International Journal of Molecular Sciences (Sep 2023)

Direct Patlak Reconstruction of [<sup>68</sup>Ga]Ga-PSMA PET for the Evaluation of Primary Prostate Cancer Prior Total Prostatectomy: Results of a Pilot Study

  • Sazan Rasul,
  • Barbara Katharina Geist,
  • Holger Einspieler,
  • Harun Fajkovic,
  • Shahrokh F. Shariat,
  • Stefan Schmitl,
  • Markus Mitterhauser,
  • Rainer Bartosch,
  • Werner Langsteger,
  • Pascal Andreas Thomas Baltzer,
  • Thomas Beyer,
  • Daria Ferrara,
  • Alexander R. Haug,
  • Marcus Hacker,
  • Ivo Rausch

DOI
https://doi.org/10.3390/ijms241813677
Journal volume & issue
Vol. 24, no. 18
p. 13677

Abstract

Read online

To investigate the use of kinetic parameters derived from direct Patlak reconstructions of [68Ga]Ga-PSMA-11 positron emission tomography/computed tomography (PET/CT) to predict the histological grade of malignancy of the primary tumor of patients with prostate cancer (PCa). Thirteen patients (mean age 66 ± 10 years) with a primary, therapy-naïve PCa (median PSA 9.3 [range: 6.3–130 µg/L]) prior radical prostatectomy, were recruited in this exploratory prospective study. A dynamic whole-body [68Ga]Ga-PSMA-11 PET/CT scan was performed for all patients. Measured quantification parameters included Patlak slope (Ki: absolute rate of tracer consumption) and Patlak intercept (Vb: degree of tracer perfusion in the tumor). Additionally, the mean and maximum standardized uptake values (SUVmean and SUVmax) of the tumor were determined from a static PET 60 min post tracer injection. In every patient, initial PSA (iPSA) values that were also the PSA level at the time of the examination and final histology results with Gleason score (GS) grading were correlated with the quantitative readouts. Collectively, 20 individual malignant prostate lesions were ascertained and histologically graded for GS with ISUP classification. Six lesions were classified as ISUP 5, two as ISUP 4, eight as ISUP 3, and four as ISUP 2. In both static and dynamic PET/CT imaging, the prostate lesions could be visually distinguished from the background. The average values of the SUVmean, slope, and intercept of the background were 2.4 (±0.4), 0.015 1/min (±0.006), and 52% (±12), respectively. These were significantly lower than the corresponding parameters extracted from the prostate lesions (all p p > 0.05). Spearman correlation coefficient analysis demonstrated a strong correlation between static and dynamic PET/CT parameters (all r ≥ 0.70, p 68Ga]Ga-PSMA-11 perfusion and consumption and the aggressiveness of the primary tumor was observed. This suggests that the association between SUV values and GS may be more distinctive when distinguishing clinically relevant from clinically non-relevant PCa.

Keywords