Hindawi

Contrast Media & Molecular Imaging

Volume 2022, Article ID 8981078, 12 pages
https://doi.org/10.1155/2022/8981078

Research Article
FOXF1 Was Identified as a Novel Biomarker of Infantile
Hemangioma by Weighted Coexpression Network Analysis and
Differential Gene Expression Analysis

Yuchao Zhang and Ping Wang

Department of Vascular Surgery, The Affiliated Huaian No. 1 People’s Hospital of Nanjing Medical University,
Huaian 223300, China

Correspondence should be addressed to Ping Wang; drwangping@126.com

Received 8 June 2022; Revised 5 July 2022; Accepted 8 July 2022; Published 30 July 2022

Academic Editor: Mohammad Farukh Hashmi

Copyright © 2022 Yuchao Zhang and Ping Wang. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Background. The most frequent benign tumor in newborns is infantile hemangioma. The majority of infantile hemangiomas has a
favorable prognosis and generally fades away on their own. Some people, however, do experience major consequences.
Transcriptome alterations in infantile hemangiomas are yet unclear. The use of transcriptome analysis to uncover diagnostic
markers for infantile hemangioma has clinical implications. Methods. The dataset GSE127487 for infantile hemangioma was
obtained from the GEO database. The gene set most related with infantile hemangioma was investigated using weighted
coexpression network analysis. Differential expression analysis was performed to see whether genes were up or downregulated in
infantile hemangiomas. The enrichment of gene sets in pathways or functions is determined via enrichment analysis. Hub genes
were discovered via protein-protein interaction network analysis. The relationship between hub genes and immune cells was
investigated using immunomicroenvironment analysis. Results. Turquoise and Pink modules were revealed to be the most linked
with infantile hemangioma in a weighted coexpression network analysis (p < 0.001). The genes in the two modules were mostly
concentrated in actin filament organization, embryonic organ development, reproductive structure development, cell substrate
adhesion, extracellular matrix organization, and so on, according to GO enrichment analysis (p < 0.05). These gene enrichment
pathways comprised the PI3K-Akt signaling pathway, human papillomavirus infection, focal adhesion, and hepatitis C pathways,
according to KEGG enrichment analyzes (p < 0.05). Differential expressed gene analysis showed 43 upregulated and 21
downregulated genes in infantile hemangiomas. We found the gene set most associated to infantile hemangioma by intersecting
the elevated genes with the genes acquired by WGCNA, with FOXF1 serving as the hub gene. FOXF1 was linked to the degree of
monocyte infiltration, according to immunocorrelation analysis (p < 0.05). B cell memory, dendritic cells resting, macrophage
M0, neutrophils, and T cells helper are all negatively connected (p < 0.05). In the FOXF1 hyperexpression group, GSEA analysis
revealed that cholesterol homeostasis and cell cycle-associated pathways G2M checkpoint were primarily activated (p < 0.05).
Conclusion. FOXF1 was found to be a reliable biomarker of infantile hemangiomas in our research of transcriptome changes in
infantile hemangiomas.

