A non-invasive 25-Gene PLNM-Score urine test for detection of prostate cancer pelvic lymph node metastasis

Jinan Guo ^1,2,3,4,5, Liangyou Gu ⁶, Heather Johnson ⁷

¹Department of Urology, Shenzhen People's Hospital (The Second Clinical Medical College, Jinan University), Shenzhen, China.
²Shenzhen Clinical Research Centre for Geriatrics, Shenzhen People's Hospital, Shenzhen, China.
³Shenzhen Urology Minimally Invasive Engineering Center, Shenzhen, China.
⁴Shenzhen Public Service Platform on Tumor Precision Medicine and Molecular Diagnosis, Clinical Medicine Research Centre, Shenzhen, China.
⁵The First Affiliated Hospital, Southern University of Science and Technology, Shenzhen, China.
⁶Department of Urology, The Third Medical Centre, Chinese PLA General Hospital, Beijing, China.
⁷Olympia Diagnostics, Inc., Sunnyvale, CA, USA.
⁸Department of Urology, The First affiliated Hospital of Guangzhou Medical University, Guangzhou, China.
⁹The University of Hong Kong-Shenzhen Hospital, Shenzhen, China.
¹⁰Department of Bio-diagnosis, Institute of Basic Medical Sciences, Beijing, China.
¹¹Department of Urology, Foshan First People's Hospital, Foshan, China.
¹²Department of Urology, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China.
¹³Clinical Laboratories, San Francisco General Hospital, San Francisco, CA, USA.
¹⁴Kinetic Reality, Santa Clara, USA.
¹⁵Department of Translational Medicine, Lund University, Clinical Research Centre, Malmö, Sweden.
¹⁶Department of Urology, Shenzhen People's Hospital (The Second Clinical Medical College, Jinan University), Shenzhen, China. zou.chang@szhospital.com.
¹⁷Shenzhen Urology Minimally Invasive Engineering Center, Shenzhen, China. zou.chang@szhospital.com.
¹⁸Shenzhen Public Service Platform on Tumor Precision Medicine and Molecular Diagnosis, Clinical Medicine Research Centre, Shenzhen, China. zou.chang@szhospital.com.
¹⁹The First Affiliated Hospital, Southern University of Science and Technology, Shenzhen, China. zou.chang@szhospital.com.
²⁰Key Laboratory of Medical Electrophysiology of Education Ministry, School of Pharmacy, Southwest Medical University, Luzhou, China. zou.chang@szhospital.com.
²¹Department of Molecular Biology, Umeå University, Umeå, Sweden. jenny.persson@umu.se.
²²Department of Biomedical Sciences, Malmö University, Malmö, Sweden. jenny.persson@umu.se.

⁰Department of Urology, Shenzhen People's Hospital (The Second Clinical Medical College, Jinan University), Shenzhen, China.

Prostate Cancer and Prostatic Diseases +

|

February 2, 2024

View abstract on PubMed

Summary

A new urine test, the 25G PLNM-Score, accurately detects prostate cancer pelvic lymph node metastasis (PLNM). This non-invasive tool helps avoid unnecessary surgeries and improves treatment decisions for patients.

Area Of Science

Urology
Oncology
Bioinformatics

Background

Prostate cancer patients with pelvic lymph node metastasis (PLNM) face poor prognoses.
Current nomograms for PLNM risk assessment have limited accuracy, leading to unnecessary pelvic lymph node dissection (PLND) and potential side effects.
Accurate identification of PLNM is crucial, but existing tests, including imaging, are insufficient.

Purpose Of The Study

To develop a highly accurate, non-invasive gene expression-based algorithm for detecting PLNM.
To create a classifier that can reliably identify patients with and without PLNM before PLND.

Main Methods

A random forest machine learning algorithm was employed to develop a PLNM classifier using urine samples.
The algorithm was trained on a multi-center retrospective cohort (n=413) and validated in a prospective cohort (n=243).
Discriminant analyses compared the algorithm's performance against the MSKCC nomogram score.

Main Results

The developed 25G PLNM-Score demonstrated high accuracy with an AUC of 0.93 in both retrospective and prospective cohorts.
The 25G PLNM-Score significantly improved biochemical recurrence-free and distant metastasis-free survival outcomes.
The algorithm spared 96% (retrospective) and 80% (prospective) of unnecessary PLND with minimal missed PLNM (0.51% and 1%, respectively), outperforming the MSKCC score.

Conclusions

The 25G PLNM-Score is a novel, highly accurate, non-invasive urine test for identifying PLNM prior to PLND.
This algorithm offers significant clinical benefits by avoiding unnecessary surgeries and enhancing treatment decision-making in prostate cancer management.