# Cuproptosis-related gene PROK1 predicts the diagnosis and prognosis of prostate cancer based on multiple machine learning

**Authors:** Xin Qin, Qinghua Wang, Wei Jiang, Yan Zhao, Haopeng Li, Tong Zi, Yaru Zhu, Xilei Li, Chengdang Xu, Tao Yang, Xinan Wang, Yicong Yao, Xi Chen, Juan Zhou, Gang Wu

PMC · DOI: 10.7150/jca.113505 · Journal of Cancer · 2026-01-01

## TL;DR

This study identifies PROK1 as a key cuproptosis-related gene that helps predict prostate cancer diagnosis and prognosis using machine learning models.

## Contribution

The study introduces PROK1 as a novel cuproptosis-related gene for PCa diagnosis and prognosis using machine learning.

## Key findings

- PROK1 expression significantly affects prostate cancer cell proliferation and invasion.
- Machine learning models based on cuproptosis-related genes show strong diagnostic and prognostic performance.
- Two clusters with distinct clinical and immune features were identified using cuproptosis gene expression.

## Abstract

Cuproptosis, a newly identified form of cell death, influences the development, progression, and prognosis of prostate cancer (PCa). Identifying key genes associated with cuproptosis and developing robust predictive models through machine learning approaches are crucial for personalized PCa treatment. In our study, multiple machine learning methods and their combinations were employed for the construction of diagnostic and prognostic models for PCa, which were then validated in multiple external independent cohorts. The model key gene, PROK1, was selected for further analysis, and its expression was compared in clinical samples and cell lines. Additionally, the anticancer effect of PROK1 was explored through regulating the expression of PROK1. Most cuproptosis-related genes (CRGs) showed differential expression between PCa and normal prostate tissues. The two clusters derived from the Consensus Clustering method, based on cuproptosis gene expression characteristics, exhibit distinct clinical features and immune microenvironment infiltration patterns. Models constructed based on machine learning methods showed promising diagnostic capabilities for PCa and were associated with the prediction of biochemical recurrence-free survival and disease-free survival of patients. Inhibiting PROK1 expression promoted PCa cell proliferation and invasion, while its overexpression had the opposite effect. Furthermore, pathway exploration revealed that PROK1 inhibits tumor growth by mediating apoptosis under copper ion stress. Its association with cuproptosis warrants further investigation to elucidate the precise mechanism.

## Linked entities

- **Genes:** PROK1 (prokineticin 1) [NCBI Gene 84432]
- **Diseases:** prostate cancer (MONDO:0005159)

## Full-text entities

- **Genes:** PROK1 (prokineticin 1) [NCBI Gene 84432] {aka EGVEGF, PK1, PRK1}
- **Diseases:** tumor (MESH:D009369), PCa (MESH:D011471)
- **Chemicals:** copper (MESH:D003300)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12825426/full.md

## Figures

14 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12825426/full.md

## References

53 references — full list in the complete paper: https://tomesphere.com/paper/PMC12825426/full.md

---
Source: https://tomesphere.com/paper/PMC12825426