1. Introduction mixed [5]. IH is classified as focal, multifocal, segmental, and

Infantile hemangioma (IH) is the most frequent tumor in
newborns, affecting roughly 5–10% of them [1–3]. IH usually
occurs in the first few weeks of life and spreads quickly
during childhood [4]. According to the tissue level of
growth, IH can be characterized as superficial, deep, or
so on, depending on the pattern of growth [6]. The peri-
neum, face, limbs, and trunk are common IH locations [7].
Although most IH is benign and has a fair prognosis, there
are a number of problems that can have a significant impact
on a child’s look and quality of life, and extreme compli-
cations can even result in death [8]. Therefore, it is critical to
2 Contrast Media & Molecular Imaging
stratify the risk of IH patients and look for new biomarkers
to help with diagnosis and treatment.
The genomes and transcriptome alterations of IH are yet
unclear. Using multiomics approaches to investigate genetic
changes in IH can serve as a starting point for the discovery
of new IH indicators and the investigation of the immu-
nological microenvironment [9]. In recent years, RNA-se-
quencing has been utilized to investigate genetic alterations
in a variety of disorders. Many biological and medical re-
search studies are increasingly relying on bioinformatics as a
result of the rise of bioinformatics [10]. Understanding gene
expression regulatory mechanisms is also an important
aspect of bioinformatics [11]. The diagnosis and treatment of
human diseases are defined in terms of the role of bio-
molecules in gene regulation. Bioinformatics has emerged as
a critical component in the advancement of life science as a
whole, as well as the frontier of life science research. GEO is
currently one of the most widely used databases in bio-
informatics analysis, with gene expression data given by
research organizations all over the world, primarily gene
chip and high-throughput sequencing data. We used the
GEO database to download IH dataset for analysis in this
study. Dataset GSE127487, published in 2019, containing 23
IH samples and 5 normal controls, was downloaded for
analysis. In the original study, transcriptomic sequencing
data were used to reveal potential biomarkers of propranolol
for IH: EPAS1, LASP1, SLC25A23, MYO1B, and ALDH1A1.
We used this transcriptome data to perform weighted
coexpression network analysis and differential expression
analysis and identified some new biomarkers for IH [12].
2. Methods
2.1. Data Download and Processing. The Gene Expression
Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/
geo/) is a comprehensive website that collects transcriptome
data from a wide range of diseases and clinical data, making
it easy for researchers to examine. GSE127487, a dataset
encompassing 23 infantile hemangiomas and 5 normal
control samples, was retrieved from this database. GPL13607
annotated gene names and averaged several probes be-
longing to a gene name. The data from the transcriptome
were standardized and log2 transformed.
2.2. Weighted Coexpression Network Analysis (WGCNA).
Genes with comparable biological roles are classified into
clusters in WGCNA analysis, which is based on grouping of
gene expression patterns of each sample. The soft threshold
range in this study is 0–10 as a continuous integer and 10–20
as an integer with step 2. The pickSoftThreshold function of
the function WGCNA package can be used to find the best
soft domain value. To execute gene clustering, set the
minimum number of module genes to 200 and deepSplit to
2.
2.3. Enrichment Analysis of Gene Ontology (GO). Cellular
component (CC), molecular function (MF), and biological
process (BP) are the three sections of the GO database. By
comparing the given gene list with the gene set of biological
functions, the GO database was used to examine the bio-
logical functions enriched.
2.4. Kyoto Encyclopedia of Genes and Genomes (KEGG) En-
richment Analysis. The gene list provided was matched with
genes in human pathways collected by the KEGG database,
and the enriched pathway was studied.
2.5. Analysis of Differentially Expressed Genes (DEGs).
The “LIMMA” software was used to look for differentially
expressed genes in the control and infantile hemangioma
groups, with p value < 0.05 and | logFC | > 0.5. For visual-
ization, the R packages heatmap and ggplot2 are utilized.
2.6. Protein-Protein Interaction Network Analysis.
STRING<>(https://cn.string-db.org/) is a popular protein
interaction analysis website. The site analyzes the organism
by inputting a list of genes into a box and selecting Homo
sapiens as the target species and then downloading and
visualizing the results with Cytoscape software. The cyto-
Hubba plug-in was then used to look for other hub genes,
which were ordered using the MCC technique.
2.7. Immune Cell Infiltration Analysis. The CIBERSORT
method is the most widely used method for measuring
immune infiltration, and it was utilized to quantify the
degree of immune cell infiltration in infantile hemangioma
and the control group in this study.
2.8. Gene Set Enrichment Analysis (GSEA). GSEA is a widely
used enrichment analysis method that compares genes to a
set of genes. GSEA examines if all of the genes in a gene set
are clustered at the top or bottom of the multiples of dif-
ference list.
3. Results
3.1. Weighted Coexpression Network Analysis (WGCNA).
WGCNA analysis was performed in this work to look for
gene sets that change in conjunction with infantile hem-
angioma. The best soft domain value found from function
computation is 10, as shown in Figure 1(a). The data adhere
to the power law distribution when the soft domain value is
set to this value, R2 > 0.8, and when the soft domain value is
more than 10, mean connectivity gradually becomes stable.
All genes were coclustered into 13 nongray modules, as
shown in Figures 1(b) and 1(c), with turquoise and pink
being the most closely connected to infantile hemangioma.
There was a high positive association between gene signif-
icance for body weight and module membership in tur-
quoise and pink modules (COR � 0.92 and p < 0.001 and
COR � 0.78 and p < 0.001), as shown in Figures 1(d) and
1(e). The following analysis comprised genes from the
turquoise module, which contains 2,495 genes, and the pink
module, which has 673 genes.
Contrast Media & Molecular Imaging 3

