patients in our cohort. Three harboured an IRF6 (NM_006147.4) variant, while the remaining index patient had a variant in GRHL3 (NM_198173.3). All variants were classified as likely pathogenic or pathogenic according to the American College of Medical Genetics and Genomics guidelines1. Two of the IRF6 variants were missense variants, while the remaining one was a splice site variant. The only GRHL3 variant identified in our cohort was a nonsense variant. All other variants in our cohort were novel, except IRF6 c.52G>A p.(Val18Met) which has been previously reported.2

The molecular findings of our patients are summarised in the Table and online supplementary Figure 2.

Orofacial cleft is a prevalent congenital defect with an estimated global occurrence of 1 in 500 to 1000 births and a comparatively higher incidence in Asia according to 2019 Global Burden of Disease data3. It is linked to both environmental and genetic factors. Most cases are non-syndromic, presenting a reduced risk of a genetic disorder, with orofacial clefts manifesting as isolated structural anomalies. Nevertheless, syndromic cases comprise approximately 30% of all cases, with an associated higher risk in patients with central cleft lip and/or palate (CL/P) or isolated cleft palate.4 Cases may be familial or non-familial. According to a local study,4 the diagnostic yield of genomic variants, including variants of uncertain significance, for non-syndromic orofacial clefts was only about 4%. In contrast, for syndromic cases, the detection rate of genomic variants is as high as around 70%. Among these, most variants have been detected by karyotyping or chromosomal microarray.5 In view of this finding, genetic testing (ie, chromosomal microarray) is conventionally offered as first-line screening in patients with syndromic orofacial clefts. Whole exome sequencing is less commonly performed in Hong Kong’s public clinical sector, although there is no international consensus. Clinical and molecular findings from our four families with VWS suggest that whole exome sequencing may be helpful in making a genetic diagnosis in patients with orofacial clefts. This is advantageous for both the patient and their family, as reproductive options such as preimplantation genetic diagnosis or prenatal diagnosis can be provided for at-risk individuals.

Van der Woude syndrome has historically been linked to pathogenic variants in IRF6, which encodes a transcription factor essential for the differentiation of skin, as well as breast and oral epithelium. Abnormal differentiation of the epidermis or oral periderm may be implicated in the pathogenesis of CL/P. The subsequent increase in the number of patients with CL/P has been accompanied by the discovery of a second gene, GRHL3. In vivo studies have demonstrated that GRHL3 regulates the epidermal permeability barrier through action downstream of IRF6, explaining the phenotypic convergence.6 Although it has been postulated that GRHL3 is more likely associated with cleft palate and less likely with cleft lip, CL/P and lip pits, the number of affected patients remains too small to draw definitive conclusions about genotype-phenotype correlations. In our cohort, all three individuals with a GRHL3 variant had cleft palate (mostly submucous), no cleft lip, and only one had lower lip pits. These findings appear to align with previous reports.2 4 5

According to the literature, approximately 60% of VWS cases show familial occurrence.7 All index individuals in our cohort had an affected parent. The concurrence of lower lip pits and CL/P has been reported as the most common clinical features among patients with VWS, affecting 80% of cases.8 In total, 75% and 62.5% of our patients had cleft palate or lower lip pits, respectively. Cleft lip and/or alveolus was evident in 50% of individuals. These findings are comparable with figures described in previous studies.2 4 5 Disease-causing variants in both genes have also been associated with dental anomalies, including hypodontia, dental aplasia, and malocclusion. Bifid uvula, hypodontia and ankyloglossia were also observed in our cohort. It is known that individuals with VWS exhibit highly variable expressivity, ranging from isolated lower lip pits to bilateral CL/P. As highlighted by Family 4 in our cohort, lower lip pits were identified only in individual II:2, but not in her elder sibling (II:1) or father (I:1). This again demonstrates the broad intrafamilial variability.

An enriched prevalence of vitiligo was also observed in our cohort. Although not previously reported in patients with VWS, two patients (one harbouring an IRF6 and one a GRHL3 variant) in our cohort had vitiligo. In individual F1 I:1, vitiligo was diagnosed during adulthood; in individual F4 II:2, at 16 years of age. Our observation points to a possible disease association, with the argument that interferon regulatory factors play a significant role in the immune system by functioning as major transcriptional regulators of type I interferon.9 Further research has revealed that IRF6 regulates a subset of toll-like receptor 3 responses in human keratinocytes and may play a role in keratinocyte and/or immune cell functions during cell damage and wound healing. Alternatively, GRHL3 is a transcription factor critical for epidermal differentiation and skin barrier repair. A possible association of dysfunction in interferon regulatory factors or GRHL3 with autoimmune dermatological conditions such as vitiligo cannot be excluded. Further studies are required to confirm a potential correlation.

