Keywords:
Neoplasia, Contrast agent-intravenous, MR-Diffusion/Perfusion, MR, Oncology, Genital / Reproductive system male, Genitourinary, Prospective, Observational, Performed at one institution
Authors:
S. cipollari1, G. Anello1, F. Giganti2, M. Pecoraro1, D. Fierro1, R. Campa3, M. Del Monte1, C. Catalano1, V. Panebianco1; 1rome/IT, 2London/UK, 3Roma/IT
DOI:
10.26044/ecr2020/C-11563
Results
A 4-point score was elaborated to define the likelihood of local recurrence, based on the findings on DCE, T2W and DWI. In addition, we described the imaging characteristics of other tissues that might lead to a misdiagnosis, consisting with fibrosis, residual gland or inflammation.
Following the evaluation of the finding in each sequence, evaluated as "positive" or "negative" considering the presence or absence of several suspect characteristics in each sequence, an algorithm was developed with assessment categories (from 1 to 4) that reflect the increasing probability of PCa recurrence. Imaging results were validated with standard of reference consisting of at least one of the following: 1) concordant findings on following imaging with the same technique or with other correlative imaging techniques 2) PSA follow-up before and after treatment directed to the suspicious area at imaging (undetectable PSA after therapy is interpreted as evidence of disease in the treated lesions and absence of disease in another site). Patients who had been assigned the score of 3 and 4 (respectively n=18, 27.7%; n=20, 30.8%) underwent salvage RT, with undetectable PSA after treatment in all cases. Patients with score 2 (n=13, 20%) underwent PET with PSMA, with detection of metastatic lymph nodes in 3 patients, a distant metastasis in 5 patients and local recurrence in 3 cases. The fourteen patients (21.5%) who were assigned score 1 are still under observation (follow up from 5 to 8 months) with PSA control every 2 months, with currently PSA stability.
The sensitivity of the final score assigned to each finding on the basis of the mpMRI characteristics was 100%, while the specificity was 91.7%. The PPV was 95% and the NPV 100%. The diagnostic accuracy was 97%. Patients risk stratification based on GS and PSA-DT could not be performed
because patients underwent mpMRI shortly after (within 12 months) the onset of BCR, therefore it was not possible to assess whether PSA-DT was higher or lower than one year. All the findings we reported were in line with the literature confirming that the DCE is the dominant sequence for PCa local recurrence detection in patients treated with radical prostatectomy.
Limitations of our study include the absence of histological report, of external validation and small study population.