Scale independence Mean connectivity

Scale Free Topology Model Fit, signed R^2

16 18 20 1
12 14
8 910 3000
0.8 67
5
4

Mean Connectivity
0.6
3 2000

0.4
2
2 1000
0.2
500 3
4
567
0.0 1 0 8 91012 14 16 18 20

5 10 15 20 5 10 15 20
Soft Threshold (power) Soft Threshold (power)
(a)
Gene dendrogram and module colors
1.0

0.9

0.8

0.7
Height

0.6

0.5

0.4

0.3
Dynamic
Tree Cut

(b)
Figure 1: Continued.
4 Contrast Media & Molecular Imaging

Module−trait relationships
−0.9 0.9
MEturquoise
(1e−10) (1e−10) 1
−0.63 0.63
MEblack
(3e−04) (3e−04)
−0.62 0.62
MEbrown
(4e−04) (4e−04)
0.4 −0.4
MEgreen
(0.04) (0.04) 0.5
0.61 −0.61
MEpink
(6e−04) (6e−04)
0.32 −0.32
MEyellow
(0.09) (0.09)
−0.11 0.11
MEsalmon
(0.6) (0.6) 0
−0.56 0.56
MEpurple
(0.002) (0.002)
−0.18 0.18
MEtan
(0.4) (0.4)
−0.25 0.25
MEblue
(0.2) (0.2)
0.26 −0.26 −0.5
MEred
(0.2) (0.2)
0.45 −0.45
MEgreenyellow
(0.02) (0.02)
0.35 −0.35
MEmagenta
(0.07) (0.07)
0.21 −0.21 −1
MEgrey
(0.3) (0.3)
Control Hemangioma
(c)
Module membership vs. gene significance Module membership vs. gene significance
cor=0.92, p<1e−200 cor=0.78, p=9.9e−139
0.8
Gene significance for body weight

Gene significance for body weight

0.8
0.6

0.6
0.4
0.4

0.2 0.2

0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8
Module Membership in turquoise module Module Membership in pink module
(d) (e)

Figure 1: Weighted coexpression network analysis (WGCNA). (a) The best soft domain value found from function computation is 10. (b)-
(c) All genes coclustered into 13 nongray modules with turquoise and pink being the most closely connected to infantile hemangioma. (d)-
(e) Association between gene significance for body weight and module membership in turquoise and pink modules (COR � 0.92 and
p < 0.001 and COR � 0.78 and p < 0.001).
Contrast Media & Molecular Imaging 5

actin filament organization Count

60
embryonic organ development
70
reproductive structure development 80
90
cell−substrate adhesion
100
extracellular matrix organization 110
extracellular structure organization
p.adjust
regulation of small GTPase mediated signal transduction

positive regulation of ion transport 2e−05

cell−matrix adhesion 4e−05
regulation of cell shape 6e−05
0.020 0.025 0.030 0.035
GeneRatio
(a)

PI3K−Akt signaling pathway

Human papillomavirus infection Count
40
Focal adhesion
60
Hepatitis C 80
Breast cancer
p.adjust
Measles
Prostate cancer 0.005
0.010
Endocrine resistance
0.015
Small cell lung cancer
Non−small cell lung cancer
0.02 0.03 0.04 0.05 0.06
GeneRatio
(b)

Figure 2: (a) GO and (b) KEGG enrichment analyses.