All IRF6 and GRHL3 variants found in our cohort, except one, were novel. In previous VWS reports, missense variants in IRF6 were mainly clustered in exons 3, 4, 7, and 9, whereas truncating variants were evenly distributed across the whole gene.10 One novel IRF6 variant in our cohort was a missense variant in exon 4, while another was a splice site variant in intron 2. No specific genotype-phenotype correlation was established in our current study due to the limited sample size.

Due to the highly variable expressivity and incomplete penetrance, it is essential for clinicians to remain vigilant in diagnosing individuals with a relatively mild phenotype. Referral to clinical geneticists for consideration of genetic testing is beneficial for affected individuals with familial occurrence of CL/P or when other features suggest a syndromic diagnosis. In view of the limited number of individuals identified in our cohort, future studies may be needed to establish clearer genotype-phenotype genotypephenotype correlations, explore a potential association with vitiligo, and evaluate the diagnostic efficacy of whole exome sequencing.

Concept or design: LT Leung, SKL Ho.
Acquisition of data: WC Yiu, M Ou, JYY Poon, SSW Cheng, IFM Lo, HM Luk.
Analysis or interpretation of data: LT Leung, SKL Ho, WC Yiu.
Drafting of the manuscript: LT Leung, SKL Ho, IFM Lo, HM Luk.
Critical revision of the manuscript for important intellectual content: IFM Lo, HM Luk.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors disclosed no conflicts of interest.

The authors thank the patients and their family for their support.

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

This study was conducted in accordance with the Declaration of Helsinki. The patients/their legal guardian provided written informed consent for participation and publication of this case report.

The supplementary material was provided by the authors and some information may not have been peer reviewed. Accepted supplementary material will be published as submitted by the authors, without any editing or formatting. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by the Hong Kong Academy of Medicine or the Hong Kong Medical Association. The Hong Kong Academy of Medicine and the Hong Kong Medical Association disclaim all liability and responsibility arising from any reliance placed on the content.

1. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in medicine. 2015 May;17(5):405-23. Crossref

2. Kondo S, Schutte BC, Richardson RJ, et al. Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Genet 2002;32:285-9. Crossref

3. Wang D, Zhang B, Zhang Q, Wu Y. Global, regional and national burden of orofacial clefts from 1990 to 2019: an analysis of the Global Burden of Disease Study 2019. Annals of Medicine. 2023 Dec 12;55(1):2215540. Crossref

4. Chan KW, Lee KH, Pang KK, Mou JW, Tam YH. Clinical characteristics of children with orofacial cleft in a tertially centre in Hong Kong. HK J Paediatr (new series) 2013;18:147-51.

5. Li YY, Tse WT, Kong CW, et al. Prenatal diagnosis and pregnancy outcomes of fetuses with orofacial cleft: a retrospective cohort study in two centres in Hong Kong. Cleft Palate Craniofac J 2024;61:391-9. Crossref

6. De La Garza G, Schleiffarth JR, Dunnwald M, Mankad A, Weirather JL, Bonde G, Butcher S, Mansour TA, Kousa YA, Fukazawa CF, Houston DW. Interferon regulatory factor 6 promotes differentiation of the periderm by activating expression of Grainyhead-like 3. Journal of investigative dermatology. 2013 Jan 1;133(1):68-77. Crossref

7. Children's Hospital of Philadelphia. Van der Woude Syndrome [Internet]. Philadelphia: CHOP; 2024 Mar 31 [cited 2025 Nov 25]. Available from: https://www.chop.edu/conditions-diseases/van-der-woude-syndrome

8. Hersh JH, Verdi GD. Natal teeth in monozygotic twins with Van der Woude syndrome. Cleft Palate Craniofac J 1992;29:279-81. Crossref

9. Honda K, Takaoka A, Taniguchi T. Type I interferon [corrected] gene induction by the interferon regulatory factor family of transcription factors. Immunity 2006;25:349-60. Crossref

10. Leslie EJ, Standley J, Compton J, Bale S, Schutte BC, Murray JC. Comparative analysis of IRF6 variants in families with Van der Woude syndrome and popliteal pterygium syndrome using public whole-exome databases. Genetics in Medicine. 2013 May;15(5):338-44. Crossref

