Transcriptomic insights into nerve growth factor–associated appositional bone formation in the reparative zone in osteonecrosis of the femoral head

Abstract

Background

In osteonecrosis of the femoral head, appositional bone formation occurs in the reparative zone, where new trabeculae surround necrotic bone, leading to trabecular thickening. However, the underlying molecular mechanisms remain unclear. We previously demonstrated a significant correlation between trabecular thickening and nerve growth factor (NGF) expression in the reparative zone of stage 3 osteonecrosis of the femoral head. This study explored NGF’s molecular relationship with appositional bone formation.

Methods

RNA sequencing was performed on reparative zone samples from four stage 3 osteonecrosis of the femoral head femoral heads. One to three cylindrical samples (6 mm in diameter) were harvested per head. As controls, three samples were collected from the reparative zone of a single stage 1 osteonecrosis of the femoral head without histological evidence of appositional bone formation. Differentially expressed genes were identified using the DESeq2. Enrichment analysis was performed using the Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases via DAVID. NGF expression was evaluated using immunohistochemistry.

Results

Stage 3 samples showed the upregulation of bone formation–related genes, including COL1A1, RUNX2, BMP2, BGLAP, and NGF. KEGG analysis revealed the enrichment of bone-related signaling pathways, such as Wnt, TGF-β, and PI3K-Akt. NGF expression was observed in the reparative zone of stage 3 femoral heads, which exhibited appositional bone formation, but not in stage 1.

Conclusion

These findings indicate that the reparative zone exhibiting appositional bone formation shows transcriptional activation of bone formation, accompanied by increased NGF expression. NGF may be involved in the molecular mechanisms underlying appositional bone formation in osteonecrosis of the femoral head, and serve as a potential regulator of this process.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-025-06549-7.

Background

Recent reviews have summarized current advances in the clinical understanding, management, and pathophysiology of osteonecrosis of the femoral head [1–8]. In early-stage osteonecrosis of the femoral head (ONFH), necrotic lesions are detectable on magnetic resonance imaging, whereas radiographic findings often remain inconclusive. As the disease progresses, a sclerotic band typically forms at the boundary between the necrotic and viable bones, corresponding to the reparative zone [9, 10]. This zone is histologically characterized by appositional bone formation, in which newly formed trabeculae surround necrotic trabeculae [11, 12]. Previous studies have suggested that the resulting sclerotic area may alter the stress distribution, concentrating the mechanical load between the sclerotic region and the adjacent necrotic area, potentially contributing to femoral head collapse [13, 14]. Despite its clinical importance, the mechanisms underlying appositional bone formation remain unclear.

Nerve growth factor (NGF) is a neurotrophin that plays a critical role in the skeletal and neural systems. In addition to its involvement in pain signaling, studies have shown that NGF promotes bone formation by modulating osteoblast activity and enhancing vascularization and neurogenesis, which are increasingly recognized as contributors to bone remodeling and regeneration [15–19]. Similarly, vascular endothelial growth factor (VEGF) regulates angiogenesis and promotes bone formation by coupling vascular and osteogenic processes during bone repair [20–22]. Our previous histological analyses demonstrated that both NGF and VEGF were significantly increased in the reparative zone of ONFH. NGF was localized to osteoblast-like and fibroblast-like cells, whereas VEGF was expressed in osteoblast-like cells and vascular-rich fibrous tissue. These findings suggest that neurogenic and angiogenic activities coexist in the reparative zone during the repair process. Moreover, NGF expression correlated with trabecular thickening, indicating its potential involvement in appositional bone formation in ONFH [23]. However, the molecular mechanisms by which NGF contributes to this process remain unclear.

Although RNA sequencing (RNA-seq) has been used to investigate gene expression in ONFH, most studies have focused on comparisons between ONFH and other diseases such as osteoarthritis or femoral neck fractures [24–26]. Consequently, the gene expression patterns in different anatomical regions or stages of ONFH remain largely unexplored.

To address this gap in different anatomical regions and stage-specific gene expression, we performed RNA-seq to analyze the gene expression profiles of the reparative zones in stages 3 and 1 ONFH, focusing on the genes involved in bone formation and the potential role of NGF.

Methods

This retrospective study received approval from our institutional review board (approval date: 19 August 2022; approval no.: 2022‐86) and adhered to the Declaration of Helsinki. Informed consent was acquired from all participants.

