Skip to main content
NCBI home page
EJNMMI Research logoLink to EJNMMI Research
. 2026 Jul 4;16:139. doi: 10.1186/s13550-026-01471-3

Identifying genetics biomarkers in correlation with pathological and PSMA PET/CT characteristics in prostate cancer

Lili Qu 1, Kaiyue Li 1, Muwen Wang 2, Yadi Xiao 3, Xin Jin 4, Hang Zhou 1, Lujie Yuan 1, Yuekai Li 1, Shiwei Wang 5, Rui Li 5, Marcus Hacker 7, Xin Li 1,✉, Xiang Li 6,7,8,✉
PMCID: PMC13612772  PMID: 42400802

Abstract

Background

Prostate cancer is a prevalent disease with diverse tumor characteristics that complicate treatment. The integration of spatial patterns from prostate-specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT), pathology and expanding genomic data represents a groundbreaking advancement in histo-imaging genomics. The aim of this study was to elucidate the internetwork mapping between genetic biomarkers and PSMA PET/CT imaging in prostate cancer patients.

Results

mRNA sequencing and clinical data from 433 prostate cancer patients were retrieved from The Cancer Genome Atlas (TCGA) database. Differential gene expression between the Gleason score (GS) > 7 and GS ≤ 7 groups was analyzed. Feature selection was performed following the univariate and multivariate logistic regression analyses. A GS predictive model was developed using multivariate logistic regression. Additionally, local samples and images from 27 patients were collected. PSMA PET/CT imaging was performed before radical prostatectomy, and mRNA sequencing of prostate cancer lesions was conducted using next-generation sequencing. Differentially expressed genes identified from the TCGA dataset were subsequently analyzed for correlations with PET-related metrics in the local dataset by utilizing Pearson correlation analysis.Out of the TCGA dataset, 174 genes exhibited differential expression. After feature selection, 53 genes remained. In the local dataset, ten genes (EFNA2, CACNA1I, CA1, MYBPC3, CYP1A1, TLCD3B, LRTM2, GBX2, SPSB4, and GDF3) demonstrated significant associations with PET-related metrics. When comparing the differential expression of genes between the GS>7 and GS≤7 groups, six genes (STMN2, CYP1A1, THRSP, LIPC, GBX2, and SPSB4) in the GS>7 group and eight genes (FBXL16, KLK14, DIRAS2, TERB2, PRAME, UTS2B, UGT2B15, and LINC02798) in the GS≤7 group were significantly correlated with PET-related parameters.

Conclusions

This study identified genetic markers significantly correlated with PSMA PET/CT imaging features in prostate cancer patients. These findings may provide a valuable foundation for optimizing prostate cancer diagnostic procedures and tailoring therapeutic approaches based on genetic and imaging biomarkers.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13550-026-01471-3.

Keywords: Prostate cancer, PSMA PET/CT, Radiogenomics, mRNA, Gleason score

Introduction

Prostate cancer remains a leading cause of male cancer mortality worldwide, with over 1.4 million new cases annually [1, 2]. Due to the lack of obvious early symptoms [3], current diagnostic methods, which combine clinical assessment, imaging tests, and invasive biopsies, carry risks of complications and may overlook key tumour characteristics due to sampling limitations. Therefore, routine screening using biomarkers and imaging tests is crucial for early detection and assessment.

Currently, the recognition and management of prostate cancer mainly rely on clinical, histological, and imaging parameters [4, 5], but the emergence of next-generation sequencing has revolutionized our understanding of its molecular complexity [6–8]. It is now evident that prostate cancer is not a singular entity, but rather a conglomerate of various subtypes, each with distinct genetic signatures and potential therapeutic vulnerabilities [9]. Genomic biomarker and pathway analyses have been conducted. The results not only aid in identifying potential therapeutic targets but also offer insights into prognosis, therapeutic responsiveness, and disease monitoring [10]. Numerous genomic classifiers are available for patients with localized prostate cancer to assess the risks of distant metastasis or prostate cancer-induced mortality. Analyses of gene expression from prostate cancer biopsies or prostatectomy specimens can predict metastatic risks and disease-specific fatalities [11].

In recent decades, there have been transformative advancements in positron emission tomography/computed tomography (PET/CT) for diagnosing prostate cancer. Among these advancements, prostate-specific membrane antigen (PSMA) has become a pivotal diagnostic tool [12, 13]. Clinically, high PSMA expression is associated with aggressive prostate cancer and advanced disease, and higher tumor uptake often predicts poorer survival outcomes, highlighting the close association between PSMA as an imaging biomarker and tumor biology [14–16]. The integration of imaging and pathology offers a unique “window” into the tumor microenvironment, potentially bridging the gap between gross anatomical changes and underlying molecular events [17]. This integrated approach serves dual purposes. First, it identifies regions of interest that may better represent the overall genetic profile of the tumor, given the heterogeneity often observed within a single tumor mass. Guiding biopsies to these areas enables a more accurate genetic signature of the tumor. Second, when genomic data are integrated with imaging and histological findings, it becomes possible to “visualize” the phenotypic manifestations of underlying genomics, offering an unparalleled understanding of how genetic alterations drive disease progression and behavior. The fusion of medical imaging and genomics, termed imaging genomics, is burgeoning [18]. This technique enables exploration of the nexus between the imaging attributes of lesions and their genomic/molecular aspects, providing extensive disease insight [9]. This development has amplified the appeal of both radiomics and radiogenomics in prostate cancer research.

