Plasma Cell-Free DNA After Embolization: A Novel, Sensitive Method for the Molecular Diagnosis of Venous Malformations : Journal of Vascular Anomalies

Secondary Logo

Journal Logo

Advanced Search
Original Article

Plasma Cell-Free DNA After Embolization

A Novel, Sensitive Method for the Molecular Diagnosis of Venous Malformations

Sun, Yia; Cai, Rena; Wang, Zhenfenga; Wang, Deminga; Zhao, Xiongb; Yue, Xiaojieb; Gu, Haob; Shi, Haoc; Liu, Yund; Fan, Xindonga; Su, Lixina

Author Information

aDepartment of Interventional Therapy, Multidisciplinary Team of Vascular Anomalies, Shanghai Ninth People’s hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, People’s Republic of China

bDepartment of Burn and Plastic Surgery, The Children’s Hospital, Zhejiang University School of Medicine, National Clinical Research Center for Child Health, Hangzhou, People’s Republic of China

cDepartment of Vascular Surgery, Gansu Province People’s Hospital, Gansu, People’s Republic of China

dState Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China

Received: 26 July 2022; Accepted 2 November 2022

Published online 28 February 2023

Sun, Cai, Wang are co-first authors.

Funding: Transverse Research Project of Shanghai Ninth Prople’s Hospital (No. JYHX2022004), Peripheral Arteriovenous Malformations Biological Sample Bank Construction Project of Shanghai Ninth People’s hospital (YBKB202104) and Sailing Project of Shanghai Rising-Star Program (22YF1422400).

The authors declare that they have no conflicts of interest with regard to the content of this report.

Data available statement: All supporting data of this article are included in the submitted manuscript.

STREGA statement: the authors have followed STREGA reporting guidelines.

Correspondence: Lixin Su, Department of Interventional Therapy, Multidisciplinary Team of Vascular Anomalies, Shanghai Ninth People’s hospital, Shanghai Jiao Tong University, Shanghai, People’s Republic of China ([email protected]); Xindong Fan, Department of Interventional Therapy, Multidisciplinary Team of Vascular Anomalies, Shanghai Ninth People’s hospital, Shanghai Jiao Tong University, Shanghai, People’s Republic of China ([email protected]); Yun Liu, State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, People’s Republic of China ([email protected]).

This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

Journal of Vascular Anomalies 4(1):p e054, March 2023. | DOI: 10.1097/JOVA.0000000000000054
  • Open

Abstract

Introduction

Venous malformation (VM), comprised of abnormally formed vascular apparatuses, is the most common low-flow vascular malformation.1,2 Depending on their size and location, VMs can result in various symptoms, such as cosmetic complaints, pain, pressure, bleeding, life-threatening expansion, or obstruction of vital structures.3,4 Treatment options for VMs, including surgical excision, sclerotherapy, systemic treatment with drugs or combination of these procedures, depend on the location, type, and extent of the lesions.5–8 Abnormal signaling processes during embryogenesis are thought to result in persistent vascular plexus cells with a certain degree of differentiation, leading to VM.9

In the last 2 decades, genetic research has attempted to identify causative variants and elucidate the pathogenesis of VM.9 VMs are historically known to be associated with either TEK or PIK3CA variants.10–13 Activating TEK variants in VM were first reported by Vikkula et al.13 In 2015, Limaye et al12 detected PIK3CA variants in 27 individuals with VM. However, the sensitivity of variant detection may be reduced by tissue heterogeneity.14 In addition, genetic analysis may be hindered by difficulties in tissue biopsies that are unsafe, impracticable, or otherwise unsuccessful. More sensitive and noninvasive diagnostic methods may provide adequate information for molecular characterization. Cell-free DNA (cfDNA) may enable a precision diagnostic approach in various clinical fields, obviating the costs, and complications associated with tissue biopsy.15–17 The clinical utility of cfDNA in the serum and plasma has been explored in many disciplines of medicine, including vascular anomalies.18–20 However, due to the limited number of cases in previous studies, the sensitivity and reliability of cfDNA detection for genetic variants in VMs has not been identified. In addition, the optimal biological fluid for cfDNA detection is still unclear.