3.2. GO and KEGG Enrichment Analyses. As shown in
Figure 2(a), infantile hemangioma-related genes acquired by
WGCNA were primarily enriched in actin filament orga-
nization, embryonic organ development, reproductive
structure development, cell substrate adhesion, extracellular
matrix organization, extracellular structure organization,
regulation of small GTPase-mediated signal transduction,
positive regulation of ion transport, cell matrix adhesion, cell
shape regulation, and so on, as determined by GO en-
richment. KEGG enrichment analysis revealed that the
PI3K-Akt signaling pathway, human papillomavirus infec-
tion, focal adhesion, and hepatitis C were the most enriched,
as shown in Figure 2(b).
gene expression in infantile hemangiomas and controls. The
volcano map, as shown in Figure 3(b), illustrates the dis-
tribution of differential genes, with HSD17B2, IDO1,
PCDH12, FCN3, and LIN28B being the most significantly
upregulated genes with the lowest p value. SERPINA3, SLPI,
TC2N, TFF3, and MUC7 were the five genes that were
significantly downregulated. 36 genes were determined to be
most linked with infantile hemangioma when the turquoise
and pink modules were intersected with these upregulated
genes. The interaction between these proteins is shown in
Figure 3(c). FOXF1 was identified as the top hub gene by the
CytoHubba plug-in.

3.4. Expression and Diagnostic Value of FOXF1 in Infantile

3.3. Differential Expression Gene Analysis and Protein-Protein
Interaction Network Construction. By using the differential
analysis of limma program, a total of 64 differential genes
were screened from infantile hemangioma and the control
group, of which 43 were upregulated and 21 were down-
regulated. The heat map in Figure 3(a) shows differential
Hemangioma. FOXF1 was found to be strongly expressed in
infantile hemangiomas (p < 0.001), as shown in Figure 4(a).
ROC and PRC analyses were used to further investigate the
diagnostic significance of this gene in infantile hemangioma.
The AUC of ROC analysis was 0.97, while the AUC of PRC
analysis was 0.99, indicating that FOXF1 is a reliable marker
6 Contrast Media & Molecular Imaging

Type Type
4
DPY19L2 Control
HIGD1B
LRRC36 Hemangioma
DNAH2
KL 2
TFPI2
GPR4
ITGA1
TRPC6
ZBTB46 0
MAEL
ADRB1
HOXA13
ZNF366
PCDH12 −2
SHROOM1
HSD17B2
IDO1
CLEC1A
APLN
FCN3 −4
HS3ST3A1
P2RY1
TCF15
HOXB7
HLX
TBX2
TMEM88
ABLIM3
ADRA1B
NDUFA4L2
COX4I2
PCDH17
CLIC5
TMSB15A
LIN28B
NETO2
SSTR5
IDO2
TNNT3
DLK1
FOXF1
SLAMF1
DAPL1
CLDN1
KRT15
MAL2
FOXQ1
DEFB1
OVOL2
SCNN1A
CDH1
AZGP1
SLPI
MUC7
PIP
WDR72
PIGR
SCGB3A1
TC2N
PITX1
ADH1A
SERPINA3
TFF3
(a)

10.0 HSD17B2
IDO1
PCDH12 FCN3

LIN28B
7.5
−Log10 (FDR q−value)

−Log10_q−value

7.5
SERPINA3
5.0
5.0
SLPI TC2N
2.5

MUC7 TFF3
2.5

0.0

−0.5 0.0 0.5 1.0

Log2FC
(b)
Figure 3: Continued.
Contrast Media & Molecular Imaging 7

(c)

Figure 3: Differential expression gene analysis and protein-protein interaction network construction. (a) The heat map of differential gene
expression in infantile hemangiomas and controls. (b) The volcano map. (c) The interaction between these proteins. FOXF1 was identified as
the top hub gene by the CytoHubba plug-in.

