Sirolimus for the treatment of vascular tumors and malformations in children: A 10-year single-institution retrospective study

Abstract

BACKGROUND

Vascular anomalies, including vascular tumors and vascular malformations, are common pediatric conditions that may cause significant morbidity. Although standard treatments are effective in many cases, a subset of patients experiences inadequate responses or has unresectable disease, necessitating alternative therapeutic strategies. Sirolimus has emerged as a targeted treatment option for selected vascular anomalies.

AIM

To evaluate treatment outcomes and the safety of sirolimus in children with vascular anomalies.

METHODS

We conducted a retrospective review of medical records of pediatric patients with vascular anomalies who received sirolimus at Chiang Mai University Hospital between 2016 and 2025. Clinical characteristics, treatment regimens, therapeutic responses, and adverse events were analyzed.

RESULTS

Eleven children were included in the study. Five patients had vascular tumors, including two kaposiform hemangioendotheliomas, one composite hemangioendothelioma, and two hemangiomas, while six patients had vascular malformations. Sirolimus was administered as salvage therapy in nine patients. After sirolimus treatment, 10 patients (90.9%) achieved partial response, and one patient had stable disease. At a median follow-up of 30 months, all patients were alive with disease. Sirolimus-related adverse events occurred in four patients, with proteinuria being the most common. Other events included hypokalemia, elevated serum creatinine, and hypercholesterolemia. No serious infections were observed.

CONCLUSION

Sirolimus demonstrated a high response rate and an acceptable safety profile in children with vascular anomalies. These findings support the use of sirolimus as an effective treatment option for refractory or unresectable cases, including complex entities such as composite hemangioendothelioma.

Key Words: Sirolimus; Vascular anomalies; Vascular malformations; Vascular tumors; Children

Core Tip: Sirolimus is increasingly used as targeted therapy for pediatric vascular anomalies, yet real-world clinical data remain limited. In this retrospective study of children with vascular tumors and malformations, sirolimus demonstrated a high response rate and an acceptable safety profile across diverse disease subtypes. Rapid resolution of Kasabach-Merritt phenomenon was observed in patients with kaposiform hemangioendothelioma receiving combination therapy. However, disease progression occurred after treatment discontinuation, highlighting the need for careful post-therapy monitoring. These findings support sirolimus as an effective treatment option in refractory or unresectable pediatric vascular anomalies and emphasize the importance of optimizing treatment duration and follow-up strategies.

INTRODUCTION

Vascular anomalies comprise a broad and heterogeneous spectrum of disorders, including vascular tumors, such as infantile hemangioma and kaposiform hemangioendothelioma (KHE), and vascular malformations, including lymphatic malformations, venous malformations, and complex combined lesions. These conditions commonly present during the neonatal and infantile periods and may lead to substantial morbidity. Although standard treatment modalities, including surgery, sclerotherapy, and corticosteroids, are effective in many cases, a subset of patients demonstrates suboptimal or limited clinical responses[1,2].

Growing evidence indicates that dysregulation of the phosphatidylinositol 3-kinase/protein kinase B/mammalian target of rapamycin pathway (mTOR) signaling pathway plays a key role in the pathogenesis of several vascular anomalies. This has provided a strong biological rationale for the use of sirolimus, an mTOR inhibitor, as targeted therapy[3,4]. A systematic review by Freixo et al[5] demonstrated that sirolimus was highly effective in the treatment of both vascular tumors and vascular malformations. Triana et al[3] reported significant or partial clinical responses in approximately 80% of 41 pediatric patients treated with an oral sirolimus regimen, with acceptable tolerability. Subsequently, Neirotti et al[6] described clinical improvement or stability in 86% of 14 children when sirolimus was administered in combination with infection prophylaxis and therapeutic drug monitoring, further supporting its real-world efficacy and manageable safety profile. Common adverse events associated with sirolimus include dyslipidemia, mucositis, and hematologic toxicity, which are generally reversible[7]. However, severe adverse events, particularly infections, have been reported in up to 15% of cases[8].