A 67-year-old Asian female with no family history of heart disease was admitted to the Department of Psychosomatic Diseases at our hospital on 30 November 2024, primarily for intermittent anxiety and a 2-week history of general malaise but also headache, stomachache, backache, a sensation of choking and throat pain, dyspnoea, and, occasionally, a feeling of impending death. After attending a psychiatric hospital she was diagnosed with anxiety and prescribed oral flupentixol and melitracen tablets, zaleplon and oxazepam. Upon admission to the department, symptomatic interventions including anxiolytics and sleep aids were administered. On 3 December 2024 at 9:01 am, the patient abruptly encountered back pain. During the episode, the patient appeared to be choking but reported no perspiration or chest pain. Urgent electrocardiogram revealed ST segment elevation of 0.2 to 0.25 mV in the augmented vector right lead and depression of 0.2 to 0.35 mV in lead I, augmented vector left lead, lead II, and leads V1 to V5 (Fig 1). The high-sensitivity troponin was 0.061 ng/mL (normal, <0.016 ng/mL). In view of the suspected acute myocardial infarction, the patient was transferred to the Coronary Care Unit. Emergency coronary angiography revealed 90% stenosis in the proximal segment of the right coronary artery (RCA). The left anterior descending artery (LAD) and left circumflex artery were obscured in various angiographic views, while the blood flow in the distal segment of the RCA could be observed. Intravascular ultrasound examination of the RCA revealed a minimum lumen area of 2.59 mm² at the lesion site, with a plaque burden of 91%. Subsequent to balloon dilation, a single everolimus-eluting stent (PROMUS Element Plus; Boston Scientific, Marlborough [MA], United States) was implanted. Postoperatively there was no residual stenosis at the RCA lesion site, and distal blood flow was classified as TIMI III (the Thrombolysis in Myocardial Ischemia Trial III) [Fig 2a-d]. Postoperatively, the patient reported no pain in the precordial region or back.

Coronary computed tomography angiography (CTA) [Fig 2e and f] revealed that the left main trunk, LAD, and left circumflex artery did not arise from the left coronary sinus. The RCA originated from the right coronary sinus. These results aligned with the findings of coronary angiography. Echocardiography revealed that the left ventricular ejection fraction was 60% with mild mitral regurgitation and reduced left ventricular diastolic function. The patient’s condition improved after treatment and she was discharged home on 10 December. At 1-month follow-up, the patient reported no chest tightness, shortness of breath, or precordial pain, and ECG showed no signs of ischaemia.

A single coronary artery (SCA) is an uncommon coronary artery anomaly (CAA). The congenital absence of the left coronary artery (LCA) is an uncommon subtype of SCA, occurring with an incidence rate of 0.024%, with no discernible gender disparity.1 In 1979, Lipton classified SCA into types I, II, and III based on coronary origin, branching pattern, and disease course.2 In our patient, the single RCA was type R-I. As myocardial ischaemia is the aetiology of cardiovascular events induced by CAAs, coronary angiography is vital. Prior to percutaneous coronary intervention, a meticulous assessment of the surgical treatment strategy is essential.

The SCA form of CAA typically presents with no clinical symptoms and lacks specificity. Our patient exhibited cardiac-related symptoms, including backache, a sensation of choking, throat pain, and dyspnoea. Nonetheless the simultaneous occurrence of symptoms such as headache, stomachache and general malaise prompted a diagnosis of anxiety disorder. Coronary angiography was conducted to assess the extent of vascular stenosis but revealed the absence of an LCA, complicated by 90% stenosis in the proximal segment of the RCA. The lesion in the proximal segment of the patient’s RCA was comparable to that in the left main trunk. Thereafter, under intravascular ultrasound guidance, the coronary artery lesions were assessed, and a therapeutic approach was planned. Coronary CTA is crucial for identifying aberrant openings and congenital anomalies and unequivocally confirmed the congenital absence of the LAD in our patient. Secondary prevention of coronary heart disease is essential for these patients, ensuring proper maintenance of the lumen in the distal vessels of the RCA that supply the anterior and lateral walls.

The congenital absence of the LCA is an uncommon condition with an often non-specific clinical presentation. Clinically, if patients exhibit symptoms of angina pectoris or electrocardiogram alterations indicative of ischaemia, coronary CTA or coronary angiography should be promptly conducted to exclude congenital cardiovascular malformations. This enables clinicians to accurately diagnose and implement appropriate management.

