Brain metastases occur in 20-40% of all cancer patients¹. Stereotactic radiosurgery (SRS), which involves delivering a high focal dose of radiation, is commonly applied to patients with a limited number of brain metastases. Current response evaluation techniques (i.e. RANO-BM²), that correlate response with tumor size change, are unable to assess response to SRS early after treatment. SRS induces DNA damage in cells leading to apoptosis which occurs within hours post-treatment³. Quantitative MRI techniques that probe metabolic and micro-structural changes in the tumor are potential biomarkers of early treatment response. We hypothesize that by measuring, a) concentration and exchange of protons associated with intracellular proteins and peptides with chemical exchange saturation transfer (CEST), and b) intracellular-to-extracellular water exchange rate with relaxometry, it is possible to provide accurate assessment of tumor response to SRS as early as one-week post-treatment.

Acquisition: $$$12$$$ patients with metastatic brain tumors were scanned on a 3T Philips Achieva MRI system under Sunnybrook Ethics Board approved protocols. Each patient was scanned before SRS, one-week post-SRS, and one-month post-SRS.

Relaxometry: DCE-MRI was acquired using 3D-SPGR ($$$TR/TE=4/2ms$$$, $$$FA=15^o$$$, $$$FOV=25.6\times25.6cm$$$, Matrix=$$$256\times256\times20$$$, Slice Thickness=$$$8mm$$$). Pre-contrast $$$T_{1}/B_{1}$$$ mapping was performed using Method of Slopes⁴ with $$$FA=3,14,130,150^o$$$.

CEST: Single-shot EPI CEST spectra were obtained with: $$$64$$$ offset frequencies between $$$-750Hz$$$ and $$$750Hz$$$ with a reference image at $$$100kHz$$$, Saturation pulse duration=$$$750ms$$$, amplitude=$$$0.52µT$$$. $$$TR/TE=1532/29.84ms$$$, $$$FOV=20\times20cm$$$, Matrix=$$$80\times80$$$, Slice Thickness=$$$3mm$$$. The experiment was repeated $$$5$$$ times, total scan time was $$$12$$$ minutes.

Tumor volume was assessed from post-Gd $$$T_{1}$$$-weighted MRI and was used for determining response. Using RANO-BM criteria, patients with partial response were considered responders and patients with stable disease or progressing disease were considered non-responders.

Analysis: Relaxometry: A three water compartment model of tissue longitudinal relaxation was used in this study^5–7. Each compartment (vascular, $$$V$$$, extracellular extravascular, $$$E$$$, and intracellular, $$$I$$$) in a voxel was assumed to contain a fraction of its total water content proportional to the compartment volume fraction $$$(M_{0,V},M_{0,E},M_{0,I})$$$, such that $$$M_{0,V}+M_{0,E}+M_{0,I}=1$$$. Water was assumed to move from intracellular to extracellular extravascular compartment with exchange rate constant $$$k_{IE}$$$. The water exchange between vascular and extracellular extravascular compartments was assumed to be negligible. Assuming negligible $$$T_{2}$$$ decay, Bloch equations describing longitudinal magnetization recovery from perturbation in each compartment are: $$\begin{cases}\frac{\text{d}M_{Z,V}(t)}{\text{d}t}=R_{1,V}(M_{0,V}-M_{Z,V}(t))\\\frac{\text{d}M_{Z,I}(t)}{\text{d}t}=R_{1,I}(M_{0,I}-M_{Z,I}(t))-k_{IE}M_{Z,I}(t)+k_{EI}M_{Z,E}(t)\\\frac{\text{d}M_{Z,E}(t)}{\text{d}t}=R_{1,E}(M_{0,E}-M_{Z,E}(t))-k_{EI}M_{Z,E}(t)+k_{IE}M_{Z,I}(t)\end{cases}$$ The vascular enhancement of DCE-MRI was separated using independent component analysis and then, the Bloch equations were fit to the DCE-MRI data and the model parameters $$$(k_{IE},M_{0,V},M_{0,E},M_{0,I})$$$ were calculated.

CEST: Images of each CEST offset frequency were divided by the reference image. Correction for $$$B_{0}$$$ inhomogeneities was performed by determining the minimum of water peak through fitting a Lorentzian-shape to each voxel and then shifting this peak to $$$0Hz$$$ offset. Parametric maps were then constructed by using a peak-fitting algorithm that decomposed each CEST spectrum into the following 5 Lorentzian-shaped peaks⁸: direct effect ($$$0$$$), amide ($$$3.5ppm$$$), amine ($$$2ppm$$$), Nuclear Overhauser Effect (NOE) ($$$-3.5ppm$$$) and magnetization transfer.

CEST and relaxometry were applied to all $$$12$$$ patients and model parameters were evaluated for pre-treatment and one-week post-treatment scans. The changes in each model parameter between pre-treatment and one-week post treatment scan were correlated to the changes in tumor volume between pre-treatment and one-month post-treatment scans. The intracellular-to-extracellular water exchange rate constant ($$$k_{IE}$$$) provided the highest correlation with tumor volume change for relaxometry (Fig.1). For CEST, the peak amplitude of NOE at the ipsilateral normal appearing white matter (iNAWM) provided the highest correlation with tumor volume change (Fig.2). The correlation between $$$k_{IE}$$$ and peak amplitude of NOE (iNAWM) is shown in Fig.3.Amongst all relaxometry parameters, early changes in $$$k_{IE}$$$ (shown in Fig.1) demonstrated the strongest correlation with tumor volume change one-month post-treatment $$$(R=-0.65,p=0.02)$$$. This could be an indicative of increased cell membrane permeability and surface-to-volume ratio in apoptotic cells⁷. As shown in Fig.2 there was a high positive correlation between the early changes in NOE (iNAWM) peak amplitude and tumor volume change one-month post-treatment $$$(R=0.67,p=0.02)$$$. This could be suggestive of an increase in cytoplasmic lipids which occurs with apoptosis⁹. Interestingly NOE change in iNAWM had stronger correlation compared to NOE of the tumor itself $$$(R=0.29,p=0.35)$$$. Fig.3 shows that there was a strong negative correlation between changes in $$$k_{IE}$$$ and NOE (iNAWM) peak amplitude $$$(R=-0.66,p=0.02)$$$ suggesting they could provide complementary information about treatment effects. Moreover, as can be seen in Fig.1&2, $$$k_{IE}$$$ provided a better separation of the responders from non-responders, while early NOE changes were better predicting the tumor volume change one-month post-treatment, suggesting that a combination of the two metrics had the potential to provide more accurate assessment of brain metastases response to SRS.