Over the past decade, several observational studies and systematic reviews have reported promising outcomes in children with vascular anomalies treated with sirolimus[3,4]. However, evidence specific to pediatric patients in real-world clinical settings remains limited, particularly across diverse disease subtypes, treatment indications, dosing strategies, and outcome measures. To address this gap, this study aimed to describe the clinical characteristics and treatment outcomes of children with vascular anomalies treated with sirolimus in a real-world clinical setting, including both newly diagnosed and relapsed or refractory cases. Because sirolimus is used across multiple vascular tumor and vascular malformation subtypes in routine clinical practice, both entities were included in this single-institution cohort. Secondary objectives were to evaluate the safety profile of sirolimus based on treatment-related adverse events and to assess treatment responses across different vascular anomaly subtypes.

MATERIALS AND METHODS

Patient selection

A retrospective medical record review was conducted among children diagnosed with vascular anomalies who were treated with sirolimus at Chiang Mai University Hospital between January 1, 2016, and November 15, 2025. The study was approved by the Institutional Review Board/Ethics Committee, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand (approval No. PED-2568-0835). Eligible patients were children diagnosed with vascular anomalies based on clinical features and/or imaging studies and/or histopathology, with available baseline clinical data and at least one documented follow-up assessment after initiation of sirolimus therapy. Patients were excluded if they were not treated at Chiang Mai University Hospital, received sirolimus for indications unrelated to vascular anomalies, or lacked adequate baseline or follow-up data.

Data collection

Demographic data included age, sex, weight, height, and body surface area. Disease-related characteristics comprised vascular anomaly subtype, lesion location, and associated disease complications. Treatment-related variables included prior therapies, sirolimus dosing (initial dose of 0.8 mg/m²/dose administered every 12 hours), therapeutic drug monitoring targeting trough levels of 5-15 ng/mL, duration of sirolimus therapy, treatment outcomes, and sirolimus-related adverse events graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 6.0[9].

Treatment response was evaluated according to modified disease response criteria based on a previously published phase II study of sirolimus for complicated vascular anomalies. Complete response was defined as no evidence of disease on radiologic imaging, absence of disease-related organ dysfunction, and normalization of quality-of-life criteria. Partial response (PR) was defined as either > 20% reduction in the size of the target vascular lesion on radiologic imaging, improvement in target organ dysfunction by at least one grade, or improvement in quality-of-life measures. Progressive disease (PD) was defined as > 20% increase in the size of the target vascular lesion on radiologic imaging, worsening of target organ dysfunction by at least one grade, or deterioration in quality-of-life measures. Stable disease (SD) was defined as a disease that did not meet the criteria for complete response, PR, or PD[10]. However, quality-of-life instruments were not systematically collected in this retrospective cohort. For safety monitoring, clinical and disease assessments were performed together with laboratory investigations at follow-up visits. Laboratory evaluations included complete blood count, renal function tests, liver function tests, electrolytes, lipid profile, urinalysis, and urine protein-to-creatinine ratio.

Sirolimus discontinuation was individualized based on treatment response, disease stability, symptom control, and physician judgment in discussion with caregivers. In general, treatment was continued for at least 12 months once disease stabilization and symptom resolution were achieved. After consensus with caregivers, sirolimus was discontinued without tapering. Following discontinuation, patients underwent regular disease monitoring with clinical assessment and radiologic imaging every 3–6 months.

Statistical analysis

Statistical analyses were conducted using Stata v.16.1 software (Stata Corporation LLC, College Station, United States). Categorical variables were summarized as counts and percentages. Continuous variables were presented as medians with interquartile ranges (IQR). Kaplan-Meier survival analysis was utilized to evaluate the progression-free survival rates. Patients without disease progression were censored at the last follow-up.

