Innovative Value of Amide Proton Transfer Imaging in the Diagnosis and Treatment of Brain Gliomas: From Basic Principles to Clinical Applications

Abstract

Conventional magnetic resonance imaging (MRI) exhibits notable limitations in the diagnosis, grading, and therapeutic assessment of gliomas, making it insufficient to meet the demands of precision medicine. As a chemical exchange saturation transfer MRI technique, amide proton transfer (APT) imaging enables molecular-level visualization by detecting the chemical exchange of amide protons in endogenous mobile proteins and peptides. Previous studies have demonstrated that APT imaging provides substantial advantages over conventional MRI in the diagnosis, grading, and treatment monitoring of gliomas. This review systematically summarizes the development of APT imaging technology, emphasizing its innovative clinical applications, including preoperative grading, differentiation of postoperative recurrence, and dynamic evaluation of radiotherapy and chemotherapy efficacy. Furthermore, it discusses current challenges and future directions for clinical implementation, aiming to offer new perspectives for advancing precision medicine in glioma management.

Keywords

APT CEST-MRI clinical application diagnosis gliomas

1. Introduction

Magnetic resonance imaging (MRI) is the preferred modality for evaluating brain gliomas. Conventional sequences, such as T1-weighted imaging (T1WI), T2-FLAIR, and contrast-enhanced imaging, can reflect tumor water content and microstructural alterations; however, despite their high spatial resolution, they have limited capacity for revealing underlying molecular and metabolic changes.¹ Perfusion-weighted imaging (PWI), including dynamic susceptibility contrast (DSC), dynamic contrast-enhanced (DCE) MRI, and arterial spin labeling (ASL), provides hemodynamic information related to tumor angiogenesis and perfusion. Nevertheless, DSC relies on exogenous contrast agents, while DCE requires prolonged acquisition times, thereby limiting their clinical applicability.² Diffusion-weighted imaging and the apparent diffusion coefficient (ADC) characterize cellular structural density based on water molecule diffusion, yet their tissue specificity remains limited.³ Magnetic resonance spectroscopy (MRS) can assess intratumoral metabolic profiles without requiring biopsy; however, spectral overlap among different lesions and the relatively time-consuming examination process constrain its widespread use.⁴ Furthermore, functional MRI enables precise mapping of eloquent brain regions and is primarily utilized in preoperative surgical planning; however, its diagnostic and grading value in tumors remains limited.⁵

Therefore, innovative imaging approaches are essential to address these limitations. Among them, amide proton transfer (APT) imaging, first introduced in the early 2000s,⁶ has emerged as a promising molecular MRI technique for brain tumors.APT imaging specifically detects the chemical exchange between amide protons in proteins and peptides and water protons, thereby reflecting alterations in tissue protein concentration and pH.^7,8 The signal intensity of APT imaging is significantly correlated with the malignancy of brain gliomas: regions exhibiting high APT signals correspond to areas of active tumor proliferation, likely related to enhanced tumor cell growth and elevated protein synthesis, whereas regions with low APT signals indicate necrosis or peritumoral edema.^8,9 Compared with conventional MRI, APT imaging demonstrates substantial potential in diagnostic classification, treatment response evaluation, and survival prognosis of brain gliomas. This review systematically summarizes the potential clinical applications of magnetic resonance APT technology in the assessment and management of brain gliomas.

2. The Principle and Clinical Applications of APT Imaging

2.1. The Principle of APT Imaging

Chemical exchange saturation transfer (CEST) MRI is an advanced imaging modality that employs endogenous molecular contrast. This technique detects the chemical exchange between exchangeable protons in functional groups, such as amide (-NH), amino (-NH2), hydroxyl (-OH), and guanidino (-NHC(=NH)NH2), and free water protons.⁷ These endogenous small molecules exhibit characteristic chemical shifts near the water resonance frequency, enabling the selective detection of peptides, cytoplasmic proteins, carbohydrates (e.g., glucose), neurotransmitters (e.g., glutamate), and energy metabolites (e.g., creatine) through frequency-selective saturation pulses.¹⁰ Consequently, CEST MRI can noninvasively reflect the pathophysiological changes associated with brain gliomas and other diseases.

Amide proton transfer (APT) imaging represents a specialized branch of CEST technology. By detecting the chemical exchange between amide protons at 3.5 ppm and water protons, APT imaging reflects variations in mobile protein and peptide concentration as well as tissue pH¹¹ (Figure 1). This technique reveals molecular-level tumor heterogeneity, such as that observed in brain gliomas. Beyond its original role in brain tumor diagnosis, APT imaging has been extended to the study of solid tumors, including breast and prostate cancers, providing novel imaging biomarkers for early diagnosis, tumor grading, and treatment monitoring.

Figure 1.

Roadmap of APT imaging technology. Adapted with the permission of [Elsevier] and [Vinogradov et al], J Magn Reson, [2013].¹² (A): Chemical exchange process—dynamic exchange between amide protons and water protons. (B): Pulse sequence—utilizes presaturation pulses (typically 2–4 s, power 1–3 μT) with frequency offsets of ±3.5 ppm (referenced to the water resonance at 0 ppm), followed by a readout sequence (commonly fast spin-echo or gradient-echo). (C): Z-spectrum analysis—asymmetric analysis relative to the water resonance frequency (MTR_asym) is employed to minimize the effects of direct water saturation and to extract APT-specific contrast

2.2. Clinical Applications of APT Imaging

Currently, APT-related techniques include three-dimensional acquisition strategies, implementations at different magnetic field strengths (e.g., 3T and 7T), accelerated imaging schemes, and quantitative modeling approaches such as multi-pool model (MPM)-based CEST imaging.

Detailed parameters are listed in Table 1 (Comparison of APT Imaging Technology Parameters and Optimization). APT imaging has evolved from a single weighted-contrast approach into a comprehensive suite of techniques that integrate various field strengths, acquisition speeds, dimensional scales, and quantitative accuracies, thereby meeting diverse clinical and research needs. The selection of an appropriate imaging protocol depends on the specific clinical objective (routine diagnosis versus advanced research), available hardware (clinical 3T or ultra-high-field 7T systems), and the balance between scan duration and quantitative precision.

Table 1.

APT Imaging Technical Parameters and Optimization Comparison

Parameter category	3D APT imaging^13-16	3T APT-weighted (APTw) imaging protocols^17,18	Ultra-high-field (7T) APT imaging^19,20	Accelerated APT acquisition strategies²¹	Multi-pool model (MPM)-based CEST imaging^22,23	Deep learning–based CEST reconstruction and quantification approaches^24,25
Saturation Power	2μT	0.6–3.7µT	∼ 0.83 μT	1μT	2μT	0.6µT
Saturation Duration	1.5s/500ms/200 ms	3.5s	4s	δ=250ms	-	Tsat=3.2s
Offset Frequency	±2.7-±4.3/-1560ppm	±3.5/±1.8ppm	-5 ppm to +9 ppm/±300 ppm	Z-Spectrum ±8 ppm (0.5 ppm Step Size); Fine Segment ±3–4 ppm (0.1 ppm Step Size)	RF Offset ±7.0 ppm; APT 3.5 ppm; NOE -3.5 ppm	±100 ppm/Reference Point: -300 ppm
Pulse Parameters	Alternating Dual-Source Gaussian Pulses (CW/Pulsed Mode)	80 Gaussian Pulses, 20ms Pulse Width	200 Sinc Pulses, 10ms Pulse Width, 180° Flip Angle	Pulse Shape: Rectangular Pulse	-	80 Gaussian Pulses, 20ms Pulse Width, 20ms Interval
Other parameters	SENSE = 1/1.6; Acquisition Time ≈ 7 min;TSE factor: 22 (in the RL direction); EPI factor: 7 (in the SI direction)	In-Plane Resolution 1.7 × 1.7 mm²; Matrix Size 128 × 104	Sequence: snapshot-CEST (Gradient Echo + Spiral k-Space)	Dummy Pulse Number: 16; Scan Time: 4 min 32 sec (48% Acceleration)	Magnetization Transfer Rate: APT/NOE Pool 30 Hz, Bound Water Pool 2 Hz	Single Pulse Length: 20 ms; Inter-Pulse Delay: 20 ms
TR/TE	5105/7.3ms or 2000/8ms/TR = 2.5s	2.75/5.5ms or 1.99/3.9ms	4 ms/2 ms (Minimum TR)	Dual TE: 22.08/23.28ms	3034/26ms	4/2ms
Sampling Strategy	Spectral Presaturation with Inversion Recovery (SPIR)	3D Relaxation-Compensated APT	Z-Spectrum Acquisition	Dual TE Acquisition Combined with Dixon Technique	-	Wide-range, likely non-uniform sampling
Advantages	Reduction of Respiratory Motion Artifacts; Multi-Slice Coverage of the Entire Lesion;Excellent system compatibilityand Clinically feasible scan time	Stable Technology, Simple Process, and High Applicability	High SNR; High Chemical Shift Separation; Ultra-High Field Strength; Large Spectral Dispersion; High Sensitivity	Measurement Speed Increased by 50%; Tumor-to-White Matter CNR Improved to 1.8; Enhanced Magnetic Field Stability	Enhanced Specificity (for Distinguishing Tumor/Edema); Capable of Separating Multiple Effects	Deep Learning-Based Fast Fitting, Enabling Rapid and Robust Parameter Estimation and Online Image Reconstruction
Disadvantages	Motion Artifacts; High sequence complexity	B0 Field Inhomogeneity; B1 Field Inhomogeneity	Ghost Artifact	Motion Artifacts; High Field Strength Required; SNR Optimization Needed	Complex Computation; Long Scan Time	B0/B1 Inhomogeneity
B0/B1 Correction Measures	Dixon-Based Automatic B0 Mapping and Correction; Second-Order Linear Shimming;	WASABI	B0/B1 Inhomogeneity Correction Using the WASABI Method	Real-Time Dynamic Shimming	ΔfB0 Measurement Using the Water Saturation Shift Reference Method	Deep Learning - Assisted B0/B1 Shimming