for the diagnosis of infantile hemangioma, as shown in
Figure 4(b).
3.5. Immune Cell Infiltration Analysis. The CIBORSORT
algorithm was used to assess the degree of immune cell
infiltration in infantile hemangiomas. Mast cells make up a
significant fraction, as shown in Figure 5(a). Mast cells
resting identified higher infiltration in the FOXF1 high
expression group, as shown in Figure 5(b). Whereas,
macrophages M0 had a lower degree of infiltration
(p < 0.05). FOXF1 is favorably connected with monocyte
infiltration degree, whereas it is negatively correlated with
B cell memory, dendritic cells resting, macrophage M0,
neutrophils, and T cells follicular helper (p < 0.05),
according to correlation analysis (Figures 5(c)–5(h)).
3.6. GSEA Analysis. The cholesterol homeostasis and cell
cycle-associated pathways G2M CHECKPOINT were pri-
marily activated in the FOXF1 hyperexpression group, as
shown in Figures 6(a) and 6(b).
8 Contrast Media & Molecular Imaging

9 ***

FOXF1 _expression
7

Control IH

Group
Control
Hemangioma
(a)
ROC − P: 23, N: 5 Precision−Recall − P: 23, N: 5

1.00 1.00

0.75 0.75
Sensitivity

Precision

0.50 0.50

0.25 0.25

0.00 0.00

0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
1-specificity Recall
(b)

Figure 4: Expression and diagnostic value of FOXF1 in infantile hemangioma. (a) FOXF1 found to be strongly expressed in infantile
hemangiomas (p < 0.001). (b) The AUC of ROC analysis is 0.97 and PRC analysis is 0.99.

4. Discussion
The most common benign tumor in children, infantile
hemangioma (IH), has a distinct development pattern [13].
IH grows rapidly before the age of eight weeks and is es-
sentially mature before the age of five months, with the
majority of spontaneous regression [13]. For the treatment
of early IH, beta-blockers are suggested [14]. However, beta-
blockers have limited clinical efficacy in some patients [14].
Furthermore, some children (approximately 10%) still suffer
evident clinical consequences, such as blindness, ulceration,
infection, and even heart failure, which cause many chil-
dren’s lives to be ruined and even harm them for the rest of
their lives [15]. As a result, early and reliable biomarkers for
IH are needed to guide IH diagnosis and treatment.
Bioinformatics analysis was employed to investigate
transcriptome changes in IH in depth in this work. To
begin, weighted coexpression network analysis (WGCNA)
was utilized to look at the modules that were most closely
linked to IH. Turquoise and pink modules were found to
have a strong correlation with IH. Then, we did an en-
richment analysis of those genes and discovered that they
were mostly involved in actin filament organization, em-
bryonic organ development, reproductive structure de-
velopment, cell substrate adhesion, extracellular matrix
organization, extracellular structure organization,
Contrast Media & Molecular Imaging 9

100%
B cells naive
B cells memory
Plasma cells
80% T cells CD8
T cells CD4 memory resting
T cells CD4 memory activated
T cells follicular helper
Relative percentage T cells regulatory (Tregs)
60% T cells gamma delta
NK cells resting
NK cells activated
Monocytes
Macrophages M0
40% Macrophages M1
Macrophages M2
Dendritic cells resting
Dendritic cells activated
20% Mast cells resting
Mast cells activated
Eosinophils
Neutrophils
0%
(a)
Fraction

0.0

0.1

0.2

0.3

0.4

0.5

0.6
P=0.25
High_FOXF1
Low_FOXF1

B cells naive
P=0.258

B cells memory
P=0.321

Plasma cells
P=0.271

T cells CD8 P=0.376

T cells CD4 memory resting

P=0.757

T cells CD4 memory activated

P=0.082

T cells follicular helper

P=0.534 P=1

T cells regulatory (Tregs)

