Performance of amide proton transfer imaging to differentiate true progression from therapy-related changes in gliomas and metastases

Abstract

Objectives: Differentiating true progression or recurrence (TP/TR) from therapy-related changes (TRC) is complex in brain tumours. Amide proton transfer-weighted (APT) imaging is a chemical exchange saturation transfer (CEST) MRI technique that may improve diagnostic accuracy during radiological follow-up. This systematic review and meta-analysis elucidated the level of evidence and details of state-of-the-art imaging for APT-CEST in glioma and brain metastasis surveillance.

Methods: PubMed, EMBASE, Web of Science, and Cochrane Library were systematically searched for original articles about glioma and metastasis patients who received APT-CEST imaging for suspected TP/TR within 2 years after (chemo)radiotherapy completion. Modified Quality Assessment of Diagnostic Accuracy Studies-2 criteria were applied. A meta-analysis was performed to pool results and to compare subgroups.

Results: Fifteen studies were included for a narrative synthesis, twelve of which (500 patients) were deemed sufficiently homogeneous for a meta-analysis. Magnetisation transfer ratio asymmetry performed well in gliomas (sensitivity 0.88 [0.82-0.92], specificity 0.84 [0.72-0.91]) but not in metastases (sensitivity 0.64 [0.38-0.84], specificity 0.56 [0.33-0.77]). APT-CEST combined with conventional/advanced MRI rendered 0.92 [0.86-0.96] and 0.88 [0.72-0.95] in gliomas. Tumour type, TR prevalence, sex, and acquisition protocol were sources of significant inter-study heterogeneity in sensitivity (I² = 62.25%; p < 0.01) and specificity (I² = 66.31%; p < 0.001).

Conclusion: A growing body of literature suggests that APT-CEST is a promising technique for improving the discrimination of TP/TR from TRC in gliomas, with limited data on metastases.

Clinical relevance statement: This meta-analysis identified a utility for APT-CEST imaging regarding the non-invasive discrimination of brain tumour progression from therapy-related changes, providing a critical evaluation of sequence parameters and cut-off values, which can be used to improve response assessment and patient outcome.

Key points: Therapy-related changes mimicking progression complicate brain tumour treatment. Amide proton imaging improves the non-invasive discrimination of glioma progression from therapy-related changes. Magnetisation transfer ratio asymmetry measurement seems not to have added value in brain metastases.

Keywords: Brain metastasis; Glioma; Magnetic resonance imaging; Molecular imaging; Therapy response.

Fig. 1

Flow diagram describing the study screening and selection

Fig. 2

Overview of risk of bias/applicability assessment according to modified Quality Assessment of Diagnostic Accuracy Studies-2 criteria

Fig. 3

Forest plot of the diagnostic performance of amide proton transfer imaging alone for distinguishing tumour progression and recurrence from therapy-related changes in gliomas (black) and brain metastases (maroon), showing substantial heterogeneity between sensitivities (left) and specificities (right)

Fig. 4

Forest plot of the diagnostic performance of amide proton transfer imaging in combination with arterial spin labelling (ASL) or dynamic susceptibility contrast (DSC) imaging and/or other imaging techniques for distinguishing tumour progression and recurrence from therapy-related changes showing mild heterogeneity between specificities (left)

Fig. 5

Hierarchical summary receiver operating characteristics (HSROC) curves of the diagnostic performance of amide proton transfer imaging alone (left) in gliomas (blue x) and brain metastases (red x), and in combination with other magnetic resonance imaging techniques (right) in gliomas for distinguishing tumour progression and recurrence from therapy-related changes. Summary area under the curve estimate was 0.88 [0.85–0.91] (left) and 0.94 [0.92–0.96] (right)

Fig. 6

Images illustrating APT-weighted imaging (MTRasym at 3.5 ppm) in two patients after treatment with radiotherapy. Contrast enhancement on T1-weighted images was seen in patient 1, 62 months after radiotherapy treatment and resection for grade 2 astrocytoma (A). The additional increased MTRasym in the same patient (C) illustrates tumour recurrence, which was confirmed as grade 4 glioblastoma after repeat surgery. Contrast enhancement on T1-weighted images was seen in patient 2, 14 months after chemoradiotherapy of grade 3 astrocytoma, with regional anaplastic oligodendroglioma (B). The additional MTRasym in patient 2 (D) illustrates low values, indicating treatment effect, which was confirmed as radiation necrosis with histopathology after repeat surgery. Image reproduced via open access CC license and permission from ref. [11]