At present, 3T APT-weighted (APTw) imaging protocols have become the standard for routine clinical application due to their technical maturity and accessibility. Accelerated APT acquisition strategies, such as fast multislice imaging utilizing a CEST-EPI readout, and deep learning–based reconstruction and quantification approaches (e.g., DeepCEST) represent key developments aimed at overcoming limitations to broader clinical adoption. In contrast, ultra-high-field (7T) APT imaging and quantitative multi-pool CEST modeling constitute the research frontier, addressing more complex scientific questions while establishing validation standards and optimization benchmarks for other methods.Moving forward, as MRI hardware, sequence algorithms, and artificial intelligence (AI) become more integrated, APT imaging is expected to advance toward faster, more precise, and more intelligent applications. Ultimately, it is anticipated to develop into a robust and comprehensive quantitative imaging biomarker platform applicable across the continuum of precision diagnosis and treatment for tumors, neurological disorders, and other major diseases.

3. Clinical Translation and Application Scenarios

3.1. APT in the Precise Delineation of Tumor Boundaries

APT imaging provides a distinct advantage in differentiating tumor infiltration from peritumoral edema by specifically detecting metabolic alterations in proteins and peptides. Gliomas are characterized by strong invasiveness, and peritumoral edema frequently contains infiltrating tumor cells. Clinical evidence indicates that approximately 80% of recurrent brain tumors arise at the surgical margin, emphasizing that accurately defined resection boundaries may reduce postoperative recurrence. Therefore, precise differentiation between tumor infiltration and peritumoral edema is essential for guiding patient management and prognostic assessment.²⁶ In earlier investigations employing the rat 9L brain tumor model, 9L cells were implanted into the right forebrain of rats. Several days later, tumors, peritumoral regions, and contralateral normal tissues were imaged and subsequently analyzed through histological staining. The results demonstrated that APT imaging effectively distinguished tumor tissue from edema by identifying elevated protein and peptide content in tumors.²⁷ In a preclinical study involving ten patients with brain tumors, APT imaging was compared with conventional MRI. The findings revealed that tumor regions exhibited approximately a 4% difference in water signal relative to edema and apparently normal white matter.²⁸ In highly invasive glioblastoma multiforme (GBM), three-dimensional fast spin echo-APT imaging demonstrated APT signal extension well beyond the gadolinium-enhanced tumor core. This technique accurately discriminated areas of tumor cell infiltration from regions with minimal neoplastic involvement, thereby providing neurosurgeons with an objective reference for achieving supra-total resection.¹³ In summary, APT imaging offers a robust, noninvasive method for distinguishing tumor infiltration from peritumoral edema by capturing molecular-level variations in protein and peptide metabolism. This capability enhances diagnostic precision, facilitates more accurate surgical planning, and supports prognostic evaluation, underscoring its substantial clinical value in glioma management.

3.2. Diagnosis and Grading

It should be noted that glioma grading in many earlier APT imaging studies was based on the World Health Organization (WHO) 2016 classification, in which diffuse gliomas were primarily categorized as grades II–IV according to histopathological features.²⁹ With the introduction of the WHO 2021 classification, molecular markers—such as isocitrate dehydrogenase (IDH) mutation status,Chromosome 7/10 alteration and 1p/19q codeletion—have become integral components of tumor classification and grading, reflecting a paradigm shift toward molecularly defined glioma entities.³⁰ However, in current clinical practice, particularly at the time of initial diagnosis, therapeutic decisions are often made before complete molecular information becomes available. Therefore, in current clinical practice, a hybrid diagnostic framework that integrates traditional histopathological grading (WHO 2016) with emerging molecular classification (WHO 2021) is still widely adopted. In this transitional context, noninvasive imaging biomarkers capable of reflecting tumor aggressiveness and underlying molecular characteristics remain highly valuable.

Conventional MRI sequences, such as Gd-T1w, T2w, and FLAIR, continue to exhibit limited specificity in glioma diagnosis. For example, contrast-enhanced T1w imaging often fails to clearly distinguish highly cellular infiltrative regions from necrotic tissue,³¹ and T2w imaging struggles to separate infiltrative tumors from peritumoral vasogenic edema.³² Moreover, approximately 10% of GBM cases and 30% of anaplastic astrocytomas do not exhibit gadolinium enhancement on Gd-T1w imaging.³³ These limitations significantly hinder accurate diagnosis and grading of gliomas.

As a molecular imaging technique, APT imaging sensitively reflects tumor heterogeneity and malignancy by detecting amide proton exchanges in tumor proteins and peptides. Multiple clinical studies have confirmed its distinct diagnostic and grading advantages (Figure 2C for diagnosis, Figure 2A and B for grading). In a cohort of 46 glioma patients across various grades, APT signals demonstrated a clear upward trend corresponding to tumor malignancy: Grade II: 0.84 ± 0.60%, Grade III: 1.55 ± 0.87%, Grade IV: 2.53 ± 0.70% (P for trend < 0.001). Combining APT imaging with the ADC derived from diffusion tensor imaging (DTI) substantially improved glioma grading accuracy, with the area under the receiver operating characteristic curve (AUC; area under the curve) increasing from 0.888 to 0.910 (P = 0.007).³⁴ Another clinical study validated this correlation, showing mean APT values of 2.1 ± 0.4% for Grade II gliomas (n = 8), 3.2 ± 0.9% for Grade III gliomas (n = 10), and 4.1 ± 1.0% for Grade IV gliomas (n = 18).³⁵ These findings, confirmed by several independent studies, consistently demonstrate that APT signal intensity increases with glioma grade.^9,36-38

Figure 2.

APTw signal intensity correlates with glioma grade, showing focal hyperintensity within the lesion. Reprinted with permission of [Elsevier] and [Jiang SS, et al], Eur J Cancer, [2017].³³ (A and B): Correlation between APTw imaging intensity and pathological grade. (C): APTw MRI displays hyperintensity within part of the lesion

APT imaging exhibits significant potential in clinical decision-making. Studies indicate that APTw imaging can identify tumor regions with higher cellular density and proliferative activity, thereby improving localization of high-grade lesions and enhancing the precision of biopsy sampling.³³ This contributes decisively to individualized treatment planning and prognosis assessment. At present, APT signal enhancement is believed to result primarily from elevated intracellular protein and peptide concentrations within tumor cells, although this mechanism warrants further validation through animal studies. Additionally, potential confounding factors require in-depth investigation. Despite the promising applications of APT imaging, current studies face several limitations, including small sample sizes (typically fewer than 50 cases), predominance of single-center data, and a lack of standardized imaging protocols. Future research should emphasize multicenter, large-sample clinical studies to further validate APT imaging’s diagnostic accuracy and to establish standardized criteria for glioma grading.