T cells gamma delta

P=0.678

NK cells resting
P=0.871

NK cells activated
P=0.505

Monocytes
P=0.017

Macrophages M0
P=0.836

Macrophages M1
P=0.511 P=0.201

Macrophages M2

Dendritic cells resting

P=0.87

Dendritic cells activated

P=0.031

Mast cells resting

P=0.326 P=1P=0.165

Mast cells activated

Eosinophils

Neutrophils

(b)
Figure 5: Continued.
10 Contrast Media & Molecular Imaging

R = –0.53, p = 0.0035 R = –0.55, p = 0.0027 R = –0.4, p = 0.033

0.020 0.20

Dendritic cells resting

0.2

Macrophages M0
0.015 0.15
B cells memory

0.010 0.10
0.1
0.005
0.05

0.000 0.0
0.00

5 6 7 8 5 6 7 8 5 6 7 8
FOXF1 expression level FOXF1 expression level FOXF1 expression level

R = 0.38, p = 0.047 R = –0.46, p = 0.014 R = –0.69,, p = 4.7e−05

0.15 0.075

T cells follicular helper

0.02
Neutrophils
Monocytes

0.10 0.050

0.01
0.05 0.025

0.00
0.00 0.000

5 6 7 8 5 6 7 8 5 6 7 8
FOXF1 expression level FOXF1 expression level FOXF1 expression level
(c)

Figure 5: Immune cell infiltration analysis. (a) The degree of immune cell infiltration in infantile hemangiomas. (b) Mast cells resting
identified higher infiltration in the FOXF1 high expression group. Whereas, macrophages M0 had a lower degree of infiltration (p < 0.05).
(c)–(h) Immune correlation analysis.

regulation of small GTPase mediated signal transduction,
positive regulation of ion transport, cell matrix adhesion,
cell shape regulation, and so on. The PI3K-Akt signaling
pathway, human papillomavirus infection, focal adhesion,
and hepatitis C, among others, are heavily represented in
the pathways of these genes. This provides insights into the
mechanisms of these genes and their interactions in in-
fantile hemangiomas. FOXF1 was identified as a critical
gene for IH after further protein-protein interaction net-
work research. FOXF1 was considerably upregulated in IH,
according to expression analysis, and the ROC curve
revealed that FOXF1 had a very high AUC value and good
diagnostic accuracy. FOXF1 was shown to be connected
with various immune cells in IH, most of which were
negatively correlated, but favorably correlated with
monocytes, according to a subsequent investigation of the
immunological microenvironment. This serves as a starting
point for learning more about FOXF1’s significance in the
immunological background of IH. Finally, gene set en-
richment analysis (GSEA) revealed significant cholesterol
homeostasis and G2M checkpoint pathway enrichment in
the FOXF1 high expression group.
FOXF1 belongs to the forkhead-box (FOX) family of
proteins and is involved in cell proliferation, differentia-
tion, embryogenesis, and longevity [16–18]. The FOX
family is also involved in transcription and DNA repair
control. The importance of FOXF1 in many human diseases
has been preliminarily elucidated, particularly in various
malignancies, at this time. Milewski et al. found that FOXF1 is
downstream of PAX3-FOXO1 in rhabdomyosarcoma, pro-
moting tumor development, and FOXF1 may be a potential
therapeutic target for rhabdomyosarcoma [19]. Zhao et al.
discovered that epigenetic activation of FOXF1 allows cis-
platin-resistant NSCLC to acquire stem cell characteristics
[20]. FOXF1 increases tumor spread in colorectal cancer by
activating SNAI1 and causing epithelial-mesenchymal tran-
sition (EMT), according to Wang et al. [21]. Not only that,
FOXF1 has also been shown to play an important role in
benign disease. FOXF1 knockdown enhances Wnt/β-cate-
cholamine activation, which enables bone marrow-derived
mesenchymal stem cell osteogenesis and protects bone loss
after oophorectomy, according to Shen et al. [22]. Sturtzel et al.
discovered that FOXF1 is involved in the germination of
endothelial progenitor cells and linked blood vessels, which
may be implicated in tissue neovascularization [23]. Our study
reveals the value of FOXF1 in infantile hemangiomas, where
the role of FOXF1 was previously unknown. FOXF1 has a high
AUC value in infantile hemangioma, and the AUC value in
both ROC curve and PRC curve is greater than 0.9, indicating
that FOXF1 is very robust in the diagnosis of infantile
Contrast Media & Molecular Imaging 11