Sample collection

Femoral head specimens were obtained from patients diagnosed with ONFH who underwent total hip arthroplasty at our institution. According to the Japanese Investigation Committee staging system [9], samples were collected from the reparative zones of stages 3 and 1 ONFH femoral heads. Immediately after surgical resection, the femoral heads were cooled on ice and sectioned into 5-mm thick coronal slices. Cylindrical tissue samples (6 mm in diameter) were harvested from both the reparative and viable zones of each femoral head using a single-use Osteochondral Autograft Transfer System (OATS) bone graft-harvesting instrument (6 mm, Arthrex Inc., Naples, FL, USA) (Fig. 1). One to three samples per zone were snap-frozen in liquid nitrogen. Nine reparative zone samples were obtained from four stage 3 femoral heads, and three samples were obtained from a single stage 1 femoral head. Among the four femoral heads with stage 3 ONFH, three were obtained from male patients and one from a female patient, with a mean age of 54.2 years. The etiologies of stage 3 ONFH were alcohol abuse in two cases and corticosteroid use in two cases. A single stage 1 femoral head was obtained from a 79-year-old female patient with corticosteroid-associated ONFH.

Fig. 1.

Representative images of the sampling procedure for RNA extraction. The resected femoral heads were sectioned coronally, and cylindrical tissue samples (6 mm in diameter) were harvested from the reparative zone using a bone graft-harvesting instrument. These samples were then used for RNA sequencing. a Photograph of a coronal slice before sampling. b Photograph taken after cylindrical sample extraction from the reparative zone. N, R, and V indicate the necrotic, reparative, and viable zones, respectively. c Hematoxylin and eosin (H&E) staining of the extracted samples confirmed that the tissue was obtained from the reparative zone. Scale bars: 5 mm

RNA extraction and sequencing

The frozen tissue samples were quickly pulverized after removal from liquid nitrogen and immersed in TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA) for RNA extraction. Total RNA was purified using an RNeasy Mini Kit (Qiagen, Hilden, Germany). RNA quality and concentration were assessed using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and an Agilent 2200 TapeStation system (Agilent Technologies, Santa Clara, CA, USA) with RNA ScreenTape. Samples that passed quality control were used for library preparation. Ribosomal RNA was depleted using the MGIEasy rRNA Depletion Kit v1.3 (MGI Tech Co., Ltd., Shenzhen, China), and strand-specific libraries were constructed using the MGIEasy Fast RNA Library Prep Set (MGI Tech Co., Ltd., Shenzhen, China). Sequencing was performed on the DNBSEQ-G400 FAST platform (MGI Tech Co., Ltd., Shenzhen, China) with 150-bp paired-end reads.

Data processing and differential gene expression analysis

Raw sequencing reads were trimmed using Trimmomatic v0.38, and the quality was assessed using FastQC. Trimmed reads were aligned to the human reference genome (GRCh38) using HISAT2 v2.1.0. Gene-level read counts were obtained using featureCounts, and differential gene expression analysis was performed using the DESeq2 package (v1.38.3) in R. Differentially expressed genes (DEGs) were defined as those with an adjusted p-value < 0.05 (Benjamini–Hochberg correction) and an absolute log₂ fold change ≥ 1. To minimize the influence of limited biological replicates, particularly because stage 1 samples were obtained from a single femoral head, DESeq2 was used to account for variance estimates, even in small-sample settings [27].

Functional enrichment analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using DAVID Bioinformatics Resources (version 6.8). GO terms were analyzed across the biological process, molecular function, and cellular component categories. GO terms with a Benjamini-adjusted p-value < 0.01 and KEGG pathways with p < 0.05 were considered significantly enriched. Upregulated and downregulated DEGs were analyzed separately. Functional enrichment was performed using the DEGs, as defined in data processing and differential gene expression analysis section.