3.3. The Association Between APT and Molecular Subtyping

Molecular biomarkers, such as isocitrate dehydrogenase (IDH) mutation, O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation, and 1p19q codeletion, are closely associated with glioma prognosis.³⁹ A thorough understanding of these molecular characteristics is critical for accurate diagnosis, therapeutic planning, and prognostic evaluation. Conventional MRI primarily detects hydrogen nuclei in tissues and therefore cannot distinguish between specific metabolites such as lactate, proteins, and 2-hydroxyglutarate (2-HG). Although mass spectrometry provides exceptional sensitivity (down to picomolar levels) and comprehensive metabolite profiling, it requires the destruction of tissue samples. MRS can detect metabolites at concentrations above 0.1–1 mM; however, in vivo spectra obtained at clinical magnetic field strengths often cannot resolve metabolites with closely overlapping resonance frequencies.⁴⁰ While MRS can assess IDH mutation status, the technique requires a large tumor volume and is time-consuming.⁴¹ APT imaging indirectly detects molecules in the millimolar range by exploiting the chemical exchange between protein or peptide amide protons and water protons, thereby eliminating the need for exogenous contrast agents. This property may partially overcome the limitations of conventional MRI in molecular characterization, particularly in the assessment of IDH mutation status and other molecular subtypes.

In one clinical study, a retrospective analysis of 27 pathologically confirmed low-grade glioma patients was performed. Based on multi–region-of-interest analysis, the maximum (2.03 ± 0.72 vs. 0.99 ± 0.33) and minimum (0.99 ± 0.47 vs. 0.59 ± 0.32) APTw values were significantly higher in IDH–wild-type gliomas than in IDH–mutant gliomas (P < 0.001 and P = 0.02, respectively).⁴² Consistent findings from multiple independent studies reported that IDH–wild-type gliomas exhibited higher APTw values compared to IDH–mutant gliomas,^43,44 These results align with proteomic analyses showing that, relative to HRAS IDH1–wild-type glioma cells, IDH1–mutant cells exhibit globally reduced protein expression.⁴⁵

MGMT promoter hypermethylation, a common epigenetic modification, downregulates MGMT protein expression by inhibiting gene transcription.⁴⁶ This suppression reduces the DNA repair capability of tumor cells, enhancing their sensitivity to alkylating agent chemotherapy. Clinical data demonstrate that glioblastoma patients with unmethylated MGMT promoters show significantly higher APTw signal values than those with methylated promoters, including higher mean values (2.54 ± 0.41 vs. 2.01 ± 0.42; P = 0.022) and greater variance (1.01 ± 0.34 vs. 0.59 ± 0.24; P = 0.011).⁴⁷ From the perspective of the WHO 2021 classification, APT imaging provides a noninvasive surrogate for key molecular features such as IDH mutation and MGMT promoter methylation. This capability bridges imaging phenotypes with molecular pathology, and although current standard therapies like the Stupp regimen are not yet fully stratified by these subtypes, APT imaging holds significant potential for enabling earlier molecular risk stratification and guiding the development of future personalized treatment strategies.

3.4. Differentiation Among Pseudoprogression, Radiation Necrosis, and True Progression

Following chemoradiotherapy (CRT) for glioblastoma, three imaging manifestations—pseudoprogression (PsP), radiation necrosis (RN), and true progression (TP)—must be differentiated to guide treatment and prognosis accurately.

APT imaging provides a distinct advantage in distinguishing PsP from TP (Figure 3A and B). PsP results from treatment-induced vascular permeability and is characterized by increased enhancement and edema that can mimic tumor recurrence.⁴⁸ Unlike TP, PsP is typically not accompanied by progressive neurological decline, and lesions tend to resolve spontaneously.⁴⁹ In a study involving 32 glioma patients with suspected recurrence after CRT, 3D APT imaging revealed that regions corresponding to PsP exhibited lower cell density and cytoplasmic damage, resulting in fewer mobile cytoplasmic proteins and peptides and consequently reduced APTw signal intensity (n = 12, mean APTw = 1.56% ± 0.42%). In contrast, TP regions displayed higher APTw signal intensity (n = 20, mean APTw = 2.75% ± 0.42%) due to increased cellularity and abundant cytoplasmic content (P < 0.001).⁵⁰

Figure 3.

APTw MRI differentiation between tumor progression and PsP.Reprinted with permission of [John Wiley and Sons] and [Zhou J, et al], J Magn Reson Imaging, [2019].³² (A): Conventional and APTw MRI and histology from a patient with tumor progression. The gadolinium-enhancing regions on Gd-T1w appear hyperintense on APTw imaging compared with the contralateral brain. H&E staining demonstrates spindle mesenchymal cell proliferation with segregated glial cells. (B): Conventional and APTw MRI and histology from a patient diagnosed with PsP. The gadolinium-enhancing lesion appears isointense on APTw imaging, with punctate hyperintensity scattered within the lesion. H&E staining reveals large necrotic areas with scattered degenerating tumor and inflammatory cells

RN typically develops 6–24 months after radiotherapy and correlates with radiation dose, exposure volume, and fractionation regimen. In RN, extensive coagulative necrosis, destruction of normal tissue components, cellular degradation, and vascular damage lead to reduced APTw signal intensity. The high-power magnetization transfer ratio (MTR) demonstrates superior diagnostic performance, with an AUC of 0.88. Additionally, combining low-power CEST parameters further improves diagnostic discrimination compared with single-parameter evaluation.⁵¹ Integrating magnetization transfer (MT) and CEST imaging with DSC-MRI enhances differentiation between RN and TP—high perfusion indicating TP and low perfusion suggesting RN.⁵¹ In summary, APT imaging enables accurate, noninvasive differentiation among PsP, RN, and TP in glioblastoma by detecting molecular alterations in proteins and peptides. When combined with complementary modalities such as MT and perfusion imaging, it enhances diagnostic accuracy and supports refined assessment of treatment efficacy.

3.5. Efficacy Monitoring

A high APT signal serves as a critical predictor of poor overall survival (OS) and progression-free survival (PFS) in patients with high-grade gliomas.³⁹ In a clinical study, glioma patients underwent 3T CEST-MRI 4–6 weeks after completing radiotherapy. Statistical analysis revealed that Lorentzian fitting parameters (PeakArea_APT: HR = 0.39, p = 0.03) and asymmetry analysis (APTw_asym: HR = 2.63, p = 0.02) were significantly correlated with PFS.⁵²

Dynamic monitoring was conducted before and after radical CRT in 12 inoperable glioma patients, with APT imaging performed before CRT, immediately after completion, and 6 weeks post-treatment. The results demonstrated that the relayed nuclear Overhauser effect (rNOE) signal immediately after CRT significantly differentiated the disease-stable group (1.090 ± 0.110) from the progression group (0.808 ± 0.155, P = 0.015). This approach enabled therapeutic efficacy prediction at least four weeks earlier than evaluations based on the Response Assessment in Neuro-Oncology criteria, potentially revealing pathophysiological alterations preceding morphological changes.⁵³ Temozolomide (TMZ), the standard adjuvant therapy for glioblastoma, has been the focus of several recent studies evaluating treatment response. In a mouse experiment, orthotopic glioma models treated with an 80 mg/kg regimen (3 days on/4 days off) showed that APT signal intensity significantly decreased in TMZ-sensitive GBM lines, while morphological parameters such as tumor volume remained largely unchanged. In contrast, drug-resistant lines exhibited synchronous increases in both APT signal intensity and Ki67 expression. These findings indicate that metabolic alterations occur at early stages of chemotherapy, preceding detectable morphological changes.⁵⁴

Furthermore, a mouse model study demonstrated that APT hypointense regions corresponded to necrotic areas on H&E staining, confirming the sensitivity of APT imaging in reflecting TMZ treatment response.⁵⁵ These results suggest that APT imaging may serve as a sensitive biomarker for early assessment of therapeutic efficacy. Glioblastoma is pathologically characterized by vascular endothelial growth factor-mediated abnormal angiogenesis. Antiangiogenic agents such as bevacizumab are increasingly employed in recurrent GBM. Studies have shown that APT signal intensity markedly decreases 4–6 weeks after initiating bevacizumab therapy, and this reduction strongly correlates with improved treatment response and prolonged 12-month PFS, indicating its potential as an early predictor of therapeutic efficacy.⁵⁶ In malignant central nervous system tumors, disease progression is associated with reduced creatine metabolism. CEST-MRI has been used to detect enhanced creatine metabolism, which correlates with significant intratumoral T-cell infiltration following immune checkpoint inhibitor (ICI) therapy. This metabolic shift represents a specific marker of effective anticancer immunotherapy.^57,58 Overall, APT imaging demonstrates substantial value in evaluating therapeutic efficacy and predicting prognosis in gliomas. Both clinical and preclinical studies have confirmed that dynamic changes in APTw signals before and after treatment closely correlate with therapeutic outcomes. Its predictive utility is primarily evident in prognosis estimation, early evaluation of treatment efficacy, and ongoing response monitoring. As a highly sensitive functional imaging biomarker, APT imaging enables ultra-early assessment of treatment response and prognostic stratification, offering vital guidance for individualized treatment adjustments and serving as an important reference for endpoint evaluation in clinical trials.