Recent studies have shown that GS > 7 prostate tumors exhibit significantly higher PSMA-TV and TL-PSMA than GS ≤ 7 groups, suggesting distinct imaging-genomic associations across tumor grades [19, 20]. This study aims to explore the relationship between mRNA expression profiles and PSMA imaging parameters, providing a novel imaging-genomic framework for prostate cancer risk stratification and personalized treatment. By aligning imaging-genomic biomarkers with prostate cancer outcomes, we endeavored to elucidate the link between invasive gene activity and noninvasive imaging properties. With the goal of achieving a comprehensive understanding of prostate cancer genomics, we also aimed to refine genomics-imaging classifications, paving the way for personalized treatments optimized for individual patients, thereby enhancing patient prognoses.

Methods

Patients and datasets

Transcriptome data, along with clinical and pathological information of 1000 prostate tumor samples, were retrieved from The Cancer Genome Atlas (TCGA) database. RNA sequencing data were normalized to fragments per kilobase of transcriptome per million reads. After filtering out genes not expressed in more than 80% of the samples, a gene expression matrix consisting of 19,297 genes and 443 samples was obtained (Fig. 1a).

Fig. 1.

Fig. 1

Open in a new tab

Flowchart of this study. (a) TCGA-PRAD genomics dataset. (b) Local radiogenomics dataset

To explore the relationship between mRNA expression and imaging parameters, 27 prostate cancer patients from Qilu Hospital of Shandong University were prospectively enrolled, and their data were used as the local dataset for this study. Written informed consent was obtained from all patients. All 27 patients underwent 18F-DCFPyL-PET/CT imaging, followed by radical prostatectomy. Finally, 29 tumor lesions were obtained for mRNA sequencing and image analysis (Fig. 1b).

PSMA PET/CT acquisition

All the patients underwent 18F-DCFPyL PET/CT imaging without fasting. All PET/CT examinations were performed on a whole-body PET/CT scanner (General Electric, Discovery STE16). 18F-DCFPyL, the PSMA-targeting tracer produced by the same PET/CT center, was intravenously injected into patients at a standardized dose of 3.7 MBq/kg. After 18F-DCFPyL administration, the patients rested in a quiet room for about 60 min before scanning. The image acquisition ranged from parietal to femoral.

Sixteen row helical CT scan was used with the following conditions: tube voltage (120 kV in the body and 160 kV in the brain), tube current (110 mA in the body and 260 mA in the brain), 3.25 mm thick. PET collection time of every bed was three minutes, the whole body scanning needed six to eight beds. Using the viewpoint method, the scan data was reconstructed into CT, PET and PET/CT fusion images. PET matrix was 128 × 128, and CT matrix was 512 × 512. The voxel size was 3.75 × 3.75 × 3.75 mm3 .

Region of interest (ROI) segmentation and feature extraction

Image interpretation and region of interest (ROI) segmentation were performed by two board-certified radiologists with specialized expertise in prostate cancer imaging, each possessing over 10 years of experience in PSMA PET/CT interpretation (averaging > 100 prostate cancer cases annually). All imaging procedures adhered to the European Association of Nuclear Medicine (EANM) guidelines for PSMA PET/CT in prostate cancer, ensuring standardization and reproducibility [21]. Using Evomics software (version 1.0), they manually delineated tumor volumes on PET images with reference to corresponding CT anatomical data. For each ROI file, the maximum standard uptake value (SUVmax), mean SUV (SUVmean), peak SUV (SUVpeak), lean body mass-corrected SULpeak, PSMA-derived tumor volume (PSMA-TV), and total lesion PSMA (TL-PSMA) were calculated. Among these, PSMA-TV and TL-PSMA are the primary analytical indicators, as they directly reflect the PSMA metabolic load and spatial distribution of tumors and are more strongly associated with tumor invasiveness and genomic characteristics [22–24]. Other parameters are used as secondary indicators for supplementary verification.

Differential expression analysis

Patients from the TCGA database were categorized into two groups according to their Gleason score. A Gleason score of 7 or lower was classified as the GS ≤ 7 group, while a Gleason score greater than 7 was classified as the GS > 7 group. Differential expression of mRNAs between the GS ≤ 7 and GS > 7 groups was analyzed using the R (v4.1.1) package edgeR (v3.34.1). Genes with a |log2 fold change (FC)| of at least 1.5 and a false discovery rate (FDR) cutoff of 0.05 were classified as differentially expressed genes (DEGs). A volcano map of DEGs was generated using the R package ggplot2 (v3.3.5).