All authors contributed to the concept or design of the study, acquisition of data, analysis or interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

The study was funded by:
(1) Study of the Mechanism of Yangyin Shuxin Formula Inhibiting Calcium Overload in Cardiomyocytes through PI3K/IP3R Pathway, Improving Ejection Fraction, and Preserving Diastolic Function in Heart Failure, Key Research Project of Traditional Chinese Medicine in Tianjin (Ref No.: A0101); and
(2) Study on the immune-inflammatory mechanism of optimizing the polarization of macrophages mediated by IL-17 in Xinshengmaisan targeting myocardial fibrosis, Research Fund of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine (Ref No.: XB2024006).
The funders had no role in the study design, data collection/ analysis/interpretation, or manuscript preparation.

The study was approved by the Institutional Review Board of The First Teaching Hospital of Tianjin University of Traditional Chinese Medicine, China (Ref No.: TYLL2024[Z]). Written informed consent was obtained from the patient for all treatments and procedures, and for the publication of this case report (including the accompanying clinical images).

1. Sampath A, Chandrasekaran K, Venugopal S, et al. Single coronary artery left (SCA L)–right coronary artery arising from mid-left anterior descending coronary artery: new variant of Lipton classification (SCA L-II) diagnosed by computed tomographic angiography. Echocardiography 2020;37:1642-5. Crossref

2. Lipton MJ, Barry WH, Obrez I, Silverman JF, Wexler L. Isolated single coronary artery: diagnosis, angiographic classification, and clinical significance. Radiology 1979;130:39-47. Crossref

A 49-year-old woman presented to our hospital with a gastric submucosal mass in March 2024. She was asymptomatic with no abdominal signs, and laboratory tests revealed no significant abnormalities. Abdominal contrast-enhanced computed tomography (CE-CT) identified a 3.2 × 2.2 cm² rounded mass on the greater curvature of the gastric antrum. The mass protruded into and out of the cavity, displaying a clear boundary, slightly uneven density, evident inhomogeneous enhancement, and marked vascularity, indicative of abundant blood supply. Endoscopic ultrasound (EUS) examination revealed a smooth-surfaced spherical bulge originating from the muscularis propria layer. It appeared hypoechoic with areas of hyperechoic signals, protruding into and out of the lumen, measuring 24.5 × 20.3 mm² in cross-section (Fig 1).

The tentative diagnosis was gastric stromal tumour. The patient underwent laparoscopy endoscopy cooperative surgery (LECS). During the operation, the tumour was observed to have an irregular shape, with no invasion or adhesion to surrounding organs and no enlarged lymph nodes in the abdominal cavity. The surgical specimen revealed a tan-white solid tumour measuring 3.2 × 2.5 × 2 cm³, with a complete capsule. The margins of the specimen were clean. Postoperatively, routine symptomatic supportive treatments such as fasting, gastrointestinal decompression, anti-infection therapy, acid suppression, and nutritional support were provided. The patient recovered uneventfully.

Histological examination revealed the tumour to be multifocally centred in the muscularis propria, arranged in nests and cribriform patterns, with intraluminal dense eosinophilic material. The cells were oval, with a few fusiform in shape (Fig 2). Immunohistochemistry revealed tumour cells to be positive for CD56, cytokeratin, CD34, CD10, and vimentin. The Ki-67 proliferation index was 3%. We performed whole-transcriptome messenger RNA sequencing of tumour tissue, which revealed a fusion between MALAT1 (exon 1) and GLI1 (exon 3). Based on the above evidence, the patient was definitively diagnosed with gastroblastoma.

Gastroblastoma is a rare biphasic gastric tumour characterised by its distinctive biphasic epithelial-mesenchymal morphology. Diagnosis from biopsy specimens is typically challenging and often requires additional immunohistochemistry.1 In 2009, Miettinen et al1 reported the first case of gastroblastoma. At the time of writing, only 27 cases had been reported.2

According to the review by Luo et al,2 the 27 patients ranged from 5 to 74 years, with a mean age of 35 years. There were 14 male and 13 female patients. Tumour size ranged from 1.3 to 15 cm, with an average of 5.7 cm. Most lesions occurred in the gastric antrum. Clinical manifestations were non-specific, and the tumours primarily involved the muscularis propria. Computed tomography (CT) and EUS were the most commonly used diagnostic methods, typically revealing a mixed solid-cystic mass with heterogeneous hypoechoic areas, occasionally accompanied by ulcers. Two patients underwent preoperative EUS-guided fine-needle aspiration. In our case, the preliminary diagnosis was gastrointestinal stromal tumour based on EUS and CE-CT findings. Endoscopic ultrasound–guided fine-needle aspiration was not performed to avoid damaging the tumour and increasing the risk of metastasis. No clear associated with systemic conditions was observed.