RESULTS

Demographic and baseline disease characteristics

Eleven children with vascular anomalies were included in the study, of whom 72.7% were male. The median age at diagnosis was 1 month (IQR, 0-4 months). Five patients had vascular tumors, including two cases of KHE, one composite hemangioendothelioma (CHE), and two hemangiomas. The remaining six patients were diagnosed with venolymphatic malformations. Regarding lesion location, five patients (45.5%) had lesions involving the head and neck, four patients (36.4%) had lesions involving the trunk or buttock, and two patients (18.2%) had extremity involvement. Initial clinical manifestations included palpable mass (72.7%), Kasabach-Merritt phenomenon (KMP) (18.2%), proptosis (9.1%), flexion contracture (9.1%), and chylothorax (9.1%). Histopathological confirmation was available in 54.5% of cases, including two KHE, one CHE, one non-involuting congenital hemangioma, and two lymphatic malformations. The remaining cases were diagnosed based on clinical and imaging findings.

Treatment and outcomes

Vascular tumor group: All five patients with vascular tumors received sirolimus as salvage therapy. Prior treatments included partial surgical resection (20.0%), propranolol (40.0%), chemotherapy (60.0%), and radiotherapy (20.0%). Responses to these prior therapies included PR in 20%, SD in 40%, and PD in 40%. For sirolimus treatment, the median age at initiation was 60 months (IQR, 8-120 months). Sirolimus monotherapy was administered in three cases, whereas combination therapy with sirolimus and prednisolone was used in two patients with KHE. All patients achieved PR with clinical improvement, and the median lesion size reduction from baseline was 55.5% (IQR, 33.3%-56.8%). In both KHE patients with KMP, complete resolution of KMP was observed within 1 month following combination therapy with sirolimus and prednisolone.

The median sirolimus dose was 1.6 mg/m²/day (IQR, 1.6-2.0 mg/m²/day), and the median sirolimus trough level was 10.4 ng/mL (IQR, 9.6-11.5 ng/mL). The median treatment duration was 20 months (IQR, 9-31 months). Sirolimus was discontinued in three patients due to disease stabilization. The CHE patient continued to have SD until the last follow-up visit. However, all KHE patients subsequently developed disease progression, including progressive functional impairment due to iliopsoas involvement in one case and severe shoulder pain in another. Reintroduction of sirolimus resulted in clinical improvement and radiologic PR in both KHE cases.

Vascular malformation group: Four patients with vascular malformations (66.7%) received sirolimus as salvage therapy, whereas two patients received sirolimus as first-line treatment. Prior treatments included partial surgical resection (50.0%), bleomycin sclerotherapy (33.3%), and propranolol (16.7%). Responses to prior therapies included PR in 50%, SD in 25%, and PD in 25%. For sirolimus treatment, the median age at initiation was 18.5 months (IQR, 4-36 months). Sirolimus monotherapy was administered in all cases. Five patients (83.3%) achieved PR with clinical improvement, with a median lesion size reduction from baseline of 20.5% (IQR, 15.4%-25.6%). One patient had clinical improvement but a 15.4% increase in lesion size, which was classified as SD.

The median sirolimus dose was 1.2 mg/m²/day (IQR, 0.5-1.6 mg/m²/day), and the median sirolimus trough level was 3.9 ng/mL (IQR, 2.4-9.2 ng/mL). The median treatment duration was 7 months (IQR, 2-30 months). Sirolimus was discontinued in two patients, one with SD and another who achieved PR. Both patients subsequently underwent sclerotherapy as further management. When combining both vascular tumor and vascular malformation groups, at a median follow-up of 30 months (IQR, 3–42 months), all patients were alive with disease. The 2-year progression-free survival is shown in Figure 1. Clinical characteristics and treatment outcomes are summarized in Table 1.

Figure 1 Kaplan-Meier analysis of progression-free survival in children with vascular tumors and vascular malformations treated with sirolimus. Two progression events occurred in patients with kaposiform hemangioendothelioma after discontinuation of sirolimus. VT: Vascular tumors; VM: Vascular malformations.