4. Challenges and Controversies

4.1. Pathophysiological Confounders of APT Signal

Early studies attributed the enhancement of APT signals (relative to normal brain tissue) primarily to elevated concentrations of proteins and peptides in brain tumors²⁷ and to increased intracellular pH within tumor cells.^59,60 These findings were later validated in mouse model experiments ^54,61 and were consistent with the higher protein concentrations in tumor tissues identified by proteomics and in vivo spectroscopic MRI.^62,63

With ongoing technological advancements, additional factors have been recognized as contributors to APT signal elevation in tumors. One major factor is the reduction in semi-solid magnetization transfer asymmetry (MTC_asym) at 3.5 ppm. During Z-spectrum asymmetry analysis, which detects APT signals by comparing the difference between frequencies on either side of the water resonance, magnetization transfer (MT) asymmetry induces a baseline shift in the Z-spectrum, thereby complicating APT signal interpretation.^64,65 Both the nuclear Overhauser effect (NOE) and reduced MTC_asym can enhance the obtained APTw signals. According to the equation MTR_asym (3.5 ppm) = APTR + MTR′_asym (3.5 ppm) = APTR − NOER (−3.5 ppm), the APTw signal derived from MTR_asym actually represents a composite of multiple effects, including aliphatic NOE and MTC_asym contributions. In tumor tissues, reductions in NOE and MTC_asym paradoxically increase APTw contrast.^65,66 rNOE-suppressed amide CEST can further enhance APT contrast by clearly visualizing tumor regions and improving the distinction between normal white matter and tumor tissue.⁶⁷ Liquefactive necrosis also leads to elevated APTw signals, as the microenvironment promotes greater protein and peptide mobility, resulting in stronger APTw contrast.³² Similarly, hyperacute hemorrhage produces high APTw signals, primarily due to abundant hemoglobin, plasma proteins, and other mobile protein and peptide constituents within the hematoma.⁶⁸ Additionally, factors such as direct water saturation (spillover effect), MTC dilution effects,⁶⁹ and alterations in T1-weighted signals^70,71 may also influence APT signal intensity.

4.2. Technical Challenges

As an emerging molecular imaging technique, APT imaging exhibits unique advantages in glioma diagnosis and treatment, effectively compensating for the limitations of traditional MRI methods. However, several technical challenges persist. Magnetic field (B0) inhomogeneity can lead to significant shifts in the Z-spectrum along the frequency axis, while radiofrequency field (B1) inhomogeneity introduces spatial heterogeneity in the CEST effect, altering Z-spectrum morphology. These distortions typically manifest as nonlinear frequency shifts.⁷² Correction strategies for B0 and B1 inhomogeneities are summarized in Table 1. Additional challenges include motion-related artifacts, limited spatial resolution, low signal-to-noise ratio (SNR), extended scan duration, and stringent hardware requirements, all of which constrain APT’s clinical translation. Z-spectrum frequency-domain motion correction (MOCOΩ), a deep learning (DL)-based approach, effectively mitigates motion artifacts and maintains high spectral quality even under moderate-to-severe patient movement.⁷³ Moreover, three-dimensional (3D) CEST imaging reduces direct saturation artifacts through motion parameter–weighted averaging.⁷⁴ At a field strength of 3T, whole-brain 3D relaxation-compensated APT and rNOE CEST-MRI protocols can reduce total acquisition time to under seven minutes by selectively minimizing spectral acquisition volume, accelerating 3D readout, and optimizing presaturation parameters.¹⁷ Similarly, the fast CEST echo-planar imaging sequence significantly shortens scan duration.²¹ Despite these advancements, low spatial resolution and inadequate SNR continue to restrict APT imaging performance in clinical settings, limiting its full integration into precision medicine. Recent developments in MRI physics and engineering, particularly the introduction of ultra-high-field systems (≥7T), have alleviated some of these constraints. These systems enhance CEST detection sensitivity, improve spectral separation, and enable more accurate quantification of APT effects, paving the way for broader clinical implementation.⁷⁵

Currently, clinical APT imaging still lacks a unified standard for determining key acquisition parameters such as total saturation duration and power. The limitations of clinical MRI hardware—specifically, radiofrequency amplifier capacity and coil performance—restrict the use of multi-second saturation pulses and introduce B1 inhomogeneity, directly affecting CEST spatial uniformity.⁷⁶ Overcoming these issues is critical to establishing standardized imaging protocols that support multicenter studies and facilitate clinical adoption. In the long term, large-scale multicenter clinical trials are essential to compare and harmonize data across different scanner platforms and acquisition settings. Such studies will be instrumental in defining widely applicable imaging standards and promoting the transition of APT imaging from research laboratories to routine clinical practice.

5. Future Perspectives

5.1. Technological Integration

With the increasing clinical application of APT imaging, its independent diagnostic and therapeutic value in gliomas has been preliminarily established. However, because APT signals are highly sensitive to magnetic field homogeneity and tissue microenvironmental conditions, a single acquisition sequence often fails to comprehensively and stably characterize tumor biology. Consequently, integrating APT with complementary imaging and correction techniques has become a central research focus. The combination of field-shimming and fat-suppression methods markedly enhances image stability and reproducibility. For example, using water saturation shift referencing for accurate B0 correction^77,78 and polynomial fitting for B1 correction⁷⁹ effectively mitigates the influence of radiofrequency field inhomogeneity on APT signals. Techniques such as modified Dixon,⁸⁰ short tau inversion recovery,^81,82 and 1-2-1 spectral–spatial RF pulses⁸³ efficiently suppress fat-related artifacts, leading to substantial improvements in image clarity. Collectively, these advances establish a strong foundation for multicenter and multi-platform implementation of APT imaging.

As APT imaging continues to integrate with multimodal MRI and functional imaging, its diagnostic and therapeutic utility in brain gliomas has expanded significantly. For example, combining APT with DTI yields complementary insights into tumor protein metabolism and microstructure, offering a more comprehensive understanding of metabolism–perfusion coupling and invasive behavior, thereby improving glioma grading accuracy (P = 0.007).³⁴ The integration of APT with intravoxel incoherent motion MRI allows simultaneous evaluation of microscopic diffusion and metabolic activity, achieving excellent diagnostic performance for glioma grading (AUC = 0.986).⁸⁴ Similar multimodal strategies have also demonstrated potential in studies of hepatocellular carcinoma.^85,86 Furthermore, combining APT with PWI enables concurrent assessment of protein metabolism and hemodynamic parameters, elucidating the coupling between vascular supply and metabolic activity, and significantly improving differentiation between TP and treatment-related changes (AUC = 0.951).⁸⁷ Integration of APT with MRS additionally allows molecular-level validation of the correlation between APT signals and glutamine or protein concentrations.⁸⁸ The key advantage of multimodal APT imaging lies in its capacity to merge complementary biological information (e.g., perfusion and metabolism) and provide technical synergy (e.g., compensating for single-modality limitations). This integration significantly reduces false-positive and false-negative rates and enhances diagnostic accuracy.⁸⁹ Nonetheless, major challenges remain, such as complex multimodal data processing, time-intensive image interpretation, and the absence of standardized frameworks with defined diagnostic thresholds. These issues can introduce measurement variability and increase diagnostic uncertainty. In certain cases, multimodal sensitivity and specificity may even be inferior to those of optimized single-modality protocols. Additionally, the predominance of small-sample studies limits the generalizability and clinical translation of existing models.

Future research should prioritize identifying optimal modality combinations and parameter integration strategies, developing standardized diagnostic metrics (e.g., scoring systems or volume-weighted models), and expanding cohort sizes to validate machine learning algorithms.⁹⁰ Establishing multicenter technical standards and clinical guidelines will be crucial to improve inter-institutional consistency and reproducibility. Moreover, incorporating longitudinal datasets and prognostic biomarkers for dynamic monitoring will further advance the clinical application of APT-based multimodal imaging.