In this prospective, multicenter and cross-sectional study with a total of 121 VM patients, a panel of 40 vascular anomaly related genes was assessed using next-generation sequencing (NGS). We collected peripheral blood samples, cfDNA samples isolated from peripheral plasma before and 1 hour after ethanol embolization in 24 patients in a pilot study group and paired plasma cfDNA samples from the lesion site in 7 of those patients to determine the optimal biological fluid for cfDNA detection. The variant profiles were then analyzed from 97 patients in an independent validation group with the selected biological fluid, namely, peripheral postembolization plasma cfDNA. Our findings suggested that peripheral postembolization plasma cfDNA sequencing may potentially be used as an alternative, reliable, and sensitive method for the genetic diagnosis of VM.

Methods

Study design

This nonexperimental prospective, multicenter, and cross-sectional study was conducted in collaboration among 3 tertiary referral centers. The study consisted of 2 phases. A pilot study with 24 VM patients conducted in the Principal Investigation Center at the Department of Interventional Therapy, Multidisciplinary Team of Vascular Anomalies at Shanghai 9th People’s Hospital to elucidate the concordance of variants between different biological specimens and to compare the diagnostic performance of peripheral pre-embolization plasma cfDNA, peripheral postembolization plasma cfDNA, and lesion plasma cfDNA samples. The optimal biological samples were selected for further examination. An independent validation group with peripheral postembolization plasma cfDNA samples was then collected from 97 VM patients enrolled from each participating center for assay development and validation. An overview of the study design is summarized in Figure 1.

F1
Figure 1.:
Overview of the study design. This prospective study consisted of 2 phases as follows: a pilot study to select the optimal biological samples for molecular alterations of VM; and a subsequent independent validation cohort study for further assay development and validation. VM indicates venous malformation.

Diagnosis and ethanol embolization

A total of 121 patients with VM were enrolled from each participating center from 2020 to 2021. All patients were assessed by at least 2 radiologist (ZFW and DMW) and met the diagnostic criteria of VM in the typical clinical findings with confirmation by magnmetic resonance imaging with spin-echo T1- and T2-weighted sequences. The VM lesions were observed as isointensity on T1-weighted image and homogeneous hyperintensity on T2-weighted images.4 The study was authorized and approved by our hospital’s Committee on Clinical Investigation. Written informed consent was obtained from all involved patients.

Percutaneous ethanol embolization was performed under digital subtraction angiography guidance using intravenous general anesthesia for all patients. Cardiopulmonary functions were monitored simultaneously. Seven-gauge butterfly needles were punctured according to the position and depth of VM lesion based on magnmetic resonance imaging. Once the blood returned, direct puncture phlebography was conducted until the contrast filled the draining veins. Slowly ethanol was then administered through the needle’s tube several times with each injection for 0.5–1.5 mL. The injection amount of ethanol was about 1/2–1/3 of the contrast. The maximum dose of ethanol in 1 single procedure was 1 mL/kg. We obtained postembolization blood samples 1 hour after ethanol injection (approximately the duration of anesthesia resuscitation). After procedure, vitals including heart rate, blood pressure, and oxygen saturation were closely monitored.

Sample collection

Peripheral blood samples and cfDNA samples isolated from peripheral plasma before and 1 hour after ethanol embolization were collected from 24 patients in a pilot study group, and paired plasma cfDNA samples were collected from the lesion site in 7 of those patients. Only peripheral postembolization plasma cfDNA was collected and analyzed from 97 patients in the independent validation group.

All specimens were preserved for DNA extraction and NGS. Clinical information, including anatomic localization, lesion volume assessed with magnmetic resonance imaging and absolute ethanol usage, was collected and assessed.

DNA extraction and NGS

DNA were extracted from dissected structures at 56°C using a QIAamp DNA Mini Kit (QIAGEN, IAGEN, Hilden, Germany) with overnight proteinase K incubation for approximately 14 hours. We verified DNA concentrations with NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE). Sample collection and DNA library construction were performed according to the protocol.