Gastroblastoma rarely expresses markers typically positive in gastrointestinal stromal tumours, solitary fibrous tumours, gastric schwannomas, or mesotheliomas. It also generally lacks expression of markers characteristic of gastric neuroendocrine tumours or leiomyomas.3 In the 27 reported cases,2 most tumour cells were positive for vimentin, CD56, CD10, and PCK. The MALAT1-GLI1 fusion gene was detected in nine patients and is considered a valuable diagnostic marker for this tumour.2

Only three cases presented with organ or lymph node metastasis.2 One patient had peritoneal, liver, and pelvic metastases, as well as bladder adhesion and lymph node metastasis at diagnosis. Following partial gastrectomy, no recurrence or metastasis was observed at 3-month follow-up. Another patient had two lymph node metastases at the splenic hilum prior to surgery. After partial gastrectomy and splenectomy, local recurrence was noted at 6 months, and surgical debulking was performed. Another patient had liver metastasis preoperatively, but no postoperative follow-up information was available. Among the other surgically treated patients, only one experienced local recurrence. These findings suggest that while gastroblastoma may exhibit indolent behaviour, preoperative metastasis or invasion may be associated with a poor prognosis.

Surgery remains the mainstay of treatment. Of the 27 cases,2 23 underwent surgical resection and three received endoscopic treatment. The mean tumour size in the surgical group was 5.80 cm, compared to 1.91 cm in the endoscopic treatment group, suggesting that tumour size influenced the choice of treatment modality.2 Among the endoscopically treated patients, one underwent endoscopic full-thickness resection, and two underwent endoscopic submucosal dissection.2 The choice of endoscopic resection generally depends on lesion depth, size, and location. When endoscopic resection is not feasible, laparoscopy endoscopy cooperative surgery may be considered.4 In our case, laparoscopy endoscopy cooperative surgery was used, offering the benefits of minimally invasive surgery while reducing postoperative complications. In one previous case, a patient with a 15-cm gastroblastoma underwent postoperative radiotherapy and remained disease-free after 14 years of follow-up.2 The remaining patients received routine postoperative follow-up.2 Given the potential for recurrence, particularly in cases with preoperative metastasis or invasion, adjuvant radiotherapy or chemotherapy may be considered on an individual basis. The average follow-up duration across reported cases was 31 months.2 Given its generally indolent nature, annual follow-up is recommended for most patients with gastroblastoma. For those with preoperative metastasis or invasion, more frequent follow-up every 3 to 6 months is advisable. As gastroblastoma remains rare, additional case reports and studies are needed to enhance our understanding of its biological behaviour, diagnosis, and optimal management.

Concept or design: Q Chen.
Acquisition of data: Y Sun, H Shi, X Wu, Y Wang, E Linghu.
Analysis or interpretation of data: Y Sun, H Shi, J Wang, Y Yuan, J Pang, L Wei, S Song.
Drafting of the manuscript: Y Sun.
Critical revision of the manuscript for important intellectual content: Q Chen.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This study was funded by the 14th Five-year National Key Research and Development Program of China (Project No.: 2022YFC2503600). The funder had no role in study design, data collection, analysis, interpretation, or manuscript preparation.

This study was approved by the Ethics Committee of Chinese People’s Liberation Army General Hospital, China (Ref No.: S2023-188-01). The patient was treated in accordance with the Declaration of Helsinki and provided written informed consent for all treatments, procedures and the publication of this case report.

1. Miettinen M, Dow N, Lasota J, Sobin LH. A distinctive novel epitheliomesenchymal biphasic tumor of the stomach in young adults (“gastroblastoma”): a series of 3 cases. Am J Surg Pathol 2009;33:1370-7. Crossref

2. Luo Z, Cui J, Ma F, et al. Gastroblastoma—a case report and literature review. World J Surg Oncol 2024;22:255. Crossref

3. Chen C, Lu J, Wu H. Case report: submucosal gastroblastoma with a novel PTCH1::GLI2 gene fusion in a 58-year-old man. Front Oncol 2022:12:935914. Crossref

4. Sharzehi K, Sethi A, Savides T. AGA clinical practice update on management of subepithelial lesions encountered during routine endoscopy: expert review. Clin Gastroenterol Hepatol 2022;20:2435-43.e4. Crossref

A 67-year-old woman with an unremarkable ophthalmic history presented to the ophthalmology department with a ‘gush of fluid’ and reduced vision in her left eye, from 6/20 in the Snellen test (recorded in 2023 at a private hospital), to hand movements, after her sixth cycle of combination chemotherapy (paclitaxel and carboplatin) and targeted therapy (combined pertuzumab/trastuzumab in a subcutaneous injection) for metastatic breast cancer. She had experienced bilateral eye discomfort and intermittent epiphora since the start of her combined chemotherapy/targeted therapy. She denied any topical medication use such as steroid or nonsteroidal anti-inflammatory drugs.

