In view of the "5th edition of the WHO classification of tumours of the central nervous system" published in 2021, the conventional imaging features of weighted MRI sequences (clinical standard) are often no longer sufficient to predict the biology of diffuse adult gliomas with regard to their molecular and epigenetic features. However, it is precisely these characteristics that are relevant for individual prognosis and personalised therapy options in a constantly evolving field. Furthermore, it is not possible to reliably distinguish therapy-related changes from genuine tumour progression on the basis of weighted sequences. This applies in particular to the use of combination therapies, e.g. with immunomodulators. The diagnostic disciplines are therefore faced with the challenge of keeping pace with these developments and providing new methodological approaches.
Two scientific fields are particularly promising here: firstly, the use and further development of AI for image analysis and, secondly, the use of advanced MRI techniques that can multiply the potential of AI-based analyses. Multiparametric quantitative MRI (mp-qMRI) provides quantitative parameters that map the microstructural properties of brain and tumour tissue, such as axon, myelin, iron and water concentrations, with a high degree of accuracy. Compared to the weighted MRI sequences used in routine imaging, the method is largely independent of hardware and software. Previous multi-centre mp-qMRI studies in brain tumour patients were limited by long MRI measurement times and the lack of a standardised, manufacturer-independent protocol. In the run-up to this project, an easy-to-implement protocol based on manufacturers' standard sequences (multi-echo gradient echo sequences, echoplanar imaging) was therefore developed specifically for brain tumour patients. Four quantitative parameter maps (T2*, QSM, H₂O and T1) can be generated within a clinically tolerable measurement time of 8 minutes (1). These multiparametric maps can be used in addition to standard sequences for diagnostics, therapy planning and monitoring and thus offer great potential for improving patient treatment.
(1) Thomas DC, Deichmann R, Nöth U, Langkammer C, Ferreira M, Golbach R, Hattingen E, Wenger KJ. A fast, vendor-neutral protocol for multi-centre, multi-
parametric quantitative MRI studies in brain tumour patients. Neurooncology Advances June 2024