Targeted NGS was designed across a panel of 40 vascular anomaly specific genes according to the ISSVA classification21 with a median target fold coverage of >10000×. Library construction was performed as recommended by Illumina (San Diego, CAA). DNA was sheared, purified, end-repaired, adenylated at the 3′ ends, ligated with Illumina adaptors and amplified by Polymerase Chain Reaction. A series of probes was designed and synthesized by Integrated DNA Technologies (IDT) (Coralville, IA) to target the exons and exon/intron boundaries of the genes. After target capture and purification, the quantity of the library was validated using quantitative Polymerase Chain Reaction (Kapa Biosystems, Wilmington, MA), and the integrity was validated using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara). The libraries were sequenced by HiSeq-series sequencing systems (Illumina, San Diego). A bioinformatics pipeline, including GATK (version 3.4), BWA (version 0.7.13), and VarDict, was used to analyze the variants in PIK3CA and TEK, the most common variants previously detected and reported.10,11,22,23 SnpEff and VEP software were used to annotate the variants. The results were also manually confirmed using the Integrative Genomics Viewer. Single nucleotide variants, insertions, and deletions as well as gene fusions and copy number variations were identified with a detection limit of 0.5%.

Statistics analysis

Data were analyzed using SPSS 25 (IBM Corporation, Armonk, NY). Comparison of the variant prevalence of cfDNA from 3 different biological samples was performed using the Fisher exact test. To determine the discriminative power of the cfDNA from 3 different biological samples for survival, we constructed ROC curves and calculated areas under the curve (AUCs) with 95% confidence intervals. P < .05 was considered significant.

Results

Differences in the overlap of genetic variants observed in peripheral pre-embolization plasma, peripheral postembolization plasma, and lesion plasma

Twenty-four patients were enrolled in the pilot study. Variants were detected in cfDNA samples from 21 of the 24 isolated patients, but none was identified in peripheral blood. Variants in TEK (70.8%, 17/24) and PIK3CA (16.7%, 4/24) were considered mutant hotspots. Clinical manifestations and all identified variants are shown in Table 1. Interestingly, we detected 3 variants from the TEK and PIK3CA genes in 1 patient, which is very rare. The prevalence of TEK/PIK3CA variant in the 3 types of cfDNA samples from the 24 patients were compared. As shown in Figure 2A, the TEK/PIK3CA variant prevalence of DNA isolated from peripheral pre-embolization plasma cfDNA was the lowest (12.5% [3/24]), while the highest prevalence was noted in peripheral postembolization plasma cfDNA (87.5% [21/24]) followed by lesion plasma cfDNA (71.4% [5/7]). Statistics analysis showed significant differences in TEK/PIK3CA variant prevalence among these 3 biological samples (P < .0001).

Table 1. - Clinical Manifestations and Genetic Alterations in 24 Patients Enrolled in the Pilot Study
Patient Gender Age Location Ethanol Dosage Peripheral Blood PreE-cfDNA Allele Frequency PostE-cfDNA Allele Frequency Lesion-cfDNA Allele Frequency
#1 F 40 Chest 28 mL Negative Negative Negative TEK (p.Ser586fs) 0.55% TEK (p.Ser586fs) 0.52%
#2 F 19 Lower limbs 30 mL Negative Negative Negative TEK (p.Leu914Phe) 1.60% TEK (p.Leu914Phe) 0.51%
#3 F 5 Craniofacial 10 mL Negative Negative Negative Negative Negative Negative Negative
#4 F 34 Craniofacial 15 mL Negative Negative Negative TEK (p.Leu914Phe) 1.87% Negative Negative
#5 F 33 Craniofacial 14 mL Negative Negative Negative TEK (p.Tyr1108fs) 0.59% TEK (p.Tyr1108fs) 3.39%
#6 F 37 Craniofacial 20 mL Negative Negative Negative PIK3CA (p.Glu542Lys) 11.89% PIK3CA (p.Glu542Lys) 0.96%
#7 F 43 Upper limbs 25 mL Negative Negative Negative TEK (p.Arg1003His) 30.67% TEK (p.Arg1003His) 26.23%
#8 F 19 Right body 20 mL Negative Negative Negative TEK (p.Gly1115*) 0.93% NA
#9 M 33 Left body 28 mL Negative Negative Negative TEK (p.Arg915Cys) 3.80%
#10 M 16 Left lower limb 25 mL Negative TEK (p.Leu914Phe) 1.32% TEK (p.Leu914Phe) 2.75%
#11 M 20 Left upper limb 30 mL Negative Negative Negative TEK (p.Leu914Phe) 1.79%
#12 M 5 Left upper limb 12 mL Negative Negative Negative TEK (p.Leu914Phe) 8.52%
#13 M 5 Pelvic 5 mL Negative Negative Negative TEK (p.Leu914Phe) 1.52%
#14 M 8 Lower limbs 10 mL Negative TEK (p.Leu914Phe) 0.72% TEK (p.Leu914Phe) 1.90%
#15 M 29 Upper limbs 14 mL Negative Negative Negative PIK3CA (p.Glu545Lys) 4.78%
TEK (p.Gly542Arg) 11.80%
TEK (p.Tyr1024Phe) 53.61%
#16 F 16 Lower limbs 30 mL Negative Negative Negative TEK (p.Leu914Phe) 3.83%
#17 F 3 Left upper limb 11 mL Negative Negative Negative TEK (p.Leu914Phe) 2.44%
#18 M 65 Chest 9 mL Negative Negative Negative PIK3CA (p.Cys585Phe) 0.64%
#19 M 32 Craniofacial 15 mL Negative Negative Negative Negative Negative
#20 F 58 Craniofacial 10 mL Negative Negative Negative Negative Negative
#21 F 49 Left neck 25 mL Negative TEK (p.Arg915Cys) 1.53% TEK (p.Arg915Cys) 2.73%
#22 F 28 Right foot 10 mL Negative Negative Negative PIK3CA (p.Thr229fs) 1.07%
#23 M 4 Craniofacial 6 mL Negative Negative Negative TEK (p.Leu914Phe) 1.05%
#24 F 13 Left chest and upper limb 20 mL Negative Negative Negative TEK (p.Leu914Phe) 3.27%
Abbreviation: cfDNA, cell-free DNA.

