CEST: Physics & Technology
Anagha Deshmane1
1Diagnostic & Interventional Neuroradiology, University Hospital Tuebingen, Germany

Synopsis

Keywords: Contrast mechanisms: CEST & MT

Chemical exchange saturation transfer or CEST is a family of MR technologies which can be used to interrogate protons bound in exogenous contrast agents or endogenous metabolites and macromolecules. This lecture introduces (1) the origin and mechanisms of CEST contrast, (2) models of CEST contrast using the Bloch-McConnell equations, (3) endogenous and exogenous targets available for producing CEST contrast, and (4) pulse sequences and post processing methods for acquiring and analyzing images with CEST contrast. The examples used in this lecture will focus on endogenous CEST contrast and clinical imaging techniques.

What is CEST?

Chemical exchange saturation transfer or CEST is a family of MR technologies which can be used to indirectlyinterrogate bound protons in exogenous contrast agents or endogenous metabolites and other soluble molecules. Chemical bonds to neighboring atoms alter the electron shielding of these solute protons, giving rise to a change in the resonant frequency or chemical shift. The CEST technique excites the solute protons with a long period of off-resonant saturation. Through "chemical exchange” during this long excitation, the saturation is repeatedly transferred to the larger water pool which is measured on-resonant at the proton Larmor frequency, thereby amplifying the signal originating from solute protons. In addition to concentration, the CEST signal strength is also determined by the mechanism and efficiency of saturation transfer.

Sources & Mechanisms of CEST Contrast

The main mechanism of in vivo, clinical CEST contrast is by direct proton exchange from an exchangeable bound proton to the water proton pool. Exchangeable protons giving rise to CEST contrast can be found in amides, guanidinium, hydroxyl and amine groups. Molecules containing exchangeable protons include endogenous metabolites, peptides, small molecules, or exogenous CEST agents. Depending on the molecule of origin, the exchangeable proton of interest will be characterized by different exchange rate and resonance frequency. The exchange rate increases with pH and temperature. The CEST signal also scales with the T1 of the water pool, and is therefore larger at higher field strengths where relaxation times are longer.

In biological tissues there are several other sources of protons, most notably from the backbone and functional groups of mobile proteins and macromolecules, and in phospholipid membranes. In addition to effects of direct exchange, two indirect mechanisms of magnetization transfer to the water pool provide background signal contributions. This includes the nuclear Overhauser enhancement (NOE) occurring through the backbone of mobile molecules with intermediate or short T2, and intermolecular saturation transfer occurring through the semisolid lattice with very short T2, commonly referred to as semisolid magnetization transfer contrast (ssMT or MTC).

Modelling the CEST Effect

We often model CEST effects using a simplified multi-pool system, one representing the (on-resonant) water proton pool and others representing the solute proton pool, which are coupled by exchange rates. The Bloch-McConnell equations describe how the magnetization behaves in a multi-pool system in the presence of exchange. In this model, the exchange rates couple the magnetization in two pools and alter the relaxation rates in each pool. The behavior of the magnetization depends on the resonant frequencies of each pool, and the frequency of the RF excitation.

To understand the effect of RF excitation it is useful to analyze the magnetization in the rotating reference frame. In the case of long RF excitation both precession and relaxation occur. When applying off-resonant RF excitation the effective relaxation rate in the water pool, R1rho, is altered by the solute concentration, exchange rate, and strength of the applied RF saturation. The maximum exchange-dependent contribution to relaxation, Rex, scales proportionally with applied RF power and inversely to the exchange rate, i.e., saturation efficiency decreases with faster exchange rate and adequate labelling of fast exchanging species requires stronger saturation B1 power.

Anatomy of a CEST experiment: Acquisition & Evaluation

The first critical decision in a CEST study is the B0 field strength. Higher field strengths improve spectral resolution and selectivity of CEST effects, as well as increasing the signal strength and SNR. At ultra-high fields there is however the cost of SAR constraints and field inhomogeneity. CEST at low field strengths (1.5T) is severely hampered by coalescence of effects, making the signal non-specific.

Each repetition time of a CEST imaging acquisition contains three phases:

  1. a saturation block, where off-resonant RF is applied to saturate mobile protons and allow for saturation transfer;
  2. an imaging block, where the saturated water signal is excited on-resonant and the signal Ssat sampled with a read-out (FID/1D, 2D or 3D);
  3. a recovery period, which allows the magnetization in the water pool to recover to equilibrium.
The power and duration of saturation should be selected to the target exchangeable protons and their off-resonant frequencies and exchange rates. In a clinical setting, the saturation power is deposited via a train of shorter pulses to manage SAR constraints.

The acquisition is typically repeated to cover a range of off-resonant frequency saturations. The range and spacing of off-resonant saturations depends on the expected spectral selectivity and evaluation method performed. Finally, an unsaturated reference image S0 is also acquired.

The CEST signal is calculated as a relative value Z = Ssat/S0 at for each off-resonant excitation frequency, forming a Z-spectrum. The dynamics of Z at a particular offset frequency are analyzed directly, or several metrics can be used to investigate features of the Z-spectrum. The most common evaluation approach is MTRasymmetry, in which the signal at the resonant frequency of interest is compared to the opposite side of the spectrum, which is assumed to contain the same background signal. In low-saturation power experiments where slow-exchanging contributions from ssMT and NOE cannot be neglected, spectrally-selective features of the Z spectrum can be isolated using a background reference signal from, for example, multi-Lorentzian fitting. Finally, inverse metrics isolate the exchange contribution to the Z value from spillover, MT and T1 effects.

Acknowledgements

No acknowledgement found.

References

  1. van Zijl and Yadav. Chemical exchange saturation transfer (CEST): what is in a name and what isn't? Mag Res Med 2011; doi: 10.1002/mrm.22761
  2. Wu et al. An overview of CEST MRI for non-MR physicists. EJNMMI Physics 2016. doi: 10.1186/s40658-016-0155-2
  3. Van Zijl et al. Magnetiation Transfer Contrast and Chemical Exchange Saturation Transfer MRI. Features and analysis of the field-dependent saturation spectrum. NeuroImage 2017. doi: 10.1016/j.neuroimage.2017.04.045
  4. Zhou et al. The relayed nuclear Overhauser effect in magnetization transfer and chemical exchange saturation transfer MRI. NMR in Biomedicine 2022; doi: 10.1002/nbm.4778
  5. Heo et al. Amide proton transfer imaging in stroke. NMR in Biomedicine 2022; doi: 10.1002/nbm.4734
  6. Zhou et al. Review and consensus recommendations on clinical APT-weighted imaging approaches at 3T: Application to brain tumors. Mag Res Med 2022; doi: 10.1002/mrm.29241
  7. Xu et al. Chemical exchange saturation transfer imaging of creatine, phosphocreatine, and protein arginine residue in tissues. NMR in Biomedicine 2022; doi: 10.1002/nbm.4671
  8. Zaiss et al. Theory of chemical exchange saturation transfer MRI in the context of different magnetic fields. NMR in Biomedicine 2022; doi: 10.1002/nbm.4789
  9. Zaiss et al. Inverse Z-spectrum analysis for spillover-, MT-, and T1 -corrected steady-state pulsed CEST-MRI. NMR in Biomedicine 2014. doi: 10.1002/nbm.3054

Proc. Intl. Soc. Mag. Reson. Med. 31 (2023)