BRAF/MEK inhibitors use to treat ventriculoperitoneal shunt‐associated ascites in pediatric low‐grade gliomas

To the Editor: Low-grade gliomas (LGGs), the commonest brain tumors in children, are characteristically driven by alterations in the RAS/MAPK pathway.1 This pathway can be targeted with BRAF/MEK inhibitors and achieve better progression-free survival than chemotherapy.2, 3 When hydrocephalus complicates LGGs, a ventriculoperitoneal shunt (VPS) may be needed. VPSs may be complicated by infections, malfunction,4 ascites,5 and metastatic spread.6 The mechanism underlying VPS-associated noninfectious nonmalignant ascites is not well understood. However, the high cerebrospinal fluid (CSF) protein content and the inability of the peritoneal lymphatic drainage to effectively absorb the fluid seem to be the most accepted theory.7, 8 When ascites develop, shunt diversion is usually needed and the alternatives would be ventriculo-atrial (VA),5 or ventricular-gallbladder,9 or ventriculo-pleural shunts. VA shunts have several complications (e.g., endocarditis, pulmonary hypertension, right-side heart failure),10 in addition to its challenging technical insertion in young children. There are several case reports of noninfectious nonmalignant shunt ascites in children with LGGs5, 8, 9; most needed shunt diversion. Here, we present two patients with desmoplastic infantile astrocytoma (DIA)/desmoplastic infantile ganglioglioma (DIG), who escaped shunt diversion and had quick resolution of ascites using BRAF/MEK inhibitors. Case 1 is a 5-month-old male who presented with vomiting, nystagmus, and macrocephaly. Magnetic resonance imaging (MRI) showed metastatic suprasellar mass and hydrocephalus (Figure 1: Case 1). VPS was inserted, and 6 days later ascites occurred. VPS was externalized till ascites resolved (19 days), then tumor biopsy with shunt internalization was performed. Pathology was DIA, BRAFv600E-mutated. Fourteen days later, ascites recurred, and a peritoneal drain was inserted. Dabrafenib was started, and ascites resolved (Table 1). CSF: 177–450 mg/dL Ascites: 0.5–0.9 g/dL -CT brain: no hydrocephalus -Abdominal ultrasound/CT confirmed ascites with no loculations -CSF and peritoneal samples were negative for bacterial infections and malignant cells -1st Event: temporary shunt externalization -2nd Event: peritoneal drain insertion and started compassionate dabrafenib (5.25 mg/kg/day) CSF: 45–60 mg/dL Ascites: 1.4 g/dL -CT brain: no hydrocephalus -Abdominal ultrasound/CT confirmed ascites with no loculations -CSF and peritoneal samples were negative for bacterial infections and malignant cells -1st Event: temporary shunt externalization and peritoneal drain insertion -2nd Event: peritoneal drain insertion and started compassionate combined dabrafenib (5.25 mg/kg/day) and trametinib (0.032 mg/kg/day) Case 2 is a 4-month-old male who presented with irritability, vomiting, and macrocephaly. MRI showed left hemispheric multicystic solid mass (Figure 1: Case 2). Partial tumor resection was complicated by convulsions and intratumoral bleeding. Pathology was DIG, BRAF p.G469A-mutated. VPS was inserted 6 weeks later, but needed three revisions in 2 months due to malfunction with high CSF protein (52–571 mg/dL). Chemotherapy11 was started, with variable cystic response after 6 months. Partial tumor resection and ommaya insertion were performed; thereafter, the patient was observed. Ascites developed 9 months later. VPS was externalized, and a peritoneal drain was inserted. With ascites resolution (2 weeks), VPS was internalized. A week later, ascites recurred, and peritoneal drain was inserted. Combined dabrafenib and trametinib were started and ascites resolved (Table 1). VPS-associated ascites usually needs shunt diversion to be controlled. Our two patients received targeted therapies to control their tumors, hoping that this happens quick enough and helps control their noninfectious nonmalignant ascites. Ascites was controlled quickly (within 3 weeks) and saved these children the challenges of VAS insertion, the least of which is the need for recurrent shunt revisions at their young age. This seems to support the role of high CSF protein associated with large tumor burden on the pathophysiology of ascites. Our experience suggests that trial of targeted therapy is a plausible medical alternative, particularly when there are technical challenges associated with VAS insertion. A literature review12 identified 21 children with optic pathway gliomas (OPGs) who developed ascites at a median of 5 months from shunt insertion; 19 needed shunt diversion. Ascites was controlled without shunt diversion in few patients,7 in the literature though it was not clear if these patients received any tumor-directed therapy. In other cases, the addition of chemotherapy seemed to play a role in controlling ascites. Kretschmar and Linggood6 reported an OPG and malignant ascites, which resolved with intensive chemotherapy in 1 week coinciding with tumor shrinkage. Legault et al.13 reported on a disseminated LGG (without BRAFv600E or KIAA1549-BRAF) and VPS that developed recurrent ascites while receiving sorafenib. Ascites occurred at 4 weeks, resolved with stopping sorafenib and recurred on resuming it. The tumor progressed on sorafenib, so treatment was shifted to bevacizumab/irinotecan. Ascites did not recur (>19 months). This was reported as a possible sorafenib-related complication, though it may be argued that ascites occurred secondary to tumor growth on sorafenib, which was reported to cause “paradoxical” acceleration of LGG growth.14 Solano-Páez et al.12 collected data on 19 children with non-NF1 nonmetastatic LGG (17 were OPGs) with shunt ascites. Eight patients received dabrafenib or trametinib to control their tumor growth post ascites. In addition to targeted therapy use (which was exclusive in four patients), four patients received other modalities of therapies (chemotherapy [3], surgery [1], radiotherapy [1]). Five of the eight patients had shunt diversion to control their ascites. The three patients who did not have shunt diversion were: one with BRAFv600E mutated OPG who was on dabrafenib before developing ascites, which was treated conservatively, a critically ill infant with BRAFv600E mutated OPG and diencephalic syndrome who developed tumor progression on vinblastine and then had profound clinical and radiological response on dabrafenib,15 and the third patient had KIAA1549-BRAF-OPG for which trametinib was used after ascites was drained with repeated paracentesis; however, with further tumor progression she received multiple lines of chemotherapy. In conclusion, we present two patients in whom we successfully controled shunt ascites medically with targeted therapy and were able to avoid shunt diversion. This seems an effective alternative when the underlying molecular alteration is identified and targeted therapy is available. We would like to thank Novartis for providing compassionate access to dabrafenib/trametinib to our patients. We also thank the KHCC-pharmacy personnel for facilitating drug shipments. Eric Bouffet is a member of the advisory board of Novartis and Alexion. The remaining authors declare no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.