Early detection of erlotinib treatment response in NSCLC xenograft tumors using hyperpolarized [1-13C]pyruvate MRS: comparison with FLT- and FDG-PET.

Objectives Treatment response monitoring is crucial in modern cancer therapy, and while it has long been based on the RECIST criteria of anatomical shrinkage, a new paradigm of monitoring by molecular imaging is emerging. We have evaluated a new imaging modality, hyperpolarized [1-13C]pyruvate Magnetic Resonance Spectroscopy (MRS), and compared its ability to monitoring treatment response with that of FLT- and FDG-PET in subcutaneous xenograft mouse models of non small cell lung cancers (NSCLCs) treated with the tyrosine kinase inhibitor erlotinib.

Methods Nude NMRI mice were inoculated with the HCC-827 (erlotinib sensitive) and H-1975 (erlotinib resistant) human NSCLC cell lines and randomized in groups receiving erlotinib (50 mg/kg) or vehicle. 24 hours after treatment, the mice were scanned using either hyperpolarized [1-13C]pyruvate MRS, FLT- or FDG-PET. MRS was performed on a 4.7 T preclinical scanner with a spiral readout 16x16 CSI with TR = 70ms, TE = 0.65ms and a FOV of 35mm. Anatomical and Dynamic Contrast Enhanced (DCE) perfusion imaging was also performed. PET imaging was performed on a Siemens Inveon Multimodality preclinical system. The mice were euthanized after imaging and the tumors resected for measurement of LDH activity and histological analysis.

Results The level of lactate detected by MRS in untreated HCC-827 tumors normalized to lactate level in the muscle was significantly higher than the level found in treated HCC-827 tumors (1.52 ± 0.17 [n=18] versus 0.76 ± 0.04 [n=14], p < 0.005). No significant difference was seen in the H-1975 groups (1.03 ± 0.18 and 0.95 ± 0.12, n=8 each group, p = 0.718). LDH activity normalized to total protein content in the extracted tumors was also found to be significantly different in the HCC-827 tumors, and furthermore correlated to the lactate levels found by MRS (R = 0.48, p < 0.05). Similar response was seen in PET imaging, with 54% and 38% decrease in uptake in the treated HCC-827 tumors for FLT- and FDG-PET, respectively, compared to the untreated group (p < 0.05). No significant difference in uptake between the treatment groups was observed for FLT or FDG in the H-1975 tumors. The PET results were supported by immunohistochemistry staining of tumor sections for Ki-67 (FLT-PET), glucose transporter 1 (GLUT1) and hexokinase 2 (HK2) (FDG-PET).

Conclusions Hyperpolarized [1-13C]pyruvate MRS was able to monitor the treatment response to erlotinib already 24 hours after treatment, eliciting similar response as FLT- and FDG-PET. Furthermore, LDH activity determined ex-vivo was significantly different between groups, and correlated with levels of lactate found by MRS. This suggests that hyperpolarized [1-13C]pyruvate MRS is a promising modality for clinical treatment response evaluation.

• Download figure
• Open in new tab
• Download powerpoint