Feature selection and multivariate logistic regression analysis

Univariate logistic regression was used to identify genes significantly associated with Gleason score grade (P < 0.05), followed by multivariate logistic regression to construct a model using the TCGA dataset. The model was validated with local data. A gene expression heatmap of genes significantly associated with Gleason score was generated using the R package ComplexHetmap (v2.8.0).

Correlation analysis

The correlations between the expression signatures of the differentially expressed mRNAs and the conventional imaging parameters (SUVmean, SUVmax, SUVpeak, SULpeak, PSMA-TV, and TL-PSMA) calculated from PET images were assessed using the Pearson correlation method. The significance of the correlation was evaluated by the function corr.test in R package psych (v2.2.9) with a cutoff value of 0.05. (v4.0.5). The pipeline of this study is shown in Fig. 2.

Fig. 2.

Fig. 2

Open in a new tab

Pipeline of this study

Results

Baseline clinical characteristics of the study patients

Patient data were collected from Qilu Hospital. The mean age of the 27 patients was 66 ± 5.94 years. Among these patients, 21 patients (23 lesions) had a Gleason score of 7, while 6 patients (6 lesions) had a Gleason score greater than 7. For the 21 patients with Gleason score 7, 11 patients (12 lesions) were classified as Gleason 3 + 4, and 10 patients (11 lesions) were Gleason 4 + 3. Detailed data are presented in Table 1.

Table 1.

Patient characteristics

Characteristics Patients (n = 27)
Age (mean ± SD, years) 66 ± 5.94
PSA (mean ± SD, ng/mL) 13.27 ± 7.18
Gleason score
6 0
7

3 + 4:11 (12 lesions)

4 + 3:10 (11 lesions)
8 3
9 3
10 0
Clinical TNM staging
T2a 13
T2b 3
T2c 11
Pathologic TNM staging
T2 24
T3a 2
T3b 1
Open in a new tab

SD, standard deviation; PSA, prostate specific antigen; TNM, tumor node metastasis

Differential gene expression analysis and predictive model construction for Gleason grade

Differential expression analysis between the GS > 7 group and GS ≤ 7 group was performed in the TCGA. Overall, 174 genes were found to be differentially expressed, including 123 upregulated genes and 51 downregulated genes (expression FC > 1.5 and P < 0.05) (Fig. 3). Then, feature selection was conducted using univariate logistic regression, and 53 genes remained. The remaining genes were used to construct a model to distinguish the Gleason grade by multivariate logistic regression, and the area under the curve (AUC) of this model was 0.8249.

Fig. 3.

Fig. 3

Open in a new tab

DEGs between the GS > 7 and GS ≤ 7 groups in the TCGA dataset. (a) Heatmap of DEGs screened by univariate logistic regression between the TCGA GS > 7 group and GS ≤ 7 group. Z scores indicate standardized expression levels. Darker blue indicates lower expression levels, while darker red indicates higher expression levels. (b) Volcano plot of DEGs screened by univariate logistic regression between the GS > 7 group and GS ≤ 7 group.DEGs, differentially expressed genes; TCGA, The Cancer Genome Atlas; FDR, false discovery rate; FC, fold change

Correlation between gene expression and PSMA PET/CT imaging parameters

Although TCGA provides large-scale gene expression profiles, it lacks PSMA PET/CT imaging data. Therefore, we also independently performed Pearson’s correlation analysis on the local dataset to explore the relationships between gene expression and PET-related parameters. Univariate logistic regression analysis of the 53 remaining genes identified ten genes (EFNA2, CACNA1I, CA1, MYBPC3, CYP1A1, TLCD3B, LRTM2, GBX2, SPSB4, and GDF3) as being associated with PET-related parameters. Comparison of the DEGs between the GS>7 and GS ≤ 7 groups revealed six genes (STMN2, CYP1A1, THRSP, LIPC, GBX2, and SPSB4) associated with PET-related parameters in the GS > 7 group, while eight genes (FBXL16, KLK14, DIRAS2, TERB2, PRAME, UTS2B, UGT2B15, and LINC02798) were associated with PET-related parameters in the GS ≤ 7 group (Fig. 4). Notably, CYP1A1, GBX2, and SPSB4 were found to be associated with PET-related parameters in both the overall patient groups and the GS > 7 group. These three genes in both groups were significantly related to TL-PSMA (P < 0.05). CYP1A1 and GBX2 in both groups were significantly related to PSMA-TV (P < 0.01) (Fig. 4).

Fig. 4.