Clinical findings revealed left upper lid swelling and conjunctival injection, a central corneal perforation with iris plugging and a flat anterior chamber in the patient (Fig 1a). Her right eye showed mild punctate epithelial erosions in keeping with mild dry eye disease; there was no sign of corneal melting. Corneal sensation in both eyes was intact. Schirmer’s test was not performed at first presentation due to the emergent nature of the condition.

Viral and bacterial conjunctival swabs were taken and cyanoacrylate glue and a bandage contact lens were applied immediately to the left eye corneal perforation (Fig 1b). Viral and bacterial cultures were negative. Blood tests for rheumatoid factor, anti–extractable nuclear antigen screen and anti–nuclear antigen to look for an autoimmune cause were also negative.

The combined therapy was discontinued immediately and the patient underwent a multilayer amniotic membrane transplant along with anterior chamber reformation to restore globe integrity while awaiting a subsequent corneal graft transplant (Fig 2a). One month later, with her ocular condition temporarily stabilised, she underwent radical mastectomy for her breast cancer. No further adjuvant chemotherapy was administered.

Five months later, the patient’s general health had stabilised. With the development of an intumescent cataract in her left eye, a triple procedure of penetrating keratoplasty, extracapsular cataract extraction and intrascleral haptic fixation of intraocular lens (due to zonulysis) was performed.

Fortunately, the patient’s left eye visual acuity recovered to 6/20 in the Snellen test and was expected to further improve following suture removal (Fig 2b). The patient did not develop any further corneal erosion or melting in either eye after drug discontinuation. Her right eye was well at the latest follow-up and showed no sign of ocular surface disease. The most recent Schirmer’s test in both eyes was within normal limits.

The top differential aetiologies in this case were herpetic, neurotrophic, or drug-induced. Although herpetic stromal keratitis does not usually present at such a late stage, it can occur in immunocompromised patients. Nonetheless, since the patient denied a history of herpetic infection and viral swabs were negative, this diagnosis was unlikely. Neurotrophic keratopathy was also excluded on the basis of bilaterally normal corneal sensation. In light of the timing of symptom onset following the chemotherapy/targeted therapy, combined with a negative history of ocular surface disease, we suspect the corneal perforation was most likely drug-induced.

Carboplatin has not been reported to cause ocular surface side-effects although it has been rarely linked to optic neuropathy, retinal ischaemia, and pigmentary maculopathy.1 Paclitaxel is a taxane-based chemotherapeutic agent and aside from cystoid macular oedema, has also been associated with scintillating scotoma, photopsia, dry eye, conjunctivitis, and limbal stem cell deficiency.1 2

Human epidermal growth factor receptor 2 (HER2) antibody drugs, pertuzumab and trastuzumab, are used in the treatment of HER2-positive breast cancer and were administered concurrently with chemotherapy in our patient. Given the presence of HER2 receptors on the ocular surface and lacrimal glands, ocular surface disease can occur as a side-effect.2 The ocular side-effects of trastuzumab have been well documented and include conjunctivitis, dry eye, and marginal keratitis.2 3 Nonetheless there have been limited case reports of aggressive corneal melting associated with trastuzumab monotherapy.4 Pertuzumab has been associated with epiphora and conjunctivitis, but to date there have been no reported cases of corneal melting or perforation linked to pertuzumab or the specific combination formulation of pertuzumab/trastuzumab targeted therapy. We hypothesise that the concurrent use of taxane-based chemotherapy, known for its ocular surface side-effects, in combination with HER2-targeted therapy, may have contributed to the aggressive corneal melting observed in this case. Nonetheless the possibility that the HER2 combination therapy (pertuzumab/trastuzumab) alone could have been responsible cannot be ignored.

