Somatostatin receptor subtype expression and radiomics from DWI-MRI represent SUV of [68Ga]Ga-DOTATOC PET in patients with meningioma

Abstract

Objective: This retrospective study aimed to analyse the correlation between somatostatin receptor subtypes (SSTR 1-5) and maximum standardized uptake value (SUV_max) in meningioma patients using Gallium-68 DOTA-D-Phe1-Tyr3-octreotide Positron Emission Tomography ([68Ga]Ga-DOTATOC PET). Secondly, we developed a radiomic model based on apparent diffusion coefficient (ADC) maps derived from diffusion weighted magnetic resonance images (DWI MRI) to reproduce SUV_max.

Method: The study included 51 patients who underwent MRI and [68Ga]Ga-DOTATOC PET before meningioma surgery. SUV_max values were quantified from PET images and tumour areas were segmented on post-contrast T1-weighted MRI and mapped to ADC maps. A total of 1940 radiomic features were extracted from the tumour area on each ADC map. A random forest regression model was trained to predict SUV_max and the model's performance was evaluated using repeated nested cross-validation. The expression of SSTR subtypes was quantified in 18 surgical specimens and compared to SUV_max values.

Results: The random forest regression model successfully predicted SUV_max values with a significant correlation observed in all 100 repeats (p < 0.05). The mean Pearson's r was 0.42 ± 0.07 SD, and the root mean square error (RMSE) was 28.46 ± 0.16. SSTR subtypes 2A, 2B, and 5 showed significant correlations with SUV_max values (p < 0.001, R2 = 0.669; p = 0.001, R2 = 0.393; and p = 0.012, R2 = 0.235, respectively).

Conclusion: SSTR subtypes 2A, 2B, and 5 correlated significantly with SUV_max in meningioma patients. The developed radiomic model based on ADC maps effectively reproduces SUV_max using [68Ga]Ga-DOTATOC PET.

Keywords: ADC maps; Meningioma; Radiomic features; SUVmax; [68Ga]Ga-DOTATOC PET.