Xiaodong Wang, Huilin Liu, Ke Yin, Guanyi Liao, Ping Ni, Jinjun Guo
Plasma lipidomic differences were most reproducible between patients with CRC and healthy controls, particularly between mCRC and HC. Lipid differences between mCRC and nmCRC remained exploratory and hypothesis-generating. These findings indicate systemic lipid remodeling in CRC but do not establish metastasis-specific plasma biomarkers.
BACKGROUND: Systemic lipid remodeling accompanies colorectal cancer (CRC), but whether circulating lipid alterations specifically distinguish metastatic from non-metastatic disease remains uncertain.
METHODS: This single-center prospective cross-sectional study enrolled newly diagnosed patients with non-metastatic CRC (nmCRC), metastatic CRC (mCRC), and healthy controls (HC) between September and December 2024. Two separate 1:1 nearest-neighbor propensity score matches used the same mCRC cohort as the anchor and histories of diabetes and hyperlipidemia as covariates. Matching was performed without replacement using a caliper of 0.2 on the raw propensity-score scale. Participant-level linkage generated 20 matched triplets. Fasting plasma samples underwent quantitative targeted lipidomic profiling. Primary lipid comparisons used matched-pair tests with comparison-specific Benjamini-Hochberg correction, and age- and sex-adjusted sensitivity analyses evaluated residual baseline imbalance.
RESULTS: Among 1,459 quantified lipids, 72 nominal differential lipids were identified in mCRC versus nmCRC, 186 in nmCRC versus HC, and 222 in mCRC versus HC using paired raw P<0.05. No lipid remained significant after false discovery rate correction in mCRC versus nmCRC, whereas 12 and 30 lipids remained significant in nmCRC versus HC and mCRC versus HC, respectively. After age and sex adjustment, the corresponding FDR-significant counts were 0, 10, and 9. Eight nominal candidates were shared across all three primary comparisons: BMP(14:0_20:0), BMP(14:0_20:1), Cer(d18:1/26:1), PC(18:2_18:2), PC(18:3_20:1), PC(18:3_20:4), PC(20:0_22:5), and PI(20:0_20:4).
CONCLUSION: Plasma lipidomic differences were most reproducible between patients with CRC and healthy controls, particularly between mCRC and HC. Lipid differences between mCRC and nmCRC remained exploratory and hypothesis-generating. These findings indicate systemic lipid remodeling in CRC but do not establish metastasis-specific plasma biomarkers.