It is of note that this new formulation drug is given as a fixed-dose subcutaneous injection of 1200 mg pertuzumab and 600 mg trastuzumab for one loading dose followed by 600 mg pertuzumab and 600 mg trastuzumab every 3 weeks for maintenance. The conventional intravenous dose of trastuzumab is based on body weight (8 mg/kg loading, 6 mg/kg maintenance). It is difficult to compare doses since the administration routes differ, although a phase 3 trial showed that the serum trough level of the drug is slightly lower when administered subcutaneously.5 Therefore, the severe ocular presentation in this case could not be explained by a higher dosage or bioavailability of the anti-HER2 drugs since bioavailability would be lower with the subcutaneous route.5

Furthermore, in the phase 3 trial on this new formulation combination targeted therapy drug, it was also administered along with taxane chemotherapy (similar to our patient), a common regimen in hormone receptor positive breast cancer.5 All side-effects were recorded but no ocular surface side-effects were reported.5 Hence, the potential ocular surface side-effects are not listed in the package insert for this relatively new combined subcutaneous formulation of pertuzumab and trastuzumab.

In conclusion, our patient experienced unilateral vision loss due to corneal perforation while undergoing treatment with a combination of chemotherapy and targeted therapy. Among the drugs administered, paclitaxel and trastuzumab are most frequently associated with ocular surface side-effects, and their concurrent use may have contributed to the severity of the presentation. Patients receiving combined taxane chemotherapy and HER2-targeted therapy—including newer formulations such as combined subcutaneous pertuzumab/trastuzumab—should be warned about potential ocular surface complications. Early referral for ophthalmic evaluation is recommended at the onset of any ocular symptoms to prevent serious, vision-threatening outcomes.

Concept or design: ALW Li, VWM Ho.
Acquisition of data: ALW Li, VWM Ho, DHT Wong.
Analysis or interpretation of data: ALW Li, VWM Ho.
Drafting of the manuscript: ALW Li, VWM Ho.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

As an editor of the journal, KKW Li was not involved in the peer review process. Other authors have disclosed no conflicts of interest.

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The study was approved by the Central Institutional Review Board of Hospital Authority, Hong Kong (Ref No.: IRB 2024- 707). The patient provided written informed consent for publication of this case report and unidentifiable information and images.

1. Bader A, Begemann M, Al-Obaidi A, Habib MH, Anwer F, Raza S. Ocular complications of antineoplastic therapies. Future Sci OA 2023;9:FSO871. Crossref

2. Vitiello L, Lixi F, Coco G, Giannaccare G. Ocular surface side effects of novel anticancer drugs. Cancers (Basel) 2024;16:344. Crossref

3. Kafa G, Horani M, Musa F, Al-Husban A, Hegab M, Asir N. Marginal corneal infiltration following treatment for metastatic breast cancer with triple chemotherapy of trastuzumab, pertuzumab & docetaxel. Ocul Immunol Inflamm 2023;31:431-6. Crossref

4. Barmas-Alamdari D, Chaudhary H, Baghdasaryan E, Dua P, Cheela I. Trastuzumab-induced early corneal melt in HER2-positive breast cancer: a case report and review. Am J Case Rep 2024;25:e945488. Crossref

5. Tan AR, Im SA, Mattar A, et al. Fixed-dose combination of pertuzumab and trastuzumab for subcutaneous injection plus chemotherapy in HER2-positive early breast cancer (FeDeriCa): a randomised, open-label, multicentre, noninferiority, phase 3 study. Lancet Oncol 2021;22:85-97. Crossref

A 47-year-old Chinese man presented with a 6-month history of neck pain, numbness and weakness in the left limb. In January 2019, he was admitted to the Department of Spine Surgery at our institution. His medical and family histories were unremarkable, with no prior interventions.

Neurological examination revealed deficits involving three levels of the left upper limb. Magnetic resonance imaging (MRI) revealed an extramedullary lesion at the C2-C3 level (Fig 1a to c). Following ophthalmological and dermatological assessments (including oral mucosa, external genitalia, and anus), a primary intradural extramedullary malignant melanoma (MM) was suspected. Surgical intervention was indicated to improve neurological function.

Although the tumour was in the left anterior cervical cord, its position posterior to the C2-C3 vertebral body and spindle-shaped appearance made anterior surgery less favourable, as resection of the vertebral body to access the tumour and reconstruction following removal would increase surgical trauma and risk. After discussing the risks and benefits with the patient and his family, informed consent was obtained prior to surgery.

The patient underwent posterior surgery under general anaesthesia with neuroelectrophysiological monitoring. A C2-C4 laminectomy was performed to expose the tumour, which was located on the left anterior side of the cord. The lesion was compressing and adherent to the spinal cord and dura. Using a nerve retractor and with stable intraoperative neuromonitoring, the tumour was carefully dissected and completely excised (Fig 1d to f). The dura was sutured and the laminae were re-implanted using titanium mini-plates. A drainage tube was placed, and the incision was carefully sutured.