Fig. 4

Open in a new tab

Correlations between DEGs and PET-related parameters. (a) All patients. (b) GS ≤ 7 group. (c) GS > 7 group. *P < 0.05; **P < 0.01; ***P < 0.001; P value represents the significance of the correlation. The color of the circle represents the correlation coefficient.DEGs, differentially expressed genes; PET, positron emission tomography

Protein-protein interaction network (PPI) analysis of PET-related genes and validation in the local dataset

To further explore the mechanism of DEGs in the pathogenesis of prostate cancer and explore the interactions between DEGs, a PPI network was established using genes related to PET-related parameters in all samples, as well as genes related to PET-related parameters in the GS > 7 group and GS ≤ 7 group. The PPI network was constructed based on a STRING database for functional association analysis. The proteins expressed by GBX2, GDF3, and EFNA2 jointly contributed to a shared function, the proteins expressed by FBXL16 and DIRAS2 jointly contributed to a shared function, and the proteins expressed by CYP1A1, LIPC, and THRSP jointly contributed to a shared function (Fig. 5).

Fig. 5.

Fig. 5

Open in a new tab

PPI network results of the genes related to PET-related parameters. (a) All patients. (b) GS ≤ 7 group. (c) GS > 7 group.PPI, protein‒protein interaction; PET, positron emission tomography

We also attempted to analyze the DEGs in the GS > 7 and GS ≤ 7 groups of Gleason score patients in our own local dataset, as well as their correlation with PET-related parameters. The cases representing the characteristics of this relationship are shown in Fig. 6. The results showed that 191 DEGs were associated with the image parameters (Supplemental Table 1).

Fig. 6.

Fig. 6

Open in a new tab

Case representation of the correlation between gene expression and PET-related parameters in the GS > 7 group. Upper case: positive correlation case. Lower case: negative correlation case.PET, positron emission tomography

Discussion

The advent of high-throughput arrays and next-generation sequencing technologies has ushered in a new era for genomic analysis, making it more accessible and widespread [25]. Prostate cancer, known for its intricate genetic heterogeneity, presents challenges in identifying universal genomic markers. Identifying these markers is merely a preliminary step, while establishing their clinical relevance requires rigorous trials and validation. Historically, prognosis for prostate cancer has relied on the Gleason grading system, which is based on microscopic cellular attributes from biopsy specimens [26]. However, the past decade has witnessed a paradigm shift, marked by the increased use of PET/CT in prostate cancer diagnostics [12]. Within this landscape, PSMA has become an essential diagnostic tool. Clinically, increased PSMA affinity has been linked not only to aggressive prostate cancer phenotypes but also to advanced disease progression [27]. An inverse correlation between tumoral PSMA uptake and survival outcomes further solidifies its role as a robust imaging biomarker connected to tumoral pathobiology [17]. In our cross-disciplinary field of imaging genomics, the synergy between PSMA PET/CT imaging and genomics provides profound insights, enabling researchers to explore the intricate connections between lesion imaging attributes and genomic/molecular markers. Our findings highlight the potential of integrating PSMA PET/CT imaging with genomic analysis, as evidenced by the significant correlations between mRNA expression and imaging parameters of TL-PSMA and PSMA-TV. This understanding not only deepens our understanding of the disease but also paves the way for identifying imaging biomarkers for genomic forecasting, enabling noninvasive diagnostic methods, prognostic evaluation, and treatment efficacy assessment for prostate cancer.

Our analytic methodology identified ten genes associated with PET-derived imaging metrics in prostate cancer. Among these, TLCD3B and SPSB4 were newly linked to prostate carcinoma in our dataset, despite their known roles in other malignancies. TLCD3B has been reported as an independent prognostic biomarker in diffuse large B-cell lymphoma [28], while SPSB4 mediates EphB2 ubiquitination and acts as a prognostic marker across multiple cancers, including colorectal cancer [11, 18]. Our study newly established their association with prostate cancer, although their precise functional mechanisms remain to be experimentally verified.

Several genes with known prostate cancer relevance were also validated. CA1, which is overexpressed in prostate cancer versus benign prostatic hypertrophy [29], was upregulated in high-risk patients and positively correlated with SUVmean, SUVpeak and SULpeak in our study, supporting its potential as a diagnostic and therapeutic target linked to PSMA-related metrics. CYP1A1, a cytochrome P450 family member associated with prostate cancer susceptibility [30–32], was highly expressed in high-risk patients and significantly correlated with PSMA-TV and TL-PSMA, reinforcing its role in prostate cancer progression and PSMA-associated pathways.

GBX2 was identified as a key gene for prostate cancer risk stratification, with strong correlations to PSMA-TV and TL-PSMA. GBX2 is involved in transcriptional regulation and nervous system development, and its methylation serves as a prognostic marker for prostate cancer recurrence [33–37]. Our findings further consolidate its prognostic value in prostate cancer. Additional candidate radiogenomic markers included EFNA2, CACNA1I and GDF3. EFNA2 is upregulated in prostate adenocarcinoma [38]. CACNA1I-encoded Cav3.3 promotes prostate cancer cell proliferation via the NF-κB pathway [39] and GDF3 induces epigenetic alterations linked to prostate malignancy [40]. The imaging parameters of GS > 7 were closely associated with CYP1A1, GBX2 and SPSB4, highlighting their clinical significance in clinically significant prostate cancer and potential as therapeutic targets.

