Blood Metabolites Emerge as Powerful New Signals for Predicting Cancer Outcomes

A simple tube of blood may soon tell oncologists far more about a patient’s cancer than they ever imagined possible. A comprehensive review published in Holistic Integrative Oncology by researchers at Peking University People’s Hospital maps out how plasma metabolomics, the large-scale measurement of small molecules circulating in blood, is maturing from a laboratory curiosity into a clinically meaningful tool for predicting prognosis and guiding treatment in lung cancer and other malignancies. The work, led by Yue He, Juntong Guo, and Pufan Gao under the direction of Kezhong Chen, synthesizes detection technologies, tumor biology, and clinical trial evidence into a single translational roadmap, and its central message is striking: the chemical fingerprints of metabolism that tumors leave in the bloodstream can be systematically detected, quantified, and turned into actionable medical information.

The biological logic behind this approach rests on one of the most fundamental hallmarks of cancer: metabolic reprogramming. Tumor cells do not simply grow faster than normal cells; they rewire their entire chemical economy. They devour glucose at extraordinary rates, preferentially fermenting it into lactate even when oxygen is plentiful, a phenomenon known since the 1920s as the Warburg effect. They become addicted to glutamine, which feeds carbon and nitrogen into the tricarboxylic acid cycle and into the synthesis of nucleotides needed for rapid DNA replication. They ramp up de novo lipid synthesis to build membranes for dividing cells, and they scavenge fatty acids from neighboring adipocytes when supplies run short. Each of these rewired pathways alters the types and concentrations of metabolites released into bodily fluids, and those changes can be read out through liquid biopsy, the minimally invasive sampling of blood that can be repeated over the course of treatment.

Plasma and serum have emerged as the optimal sample types for this kind of analysis, and the review quantifies just how dominant they have become. In a survey of recent lung cancer metabolomics studies, plasma or serum accounted for roughly 54 percent of samples analyzed by nuclear magnetic resonance spectroscopy, around 62 percent of those analyzed by gas chromatography-mass spectrometry, and approximately 65 percent of those analyzed by liquid chromatography-mass spectrometry, with tissue samples making up only a small fraction in each case. The reasons are practical as much as scientific. Blood draws are minimally invasive, easily repeatable, and capable of capturing metabolic features of both primary tumors and metastases, something a single tissue needle biopsy, limited by tumor heterogeneity, often cannot achieve.

The technical machinery behind plasma metabolomics has advanced considerably. The two workhorse platforms remain nuclear magnetic resonance and mass spectrometry, each with distinct trade-offs. NMR requires minimal sample preparation and delivers rapid, quantitative, highly reproducible measurements, but it suffers from relatively low sensitivity and high cost. Mass spectrometry, in its capillary electrophoresis, gas chromatography, and liquid chromatography variants, offers far greater sensitivity and specificity. Among these, LC-MS has become the preferred method for untargeted metabolomics because it can identify a wide variety of metabolites at very low concentrations. Capillary electrophoresis-mass spectrometry excels at polar and charged species such as amino acids, nucleotides, and small organic acids, and recent innovations including sheathless interfaces have boosted its sensitivity while improved throughput has made it more practical than its earlier reputation suggested.

Beyond the established platforms, the review highlights a wave of emerging techniques that push detection limits dramatically lower. High-resolution mass spectrometry, particularly Fourier transform ion cyclotron resonance, offers exceptional resolution for precise mass determination. The parallel accumulation-serial fragmentation approach, integrated with trapped ion mobility spectrometry, markedly improves peak capacity and identification depth, allowing researchers to exploit low-abundance ions in complex samples. On the NMR side, a technique called PRISE-NUS-HSQC uses probe-induced sensitivity enhancement to break through the micromolar detection ceiling that has long limited conventional NMR. Meanwhile, the molecule-resolvable surface-enhanced Raman spectroscopy method, known as MORE SERSome, combines laser desorption and ionization mass spectrometry with SERS to identify dominant metabolic molecules in serum with label-free molecular resolution. A separate technique, Catch and Display for Liquid Biopsy, enables digital counting of biomarkers on individual extracellular vesicles. Together these advances lay a technical foundation robust enough for clinical deployment.

Yet the field’s Achilles heel is metabolite instability, and the review is unusually candid about it. Metabolites in liquid samples can degrade and transform rapidly after collection, and complex matrices like plasma are vulnerable to minor metabolic fluctuations. Delayed centrifugation of whole blood significantly alters levels of glucose, potassium, and aspartate aminotransferase. Non-refrigerated storage leads to substantial degradation of specific serum and plasma components within short periods, and excessive freeze-thaw cycles can markedly change biomarker profiles. The authors point to standardized operating procedures as critically important: widely accepted protocols now call for fasting at least eight hours before blood collection, using EDTA as the anticoagulant, processing samples within thirty minutes, storing plasma at minus eighty degrees Celsius, and thawing on ice before analysis. Without such discipline, the accuracy and reproducibility that clinical medicine demands simply cannot be achieved.