Table 1 Clinical characteristics, treatment, and outcomes of pediatric vascular anomalies treated with sirolimus.

Pt	Sex	Age at dx (mo)	Age at SRL tx (mo)	VA type	Location	Clinical presentation	Prior tx (s)	Response to prior tx	SRL regimen	Duration (mo)	Best response	SRL discontinued	AEs (CTCAE v6.0)	FU (mo)
Vascular tumors
1	M	4	60	KHE	Iliopsoas	KMP, flexion contracture	Surgery, CMT	PD	+ PSL	31	PR	Yes	Hypercholesterolemia (G2)	68
2	F	3	8	KHE	Shoulder	Mass, KMP, pain	CMT	PD	+ PSL	62	PR	Yes	None	87
3	M	154	168	CHE	Paranasal sinus	Proptosis, epistaxis	CMT, radiotherapy	PR	Mono	20	PR	Yes	Proteinuria (G1), creatinine increased (G1), hypokalemia (G1)	42
4	M	1	8	IH	Face	Mass	Propranolol	SD	Mono	9	PR	No	Proteinuria (G2)	9
5	M	0	120	NICH	Face, neck	Mass	Propranolol	SD	Mono	5	PR	No	None	5
Vascular malformations
6	M	0	2	LM	Leg	Mass	None	-	Mono	30	SD	Yes	None	30
7	M	0	26	VM	Hand	Mass	Propranolol	SD	Mono	11	PR	Yes	None	33
8	M	0	4	LM	Neck	Mass	Surgery, sclerotherapy	PD	Mono	30	PR	No	Proteinuria (G3)	32
9	F	0	36	LM	Tongue	Mass, pain	None	-	Mono	2	PR	No	None	2
10	M	24	72	LM	Chest	Chylothorax, osteolytic lesion	Surgery	PR	Mono	3	PR	No	None	3
11	F	2	11	LM	Buttock	Mass with lymphatic leakage	Surgery, sclerotherapy	PR	Mono	1	PR	No	None	1

AEs: Adverse events; CHE: Composite hemangioendothelioma; CMT: Chemotherapy; CTCAE: Common Terminology Criteria for Adverse Events; dx: Diagnosis; F: Female; FU: Follow-up; G: Grade; IH: Infantile hemangioma; KHE: Kaposiform hemangioendothelioma; KMP: Kasabach-Merritt phenomenon; LM: Lymphatic malformation; M: Male; mo: Months; NICH: Non-involuting congenital hemangioma; PD: Progressive disease; PR: Partial response; PSL: Prednisolone; Pt: Patient; SD: Stable disease; SRL: Sirolimus; tx: Treatment; VA: Vascular anomaly; VM: Venous malformation.

Safety profiles

Sirolimus-related adverse events occurred in four patients, with a total of six events recorded. These included proteinuria in three cases (CTCAE grades 1-3), elevated serum creatinine in one case (CTCAE grade 1), hypokalemia in one case (CTCAE grade 1), and hypercholesterolemia in one case (CTCAE grade 1). All adverse events resolved after temporary drug interruption and dose adjustment, except for elevated creatinine, which occurred in a patient with CHE who had previously received chemotherapy, including ifosfamide and doxorubicin. No cases of mucositis or infectious complications were observed. No antimicrobial prophylaxis was used in this study.

CHE case illustration

A 13-year-old boy presented with blurred vision, proptosis, and epistaxis. Physical examination revealed a reddish, lobulated mass in the right nasal cavity originating from the superior aspect. Initial magnetic resonance imaging (MRI) of the brain and paranasal sinuses demonstrated a lobulated enhancing mass measuring 4.9 cm × 4.5 cm × 7.4 cm. The mass involved the superior nasal cavity, bilateral posterior ethmoid sinuses, anterior skull base, and anterior cranial fossa, with extension to the bilateral orbits, superior nasal structures, sphenoid bone, and tuberculum sellae, and compression of the pituitary gland without invasion.