Running Enrichment Score

pvalue p.adjust
HALLMARK_CHOLESTEROL_HOMEOSTASIS 3e–04 0.0078
0.4

0.2

0.0
Ranked List Metric

0.4
0.2
0.0
–0.2

5000 10000
Rank in Ordered Dataset
(a)
Running Enrichment Score

0.4 pvalue p.adjust

HALLMARK_G2M_CHECKPOINT 9e−04 0.0107
0.3

0.2

0.1

0.0
Ranked List Metric

0.4
0.2
0.0
−0.2

5000 10000
Rank in Ordered Dataset
(b)

Figure 6: (a)-(b) GSEA analysis.

hemangioma. FOXF1 has also been linked to a number of Conflicts of Interest

immune cells, shedding light on the pathophysiology of in-
fantile hemangiomas and the intricate interaction that occurs. The authors declare that they have no conflicts of interest.

5. Conclusion
In conclusion, our research looked at transcriptome changes
in infantile hemangioma and found FOXF1 to be a novel
biomarker of infantile hemangioma with a good diagnostic
value using several analysis approaches. Our research,
however, has certain limitations. We do not have enough
relevant in vitro and in vivo studies to verify FOXF1’s in-
fluence, but that will change in the future.
Data Availability
The data used to support the findings of this study are in-
cluded within the article.
Acknowledgments
The authors are very grateful for data provided by databases
such as TCGA and GeneCards.
References
[1] D. F. Sebaratnam, A. L. Rodrı́guez Bandera, L. C. F. Wong,
and O. Wargon, “Infantile hemangioma. Part 2: manage-
ment,” Journal of the American Academy of Dermatology,
vol. 85, no. 6, pp. 1395–1404, 2021 Dec.
[2] J. C. Lee, O. Modiri, R. W. England, C. J. Shawber, and
J. K. Wu, “Propranolol therapy in infantile hemangioma: it is
not just about the beta,” Plastic and Reconstructive Surgery,
vol. 147, no. 4, pp. 875–885, 2021 Apr 1.
12 Contrast Media & Molecular Imaging