F2
Figure 2.:
A, Bar graphs illustrating the frequency of detected variants in the peripheral pre-embolization plasma, peripheral postembolization plasma, and lesion plasma from 24 cases. The boxplot shows the number of positive mutant genes (either TEK or PIK3CA) among 24 VM patients, whose biological samples were screened using the 48-gene panel. A Fisher exact test revealed that there were significant differences in the TEK/PIK3CA variant prevalence among these 3 biological samples obtained from the same subjects (P < .0001). B, VAF identified in peripheral lesion and postembolization plasmas in 7 cases in the pilot study. The VAFs identified in peripheral postembolization plasma were comparable to those in lesion plasma and significantly higher than those in lesion plasma. cfDNA indicates cell-free DNA; VAF, variant allele frequency; VM, venous malformation.

For the 7 patients who underwent lesion plasma cfDNA analysis, the diagnostic power of the 3 types of cfDNA samples was tested. As shown in Figure 3A–C, the AUCs were 0.8571 (P = .0253), 0.5625 (P = .4579), and 0.9375 (P < .0001) for the lesion plasma, peripheral pre-embolization plasma, and peripheral postembolization plasma, respectively. Regarding the variant allele frequency of cfDNA, 5 out of 7 patients recorded higher variant allele frequencies in postembolization cfDNA than in lesion plasma cfDNA (Figure 2B).

F3
Figure 3.:
ROC curve of the 3 different methods. A, Lesion plasma in pilot group (AUC = 0.8571; P = .0253); (B) peripheral pre-embolization plasma in pilot group (AUC = 0.5625; P = .4579); (C) peripheral postembolization plasma in pilot group (AUC = 0.9375; P < .0001); (D) peripheral postembolization plasma in validation group (AUC = 0.9545; P < .0001). AUC indicates areas under the ROC curve.

These data clearly demonstrated that peripheral postembolization plasma cfDNA better reflects the genetic alterations in VM than peripheral pre-embolization plasma cfDNA and lesion plasma, suggesting that it may be better suited for diagnostic purposes.

Peripheral postembolization plasma for genetic diagnostic analysis of VM in an expanded independent validation study

The above pilot study compared the diagnostic values among different biological fluids and demonstrated the superiority of peripheral postembolization plasma. We then recruited an expanded independent group of 97 VM cases from 3 investigation centers. Variant profiles were prospectively collected and analyzed only with peripheral postembolization plasma cfDNA in the validation group.

