Abstracts

RETURN TO ABSTRACT LISTING


E1752. A Single Centre Correlation of Lesional 68Ga-PSMA PET SUV With Multiparametric Prostate MRI Apparent Diffusion Coefficient (ADC) Values
Authors
  1. David Kusuma; Royal North Shore Hospital
  2. Joanna Kao; Royal North Shore Hospital
  3. Geoffrey Schembri; Royal North Shore Hospital
  4. Michael Chew; Royal North Shore Hospital
Objective:
PIRADS is the standard for prostate cancer lesion characterisation and overall risk categorisation using multiparametric MRI (mpMRI) [1]. The combination of 68Ga-PSMA PET/CT with mpMRI has demonstrated higher sensitivity (89%) versus mpMRI alone (76%), with minimal reduction in specificity (96% vs 88%) [2]. This combination is becoming increasingly codependent. MaxSUV has been shown to be inversely proportional to apparent diffusion coefficient (ADC) values, both important and quantitative markers of prostate cancer aggressiveness, however prior studies have used only modest sample sizes (20-30 patients) [3-5]. ADC has also been shown to be more accurate in the peripheral zone (PZ) lesions than Transition zone (TZ). This study assesses the relationship between lesion PSMA SUV and ADC values overall, as well as in the PZ and TZ subgroups in 59 patients.

Materials and Methods:
Single-centre retrospective study reviewing patients from 2017-2020 undergoing both mpMRI and PSMA PET/CT for new prostate cancer. PIRADS 1 and 2 lesions were excluded due to low sample size and routine non-reporting of ADC values. Lesion concordance between modalities was performed by both visual and report interpretation.

Results:
59 patients were identified (median age 67.4 years). 76 lesions with SUVmax and ADC values were detected (66 PZ, 10 TZ). The mean SUVmax values within each PIRADS group did demonstrate an inverse trend to ADC values, however individual SUVmax to ADC values demonstrated only a weak correlation r=-0.07. This was not statistically significant (p=0.54). Sub-group analysis showed no correlation in either PZ or TZ locations (r= -0.12 and -0.23 respectively). The mean SUV was similar between each subgroup (PZ 14.6, TZ 17.8), as was ADC (PZ 0.71, TZ 0.72) with no significant difference.

Conclusion:
This study of 59 patients demonstrates no correlation between individual PSMA SUVmax and ADC values. Further subgroup analysis by location also demonstrated no significant correlation between each group for these values. This suggests each of these indices is reflecting a different facet of tumour biology.