The mechanistic core of the review explains why specific metabolites carry prognostic information at all. Glucose uptake in tumor cells is driven by glucose transporter 1 and hexokinase under the regulation of the PI3K-AKT signaling pathway, with mutations in p53, HIF-1 alpha, Kras, and c-Myc further amplifying glycolytic enzyme expression. The resulting lactate does more than fuel tumors; it chemically modifies histones through lactylation, reprogramming gene expression, and it converts anti-inflammatory type I macrophages into tumor-promoting type II macrophages while suppressing CD8 T cell and NK cell activity through glucose depletion and NFAT inhibition. Glutamine metabolism follows a similarly intricate logic, with transporter overexpression through SLC1A5, SLC7A5, and related carriers sustaining mTORC1 activation, while SLC7A11 suppresses ferroptosis to promote tumor progression. In osimertinib-resistant non-small cell lung cancer cells, AMPK pathway activation increases glutamine uptake via ASCT2, directly linking metabolic rewiring to drug resistance.

Lipid metabolism adds another layer of prognostic significance. Most cancer cells upregulate de novo lipogenesis through the ATP-citrate lyase, acetyl-CoA carboxylase, and fatty acid synthase pathway, and in lung cancer, bidirectional crosstalk between adipocytes and tumor cells establishes a metabolic symbosis in which cancer cells induce lipolysis and internalize released fatty acids via CD36 and FATP1 transporters. Elevated CD36 expression correlates with poor prognosis and metastasis in several cancers, while high expression of ALDH1L2 promotes chemoresistance in small cell lung cancer by inhibiting lipid peroxidation. Within the tumor microenvironment, PD-1 signaling suppresses phospholipid synthesis in CD8 T cells, promoting their ferroptosis, and the lipid metabolite 25-hydroxycholesterol accumulates in tumor-associated macrophages, reinforcing an immunosuppressive phenotype associated with poor survival in lung squamous cell carcinoma.

The clinical payoff is already visible in early trials. Combining the glycolysis inhibitor 2-deoxy-D-glucose with docetaxel enhanced chemotherapy efficacy in non-small cell lung cancer and thyroid cancer, though the authors note the sample size was small and larger validation is needed. In advanced renal cell carcinoma, the glutaminase inhibitor telaglenastat combined with everolimus extended progression-free survival compared with everolimus alone in the randomized ENTRATA trial. The fatty acid synthase inhibitor TVB-2640 enhanced paclitaxel efficacy in advanced solid tumors including KRAS-mutant lung cancer, with serum FASN metabolite levels serving as pharmacodynamic markers of pathway engagement. In small cell lung cancer, elevated serum lactate and lactate dehydrogenase levels were associated with reduced benefit from immunotherapy, suggesting lactate itself can predict therapeutic response. At the 2023 World Conference on Lung Cancer, a multi-omics classifier combining metabolomics, proteomics, and transcriptomics from peripheral blood demonstrated strong performance in detecting lung cancer across all stages.

The authors are careful about what remains undone. No consensus guidelines currently endorse metabolomics-based biomarkers for lung cancer, and existing expert statements emphasize potential while calling for large-scale, multicenter clinical validation. Metabolic patterns vary across tumor subtypes and disease stages, meaning prognostic analyses must account for factors such as the pronounced differences in carnitine, amino acid, and lipid profiles between early and late-stage non-small cell lung cancer, and the distinct N-acyl ethanolamine biosynthesis alterations separating adenocarcinoma from squamous cell carcinoma. Still, the trajectory is clear. With standardized sample handling protocols in place, detection technologies capable of resolving ever-lower metabolite concentrations, and clinical trials already testing whether metabolic pathway activity predicts treatment benefit, plasma metabolomics is positioned to move from the research bench toward the oncology clinic, where a routine blood draw could one day help determine which patients need more aggressive therapy and which can be spared it.

Subject of Research: Plasma metabolomics detection technology and its clinical applications in lung cancer and other malignancies

Article Title: Advances in plasma metabolomics detection technology and its clinical applications in lung cancer and other malignancies

Article References: He, Y., Guo, J., Gao, P., Li, Z., Wang, W., & Chen, K. (2026). Advances in plasma metabolomics detection technology and its clinical applications in lung cancer and other malignancies. Holistic Integrative Oncology, 5(1), Article 33. https://doi.org/10.1007/s44178-026-00248-x

Image Credits: AI Generated

DOI: 10.1007/s44178-026-00248-x

Keywords: plasma metabolomics, liquid biopsy, lung cancer, non-small cell lung cancer, metabolic reprogramming, Warburg effect, mass spectrometry, nuclear magnetic resonance, glutamine metabolism, lactate, lipid metabolism, biomarkers