Among the 97 patients with VM, 119 variants in TEK and PIK3CA variants were identified in 90.7% (88/97) of patients, and all identified variants were recorded (Figure 4). The most common variants included those in TEK (n = 79, 66.4%) and PIK3CA (n = 40, 33.6%). Most TEK variants were TEK p.L914F (n = 36, 45.6% [36/79]), p.R842H (n = 9, 11.4% [9/79]), p.R918H/S (n = 6, 7.6% [6/79]), and p.G1115fs (n = 6, 7.6% [6/79]). Most PIK3CA variants were p.E542K (n = 7, 17.5% [7/40]), p.H1047R/L (n = 5, 12.5% [5/40]), and p.E545K (n = 3, 7.5% [3/40]). Nine (9.3% [9/97]) patients had negative results (ie, none of the mentioned variants). The AUC of peripheral postembolization plasma in the independent validation group was 0.9545 (P < .0001, Figure 3D), suggesting the reliability of peripheral postembolization plasma for the molecular diagnosis of VM.

F4
Figure 4.:
Genetic variants in 97 patients enrolled in the independent validation study.

Discussion

The present study compared the diagnostic potential of the following 3 different types of cfDNA samples from the same subjects in the identification of genetic alterations in VMs: peripheral pre-embolization plasma, peripheral postembolization plasma, and lesion plasma. In total, 40 vascular anomaly related genes were analyzed in these 3 types of specimens. Variants identified in peripheral postembolization plasma were found to be the richest in comparison with lesion plasma and peripheral pre-embolization plasma samples drawn from the same cases. In a subsequent expanded and independent validation study, peripheral postembolization plasma cfDNA was successfully used for the identification of genetic alterations in 97 VMs.

VMs are historically known to be associated with either TEK or PIK3CA variants. Vikkula et al first reported that activating the TIE2 variant at position 849 (R849W) causes inherited VMs and that the TIE2 signaling pathway is critical in venous morphogenesis.13 In 2015, Limaye et al12 detected PIK3CA variants and TEK variants in 27 and 37 individuals (27/87 representing 31% and 37/87 representing 42.5%, respectively) with VM. In the present study, the most common variants included those in TEK (n = 79, 66.4%) and PIK3CA (n = 40, 33.6%), confirming the high frequency of TEK and PIK3CA variants in VMs. The mutational profile revealed a predominance of TEK variants, including p.L914F, p.R842H, p.R918H/S, and p.G1115fs, as well as PIK3CA variants, including p.E542K, p.H1047R/L, and p.E545K, consistent with previous studies.

Traditional molecular analysis of VMs with tissue samples may be limited due to tissue heterogeneity, resulting in reduced sensitivity and bias of variant detection.14 In addition, tissue biopsy may be unsafe, impractical, or otherwise unsuccessful, leading to severe complications, such as bleeding, ulcers, and coagulation disorders. Thus, genome analysis of vascular diseases requires a new approach with minimal invasion and superior sensitivity. Circulating cfDNA was initially reported by Mandel in 194824 and has gradually become an active research field in many medical disciplines. As an ideal biomarker, cfDNA is accessible, reliable, and reproducible. Recently, Zenner et al18 extracted cfDNA from plasma and detected TEK variants in one of three VM patients (33.3%, 1/3) for the first time. However, due to the small sample size, the diagnostic sensitivity of cfDNA is still unclear. In addition, cfDNA enters the circulation through various processes, such as necrosis and apoptosis. However, cell turnover in VM is not considerably increased,25 which may be the reason for the low sensitivity of variant detection with plasma cfDNA in this previous study.

We hypothesized that increasing the contact between venous endothelial cells and the systemic circulation as well as increasing the content of venous endothelial cells shed in the systemic circulation may improve the detection ability of cfDNA. Endovascular therapy with sclerosants is the mainstream treatment for VMs, in which absolute ethanol is the most effective.26 The mechanism of sclerosants is to physically or chemically cause necrosis and shedding of venous endothelial cells to further induce local coagulation and occlusion of vascular lesions.27 Absolute ethanol rapidly “erodes” endothelial cells, leading to reliable venous occlusion. Based on these studies, we collected peripheral plasma 1 hour after absolute ethanol embolization, paired lesion plasma, and peripheral plasma before embolization in VM patients, and we extracted cfDNA for genetic analysis. We compared the diagnostic abilities of these 3 samples and identified that peripheral postembolization plasma achieved the best diagnostic performance followed by lesion plasma and peripheral pre-embolization plasma. The contact between venous endothelial cells and the systemic circulation in VM lesions is higher than that in peripheral blood, which explains why the detective sensitivity of lesion plasma is higher than that of peripheral pre-embolization plasma in this study.