Histological analysis

The femoral heads were sectioned into 5-mm slices in the coronal plane, fixed in 4% paraformaldehyde (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) for 2 days, degreased in 70% ethanol, decalcified with KCX (FALMA, Tokyo, Japan) for 7 days, embedded in paraffin, and sliced into 4-μm thick sections [23]. These slides were stained with hematoxylin and eosin (H&E) and examined under a microscope (Digital Microscope; Keyence, IL, USA) to confirm the presence of the characteristic three-layered structure of ONFH, which includes necrotic, reparative, and viable zones. In stage 1 specimens, ONFH was histologically confirmed based on the presence of necrotic trabeculae with empty lacunae and necrotic bone marrow cells. Additionally, the absence of appositional bone formation was confirmed, supporting the classification of stage 1 ONFH (Fig. 2).

Fig. 2.

Hematoxylin and eosin (H&E) staining of the stage 1 osteonecrotic femoral head used for RNA-seq analysis. a At × 20 magnification, a necrotic area was observed centrally within the femoral head, surrounded by a region rich in cellular components. The area outlined by the black square is shown at higher magnification in (b). Scale bar: 5 mm. b At × 100 magnification, necrotic trabeculae with empty lacunae and necrotic bone marrow cells were observed, findings consistent with ONFH. Inflammatory cells and fibroblast-like cells were observed around the necrotic trabeculae, corresponding to the reparative zone. Scale bar: 100 μm

Immunohistochemistry

Paraffin-embedded sections were deparaffinized, rehydrated, and subjected to antigen retrieval overnight in 1 mM ethylenediaminetetraacetic acid buffer (pH 8.0). Endogenous peroxidase activity was quenched using 3% hydrogen peroxide in methanol for 30 min. Nonspecific binding was blocked with 5% normal horse serum for 30 min. The sections were incubated overnight at 4 °C with a rabbit anti-NGF primary antibody (ab52918; Abcam, Cambridge, UK; 1:250 dilution). The following day, biotinylated secondary antibodies and streptavidin–peroxidase complexes were applied sequentially using a VECTASTAIN kit (VECTASTAIN Universal Elite ABC Kit; Vector Laboratories, Burlington, Canada). Immunoreactivity was visualized using 3, 3′-diaminobenzidine and counterstained with hematoxylin [23]. NGF expression was evaluated microscopically by comparing the staining patterns between the stage 3 and stage 1 reparative zones.

Results

RNA-seq analysis suggests osteogenic gene expression characteristics in stage 3 ONFH

RNA-seq analysis comparing the reparative zones of stages 3 versus 1 ONFH revealed differences in gene expression profiles indicative of enhanced bone-related activities. RNA integrity was evaluated using the Agilent TapeStation system (Supplementary Table S1). RNA from the reparative and viable zones showed sufficient quality (RIN 4.4–7.3), whereas RNA from necrotic zones was severely degraded (RIN ≤ 3.0) and therefore excluded from further analysis. Among these, only reparative zone samples were subjected to RNA-seq analysis, as the purpose of this study was to investigate the molecular characteristics specific to the reparative zone in ONFH. Each library generated approximately 25–33 million reads with mapping rates exceeding 95%, ensuring reliable downstream analysis. A total of 2, 176 genes were upregulated and 1, 418 genes were downregulated in the stage 3 reparative zone compared to those in stage 1 (Fig. 3a).

Fig. 3.

Differential gene expression analysis and osteogenesis-related gene expression. a Volcano plot displaying differentially expressed genes (DEGs) between the reparative zones of stage 3 and stage 1 osteonecrosis of the femoral head (ONFH). b Bar plot showing the upregulated genes associated with bone formation (orange), angiogenesis (red), neurogenesis (yellow), and nerve growth factor (NGF; purple)

Notably, genes involved in osteogenesis, such as COL1A1, BGLAP, and BMP2, were significantly upregulated. Other osteoblast-related genes, including COL1A2, RUNX2, SP7 (osterix), ALPL, and IBSP, also showed increased expression, suggesting active matrix production and mineralization in the stage 3 zone. In addition, NGF was upregulated at the transcriptional level, which may have contributed to the regulation of bone formation. With respect to angiogenesis, ANGPT2 was upregulated, whereas ANGPT1 was downregulated, suggesting a proangiogenic environment in the stage 3 zone (Fig. 3b). In contrast, VEGFA and its receptors (FLT1, KDR) showed no significant differential expression (log₂FC ≈ 0.10; p = 0.78; padj > 0.05) in this comparison.

