Utility of Multiparametric Prostate MRI to Predict Regional or Distant Metastatic Disease Against Conventional Staging Using CT and Bone Scintigraphy or 68Ga-PSMA PET in Intermediate-to-High-Risk Prostate Cancer.

BACKGROUND

Multiparametric magnetic resonance imaging (mpMRI) is now the standard of care to guide prostate biopsies during workups and assessment of men with suspected prostate cancer (PCa). In addition to intraprostatic lesion detection, MRI usually covers the bony pelvis and pelvic lymph nodes, two of the commonest sites for metastatic disease. Subsequent staging has traditionally been based on further scanning using a combination of computed tomography (CT) and bone scintigraphy (BS), and more recently, positron emission tomography (PET) scanning with prostate-specific membrane antigen (PSMA) ligand. However, the value of additional staging investigations for men who are negative for metastatic disease on pelvic MRI is unclear. This study aims to evaluate the concordance of MRI findings with other imaging performed during staging.

METHODS

Patients with a Gleason score (GS) of ≥ 7 who had received both a pre-biopsy mpMRI and subsequent staging investigations from a single institution between 2019 to 2022 were identified. Imaging reports for PET, CT, and BS were used as the reference standard to evaluate MRI accuracy. PSMA-PET was considered the definitive outcome if multiple scans were performed. MRI findings were then classified as positive, negative, or equivocal. The accuracy was calculated using interpretations where equivocal cases were considered positive or negative for spread, representing a 'pessimistic' or 'optimistic' reading, respectively. A subgroup assessment of results considering only the use of CT + BS and PET was also done.

RESULTS

This study identified 214 patients for inclusion. The median age was 70 (IQR: 65-75) years, prostate-specific antigen (PSA) was 9.65 (IQR: 6.9-14.3) (ng/ml), and PSA density was 0.26 (ng/ml/cc) (IQR: 0.15-0.46). Complete conventional staging was performed for 130 patients, and PSMA-PET was performed for 102 patients. The results for the optimistic against pessimistic interpretations were the following: overall accuracy (90% vs 89%), sensitivity (0.48 vs 0.52), specificity (0.97 vs 0.95), negative predictive value (NPV) (0.84 vs 0.93), and positive predictive value (PPV) (0.71 vs 0.63). When comparing subgroup results considering only conventional imaging against only PSMA-PET, there were markedly more discordant findings in the PET group.

CONCLUSION

The impression of nodal and metastatic status through mpMRI poorly correlates with results from conventional staging and PSMA-PET. PSMA-PET more often produces discordant results to mpMRI, signifying an additive diagnostic value. MRI should not be used alone in the workup of prostate cancer in patients with a GS ≥ 7, where metastasis is a concern.