Because the lesion was unresectable, a tumor biopsy was performed. Histopathologic examination established the diagnosis of CHE comprising epithelioid hemangioendothelioma and retiform hemangioendothelioma. The patient received doxorubicin and ifosfamide chemotherapy and radiotherapy to the superior nasal cavity with a total dose of 59.4 Gy according to the COG ARST0332 protocol (A risk-based treatment strategy for non-rhabdomyosarcoma soft-tissue sarcomas in patients younger than 30 years)[11]. After completion of seven cycles of chemotherapy and radiotherapy, follow-up MRI demonstrated PR with a 52.6% volume reduction to 4.6 cm × 4.2 cm × 4.0 cm. He subsequently developed chronic kidney disease following treatment. Surgical resection was reconsidered, but was not feasible.

Given the reported activity of sirolimus in epithelioid hemangioendothelioma and the presence of this component in the tumor, salvage therapy with sirolimus (0.8 mg/m²/dose every 12 hours) was initiated after discussion with the patient and caregivers[12]. Two months after treatment initiation, laboratory evaluation revealed elevated serum creatinine, proteinuria, and hypokalemia, with a sirolimus trough level of 25.8 ng/mL. All adverse events were CTCAE grade 1. After a temporary drug interruption for two weeks and a dose adjustment to maintain sirolimus trough levels between 5 ng/mL and 15 ng/mL, only elevated serum creatinine persisted (CTCAE grade 1), which was considered likely related to prior nephrotoxic chemotherapy.

MRI findings before and 6 months after initiation of sirolimus are shown in Figure 2. The patient achieved PR with a 64.4% volume reduction to 3.4 cm × 3.0 cm × 2.7 cm compared with the MRI obtained prior to sirolimus therapy. The total duration of sirolimus treatment was 20 months. After disease stabilization confirmed by MRI for 1 year, sirolimus was discontinued. Clinical assessment and MRI surveillance every 6 months were scheduled. At 22 months after treatment discontinuation, the patient remains alive without disease progression.

Figure 2 Magnetic resonance imaging of a patient with composite hemangioendothelioma before and after sirolimus therapy. A: Baseline post-contrast T1-weighted image with fat suppression in the coronal view demonstrating a large heterogeneous mass with areas of necrosis involving the nasal cavity, sphenoid sinuses, and ethmoid sinuses (arrows); B: Follow-up magnetic resonance imaging obtained after 6 months of sirolimus therapy showing a marked reduction in tumor size (arrows).

DISCUSSION

Sirolimus is a macrocyclic lactone produced by Streptomyces hygroscopicus that inhibits phosphorylation of proteins involved in cell proliferation and survival through suppression of the mTOR complex 1 (mTORC1). Since its discovery in the early 1970s, sirolimus has been increasingly used in various clinical settings, including solid organ transplantation, lymphoproliferative disorders, and vascular anomalies[13,14]. Vascular tumors and vascular malformations are heterogeneous entities that often require a multidisciplinary treatment approach. Because dysregulation of the phosphatidylinositol 3-kinase/protein kinase B/mTOR signaling pathway plays a central role in the pathophysiology of many of these conditions, sirolimus has emerged as an important therapeutic option. In the present study, all but one patient demonstrated a favorable response to sirolimus therapy. A recent systematic review of sirolimus use in infantile hemangioma reported a response rate of 62.5%[15]. Consistent with these findings, clinical improvement was observed in one patient with infantile hemangioma and another with non-involuting congenital hemangioma who were refractory to propranolol therapy.

