Three-dimensional optical redox imaging reveals spatial heterogeneity of metabolic responses to a Vanadium-based nanotherapeutic in HeLa cervical tumor xenografts

Spatial heterogeneity in tumor metabolism complicates metabolic assessment using single tissue sections. Here, we applied cryo-micro-optical sectioning tomography (cryo-MOST) to subcutaneous HeLa cervical tumor xenografts collected before and at different times after treatment with vanadium-based nanotherapeutics (VO x ), mapping the three-dimensional (3D) distributions of reduced nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD), and the optical redox ratio (RR) at a pixel resolution of 4 μm. Compared with images downsampled to 50 μm, the original images better preserved tissue boundaries and local fluorescence features, supporting tumor delineation through comparison with adjacent hematoxylin and eosin (H&E)-stained sections. Whole-tumor reconstruction, axial profiling, and distribution analysis revealed spatial heterogeneity incompletely represented by central two-dimensional (2D) sections. Central-section measurements identified the lowest mean NADH, highest mean FAD, and lowest mean RR in the 4-, 8-, and 12-h specimens, respectively, whereas all three extrema occurred in the 8-h specimen in 3D. At RR < 0.2, the 8-h specimen had a whole-tumor low-RR volume fraction of 20.22%, compared with a central-section area fraction of 5.90%. Analysis of 21 planes per specimen further showed that section selection affected metabolic measurements and the identification of specimens with extreme values. Across the full tumor extent, agreement of all three metrics with volumetric references within ±5% at ≥95% sampling success required 13-25 randomly sampled planes or 5-9 systematically sampled planes. Sampling confined to the central region did not consistently achieve this agreement, highlighting the importance of axial coverage. Whole-tumor 3D optical redox imaging therefore links local metabolic features with volumetric measurements, reduces dependence on section selection, and provides a quantitative framework for spatial assessment of tumor metabolism following nanotherapeutic treatment.