Enrichment analyses highlight signaling pathways associated with bone formation and matrix remodeling

KEGG pathway enrichment analysis of the upregulated genes revealed the activation of signaling cascades linked to bone-related cellular functions. These included the TGF-β, Wnt, PI3K-Akt, MAPK, axon guidance, and extracellular matrix (ECM)–receptor interaction pathways, all of which are known to support osteoblast differentiation, ECM synthesis, and cellular communication (Fig. 4). Downregulated genes were enriched in the cell cycle and DNA replication pathways, suggesting a transition from proliferation to matrix organization and differentiation (Fig. 4).

Fig. 4.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of differentially expressed genes (DEGs) in the reparative zone of stage 3 vs. stage 1 ONFH. Bar plots showing the top 20 enriched KEGG pathways among upregulated genes (red) and the top five enriched pathways among downregulated genes (blue), based on -log₁₀(P-value). The upregulated pathways included axon guidance, Hippo, ECM–receptor, Wnt, PI3K-Akt, and TGF-β signaling. The downregulated pathways were primarily associated with cell cycle regulation and DNA replication

GO term analysis supported these findings. In the biological process (BP) category, enriched terms included osteoblast differentiation, angiogenesis, axon guidance, and nervous system development, indicating coordinated activation of osteogenic, vascular, and neuronal pathways in the reparative zone (Fig. 5a). In the cellular component (CC) category, upregulated genes were mainly associated with the extracellular matrix (ECM), collagen-containing ECM, and endoplasmic reticulum lumen, reflecting enhanced matrix production and protein secretion (Fig. 5b). In the molecular function (MF) category, enrichment was observed in extracellular matrix structural constituent, collagen binding, integrin binding, heparin binding, and calcium ion binding, suggesting activation of ECM-related interactions and mineralization processes (Fig. 5c).

Fig. 5.

Gene Ontology (GO) enrichment analysis of differentially expressed genes (DEGs) in the reparative zone of stage 3 vs. stage 1 osteonecrosis of the femoral head (ONFH). Upregulated (left) and downregulated (right) genes are shown in a single panel. a Biological process (BP): Enriched terms included osteoblast differentiation, angiogenesis, axon guidance, and nervous system development, indicating coordinated activation of osteogenic, vascular, and neuronal pathways in the reparative zone. b Cellular component (CC): Enriched terms included the extracellular matrix (ECM), collagen-containing ECM, and endoplasmic reticulum lumen, reflecting enhanced matrix production and protein secretion. c Molecular function (MF): Enriched terms included extracellular matrix structural constituent, collagen binding, integrin binding, heparin binding, and calcium ion binding, suggesting activation of ECM-related signaling and mineralization. Arrows indicate the key pathways emphasized in the Results section

NGF expression is observed predominantly in stage 3 reparative zones

Immunohistochemistry confirmed the transcript-level findings, demonstrating strong NGF expression in the reparative zone of the stage 3 femoral head, where appositional bone formation was evident (Fig. 6a). In contrast, no detectable NGF staining was observed in the reparative zone of the stage 1 femoral head, which lacked appositional bone formation (Fig. 6b). A high-magnification image of the boxed area in Fig. 6a further revealed that NGF-positive cells were mainly osteoblast-like cells lining newly formed trabeculae and fibroblast-like stromal cells located near vascular structures within the reparative zone (Fig. 6c).

Fig. 6.

Histological and immunohistochemical staining for nerve growth factor (NGF) in the reparative zone of osteonecrosis of the femoral head (ONFH). a In stage 3 ONFH, hematoxylin and eosin (H&E) staining (left) revealed an eosinophilic region corresponding to the reparative zone, where NGF immunostaining (right) showed strong positive staining (brown). b In stage 1 ONFH, a corresponding eosinophilic region was observed on H&E staining (left), but NGF immunostaining (right) showed no detectable signals. c High-magnification image of the boxed area in (a), showing NGF immunoreactivity in osteoblast-like cells lining newly formed trabeculae (black arrows), fibroblast-like stromal cells (black arrowheads), and endothelial-like cells within vascular structures (red arrows). Images (a) and (b) were obtained at × 20 magnification, and (c) was a × 200 magnified view of the black square indicated in (a). Scale bars: (a, b) 5 mm; (c) 50 µm

Discussion

Appositional bone formation in the reparative zone of ONFH is thought to contribute to structural changes and collapse; however, its molecular basis remains unclear. In this study, we explored the molecular differences between the reparative zones of stages 3 and 1 ONFH using RNA-seq and immunohistochemistry. Our analysis revealed increased expression of multiple osteogenic markers in stage 3, including COL1A1, RUNX2, SP7, BMP2, and BGLAP, indicating active matrix synthesis and mineralization. These findings are consistent with the histological observations showing trabecular thickening in the boundary area of stage 3 ONFH, which is thought to result from appositional bone formation [23].

