INTRODUCTION

Gliomas account for more than 80% of all malignant brain tumors. Glioma patients have a very poor prognosis and radiation therapy, together with adjuvant temozolomide (TMZ) therapy is a widely used treatment option [1]. The standard assessment of treatment response uses gadolinium enhancement on MR imaging and/or fluid-attenuated inversion recovery (FLAIR). However, such methods are not reliable in detecting pseudoprogression and pseudoresponse especially with novel chemotherapeutic drugs necessitating alternative MRI techniques reflecting tumor activity. In glioma, a metabolic shift from oxidative phosphorylation to glycolysis causes an increased production of protons and lactic acid which is eliminated out of the tumor cells to maintain a near neutral intracellular pH. This along with poor perfusion causes acidification of extracellular regions in tumors. Thus measurement of extracellular pH (pH_e) can be used as a read out of tumor activity. We measured tumor size using contrast enhanced MRI and pHe using a novel imaging method called Biosensor Imaging of Redundant Deviation in Shifts (BIRDS) using TmDOTP^5-, to compare alterations in intratumoral-peritumoral pH_e gradient (ΔpH_e) in response to therapy.

METHODS

Adult, female athymic nude rats (n =11) bearing U251 tumors were randomly assigned for 2 cycles of TMZ treatment (40 mg/kg) starting day 14, daily, for 4 days with a gap of 2 days in between consecutive cycles. Tumor size was monitored by measuring the tumor volume with contrast enhanced imaging 2 weeks (Gadavist) and 3 weeks post inoculation (TmDOTP^5-). Animals were prepared for BIRDS imaging at the end of 3 weeks in both untreated and TMZ treated groups as reported earlier [2]. This was followed by experiments for longitudinal assessment of size (using Gadavist) and ΔpH_e with BIRDS imaging using TmDOTP5-in U251 tumor bearing athymic nude rats (n=2) before and after 2 cycles of TMZ treatment using protocol described in Huang et al., 2016 [3]. In vivo CSI data were acquired on 11.7T Agilent spectrometer using a ¹H surface coil. A dual-banded refocused 90° Shinnar-Le Roux RF pulse of 35 kHz bandwidth and 90 kHz separation with 205 μs duration was used for selective excitation of the H2/H3 and H6 protons of TmDOTP5^-. The datasets were acquired as previously described [2].

RESULTS

In response to TMZ treatment, a reduction in tumor size was observed at 3 weeks in both terminal and longitudinal experiments (Figures 1 and 2). Expression of cleaved Caspase-3 (apoptosis marker) and Ki-67 (proliferation marker) in treated and untreated rats was investigated using immunohistochemistry. Increased levels of cleaved Caspase-3 (p=0.001) and low expression of Ki-67 (p=0.0268) were observed in treated rats when compared to untreated rats, suggesting the initiation of apoptosis and reduction in proliferation of U251 tumor cells upon TMZ treatment. The ΔpH_e changes in response to TMZ treatment was monitored using BIRDS with TmDOTP^5-. As shown in Figure 3, pH_e inside the tumor was acidic in untreated when compared to TMZ treated rats. In longitudinal studies an increase in pHe in intratumoral regions was observed in response to TMZ treatment (Figure 4). The pH_e was also lower in voxels surrounding the tumor, in untreated tumors while treated rats showed normalization of pH_e. The intratumoral and peritumoral pH_e values measured by BIRDS for untreated and treated tumors are represented in histograms in Figure 5A and B. The average pH_e (±SD) inside and outside the tumor in treated and untreated rats are represented in Figure 5C. Inside the tumor, average pH_e was significantly (p<0.01) lower in untreated rats when compared to treated rats while in the regions outside the tumor, no significant difference was observed between the two groups. In longitudinal studies a similar trend was observed before and after TMZ treatment (Figure 5D). The Gaussian fitting analysis (FWHM) accounting for all voxels suggested that the intratumoral-peritumoral pH_e gradient (ΔpH_e) for treated tumor was ~0.1, whereas the ΔpH_e for untreated tumor was ~0.2.

CONCLUSION

Thus, to conclude, in this study, we report that mapping intratumoral-peritumoral pH_e gradient using BIRDS serves as a useful and effective biomarker in evaluating response to TMZ therapy in gliomas.

Acknowledgements

NIH Support: •NCI (R01 CA-140102) •NIBIB (R01 EB-011968)