5.2. Radiomics and Radiogenomics

Radiomics enables high-throughput extraction of quantitative imaging features to construct multidimensional datasets that capture subtle variations in tumor morphology, texture, and function, thereby uncovering complex biological traits such as spatial heterogeneity and metabolic activity.^91,92 While radiomics primarily focuses on feature extraction to support predictive modeling, radiogenomics extends this approach by integrating imaging phenotypes with genomic data to elucidate tumor biology comprehensively, improving molecular subtyping and prognostic prediction.⁹³ This methodology has been widely applied to glioma grading,^94,95 molecular subtyping (e.g., IDH, ATRX, MGMT, and 1p/19q status),^96,97 differentiation of PsP,⁹⁸ prediction of bevacizumab response,^99,100 and survival prognosis¹⁰¹ demonstrating strong clinical potential.

Recent research integrating APT imaging with radiomics has significantly improved predictive model performance. For example, models incorporating multidimensional APT parameters with machine learning algorithms such as support vector machines (SVM) have shown outstanding discriminative accuracy in predicting IDH mutation status in grade II/III gliomas, offering a novel noninvasive approach to molecular subtyping.⁴³ Similarly, these models have achieved promising performance in grading adult-type diffuse gliomas and predicting H3K27M mutation status in brainstem gliomas.^102,103 In treatment response assessment, radiomics models incorporating APT-weighted (APTw) features have achieved an accuracy of 86.0% in distinguishing tumor recurrence from treatment-related effects, markedly outperforming conventional MRI.¹⁰⁴ Technically, compressed sensing (CS) combined with APT-SENSE has improved radiomic feature reproducibility (intraclass correlation coefficient [ICC] ≥ 0.5), particularly in regions of high tumor heterogeneity, while optimized filter selection further enhances feature robustness.¹⁰⁵

5.3. DL and AI

AI, particularly DL, is transforming the workflow and analytical capabilities of APT imaging. DL architectures, such as convolutional neural networks, fully convolutional networks, recurrent neural networks, and generative adversarial networks, significantly enhance image reconstruction quality, analytical efficiency, and automation. These models demonstrate strong clinical potential in segmentation, feature extraction, lesion detection, diagnostic classification, and prognostic prediction.¹⁰⁶ In the context of APT imaging, DL-based methods have been applied to overcome several key limitations:

Accelerated Acquisition: DL-driven CEST MR fingerprinting (MRF) enables rapid data acquisition with simultaneous multiparametric quantification, substantially reducing scan duration while generating objective and reproducible quantitative metrics.¹⁰⁷

Efficient Tissue Parameter Quantification: Magnetization transfer contrast MR fingerprinting (MTC-MRF), which integrates pseudo-random RF saturation with deep neural network modeling, permits precise quantification of MTC, CEST, and NOE signals. This approach overcomes the qualitative constraints of traditional MTR-based analysis and enables clinically feasible 3D quantitative brain imaging within practical acquisition times.¹⁰⁸

Image Reconstruction and Enhancement: DL-based super-resolution algorithms can reconstruct high-resolution CEST images from low-resolution data while preserving critical Z-spectrum information and ensuring compatibility across acquisition protocols.¹⁰⁹

Workflow Optimization and Automation: DL algorithms have been implemented for frequency selection optimization, parameter prediction, and motion artifact correction, achieving substantial reductions in acquisition time (e.g., from 5.5 minutes to 1.5 minutes) without compromising image quality.¹¹⁰

Collectively, these innovations address the inherent limitations of conventional APT quantification methods (e.g., MTR-based analysis) and provide practical solutions to enhance reproducibility and clinical scalability. The convergence of DL and APT radiomics is expected to further improve feature extraction, interpretability, and inter-center generalizability, thereby accelerating the integration of APT imaging into precision neuro-oncology.

5.4. The Potential of APT Imaging and Future Research Directions

APT imaging, as a functional MRI technique capable of noninvasively probing intracellular protein/peptide content and tissue acidity (pH), has demonstrated unique and promising potential in the prognostic evaluation of gliomas. Future research and development are likely to focus on several key directions.First, large-scale prospective studies are required to establish the independent associations between quantitative APT parameters and clinical outcomes, including overall survival (OS) and progression-free survival (PFS), thereby validating APT imaging as a robust prognostic imaging biomarker.Second, the value of APT imaging in dynamic treatment monitoring warrants further investigation. Owing to its high sensitivity to tumor microenvironmental changes, APT imaging may facilitate early differentiation between true progression and pseudoprogression. Future studies should explore whether early alterations in APT signals during radiotherapy, chemotherapy, or targeted and immunotherapies can predict ultimate treatment response and survival outcomes, enabling truly “dynamic” prognostic assessment.Third, deeper integration with artificial intelligence and multi-omics data is essential. Deep learning approaches may be employed to extract intratumoral heterogeneity features from APT images and to construct robust predictive models. In parallel, APT-based radiomics–genomics analyses should be pursued to elucidate potential links between APT imaging phenotypes and molecular pathways, such as IDH-related metabolic reprogramming.

Finally, technical standardization and clinical translation remain critical challenges. The establishment of standardized acquisition protocols and post-processing pipelines, together with the development of deep learning–based accelerated acquisition and reconstruction techniques, will be crucial for improving reproducibility and clinical accessibility, ultimately enabling the transition of APT imaging from a research tool to a practical modality for prognostic evaluation.

6. Conclusion

APT imaging enables noninvasive assessment of intratumoral protein metabolic alterations, providing distinctive value in the precision management of gliomas. It sensitively captures tumor microenvironmental changes and offers robust imaging biomarkers for clinical decision-making—particularly in molecular subtyping (e.g., IDH mutation and MGMT promoter methylation) and therapeutic monitoring. Nevertheless, several limitations remain. Most existing studies are constrained by small sample sizes (n < 50) and single-center and retrospective designs, limiting their generalizability. Moreover, the biological specificity of APT signals remains under debate: whether signal elevation primarily reflects increased tumor protein concentration or is confounded by inflammation, necrosis, hemorrhage, or treatment effects remains unclear. Significant heterogeneity across centers in sequence parameters, magnetic field strengths, and post-processing workflows further hinders cross-study comparison and the establishment of standardized diagnostic thresholds.

Although APT imaging is increasingly recognized as a promising functional biomarker, its clinical translation remains in an exploratory phase. The primary challenges include the absence of large-scale, multicenter prospective validation; the lack of standardized acquisition and analysis protocols; and incomplete understanding of the biological mechanisms underlying APT signal generation. Future research should prioritize the following areas:Large-scale multicenter prospective trials: Conduct systematic comparisons between APT findings, histopathology, and molecular profiles to establish diagnostic accuracy and clinical relevance.Technical standardization: Develop unified imaging protocols, quality control procedures, and diagnostic thresholds, emphasizing cross-scanner parameter optimization and reproducibility.Mechanistic elucidation: Employ multi-omics approaches to clarify the biological origins of APT signals and distinguish tumor-specific features from confounding physiological or treatment-related factors.Multimodal integration: Combine APT with diffusion, perfusion, metabolic, and AI-based imaging analyses to construct comprehensive, generalizable predictive models.Clinical application: Investigate the role of APT imaging in personalized therapy planning and as an imaging endpoint in clinical trials, particularly for early treatment response assessment and prognostic stratification.

In conclusion, APT imaging represents an innovative and powerful neuroimaging modality with substantial potential to advance precision glioma management. Achieving its full clinical translation will require rigorous multicenter validation, standardized methodologies, and interdisciplinary collaboration to deepen mechanistic understanding and ensure reliable implementation in clinical practice.

Footnotes

Acknowledgments

The authors would like to express their sincere gratitude to Professor Qibin Song from the Cancer Center of Renmin Hospital of Wuhan University for his invaluable support and insightful discussions throughout this study. The authors also acknowledge the use of AI-assisted technology for language polishing and refinement during the preparation of this manuscript.

ORCID iD

Changyue Yao

Ethical Considerations

This article is a review of previously published literature and does not involve any new studies with human participants or animals performed by any of the authors. Therefore, formal ethical approval was not required.

Consent to Participate

This study is a literature review and did not involve direct participation of human subjects.

Author Contributions

C.Y. drafted the manuscript. H.L., Z.R., and J.H. provided suggestions and revised the manuscript. W.H. and Q.S. conceptualized, wrote, and revised the manuscript. All of the authors reviewed and approved the final version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding was received from the Project of Hubei Provincial Natural Science Foundation in 2024 (No. 2024AFB785).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article as it is a review of existing literature.*

Author Biographies

Changyue Yao, (1996 - ), female, from Yichang, Hubei Province. Currently a master's student, mainly engaged in clinical research on thoracic tumor internal medicine treatment.