Postoperative cervical MRI confirmed complete tumour resection (Fig 2a to c). Histopathological examination (Fig 2d to f) confirmed MM with positive staining for HMB-45 (+), S-100 (+), and a Ki-67 index of 5%.

The patient underwent postoperative rehabilitation and reported good neurological recovery at the 5-year follow-up, with minimal assistance required for certain left upper limb movements. Given the favourable outcome, the patient declined adjuvant radiochemotherapy as recommended by our team. He continues to lead a relatively active and independent life.

Malignant melanoma of the spinal region is rare and may present as either a primary or metastatic lesion, with primary cases being exceptionally uncommon.1 Primary spinal MM most frequently affects the middle and lower thoracic spine, with rare involvement of the cervical region.2 This is the first case of primary intradural extramedullary MM treated in our department in recent decades. Despite the tumour being located in the upper cervical region, satisfactory clinical outcomes were achieved.

Although MRI is the recommended and effective preoperative imaging modality to diagnose spinal MM, distinguishing it from other pigmented central nervous system tumours, such as leptomeningeal melanoma or melanocytoma,3 can be challenging. A definitive diagnosis relies on histopathological confirmation. Immunohistochemical staining using antimelanoma markers such as HMB-45 and S-100 can help confirm the diagnosis.2 In our case, positive staining for both HMB-45 and S-100 was conclusively confirmed. There are currently no established guidelines for the treatment of spinal MM, but complete surgical removal is typically recommended.4 In our case, the chief surgeon carefully removed the tumour starting on the left side and gradually accessed the anterior spinal cord. Complete resection was successfully achieved without damage to the spinal cord or nerve roots. It has been reported that patients may achieve better outcomes and prognosis with postoperative adjuvant radiotherapy and chemotherapy.5 In our case, the patient was enjoying a relatively good life at the 5-year follow-up without adjuvant therapy, indicating the importance of complete resection and consistent with the literature. However, the importance of radiotherapy or chemotherapy should not be ruled out as further follow-up is necessary to assess the final prognosis. Our team still recommends that postoperative radiotherapy or chemotherapy be routinely performed to improve patient outcomes.

Our research team reports a rare case of primary intradural extramedullary MM at the C2-C3 level with successful surgery. First, complete tumour resection is crucial for improving patient prognosis and survival, emphasising the need for careful consideration during treatment planning. Second, proper handling of the anatomical relationship between the tumour and the spinal cord or nerve roots is essential to remove the tumour as completely as possible while minimising neurological injury. Third, although our patient declined adjuvant therapy and achieved a favourable outcome at the 5-year follow-up, we continue to recommend routine postoperative radiotherapy or chemotherapy to improve prognosis. A key limitation of this case is the absence of radiological imaging at final follow-up, as the patient declined hospital visits due to perceived symptom improvement, which limits our ability to definitively exclude indolent recurrence. Our case provides valuable clinical insights, and long-term follow-up remains essential for monitoring outcomes.

Concept or design: L Tang, Y Chen.
Acquisition of data: L Tang, H Liu, M Wang, Y Zhou.
Analysis or interpretation of data: L Tang, Y Chen, J He, M Chen.
Drafting of the manuscript: L Tang, H Liu.
Critical revision of the manuscript for important intellectual content: Y Chen, J Zheng, F Wang.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

The authors are grateful to the clinical staff of the Department of Spine Surgery at Suining Central Hospital for their assistance in completing this article. The authors also thank Bullet Edits Limited for providing language editing and proofreading services.

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The study was approved by Suining Central Hospital's Ethics Committee for Biomedical Research Involving Human Beings, China (Ref No.: KYLLMC20240019). Written informed consent was obtained from the patient for the publication of this case report.

1. Hastings-James R. Melanoma of the central nervous system. Radiology 1973;109:357-60. Crossref

2. Lee CH, Moon KY, Chung CK, et al. Primary intradural extramedullary melanoma of the cervical spinal cord: case report. Spine (Phila Pa 1976) 2010;35:E303-7. Crossref

3. Smith AB, Rushing EJ, Smirniotopoulos JG. Pigmented lesions of the central nervous system: radiologic-pathologic correlation. Radiographics 2009;29:1503-24. Crossref

4. Nakamae T, Kamei N, Tanaka N, et al. Primary spinal cord melanoma: a two-case report and literature review. Spine Surg Relat Res 2022;6:717-20. Crossref

5. Kim MS, Yoon DH, Shin DA. Primary spinal cord melanoma. J Korean Neurosurg Soc 2010;48:157-61. Crossref