KHE, particularly when complicated by KMP, represents one of the most life-threatening vascular tumors in infancy. Recent evidence, including a randomized controlled trial by Ji et al[16], has demonstrated that combination therapy with sirolimus and prednisolone results in more rapid and sustained resolution of KMP compared with sirolimus monotherapy, with superior early platelet recovery. In our study, both patients with KHE and KMP received combination therapy and achieved complete resolution of KMP within one month, consistent with previously published data. Notably, disease progression occurred in both KHE patients following discontinuation of sirolimus, manifesting as functional impairment and severe pain. Reintroduction of sirolimus resulted in prompt clinical improvement and radiologic PR, underscoring the potential need for prolonged or maintenance therapy in this subgroup. These observations emphasize the importance of careful treatment discontinuation strategies and close post-cessation monitoring in patients with KHE, particularly those with deep or functionally critical lesions. However, the optimal duration of therapy and criteria for safe discontinuation remain unclear and warrant further investigation.

CHE is a rare, borderline vascular tumor classified by the International Society for the Study of Vascular Anomalies as having mixed benign, intermediate, and malignant vascular components[1]. Standard treatment consists of complete surgical excision whenever feasible, while radiotherapy and chemotherapy have been used in selected unresectable or locally aggressive cases[17]. Although sirolimus is not an established therapy for CHE, prior reports have demonstrated clinical benefit in related entities, such as epithelioid hemangioendothelioma and retiform hemangioendothelioma[12]. In the present study, sirolimus provided durable disease control in a patient with unresectable CHE after first-line treatment with chemotherapy and radiotherapy, suggesting that mTOR inhibition may represent a viable therapeutic option in selected advanced or refractory cases. Given the rarity of CHE, further multicenter experience is needed to better define the role of sirolimus in this entity.

Slow-flow vascular malformations, including venous and lymphatic malformations, often present significant therapeutic challenges when lesions are extensive, infiltrative, or surgically inaccessible. In such cases, sirolimus has been increasingly used as a systemic therapy to reduce lesion volume and alleviate symptoms[18]. A systematic review by Freixo et al[5] demonstrated lesion size reduction in 88.9% of venous malformations and clinical improvement in 94.9% of lymphatic malformations treated with sirolimus. In line with these findings, the majority of patients with vascular malformations in our study achieved PRs, further supporting the effectiveness of sirolimus in this population.

With respect to safety, sirolimus was generally well-tolerated in our study. Proteinuria was the most frequently observed adverse event, consistent with reports from transplant populations, where proteinuria occurs in approximately 10% of patients and nephrotic-range proteinuria in approximately 2%[19,20]. In our study, nephrotic-range proteinuria occurred in only one case and resolved with temporary drug interruption and dose adjustment. Elevation of serum creatinine was observed in one patient with CHE who had previously received nephrotoxic chemotherapy, suggesting that prior renal injury may increase susceptibility to renal adverse effects during sirolimus therapy. Importantly, no serious infections or cases of Pneumocystis jirovecii pneumonia were observed, even in the absence of routine antimicrobial prophylaxis, consistent with safety data from the PERFORMUS trial (treatment of voluminous and complicated superficial slow-flow vascular malformations with sirolimus) in children with slow-flow vascular malformations[21].

This study has several limitations. The retrospective design, small sample size, lack of molecular data, and heterogeneity of disease subtypes limit the generalizability of the findings. In addition, statistical comparisons between the vascular tumor and vascular malformation groups were underpowered. The follow-up duration may have been insufficient to capture late relapses or long-term toxicities, and mild adverse events such as low-grade mucositis may have been underreported. Nevertheless, this study reflects real-world clinical practice and provides valuable insight into the effectiveness and safety of sirolimus across a spectrum of pediatric vascular anomalies.

CONCLUSION

This study supports the effectiveness of sirolimus in pediatric patients with vascular anomalies, with high response rates and an acceptable safety profile. Sirolimus may also provide clinical benefit in selected cases of CHE. In patients with KHE, close monitoring after treatment discontinuation is warranted due to the risk of disease progression. Larger prospective studies are needed to identify predictors of response and to determine the optimal duration of sirolimus therapy.