Weiguo Hu, male, from Wuhan, Hubei Province. Doctor, Deputy Chief Physician, Professor. Mainly engaged in clinical research on thoracic tumor internal medicine treatment.

Huali Liu, female, from Wuhan, Hubei Province. Doctor, Deputy Chief Physician, Professor. Mainly engaged in clinical research on thoracic tumor internal medicine treatment.

Zhihong Ran, male, from Guangzhou, The First Affiliated Hospital of Guangzhou Medical University. Holder of a medical postgraduate degree, with primary research focus on tumor transplantation.

Jingrui Hou, male, from Wuhan, Hubei Province. College of Basic Medical Sciences, Wuhan University, Undergraduate student.

Qibin Song, male, from Wuhan, Hubei Province. He holds a doctorate, is a chief physician, professor, and doctoral supervisor. His main research focus is on clinical studies of thoracic tumor internal medicine treatment.

References

Upadhyay

Waldman

. Conventional MRI evaluation of gliomas. Br J Radiol. 2011;84 Spec No 2(Spec Iss 2):S107-S111.

Hilario

Hernandez-Lain

Sepulveda

Lagares

Perez-Nuñez

Ramos

. Perfusion MRI grading diffuse gliomas: Impact of permeability parameters on molecular biomarkers and survival. Neurocirugia (Engl Ed). 2019;30(1):11-18.

Gihr

Horvath-Rizea

Hekeler

, et al. Diffusion weighted imaging in high-grade gliomas: A histogram-based analysis of apparent diffusion coefficient profile. PLoS One. 2021;16(4):e0249878.

Weinberg

Kuruva

Shim

Mullins

. Clinical Applications of Magnetic Resonance Spectroscopy in Brain Tumors: From Diagnosis to Treatment. Radiol Clin North Am. 2021;59(3):349-362.

Martucci

Russo

Schimperna

, et al. Magnetic Resonance Imaging of Primary Adult Brain Tumors: State of the Art and Future Perspectives. Biomedicines. 2023;11(2):364.

Zhou

Payen

Wilson

Traystman

van Zijl

. Using the amide proton signals of intracellular proteins and peptides to detect pH effects in MRI. Nat Med. 2003;9(8):1085-1090.

Dou

Lin

Ding

, et al. Chemical exchange saturation transfer magnetic resonance imaging and its main and potential applications in pre-clinical and clinical studies. Quant Imaging Med Surg. 2019;9(10):1747-1766.

Wen

Huang

, et al. MR imaging of high-grade brain tumors using endogenous protein and peptide-based contrast. Neuroimage. 2010;51(2):616-622.

Zhang

Zhu

Tain

Zhou

Cai

. Improved Differentiation of Low-Grade and High-Grade Gliomas and Detection of Tumor Proliferation Using APT Contrast Fitted from Z-Spectrum. Mol Imaging Biol. 2018;20(4):623-631.

10.

Mamoune

Barantin

Adriaensen

Tillet

. Application of Chemical Exchange Saturation Transfer (CEST) in neuroimaging. J Chem Neuroanat. 2021;114:101944.

11.

Koike

Morikawa

Ishimaru

Ideguchi

Uetani

Miyoshi

. Amide Proton Transfer-Chemical Exchange Saturation Transfer Imaging of Intracranial Brain Tumors and Tumor-like Lesions: Our Experience and a Review. Diagnostics (Basel). 2023;13(5):914.

12.

Vinogradov

Sherry

Lenkinski

. CEST: from basic principles to applications, challenges and opportunities. J Magn Reson. 2013;229:155-172.

13.

Inoue

Watanabe

Kusakabe

, et al. Three-dimensional amide proton transfer (APT) imaging appliable to navigation surgery can present comparable metabolic activity of glioblastoma to (11)C-Methionine PET. Acta Neurochir (Wien). 2025;167(1):49.

14.

Chen

Wang

, et al. Feasibility evaluation of amide proton transfer-weighted imaging in the parotid glands: a strategy to recognize artifacts and measure APT value. Quant Imaging Med Surg. 2021;11(6):2279-2291.

15.

Wang

Shen

, et al. Image quality assessment and feasibility of three-dimensional amide proton transfer-weighted imaging for hepatocellular carcinoma. Quant Imaging Med Surg. 2024;14(2):1778-1790.

16.

Zhu

Jones

van Zijl

Barker

Zhou

. Fast 3D chemical exchange saturation transfer (CEST) imaging of the human brain. Magn Reson Med. 2010;64(3):638-644.

17.

Goerke

Breitling

Korzowski

, et al. Clinical routine acquisition protocol for 3D relaxation-compensated APT and rNOE CEST-MRI of the human brain at 3T. Magn Reson Med. 2021;86(1):393-404.

18.

Wood

Derks

, et al. Reproducibility of APT-weighted CEST-MRI at 3T in healthy brain and tumor across sessions and scanners. Sci Rep. 2023;13(1):18115.

19.

Klomp

Dula

Arlinghaus

, et al. Amide proton transfer imaging of the human breast at 7T: development and reproducibility. NMR Biomed. 2013;26(10):1271-1277.

20.

Yuan

Guo

, et al. Noninvasive Delineation of Glioma Infiltration with Combined 7T Chemical Exchange Saturation Transfer Imaging and MR Spectroscopy: A Diagnostic Accuracy Study. Metabolites. 2022;12(10):901.

21.

Schüre

Pilatus

Deichmann

Hattingen

Shrestha

. A fast and novel method for amide proton transfer-chemical exchange saturation transfer multislice imaging. NMR Biomed. 2021;34(7):e4524.

22.

Takami

Norikane

Kimura

, et al. Relationship between multi-pool model-based chemical exchange saturation transfer imaging, intravoxel incoherent motion MRI, and (11)C-methionine uptake on PET/CT in patients with gliomas. Magn Reson Imaging. 2024;111:148-156.

23.

Sun

. Demonstration of accurate multi-pool chemical exchange saturation transfer MRI quantification - Quasi-steady-state reconstruction empowered quantitative CEST analysis. J Magn Reson. 2023;348:107379.

24.

Glang

Deshmane

Prokudin

, et al. DeepCEST 3T: Robust MRI parameter determination and uncertainty quantification with neural networks-application to CEST imaging of the human brain at 3T. Magn Reson Med. 2020;84(1):450-466.

25.

Schüre

Rajput

Shrestha

, et al. Toward Noninvasive High-Resolution In Vivo pH Mapping in Brain Tumors by (31)P-Informed deepCEST MRI. NMR Biomed. 2025;38(6):e70060.

26.

Zhou

Blakeley

Hua

, et al. Practical data acquisition method for human brain tumor amide proton transfer (APT) imaging. Magn Reson Med. 2008;60(4):842-849.

27.

Zhou

Lal

Wilson

Laterra

van Zijl

. Amide proton transfer (APT) contrast for imaging of brain tumors. Magn Reson Med. 2003;50(6):1120-1126.

28.

Jones

Schlosser

van Zijl

Pomper

Golay

Zhou

. Amide proton transfer imaging of human brain tumors at 3T. Magn Reson Med. 2006;56(3):585-592.

29.

Louis

Perry

Reifenberger

, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathol. 2016;131(6):803-820.

30.

Louis

Perry

Wesseling

, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro Oncol. 2021;23(8):1231-1251.

31.

Eidel

Burth

Neumann

, et al. Tumor Infiltration in Enhancing and Non-Enhancing Parts of Glioblastoma: A Correlation with Histopathology. PLoS One. 2017;12(1):e0169292.

32.

Zhou

Heo

Knutsson

van Zijl

PCM

Jiang

. APT-weighted MRI: Techniques, current neuro applications, and challenging issues. J Magn Reson Imaging. 2019;50(2):347-364.

33.

Jiang

Eberhart

Zhang

, et al. Amide proton transfer-weighted magnetic resonance image-guided stereotactic biopsy in patients with newly diagnosed gliomas. Eur J Cancer. 2017;83:9-18.

34.

Choi

Ahn

Lee

, et al. Amide proton transfer imaging to discriminate between low- and high-grade gliomas: added value to apparent diffusion coefficient and relative cerebral blood volume. Eur Radiol. 2017;27(8):3181-3189.

35.