While our study identified significant correlations between specific genes and PSMA PET parameters, the underlying biological mechanisms warrant further discussion. The association between CYP1A1 and volumetric PET parameters (PSMA-TV and TL-PSMA) may be explained through the androgen receptor (AR) signaling pathway. CYP1A1, a member of the cytochrome P450 superfamily, participates in steroid hormone metabolism, including androgen catabolism. Given that PSMA expression is regulated by AR signaling through a transcription complex containing AR, FOXA1, and HOXB13 that interacts with the FOLH1 gene promoter [41], alterations in androgen metabolism mediated by CYP1A1 could potentially influence PSMA expression levels and, consequently, PSMA tracer uptake on PET imaging. This hypothesis is supported by our observation that CYP1A1 was highly expressed in GS > 7 patients who demonstrated elevated PSMA-TV and TL-PSMA values.

Similarly, GBX2, a transcription factor involved in developmental regulation, may influence PSMA expression through epigenetic mechanisms. The FOLH1 gene (encoding PSMA) contains differentially methylated regions, where methylation status correlates with PSMA expression levels. GBX2 methylation has been established as a prognostic biomarker for prostate cancer recurrence, and its strong correlation with PSMA-TV and TL-PSMA may reflect its involvement in regulating tumor aggressiveness and potentially PSMA-associated pathways.

For CA1, the correlation with SUV parameters (SUVmean, SUVpeak, and SULpeak) may be attributed to its role in regulating tumor microenvironment pH. Carbonic anhydrases facilitate pH homeostasis in tumor cells, and altered pH can affect the cellular uptake of PSMA-targeted tracers, thereby influencing SUV measurements. Additionally, metabolic reprogramming associated with CA1 overexpression may indirectly influence PSMA expression or tracer uptake efficiency.

SPSB4, which mediates ubiquitination of cell surface receptors, may influence PET parameters through regulation of membrane protein stability and cell signaling pathways affecting tumor cell proliferation and invasion.

It is important to note that these mechanistic explanations remain hypothetical and require validation through functional studies, including in vitro experiments examining the direct effects of these genes on PSMA expression, analysis of AR signaling activity in relation to CYP1A1 expression, and correlation studies with tumor microenvironment markers [42].

The present study establishes a tangible link between genomic biomarkers and PSMA PET/CT imaging features in prostate cancer, which carries important clinical and translational implications. From a clinical perspective, the identified gene panel (including EFNA2, CACNA1I, CA1, CYP1A1, GBX2, SPSB4, and others) correlated with key PET parameters provides a novel radiogenomic signature that can complement both imaging and pathological evaluation. These biomarkers not only help improve the non-invasive risk stratification of prostate cancer before surgery but also offer potential targets for interpreting the biological heterogeneity underlying PSMA PET/CT signal differences [43, 44]. For GS > 7 patients, the combination of gene expression levels and quantitative PET metrics may enable more precise identification of aggressive disease phenotypes.

An important question arising from our study is whether genomic profiling provides additive value for clinical decision-making beyond PSMA PET quantitative parameters. Our findings suggest that while PSMA PET parameters correlate with specific gene expression patterns related to tumor proliferation and metabolism, they do not capture the full spectrum of clinically actionable genomic alterations. Specifically, DNA repair gene mutations (BRCA1/2, ATM), mismatch repair deficiency, and AR signaling alterations—which have direct therapeutic implications for PARP inhibitors, immunotherapy, and AR-targeted therapies—cannot be inferred from PSMA PET parameters alone. Therefore, PSMA PET and genomic profiling serve complementary roles: PSMA PET provides valuable information about disease extent and PSMA expression, while genomic profiling enables precision medicine approaches. Future studies should focus on developing integrated models that combine imaging and genomic data to optimize patient stratification and treatment selection.

In terms of future research, this study provides a foundational framework for the integrative analysis of histopathology, imaging, and genomics in prostate cancer. Subsequent investigations should focus on the mechanistic validation of the regulatory roles of these key genes in PSMA expression, tumor metabolism, and progression pathways using in vitro and in vivo models. Prospective multi-center studies with larger cohorts are also warranted to verify the diagnostic and prognostic performance of this radiogenomic panel in real-world clinical settings. Ultimately, we anticipate that such imaging–genomic integration will facilitate the development of non-invasive, precision-based strategies for prostate cancer, from early diagnosis and risk assessment to targeted treatment and monitoring of therapeutic response.

