Molecular Classification of Prostate Cancer Using DESNT Gene Signatures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diagnostic methods for prostate cancer lack the ability to accurately distinguish between aggressive and non-aggressive forms, leading to unnecessary treatments and side effects for patients with biologically irrelevant disease.
Innovation Solution
An algorithm-based molecular diagnostic assay using Latent Process Decomposition (LPD) and random forest classification to identify a common gene signature associated with aggressive prostate cancer (DESNT), characterized by down-regulated genes involved in cytoskeleton machinery and ion transport, allowing for accurate prediction of cancer progression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PSA-based diagnosis is used to detect prostate cancer, then cancer detection sensitivity is improved, but treatment specificity deteriorates leading to overtreatment of biologically irrelevant cases
Solution Approach 1:
The invention segments prostate cancer into distinct molecular subtypes (e.g., neuroendocrine, luminal, basal) using gene expression profiling. This segmentation allows differentiation between aggressive cancers requiring treatment and indolent cancers that are biologically irrelevant, thereby improving prognosis prediction accuracy while maintaining cancer detection sensitivity.
Solution Approach 2:
The invention changes the diagnostic parameters from单一的PSA level to a multi-gene expression profile including markers such as NKX3-1, PSA, and other lineage-specific genes. This parameter transformation enables more precise classification of cancer aggressiveness and improves prognostic accuracy beyond what PSA alone can provide.
2Reliability
If radical treatment is applied to all PSA-detected prostate cancers, then treatment coverage is improved, but patient quality of life deteriorates due to unnecessary side effects
Solution Approach 1:
The invention performs preliminary molecular classification of prostate cancer before treatment decision-making. By identifying aggressive molecular subtypes that require treatment versus indolent subtypes that do not, the method enables selective treatment approaches that maintain adequate treatment coverage for necessary cases while avoiding unnecessary radical treatments and their associated side effects in indolent cases.
3Measurement precision
If comprehensive gene expression analysis is performed to classify prostate cancer subtypes, then prognosis prediction accuracy is improved, but diagnostic complexity increases
Solution Approach 1:
The invention extracts and focuses on a specific panel of key prognostic genes from the entire genome, including markers such as NKX3-1, PSA, TMPRSS2, and other lineage-specific genes. This extraction approach maintains high prognosis prediction accuracy by analyzing only the most relevant genes, thereby reducing diagnostic complexity compared to comprehensive genome-wide analysis.
Data Source
Figure 1a
Figure 1b~1c
Figure 1d
AI summary
The present invention relates to the classification of cancers, in particular prostate cancers, using samples from patients. In particular, the invention provides methods for identifying potentially aggressive prostate cancers to determine which cancers are or will become aggressive (and hence require treatment) and which will remain indolent (and will therefore not require treatment). The present invention is therefore useful to identify patients with a poor prognosis. The specific population of cancer identified by the present invention is referred to herein as DESNT cancer. The invention also provides biomarker panels useful in the diagnosis and prognosis of cancer.