Are MRI/MRS methods complementary to FMISO-PET for evaluating hypoxia.

Hypoxia is an important prognostic factor in treatment of cancer by different methods, including radiation and chemotherapy. Our laboratory developed FMISO PET for imaging chronic hypoxia and we are exploring alternative MR-based methods for measuring other aspects of hypoxia imaging vascular parameters such as flow, volume and vascular leakage, and spectroscopic measures of biochemistry associated with hypoxia such as lactate levels and reduced thiols. Dynamic contrast enhanced (DCE) MR is a widely practiced although not well standardized method for assessing vascularity of tumors. It has not proven particularly useful in predicting overall outcome to cancer treatments such as anti-VEGF regimens. In this poster we will present imaging examples applied to patients with primary brain tumors who were imaged at baseline and on multiple occasions after initiation of therapy. Methods: MR spectroscopy and DCE imaging (256x256x16 matrix, 3.7 mm3 voxel volume, 60 time points, 5.5 sec/time point) were performed on a Philips 3T Acheiva in patients with primary brain tumors using an 8 ch head coil. The spectroscopy parameters were multi-voxel 2D/3D PRESS spectrosopic imaging,10 mm slice thickness, field of view 180 mm, repetition time 1140 msec, echo time 144 msec, 1024 complex FID points, spatial matrix 18x21x5. The acquisition box was positioned to include tumor and (whenever feasible) normal appearing contralateral brain. The MR spectroscopy data were analyzed using custom software and LCMODEL, which uses H2O as an internal reference to convert N-acetyl aspartate, choline, creatine, and lactate peak areas to concentration units. The average lactate concentration across the whole tumor was calculated at different dates coinciding with serial clinical surveillance scans, all after chemoradiation and adjuvant temozolomide and also after therapeutic doses of bevacizumab, an anti-VEGF antibody. The DCE data were acquired on the same dates and were analyzed using the extended Tofts model. Results and Conclusions: For one example patient, Ktrans values during adjuvant TMZ were relatively constant at about 0.4 and dropped precipitously after initiating bevacizumab even though the patient succumbed within weeks of the anti-VEGF treatment. The MRS results were informative in that choline, N-acetylaspartate and creatine levels held stable as did brain lactate far from the tumor, but the lactate signal in the tumor increased 20% over the value pre-bevacizumab and nearly 100% over the value in a contralateral region. These results support the potentially misleading aspect of DCE MRI in antivascular therapy and support the concept the multiple doses of the anti-VEGF antibody can exacerbate hypoxia under some conditions, inferred from the increased level of lactate after treatment. MRS and FMISO PET may be complementary in informing us about the mechanism and time of response to vascular disrupting agents. Supported by NIH Grant P01 CA042045-23 and GERRAF. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B150. Citation Format: Todd L. Richards, Mark Muzi, Kenneth Marro, James R. Fink, Jason K. Rockhill, Kenneth Krohn. Are MRI/MRS methods complementary to FMISO-PET for evaluating hypoxia. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B150.