Despite the promising and innovative nature of this approach, there are several limitations. One of the primary challenges is the inherent heterogeneity of prostate cancer, which results in a multitude of genetic variations even within the same tumor. This diversity complicates the establishment of a consistent radiogenomic profile across all patients. Additionally, our findings are primarily correlative, meaning we identified relationships between imaging parameters and genomics without pinpointing causal mechanisms. Future studies must investigate the mechanistic underpinnings of these correlations to understand how these genes influence the phenotypic presentation and progression of prostate cancer. Another limitation is the sample size. While we utilized state-of-the-art techniques, larger cohorts in future research will help to validate our findings, especially given the variability within prostate cancer. Furthermore, while we identified links between several genes and prostate cancer, the exact functional roles of these genes in disease pathogenesis remain unclear. More comprehensive molecular biology experiments are needed to determine the specific roles of these genes in prostate cancer initiation, progression, and metastasis. Additionally, this study did not include MRI PI-RADS scores or PSMA PRIMARY scores, which are clinically important parameters for prostate cancer assessment. Future studies incorporating these scoring systems would provide a more comprehensive evaluation of the relationship between imaging characteristics and genomic profiles.

Conclusion

This study provides novel insights into the imaging-pathology genomics of prostate cancer. By combining PSMA PET/CT, pathology, and genomic analysis, we have not only pinpointed potential imaging biomarkers for genomic forecasting but also identified several genes that could serve as potential therapeutic targets in prostate cancer management. Although our findings are promising, further investigations are needed to clarify the specific roles of these genes play and validate these markers in larger clinical cohorts. Continued research in this field has the potential to advance diagnostic and therapeutic strategies for patients battling prostate cancer.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1. (66.8KB, docx)

Acknowledgements

The authors wish to acknowledge all the staff of the department of Nuclear Medicine, Qilu Hospital of Shandong University, Jinan, China.

Abbreviations

PET/CT

Positron emission tomography/computed tomography

PSMA

Prostate-specific membrane antigen

FDR

False discovery rate

PPI

Protein protein interaction

IPI

International prognostic index

DLBCL

Diffuse large B-cell lymphoma

IPI-IPM

IPI-based immune prognostic model

PCa

Prostate cancer

BPH

Benign prostatic hypertrophy

BCR

Biochemical recurrence

PRAD

Prostate adenocarcinoma

CSPCa

Clinically significant prostate adenocarcinoma

Author contributions

Conception and design: Xiang Li. Acquisition of data: Xin Li. Analysis and interpretation of data: M H. Drafting of the manuscript: R L, K L. Critical revision of the manuscript for important intellectual content: S W, L Y. Statistical analysis: Y L, L Q. Administrative, technical, or material support: H Z, M W. Supervision: X J, Y X. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Qilu Hospital, Shandong University (Ethics number: KYLL-2017-573).

Informed consent

was obtained from all individual participants included in the study.

Consent for publication

The authors affirm that human research participants provided informed consent for publication of the images in Figures.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Xin Li, Email: lixin16@sdu.edu.cn.

Xiang Li, Email: xiang.li@meduniwien.ac.at.