In view of our preliminary results, we enrolled 97 VM patients from 5 tertiary centers to further validate the diagnostic sensitivity of peripheral postembolization plasma. In total, 119 variants were detected in 90.7% (88/97) of patients. The TEK/PIK3CA variant prevalence in this study was higher than that with tissue samples in previous studies.10,12–14 The limitation of the present study was that we did not match tissue samples. All patients enrolled in this study were assessed as unsuitable for surgical resection and underwent endovascular therapy, which resulted in difficulties in tissue sampling. Regardless, these results suggested that peripheral postembolization plasma cfDNA is a sensitive and safe method for molecular diagnosis. In this study, all patients were treated with absolute ethanol for endovascular therapy. It is not clear whether other sclerosants such as bleomycin and foam can cause similar detective abilities in plasma cfDNA. Further studies with other sclerosants will be needed to address these questions. In addition, considering the duration of anesthesia resuscitation, we chose 1 hour after ethanol injection as the endpoint for postoperative plasma sampling. Further studies will focused on the natural metabolism of postembolization plasma cfDNA by analyzing plasma samples collected at different postembolization time points.

In the present study, compared with lesion plasma, peripheral postembolization plasma not only had stronger diagnostic ability but was also safer. For some patients with complex VMs, it is not safe or feasible to collect blood from the lesion, which may further cause trauma and aggravate local intravascular coagulation or even induce DIC. The use of peripheral postembolization plasma for cfDNA detection is not only safe and sensitive but also combines the treatment and diagnosis of VMs, which is convenient for patients and reduces unnecessary costs, providing opportunities for molecular diagnostics in a wider range of patients in the future. Previously reported trials have demonstrated the efficacy of sirolimus, an allosteric mammalian target of rapamycin inhibitor, in treating selected vascular malformations, including VMs.7,8,28 Peripheral postembolization plasma cfDNA combines traditional endovascular therapy, molecular diagnosis, and individual targeted therapy, providing more opportunities and treatment options for the comprehensive treatment of complex VMs.

Conclusions

In conclusion, we compared the diagnostic potential of 3 different types of cfDNA samples, namely, peripheral pre-embolization plasma, peripheral postembolization plasma, and lesion plasma, from the same subjects in the identification of genetic alterations in VMs. The identified variants in the peripheral postembolization plasma were found to be the richest. Thus, peripheral postembolization plasma cfDNA realizes the combination of molecular diagnosis and treatment for VMs, suggesting that it may be a safe, sensitive, and reliable method for molecular diagnosis.

Acknowledgments

We would like to show our deepest gratitude to the patient for sharing their clinical information.