[3] A. Filoni, C. Foti, F. Ambrogio, and D. Bonamonte, “Cuta- [19] D. Milewski, S. Shukla, B. E. Gryder et al., “FOXF1 is required
neous mastocytosis and infantile hemangioma: is it only a for the oncogenic properties of PAX3-FOXO1 in rhabdo-
coincidence?” Italian Journal of Dermatology and Venereol- myosarcoma,” Oncogene, vol. 40, no. 12, pp. 2182–2199, 2021.
ogy, vol. 156, pp. 55-56, 2022. [20] J. Zhao, X. Xue, W. Fu et al., “Epigenetic activation of FOXF1
[4] S. Spring, J. Noble, and E. Pope, “Dermacase. Infantile confers cancer stem cell properties to cisplatin-resistant non-
hemangioma,” Canadian family physician Medecin de famille small cell lung cancer,” International Journal of Oncology,
canadien, vol. 55, no. 4, pp. 379–381, 2009. vol. 56, no. 5, pp. 1083–1092, 2020.
[5] L. Bonafede, M. Go, J. Cheng et al., “Periocular infantile [21] S. Wang, S. Yan, S. Zhu et al., “FOXF1 induces epithelial-
hemangioma masquerading as dacryocele,” Journal of mesenchymal transition in colorectal cancer metastasis by
American Association for Pediatric Ophthalmology and transcriptionally activating SNAI1,” Neoplasia, vol. 20, no. 10,
Strabismus, vol. 24, no. 5, pp. 326–328, 2020. pp. 996–1007, 2018.
[6] C. Giugliano, F. Reculé, K. Guler, J. T. Gantz, and T. Hasbún, [22] G. Shen, H. Ren, Q. Shang et al., “Foxf1 knockdown promotes
“Persistent nasal infantile hemangioma,” Journal of Cranio- BMSC osteogenesis in part by activating the Wnt/β-catenin
facial Surgery, vol. 29, no. 6, pp. 1509–1513, 2018. signalling pathway and prevents ovariectomy-induced bone
[7] C. Karunasumetta, C. Wongbhuddha, S. Sangkhamanon, and loss,” EBioMedicine, vol. 52, Article ID 102626, 2020.
B. Mokarat, “Infantile cardiac hemangioma causing right [23] C. Sturtzel, K. Lipnik, R. Hofer-Warbinek et al., “FOXF1
ventricular inflow obstruction,” Asian Cardiovascular & mediates endothelial progenitor functions and regulates
Thoracic Annals, vol. 27, no. 7, pp. 597–599, 2019. vascular sprouting,” Frontiers in Bioengineering and Bio-
[8] L. L. Wessman, W. S. Mori, K. Totoraitis, M. Murati, and technology, vol. 6, p. 76, 2018.
S. Maguiness, “Multifocal infantile hemangiomas associated
with concomitant giant hepatic rapidly involuting congenital
hemangioma,” Pediatric Dermatology, vol. 37, no. 5,
pp. 972-973, 2020.
[9] S. A. Kullberg, M. N. Rheault, A. Haggstrom, and
S. M. Maguiness, “Segmental infantile hemangioma and
concomitant hypertension in three African American neo-
nates,” Pediatric Dermatology, vol. 37, no. 3, pp. 524–526,
2020.
[10] R. Xu, L. Chen, W. Wei et al., “Single-cell sequencing analysis
based on public databases for constructing a metastasis-re-
lated prognostic model for gastric cancer,” Applied Bionics
and Biomechanics, vol. 2022, Article ID 7061263, 19 pages,
2022.
[11] Y. Qiu, H. Li, Q. Zhang, X. Qiao, and J. Wu, “Ferroptosis-
related long noncoding RNAs as prognostic marker for colon
adenocarcinoma,” Applied Bionics and Biomechanics,
vol. 2022, Article ID 5220368, 8 pages, 2022.
[12] H. Gomez-Acevedo, Y. Dai, G. Strub, C. Shawber, J. K. Wu,
and G. T. Richter, “Identification of putative biomarkers for
Infantile Hemangiomas and Propranolol treatment via data
integration,” Scientific Reports, vol. 10, no. 1, p. 3261, 2020.
[13] A. Achibane, H. Taouti, S. Belghamaidi, I. Hajji, and
A. Moutaouakil, “Gestion d’un hémangiome palpébral du
nourrisson,” Journal Français d’Ophtalmologie, vol. 44, no. 9,
pp. 1474-1475, 2021.
[14] C. Al-Haddad, N. A. El Salloukh, and Z. El Moussawi,
“β-blockers in the treatment of periocular infantile heman-
gioma,” Current Opinion in Ophthalmology, vol. 30, no. 5,
pp. 319–325, 2019.
[15] K. R. Satterfield and C. B. Chambers, “Current treatment and
management of infantile hemangiomas,” Survey of Ophthal-
mology, vol. 64, no. 5, pp. 608–618, 2019.
[16] A. Pradhan, A. Dunn, V. Ustiyan et al., “The S52F FOXF1
mutation inhibits STAT3 signaling and causes alveolar cap-
illary dysplasia,” American Journal of Respiratory and Critical
Care Medicine, vol. 200, no. 8, pp. 1045–1056, 2019.
[17] D. Jin and F. Han, “FOXF1 ameliorates angiotensin II-in-
duced cardiac fibrosis in cardiac fibroblasts through inhibiting
the TGF-β1/Smad3 signaling pathway,” Journal of Receptors
and Signal Transduction, vol. 40, no. 6, pp. 493–500, 2020.
[18] L. Huang, M. Tjakra, D. Luo et al., “FoxF1 is required for
ciliogenesis and distribution of sonic hedgehog signaling
components in cilium,” Current Molecular Medicine, vol. 19,
no. 5, pp. 326–334, 2019.