Quantitative MRI Relaxation Parameters for Hypoxia Assessment In 3D Cancer Cell Spheroids
Abstract
Purpose
To evaluate the relationship between quantitative MRI relaxation parameters (T1, T2, T*2, and T'2) at 3T and hypoxia level in three-dimensional (3D) cancer cell spheroids under controlled in-vitro conditions, using western blot measurements as biological ground truth.
Methods
Human colon cancer cells (HCT116) were cultured in microfluidic chips at four concentrations (0.06, 0.25, 1.5, and 6 x106 cells/mL). Increasing cell concentration resulted in progressively larger spheroids with mean diameters of approximately 400, 650, 800, and 1050 µm. As spheroid size increases, oxygen diffusion to the inner regions becomes limited, leading to natural hypoxia development over five days. On day five, all microfluidic chips were imaged using a clinical Siemens 3T MRI scanner. Multi-echo gradient echo, multi-echo spin-echo, and inversion recovery sequences were acquired to quantify T*2, T2, and T1 respectively. Parametric maps of T1, T2, T*2, and T′2 were generated using custom-developed Python scripts, previously validated using a NIST quantitative MRI phantom. Western blot analysis was used to quantify spheroid protein, Carbonic Anhydrase 9 (CA-IX), levels, which is overexpressed by hypoxic cells. MRI-derived relaxation parameters were compared with CA-IX expression to assess their relationship with spheroid hypoxia.
Results
Western blot analysis confirmed a progressive increase in CA-IX/hypoxia with increasing spheroid size. Among MRI parameters, T2 time showed a significant decrease with increasing hypoxia (Kruskal–Wallis, p = 0.001), while T1 time demonstrated a decreasing trend that did not reach statistical significance (p ≈ 0.06). No statistically significant differences were observed in T*2 or T′2.
Conclusion
In a controlled in-vitro spheroid model with biologically validated hypoxia, T2 demonstrated significant and consistent sensitivity to increasing hypoxia. These results suggest that T2 may be a more robust MRI relaxation parameter for hypoxia assessment and provide a validated experimental foundation for future translational MRI hypoxia studies, including data-driven and machine-learning–based biomarker development.