In addition to the osteogenic gene profile, KEGG and GO enrichment analyses demonstrated the upregulation of pathways involved in TGF-β, Wnt, PI3K-Akt, and MAPK signaling, all of which are known to promote osteoblast differentiation and matrix production. CC terms, such as ECM and collagen-containing ECM, were also enriched, further supporting the enhanced matrix remodeling activity. Although terms related to angiogenesis and axon guidance were also enriched, the extent to which these changes directly reflect vascular or neural remodeling remains unclear, as our study was not designed to functionally assess these processes. Consistent with this, VEGFA and its receptors (FLT1 and KDR) did not show significant differential expression (log₂FC ≈ 0.10; p = 0.78; padj > 0.05) in the present RNA-seq analysis, suggesting that angiogenic activity in the reparative zone may not primarily depend on VEGF transcriptional upregulation but could instead involve alternative or post-transcriptional mechanisms.

Among the upregulated genes, NGF was of particular interest. NGF expression was not only elevated at the transcript level in stage 3 samples exhibiting appositional bone formation but was also confirmed at the protein level via immunohistochemistry, where it was localized in the reparative zone. In our previous histological study, NGF immunoreactivity was observed in osteoblast-like cells attached to the appositional bone and in fibroblast-like cells within the stroma of the reparative zone, both of which are involved in bone repair [23]. The present transcriptomic findings showing NGF upregulation, along with the enrichment of osteogenesis-related genes and signaling pathways, further support the possible role of NGF in appositional bone formation. These findings support our histological observation that trabecular thickening in ONFH is associated with NGF expression [23] and collectively suggest that NGF may contribute to appositional bone formation in the reparative zone.

High-magnification analysis (Fig. 6c) provided further insight into the cellular localization of NGF within the reparative zone of stage 3 ONFH. NGF-positive immunoreactivity was mainly detected in osteoblast-like cells lining newly formed trabeculae, fibroblast-like stromal cells, and endothelial-like cells associated with vascular structures. These findings, together with our previous histological observations [23], indicate that NGF is locally expressed by osteogenic and stromal cells adjacent to regions of active bone formation in the reparative (boundary) zone. Given that previous studies have shown that NGF promotes osteogenesis through both neurogenic and angiogenic mechanisms [15–19], our results suggest that locally produced NGF may contribute to the coupling of osteogenic, vascular, and neural processes during the reparative phase of ONFH. Although this study analyzed tissues at a single time point and cannot establish causal relationships, the spatial association between NGF-positive cells and newly formed trabeculae supports the potential involvement of NGF in local bone formation.