Togao

Yoshiura

Keupp

, et al. Amide proton transfer imaging of adult diffuse gliomas: correlation with histopathological grades. Neuro Oncol. 2014;16(3):441-448.

36.

Sotirios

Demetriou

Topriceanu

Zakrzewska

. The role of APT imaging in gliomas grading: A systematic review and meta-analysis. Eur J Radiol. 2020;133:109353.

37.

Guo

Liu

, et al. Diagnostic performance of gliomas grading and IDH status decoding A comparison between 3D amide proton transfer APT and four diffusion-weighted MRI models. J Magn Reson Imaging. 2022;56(6):1834-1844.

38.

Durmo

Rydhög

Testud

, et al. Assessment of Amide proton transfer weighted (APTw) MRI for pre-surgical prediction of final diagnosis in gliomas. PLoS One. 2020;15(12):e0244003.

39.

Joo

Han

Ahn

, et al. Amide proton transfer imaging might predict survival and IDH mutation status in high-grade glioma. Eur Radiol. 2019;29(12):6643-6652.

40.

Viswanath

Chaumeil

Ronen

. Molecular Imaging of Metabolic Reprograming in Mutant IDH Cells. Front Oncol. 2016;6:60.

41.

Jiang

Wen

Ahn

, et al. Applications of chemical exchange saturation transfer magnetic resonance imaging in identifying genetic markers in gliomas. NMR Biomed. 2023;36(6):e4731.

42.

Jiang

Zou

Eberhart

, et al. Predicting IDH mutation status in grade II gliomas using amide proton transfer-weighted (APTw) MRI. Magn Reson Med. 2017;78(3):1100-1109.

43.

Han

Wang

Yang

, et al. Amide Proton Transfer Imaging in Predicting Isocitrate Dehydrogenase 1 Mutation Status of Grade II/III Gliomas Based on Support Vector Machine. Front Neurosci. 2020;14:144.

44.

Liu

, et al. Diagnostic performance between MR amide proton transfer (APT) and diffusion kurtosis imaging (DKI) in glioma grading and IDH mutation status prediction at 3 T. Eur J Radiol. 2021;134:109466.

45.

Doll

Urisman

Oses-Prieto

Arnott

Burlingame

. Quantitative Proteomics Reveals Fundamental Regulatory Differences in Oncogenic HRAS and Isocitrate Dehydrogenase (IDH1) Driven Astrocytoma. Mol Cell Proteomics. 2017;16(1):39-56.

46.

Sarkaria

Kitange

James

, et al. Mechanisms of chemoresistance to alkylating agents in malignant glioma. Clin Cancer Res. 2008;14(10):2900-2908.

47.

Jiang

Rui

Wang

, et al. Discriminating MGMT promoter methylation status in patients with glioblastoma employing amide proton transfer-weighted MRI metrics. Eur Radiol. 2018;28(5):2115-2123.

48.

Nichelli

Casagranda

. Current emerging MRI tools for radionecrosis and pseudoprogression diagnosis. Curr Opin Oncol. 2021;33(6):597-607.

49.

Ellingson

Chung

Pope

Boxerman

Kaufmann

. Pseudoprogression, radionecrosis, inflammation or true tumor progression? challenges associated with glioblastoma response assessment in an evolving therapeutic landscape. J Neurooncol. 2017;134(3):495-504.

50.

Blakeley

Hong

, et al. Applying amide proton transfer-weighted MRI to distinguish pseudoprogression from true progression in malignant gliomas. J Magn Reson Imaging. 2016;44(2):456-462.

51.

Mehrabian

Chan

Sahgal

, et al. Chemical Exchange Saturation Transfer MRI for Differentiating Radiation Necrosis From Tumor Progression in Brain Metastasis-Application in a Clinical Setting. J Magn Reson Imaging. 2023;57(6):1713-1725.

52.

Kroh

von Knebel Doeberitz

Breitling

, et al. Semi-solid MT and APTw CEST-MRI predict clinical outcome of patients with glioma early after radiotherapy. Magn Reson Med. 2023;90(4):1569-1581.

53.

Meissner

Korzowski

Regnery

, et al. Early response assessment of glioma patients to definitive chemoradiotherapy using chemical exchange saturation transfer imaging at 7 T. J Magn Reson Imaging. 2019;50(4):1268-1277.

54.

Sagiyama

Mashimo

Togao

, et al. In vivo chemical exchange saturation transfer imaging allows early detection of a therapeutic response in glioblastoma. Proc Natl Acad Sci U S A. 2014;111(12):4542-4547.

55.

Onishi

Sawaya

Tsuji

, et al. Evaluation of Temozolomide Treatment for Glioblastoma Using Amide Proton Transfer Imaging and Diffusion MRI. Cancers (Basel). 2022;14(8):1907.

56.

Park

Kim

Park

Jung

Kim

Heo

. Identification of Early Response to Anti-Angiogenic Therapy in Recurrent Glioblastoma: Amide Proton Transfer-weighted and Perfusion-weighted MRI compared with Diffusion-weighted MRI. Radiology. 2020;295(2):397-406.

57.

Cai

Tain

Zhou

, et al. Creatine CEST MRI for Differentiating Gliomas with Different Degrees of Aggressiveness. Mol Imaging Biol. 2017;19(2):225-232.

58.

Hoffmann

Schache

Höltke

, et al. Multiparametric chemical exchange saturation transfer MRI detects metabolic changes in breast cancer following immunotherapy. J Transl Med. 2023;21(1):577.

59.

Maintz

Heindel

Kugel

Jaeger

Lackner

. Phosphorus-31 MR spectroscopy of normal adult human brain and brain tumours. NMR Biomed. 2002;15(1):18-27.

60.

Walchhofer

Steiger

Rietzler

, et al. Phosphorous Magnetic Resonance Spectroscopy to Detect Regional Differences of Energy and Membrane Metabolism in Naïve Glioblastoma Multiforme. Cancers (Basel). 2021;13(11):2598.

61.

Ray

Simard

Larkin

, et al. Tumor pH and Protein Concentration Contribute to the Signal of Amide Proton Transfer Magnetic Resonance Imaging. Cancer Res. 2019;79(7):1343-1352.

62.

Kalinina

Peng

Ritchie

Van Meir

. Proteomics of gliomas: initial biomarker discovery and evolution of technology. Neuro Oncol. 2011;13(9):926-942.

63.

Hobbs

Shi

Homer

Harsh

Atlas

Bednarski

. Magnetic resonance image-guided proteomics of human glioblastoma multiforme. J Magn Reson Imaging. 2003;18(5):530-536.

64.

Hua

Jones

Blakeley

Smith

van Zijl

Zhou

. Quantitative description of the asymmetry in magnetization transfer effects around the water resonance in the human brain. Magn Reson Med. 2007;58(4):786-793.

65.

Zhou

Zaiss

Knutsson

, et al. Review and consensus recommendations on clinical APT-weighted imaging approaches at 3T: Application to brain tumors. Magn Reson Med. 2022;88(2):546-574.

66.

Heo

Zhang

Lee

Hong

Zhou

. Quantitative assessment of amide proton transfer (APT) and nuclear overhauser enhancement (NOE) imaging with extrapolated semi-solid magnetization transfer reference (EMR) signals: Application to a rat glioma model at 4.7 Tesla. Magn Reson Med. 2016;75(1):137-149.

67.

Zaiss

Windschuh

Goerke

, et al. Downfield-NOE-suppressed amide-CEST-MRI at 7 Tesla provides a unique contrast in human glioblastoma. Magn Reson Med. 2017;77(1):196-208.

68.

Wang

Hong

Chang

, et al. Simultaneous detection and separation of hyperacute intracerebral hemorrhage and cerebral ischemia using amide proton transfer MRI. Magn Reson Med. 2015;74(1):42-50.

69.

Zaiss

Bachert

. Chemical exchange saturation transfer (CEST) and MR Z-spectroscopy in vivo: a review of theoretical approaches and methods. Phys Med Biol. 2013;58(22):R221-R269.

70.

Lee

Heo

Zhang

, et al. Quantitative assessment of the effects of water proton concentration and water T(1) changes on amide proton transfer (APT) and nuclear overhauser enhancement (NOE) MRI: The origin of the APT imaging signal in brain tumor. Magn Reson Med. 2017;77(2):855-863.

71.

. Towards the complex dependence of MTR(asym) on T(1w) in amide proton transfer (APT) imaging. NMR Biomed. 2018;31(7):e3934.

72.

