Coupled Mass-Spectrometry-Based Lipidomics Machine Learning Approach for Early Detection of Clear Cell Renal Cell...

Area of Science:

• Biochemistry
• Oncology
• Analytical Chemistry

Background:

• Clear cell renal cell carcinoma (ccRCC) is a significant global health concern.
• Early diagnosis of ccRCC is crucial for improving patient outcomes.
• Current diagnostic methods for ccRCC have limitations, necessitating novel approaches.

Purpose of the Study:

• To discover novel lipid biomarkers for ccRCC detection.
• To develop lipid panels for discriminating ccRCC patients from healthy controls.
• To create a lipid panel for differentiating early-stage ccRCC from late-stage ccRCC.

Main Methods:

• Serum samples from ccRCC patients (stages I-IV) and controls were analyzed using ultraperformance liquid chromatography-quadrupole-time-of-flight mass spectrometry.
• Machine learning techniques, including support vector machines and LASSO, were employed for multivariate model development.
• Two discriminant lipid panels were generated for ccRCC detection and stage discrimination.

Main Results:

• A 16-lipid panel achieved 95.7% accuracy in discriminating ccRCC patients from controls in a training set and 77.1% in an independent test set.
• A 26-compound panel differentiated early-stage ccRCC from late-stage ccRCC with 82.1% accuracy in an independent test set for stage I patients.
• Thirteen specific lipid species were found at significantly lower levels in ccRCC patients compared to controls.

Conclusions:

• The identified lipid panels demonstrate potential for non-invasive ccRCC detection and early diagnosis.
• The findings support the utility of lipid profiling combined with machine learning for ccRCC biomarker discovery.
• Further validation in larger, diverse cohorts is recommended to confirm the clinical applicability of these lipid panels for ccRCC management.