This study had some limitations. The most critical limitation is that stage 1 samples were derived from a single femoral head, making it difficult to treat the results as broadly representative of the early stages of ONFH. However, stage 1 femoral heads are extremely rare and difficult to obtain, and we believe that analyzing such samples can still provide valuable biological insights. To partially address this limitation, we obtained RNA from three spatially distinct regions of the same femoral head, which allowed us to incorporate some degree of intrasample variability and assess spatial heterogeneity. While statistical inference from a single biological specimen is inherently limited, we interpreted the results from stage 1 as exploratory and hypothesis-generating rather than definitive. Importantly, the transcriptomic findings from this rare sample were consistent with the histological observations, suggesting that they may reflect meaningful early molecular changes in ONFH that warrant further validation in future studies. Although the number of biological replicates for stage 3 ONFH (n = 4) is relatively small, we minimized this limitation by utilizing DESeq2, a widely validated statistical method designed to handle low-replicate RNA-seq data. DESeq2 applies shrinkage estimators for both dispersion and fold change, enabling more reliable detection of differentially expressed genes by borrowing information across the entire dataset [28]. Previous benchmark analyses [29] have demonstrated that DESeq2 maintains high sensitivity and low false discovery rates even with limited replicates (e.g., n = 3), making it particularly suitable for studies where large sample numbers are difficult to obtain. In this study, we used DESeq2 for differential gene expression analysis to partially mitigate the limitations posed by the lack of biological replicates. Another limitation is the age difference between the stage 1 and stage 3 specimens, which may have influenced bone metabolism and osteogenic activity. The stage 1 sample was obtained from a 79-year-old patient, whereas the stage 3 specimens were collected from patients with a mean age of 54.2 years. Because bone formation capacity generally declines with age, some of the observed transcriptional differences may reflect age-related variation in osteogenic potential rather than stage-specific pathology. Nevertheless, stage 1 ONFH cases are extremely rare and difficult to obtain, and the inclusion of this specimen still provides valuable biological insight into the early reparative process. Furthermore, because of the inherently low RNA yield from necrotic regions, we were unable to reliably extract and analyze RNA from the central necrotic area; therefore, this region was excluded from the present study. In addition, this study analyzed samples obtained at a single time point, which limits the ability to infer causal relationships among NGF expression, neurogenesis, angiogenesis, and osteogenesis. However, previous studies have demonstrated that NGF can promote bone formation through neurogenic and angiogenic mechanisms [15–19]. Therefore, while our findings do not establish causality, they suggest that increased NGF expression in the reparative zone of ONFH may contribute to local bone formation through these potential pathways. To further elucidate stage-dependent molecular changes in ONFH and clarify the potential role of NGF in appositional bone formation, future studies should involve the accumulation of additional clinical specimens, particularly from early-stage femoral heads (stages 1 and 2), and the use of animal models for experimental validation.

Despite these limitations, the consistent upregulation of osteogenic genes and signaling pathways in reparative zones exhibiting appositional bone formation supports the concept that local bone formation contributes to structural changes such as trabecular thickening and sclerosis.

Conclusion

This study demonstrates that the reparative zone exhibiting appositional bone formation exhibits transcriptional and histological features consistent with those of active bone formation. NGF was upregulated at both the transcript and protein levels and was localized in the reparative zone. These findings suggest that NGF may be involved in the molecular mechanisms of appositional bone formation in ONFH and provide a basis for further understanding the local bone formation processes in the reparative zone.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ONFH

Osteonecrosis of the femoral head

NGF

Nerve growth factor

RNA-seq

RNA sequencing

OATS

Osteochondral autograft transfer system

DEGs

Differentially expressed genes

GO

Gene Ontology

KEGG

Kyoto encyclopedia of genes and genomes

H&E

Hematoxylin and eosin

ECM

Extracellular matrix

VEGF

Vascular endothelial growth factor

Author contributions

Yusuke Ayabe: research design; acquisition, analysis, and interpretation of data; drafting of the manuscript; critical revision; and approval. Soichiro Yoshino: analysis and interpretation of data, drafting of the manuscript, critical revision, and approval. Goro Motomura: research design; acquisition, analysis, and interpretation of data; drafting of the manuscript; critical revision; and approval. Ryosuke Yamaguchi: research design; acquisition, analysis, and interpretation of data; and approval. Takeshi Utsunomiya: research design; acquisition, analysis, and interpretation of data; drafting of the manuscript; critical revision; and approval. Kosei Sakamoto: research design; acquisition, analysis, and interpretation of data; and approval. Yasuharu Nakashima: research design; acquisition, analysis, and interpretation of data; and approval. All the authors have read and approved the final version of the manuscript.

Funding

This work was supported by the Ministry of Health, Labor, and Welfare Research Program on Rare and Intractable Diseases Grant (JPMH23FC1045) and Grant-in-Aid for Scientific Research (JP23K08699 and JP24K19585) from the Japan Society for the Promotion of Science.

Data availability

The RNA-seq data have been deposited in the European Nucleotide Archive (ENA) under the accession number PRJEB102753. These data will be publicly available after publication.

Declarations

Ethics approval and consent to participate

This retrospective study received approval from our institutional review board (approval date: 19 August 2022; approval no.: 2022‐86) and adhered to the Declaration of Helsinki. Informed consent was acquired from all participants.

Consent for publication

Written informed consent for publication of the clinical details and images was obtained from the patient.

Competing interests

The authors declare no competing interests.