References

  • 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49. [DOI] [PubMed] [Google Scholar]
  • 2.Gandaglia G, Leni R, Bray F, Fleshner N, Freedland SJ, Kibel A, et al. Epidemiology and prevention of prostate cancer. Eur Urol Oncol. 2021;4(6):877–92. [DOI] [PubMed] [Google Scholar]
  • 3.Dahiya V, Hans S, Kumari R, Bagchi G. Prostate cancer biomarkers: from early diagnosis to precision treatment. Clin Transl Oncol. 2024;26(10):2444–56. [DOI] [PubMed] [Google Scholar]
  • 4.Šamija I, Fröbe A. GENOMICS OF PROSTATE CANCER: CLINICAL UTILITY AND CHALLENGES. Acta Clin Croat. 2022;61(Suppl 3):86. [DOI] [PMC free article] [PubMed]
  • 5.Chandrasekar T, Yang JC, Gao AC, Evans CP. Mechanisms of resistance in castration-resistant prostate cancer (CRPC). Transl Androl Urol. 2015;4(3):365–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.McGranahan R, Herrero N, Taylor J, Swanton BS. DeconstructSigs: delineating mutational processes in single tumors distinguishes DNA repair deficiencies and patterns of carcinoma evolution. Genome Biol. 2016;17:31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Berger MF, Mardis ER. The emerging clinical relevance of genomics in cancer medicine. Nat Rev Clin Oncol. 2018;15(6):353–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Kohan A, Kulanthaivelu R, Hinzpeter R, et al. Disparity and Diversity in NSCLC Imaging and Genomics: Evaluation of a Mature, Multicenter Database. Cancers (Basel). 2023;15(7):2096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Haffner MC, Zwart W, Roudier MP, True LD, Nelson WG, Epstein JI, et al. Genomic and phenotypic heterogeneity in prostate cancer. Nat Rev Urol. 2020;18(2):79–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chakraborty G, Nandakumar S, Hirani R, Nguyen B, Stopsack KH, Kreitzer C, et al. The Impact of PIK3R1 mutations and insulin–PI3K–glycolytic pathway regulation in prostate cancer. Clin Cancer Res. 2022;28(16):3603–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Okumura F, Joo-Okumura A, Obara K, Petersen A, Nishikimi A, Fukui Y, et al. Ubiquitin ligase SPSB4 diminishes cell repulsive responses mediated by EphB2. Mol Biol Cell. 2017;28(24):3532–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Chen Y, Foss CA, Byun Y, Nimmagadda S, Pullambhatla M, Fox JJ, et al. Radiohalogenated prostate-specific membrane antigen (PSMA)-based ureas as imaging agents for prostate cancer. J Med Chem. 2008;51(24):7933–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Combes AD, Palma CA, Calopedos R, et al. PSMA PET-CT in the Diagnosis and Staging of Prostate Cancer. Diagnostics (Basel). 2022;12(11):2594. [DOI] [PMC free article] [PubMed]
  • 14. Cao G, Li Y, Wang J, et al. Gleason score, surgical and distant metastasis are associated with cancer-specific survival and overall survival in middle aged high-risk prostate cancer: A population-based study. Front Public Health. 2022;10:1028905. [DOI] [PMC free article] [PubMed]
  • 15.Swanson GP, Trevathan S, Hammonds KAP, Speights VO, Hermans MR. Gleason score evolution and the effect on prostate cancer outcomes. Am J Clin Pathol. 2021;155(5):711–7. [DOI] [PubMed] [Google Scholar]
  • 16.Chen DC, Huang S, Papa N, et al. Impact of intraprostatic PSMA maximum standardised uptake value following prostatectomy: a systematic review and meta-analysis. BJU Int. 2025;135(5):720-32. [DOI] [PubMed]
  • 17.Roberts MJ, Maurer T, Perera M, Eiber M, Hope TA, Ost P, et al. Using PSMA imaging for prognostication in localized and advanced prostate cancer. Nat Rev Urol. 2022;20(1):23–47. [DOI] [PubMed] [Google Scholar]
  • 18.Garcia-Gallastegi D, Crende P. GeromiRs Are Downregulated in the Tumor Microenvironment during Colon Cancer Colonization of the Liver in a Murine Metastasis Model. Int J Mol Sci. 2021;22(9):4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Niu XB, Li YP, Wang J, Mei XL, Zhao XY, Liu TT, et al. Diagnostic value of 18F-PSMA-1007 PET/CT for predicting the pathological grade of prostate cancer. Cancer Biol Ther. 2024;25(1):2287120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Solari EL, Gafita A, Schachoff S, Bogdanović B, Villagrán Asiares A, Amiel T, et al. The added value of PSMA PET/MR radiomics for prostate cancer staging. Eur J Nucl Med Mol Imaging. 2022;49(2):527–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Fendler WP, Eiber M, Beheshti M, Bomanji J, Calais J, Ceci F, et al. PSMA PET/CT: joint EANM procedure guideline/SNMMI procedure standard for prostate cancer imaging 2.0. Eur J Nucl Med Mol Imaging. 2023;50(5):1466–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Schmuck S, Von Klot CA, Henkenberens C, Sohns JM, Christiansen H, Wester HJ, et al. Initial experience with volumetric 68Ga-PSMA I&T PET/CT for assessment of whole-body tumor burden as a quantitative imaging biomarker in patients with prostate cancer. J Nucl Med. 2017;58(12):1961–7. [DOI] [PubMed] [Google Scholar]
  • 23.Acar E, Özdoğan Ö, Aksu A, Derebek E, Bekiş R, Çapa Kaya G. The use of molecular volumetric parameters for the evaluation of Lu-177 PSMA I&T therapy response and survival. Ann Nucl Med. 2019;33(9):681-88. [DOI] [PubMed]