Jabehdar Maralani

Chan

Lam

, et al. Chemical Exchange Saturation Transfer MRI: What Neuro-Oncology Clinicians Need To Know. Technol Cancer Res Treat. 2023;22:15330338231208613.

73.

Singh

Mahmud

Yedavalli

, et al. Learning-based motion artifact correction in the Z-spectral domain for chemical exchange saturation transfer MRI. Magn Reson Med. 2025;94(1):331-345.

74.

Breitling

Korzowski

Kempa

, et al. Motion correction for three-dimensional chemical exchange saturation transfer imaging without direct water saturation artifacts. NMR Biomed. 2022;35(7):e4720.

75.

Jacobs

Swain

Wilson

, et al. Diffuse nuclear Overhauser effect MRI contrast changes detected in multiple sclerosis subjects at 7T. Brain Commun. 2025;7(1):fcaf043.

76.

Jones

Pollard

Pagel

. Clinical applications of chemical exchange saturation transfer (CEST) MRI. J Magn Reson Imaging. 2018;47(1):11-27.

77.

Liu

Song

Chan

McMahon

. Nuts and bolts of chemical exchange saturation transfer MRI. NMR Biomed. 2013;26(7):810-828.

78.

Kim

Gillen

Landman

Zhou

van Zijl

. Water saturation shift referencing (WASSR) for chemical exchange saturation transfer (CEST) experiments. Magn Reson Med. 2009;61(6):1441-1450.

79.

Singh

Cai

Haris

Hariharan

Reddy

. On B1 inhomogeneity correction of in vivo human brain glutamate chemical exchange saturation transfer contrast at 7T. Magn Reson Med. 2013;69(3):818-824.

80.

Zhang

Keupp

Wang

, et al. Z-spectrum appearance and interpretation in the presence of fat: Influence of acquisition parameters. Magn Reson Med. 2018;79(5):2731-2737.

81.

Zhang

Tian

, et al. 3D amide proton transfer-weighted imaging may be useful for diagnosing early-stage breast cancer: a prospective monocentric study. Eur Radiol Exp. 2024;8(1):41.

82.

Meng

Wang

Sun

, et al. A comparative study of the value of amide proton transfer-weighted imaging and diffusion kurtosis imaging in the diagnosis and evaluation of breast cancer. Eur Radiol. 2021;31(3):1707-1717.

83.

Zhang

Song

Liang

, et al. Early prediction of pathological response to neoadjuvant chemotherapy of breast tumors: a comparative study using amide proton transfer-weighted, diffusion weighted and dynamic contrast enhanced MRI. Front Med (Lausanne). 2024;11:1295478.

84.

Zou

Jiang

, et al. Differentiating the histologic grades of gliomas preoperatively using amide proton transfer-weighted (APTW) and intravoxel incoherent motion MRI. NMR Biomed. 2018;31(1): doi: https://doi.org/10.1002/nbm.3850.

85.

Jia

Wang

Han

. Comparative Study of Amide Proton Transfer Imaging and Intravoxel Incoherent Motion Imaging for Predicting Histologic Grade of Hepatocellular Carcinoma. Front Oncol. 2020;10:562049.

86.

Jia

Yan

Wang

Han

. Prediction Model for Intermediate-Stage Hepatocellular Carcinoma Response to Transarterial Chemoembolization. J Magn Reson Imaging. 2020;52(6):1657-1667.

87.

Hou

Diao

, et al. Differentiation of true progression from treatment response in high-grade glioma treated with chemoradiation: a comparison study of 3D-APTW and 3D-PcASL imaging and DWI. NMR Biomed. 2023;36(1):e4821.

88.

Liu

Zhao

, et al. Amide Proton Transfer Imaging Allows Detection of Glioma Grades and Tumor Proliferation: Comparison with Ki-67 Expression and Proton MR Spectroscopy Imaging. AJNR Am J Neuroradiol. 2017;38(9):1702-1709.

89.

Booth

Wiegers

Warnert

EAH

, et al. High-Grade Glioma Treatment Response Monitoring Biomarkers: A Position Statement on the Evidence Supporting the Use of Advanced MRI Techniques in the Clinic, and the Latest Bench-to-Bedside Developments. Part 2: Spectroscopy, Chemical Exchange Saturation, Multiparametric Imaging, and Radiomics. Front Oncol. 2021;11:811425.

90.

d’Este

Nielsen

Hansen

. Visualizing Glioma Infiltration by the Combination of Multimodality Imaging and Artificial Intelligence, a Systematic Review of the Literature. Diagnostics (Basel). 2021;11(4):592.

91.

Avery

Sanelli

Aboian

Payabvash

. Radiomics: A Primer on Processing Workflow and Analysis. Semin Ultrasound CT MR. 2022;43(2):142-146.

92.

Wang

. Radiomics in glioma: emerging trends and challenges. Ann Clin Transl Neurol. 2025;12(3):460-477.

93.

Fan

Luo

, et al. Artificial intelligence-based MRI radiomics and radiogenomics in glioma. Cancer Imaging. 2024;24(1):36.

94.

Takahashi

Tanaka

, et al. Radiomics Analysis for Glioma Malignancy Evaluation Using Diffusion Kurtosis and Tensor Imaging. Int J Radiat Oncol Biol Phys. 2019;105(4):784-791.

95.

Tian

Yan

Zhang

, et al. Radiomics strategy for glioma grading using texture features from multiparametric MRI. J Magn Reson Imaging. 2018;48(6):1518-1528.

96.

Hsu

Hsieh

, et al. Machine Learning-Based Radiomics for Molecular Subtyping of Gliomas. Clin Cancer Res. 2018;24(18):4429-4436.

97.

Haubold

Hosch

Parmar

, et al. Fully Automated MR Based Virtual Biopsy of Cerebral Gliomas. Cancers (Basel). 2021;13(24):6186.

98.

Wong

Young

Guo

Wong

. Support vector machine multiparametric MRI identification of pseudoprogression from tumor recurrence in patients with resected glioblastoma. J Magn Reson Imaging. 2011;33(2):296-305.

99.

Ammari

Sallé de Chou

Assi

, et al. Machine-Learning-Based Radiomics MRI Model for Survival Prediction of Recurrent Glioblastomas Treated with Bevacizumab. Diagnostics (Basel). 2021;11(7):1263.

100.

Kickingereder

Götz

Muschelli

, et al. Large-scale Radiomic Profiling of Recurrent Glioblastoma Identifies an Imaging Predictor for Stratifying Anti-Angiogenic Treatment Response. Clin Cancer Res. 2016;22(23):5765-5771.

101.

Liu

Qian

, et al. A radiomic signature as a non-invasive predictor of progression-free survival in patients with lower-grade gliomas. Neuroimage Clin. 2018;20:1070-1077.

102.

Zhuo

Zhang

, et al. Prediction of H3K27M-mutant brainstem glioma by amide proton transfer-weighted imaging and its derived radiomics. Eur J Nucl Med Mol Imaging. 2021;48(13):4426-4436.

103.

Jiang

Guo

, et al. Amide proton transfer-weighted imaging and derived radiomics in the classification of adult-type diffuse gliomas. Eur Radiol. 2024;34(5):2986-2996.

104.

Jiang

Guo

Heo

, et al. Radiomics analysis of amide proton transfer-weighted and structural MR images for treatment response assessment in malignant gliomas. NMR Biomed. 2023;36(1):e4824.

105.

Huang

Shen

Guo

Zhou

Jiang

. Radiomic feature reliability of amide proton transfer-weighted MR images acquired with compressed sensing at 3T. Int J Imaging Syst Technol. 2024;34(2):e23027.

106.

Jiang

Yang

Wang

Sun

. Emerging role of deep learning-based artificial intelligence in tumor pathology. Cancer Commun (Lond). 2020;40(4):154-166.

107.

Perlman

Farrar

Heo

. MR fingerprinting for semisolid magnetization transfer and chemical exchange saturation transfer quantification. NMR Biomed. 2023;36(6):e4710.

108.

Kim

Schär

Park

Heo

. A deep learning approach for magnetization transfer contrast MR fingerprinting and chemical exchange saturation transfer imaging. Neuroimage. 2020;221:117165.

109.

Shen

Cheema

Xie

Ruan

. Accelerating CEST MRI With Deep Learning-Based Frequency Selection and Parameter Estimation. NMR Biomed. 2025;38(7):e70068.

110.

Pemmasani Prabakaran

Park

Lai

JHC

, et al. Deep-learning-based super-resolution for accelerating chemical exchange saturation transfer MRI. NMR Biomed. 2024;37(8):e5130.