References

1. Mulliken JB, Burrows PE, Fishman SJ. Mulliken and Young’s Vascular Anomalies: Hemangiomas and Malformations. 2nd ed. Oxford University Press; 2013.
2. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plastic Reconstructive Surg. 1982;69:412–422.
3. Seront E, Vikkula M, Boon LM. Venous malformations of the head and neck. Otolaryngol Clin North Am. 2018;51:173–184.
4. Dompmartin A, Vikkula M, Boon LM. Venous malformation: update on aetiopathogenesis, diagnosis and management. Phlebology. 2010;25:224–235.
5. Van Der Vleuten CJ, Kater A, Wijnen MH, Kool LJS, Rovers MM. Effectiveness of sclerotherapy, surgery, and laser therapy in patients with venous malformations: a systematic review. Cardiovasc Intervent Radiol. 2014;37:977–989.
6. Choi YH, Han MH, Kwon O-K, Cha SH, Chang K-H. Craniofacial cavernous venous malformations: percutaneous sclerotherapy with use of ethanolamine oleate. J Vasc Interv Radiol. 2002;13:475–482.
7. Adams DM, Trenor CC, Hammill AM, et al. Efficacy and safety of sirolimus in the treatment of complicated vascular anomalies. Pediatrics. 2016;137:e20153257.
8. Hammer J, Seront E, Duez S, et al. Sirolimus is efficacious in treatment for extensive and/or complex slow-flow vascular malformations: a monocentric prospective phase II study. Orphanet J Rare Dis. 2018;13:191.
9. Queisser A, Boon LM, Vikkula M. Etiology and genetics of congenital vascular lesions. Otolaryngol Clin North Am. 2018;51:41–53.
10. Limaye N, Wouters V, Uebelhoer M, et al. Somatic mutations in the angiopoietin-receptor TIE2 can cause both solitary and multiple sporadic venous malformations. Nat Genet. 2009;41:118–124.
11. Uebelhoer M, Boon LM, Vikkula M. Vascular anomalies: from genetics toward models for therapeutic trials. Cold Spring Harb Perspect Med. 2012;2:143–152.
12. Limaye N, Kangas J, Mendola A, et al. Somatic activating PIK3CA mutations cause venous malformation. Am J Hum Genet. 2015;97:914–921.
13. Vikkula M, Boon LM, Iii KLC, et al. Vascular dysmorphogenesis caused by an activating mutation in the receptor tyrosine kinase TIE2. Cell. 1996;87:1181–1190.
14. Soblet J, Limaye N, Uebelhoer M, Boon L, Vikkula M. Variable somatic TIE2 mutations in half of sporadic venous malformations. Mol Syndromol. 2013;4:179–183.
15. Corcoran RB, Chabner BA. Application of cell-free DNA analysis to cancer treatment. N Engl J Med. 2018;379:1754–1765.
16. Diehl F, Schmidt K, Choti MA, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med. 2008;14:985–990.
17. Diaz LA Jr, Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. J Clin Oncol. 2014;32:579–586.
18. Zenner K, Jensen DM, Cook TT, et al. Cell-free DNA as a diagnostic analyte for molecular diagnosis of vascular malformations. Genet Med. 2021;23:123–130.
19. Palmieri M, Currò A, Tommasi A, et al. Cell-free DNA next-generation sequencing liquid biopsy as a new revolutionary approach for arteriovenous malformation. JVS: Vasc Sci. 2020;1:176–180.
20. Palmieri M, Di Sarno L, Tommasi A, et al. MET somatic activating mutations are responsible for lymphovenous malformation and can be identified using cell-free DNA next generation sequencing liquid biopsy. J Vasc Surg Venous Lymphat Disord. 2021;9:740–744.
21. Blei F. ISSVA Classification of Vascular Anomalies. Berlin, Heidelberg: Springer; 2017:47–50.
22. Boscolo E, Coma S, Luks VL, et al. AKT hyper-phosphorylation associated with PI3K mutations in lymphatic endothelial cells from a patient with lymphatic malformation. Angiogenesis. 2015;18:151–162.
23. Osborn AJ, Peter D, Neilson DE, et al. Activating PIK3CA alleles and lymphangiogenic phenotype of lymphatic endothelial cells isolated from lymphatic malformations. Hum Mol Genet. 2015;24:926–938.
24. Mandel P. Les acides nucleiques du plasma sanguin chez 1 homme. CR Seances Soc Biol Fil. 1948;142:241–243.
25. Meijer-Jorna LB, van der Loos CM, de Boer OJ, van der Horst CM, van der Wal AC. Microvascular proliferation in congenital vascular malformations of skin and soft tissue. J Clin Pathol. 2007;60:798–803.
26. Berenguer B, Burrows P, Zurakowski D, Mulliken J. Sclerotherapy of craniofacial venous malformations: complications and results. Plast Reconstr Surg. 1999;104:1–11. discussion 125.
27. Sun Y, Gu H, Yang X, et al. Bleomycin polidocanol foam (BPF) stability - in vitro evidence for the effectiveness of a novel sclerosant for venous malformations. Eur J Vasc Endovasc Surg. 2020;59:1011–1018.
28. Lackner H, Karastaneva A, Schwinger W, et al. Sirolimus for the treatment of children with various complicated vascular anomalies. Eur J Pediatr. 2015;174:1579–1584.
Keywords:

plasma cell-free DNA; molecular diagnosis; venous malformation

Copyright © 2023 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of The International Society for the Study of Vascular Anomalies.