  • 24.Pozdnyakov A, Kulanthaivelu R, Jhaveri K, Berlin A, van der Kwast T, Veit-Haibach P. Histology and PSMA expression on immunohistochemistry in GS > 7 prostate cancer patients: comparison with 68Ga-PSMA PET/CT features in primary staging. Eur J Nucl Med Mol Imaging. 2024;51(4):1025–35. [Google Scholar]
  • 25. Kustrimovic N, Bombelli R, Baci D, Mortara L. Microbiome and Prostate Cancer: A Novel Target for Prevention and Treatment. Int J Mol Sci. 2023;24(2):1511. [DOI] [PMC free article] [PubMed]
  • 26.Leenders G, Verhoef EI, Hollemans E. Prostate cancer growth patterns beyond the Gleason score: entering a new era of comprehensive tumour grading. Histopathology. 2020;77(6):850–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Fanti S, Goffin K, Hadaschik BA, Herrmann K, Maurer T, MacLennan S, et al. Consensus statements on PSMA PET/CT response assessment criteria in prostate cancer. Eur J Nucl Med Mol Imaging. 2021;48(2):469–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Mu S, Shi D, Ai L, et al. International Prognostic Index-Based Immune Prognostic Model for Diffuse Large B-Cell Lymphoma. Front Immunol. 2021;12:732006. [DOI] [PMC free article] [PubMed]
  • 29.Kumar-Sinha C, Jokerst JV, Chen Z, Xu L, Nolley R, Chang E, et al. A magnetic bead-based sensor for the quantification of multiple prostate cancer biomarkers. PLoS ONE. 2015;10(9):e0139484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Mokkawes T, de Visser SP. Melatonin Activation by Cytochrome P450 Isozymes: How Does CYP1A2 Compare to CYP1A1?. Int J Mol Sci. 2023;24(4):3651. [DOI] [PMC free article] [PubMed]
  • 31.Shaik AP, Jamil K, Das P. CYP1A1 polymorphisms and risk of prostate cancer: a meta-analysis. Urol J. 2009;6(2):78–86. [PubMed] [Google Scholar]
  • 32.Li H, Xiao D, Hu L, He T. Association of CYP1A1 polymorphisms with prostate cancer risk: an updated meta-analysis. Mol Biol Rep. 2012;39(12):10273–84. [DOI] [PubMed] [Google Scholar]
  • 33.Kerstein PC, Leffler J, Sivyer B, Taylor WR, Wright KM. Gbx2 identifies two amacrine cell subtypes with distinct molecular, morphological, and physiological properties. Cell Rep. 2020;33(7):108382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Zimmer TS, Korotkov A, Zwakenberg S, Jansen FE, Zwartkruis FJT, Rensing NR, et al. Upregulation of the pathogenic transcription factor SPI1/PU.1 in tuberous sclerosis complex and focal cortical dysplasia by oxidative stress. Brain Pathol. 2021;31(5):e12949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Li X, Zhang F, Ma J, Ruan X, Liu X, Zheng J, et al. NCBP3/SNHG6 inhibits GBX2 transcription in a histone modification manner to facilitate the malignant biological behaviour of glioma cells. RNA Biol. 2021;18(1):47–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Jeyapala R, Savio AJ, Olkhov-Mitsel E, Kamdar S, Zhao F, Cuizon C, et al. GBX2 Methylation Is a Novel Prognostic Biomarker and Improves Prediction of Biochemical Recurrence Among Patients with Prostate Cancer Negative for Intraductal Carcinoma and Cribriform Architecture. Eur Urol Oncol. 2019;2(3):231–8. [DOI] [PubMed] [Google Scholar]
  • 37.Tolkach Y, Merseburger A, Herrmann T, Kuczyk M, Serth J, Imkamp F. Signatures of adverse pathological features, androgen insensitivity and metastatic potential in prostate cancer. anticancer res. 2015;35(10):5443–51. [PubMed] [Google Scholar]
  • 38.Huang S, Dong C, Zhang J, Fu S, Lv Y, Wu J. A comprehensive prognostic and immunological analysis of ephrin family genes in hepatocellular carcinoma. Front Mol Biosci. 2022;9:943384. [DOI] [PMC free article] [PubMed]
  • 39.Basson MD, Zeng B, Downey C, Sirivelu MP, Tepe JJ. Increased extracellular pressure stimulates tumor proliferation by a mechanosensitive calcium channel and PK-β. Mol Oncol. 2014;9(2):513–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Naik A, Dalpatraj N, Thakur N. Global Histone H3 Lysine 4 Trimethylation (H3K4me3) Landscape Changes in Response to TGFβ. Epigenet Insights. 2021;14:25168657211051755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Bakht MK, Yamada Y, Ku SY, Venkadakrishnan VB, Korsen JA, Kalidindi TM, et al. Landscape of prostate-specific membrane antigen heterogeneity and regulation in AR-positive and AR-negative metastatic prostate cancer. Nat Cancer. 2023;4(5):699–715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Bakht MK, Beltran H. Biological determinants of PSMA expression, regulation and heterogeneity in prostate cancer. Nat Rev Urol. 2025;22(1):26-45. [DOI] [PMC free article] [PubMed]
  • 43.Min K, Lin Q, Qiu D. Precision medicine in prostate cancer: individualized treatment through radiomics, genomics, and biomarkers. Cancer Imaging. 2025;25(1):116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Al Saffar H, Chen DC, Delgado C, et al. The Current Landscape of Prostate-Specific Membrane Antigen (PSMA) Imaging Biomarkers for Aggressive Prostate Cancer. Cancers (Basel). 2024;16(5):939. [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1. (66.8KB, docx)

Data Availability Statement

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Articles from EJNMMI Research are provided here courtesy of Springer-Verlag

RESOURCES