Targeting All Multiple Magnetic Resonance Imaging Prostate Lesions Does Not Enhance Cancer Detection: Insights from the YAU Prostate Cancer Group

Abstract

Background and objective

Although diagnostic efficacy of multiparametric magnetic resonance imaging (MRI) in identifying index lesions (ILs) in prostate cancer (PCa) patients is well established, challenges arise when multiple lesions (MLs) are present. Determination of an optimal biopsy strategy for these patients is crucial. This study aims to assess the risk of detecting PCa and clinically significant PCa (csPCa; International Society of Urological Pathology [ISUP] grade group ≥2) when targeting suspicious MRI MLs in addition to the IL.

Methods

We included 1310 biopsy-naïve patients with only a single MRI lesion (SL) and 621 men with MLs. Patients underwent a subsequent targeted biopsy (TBx) of each lesion, along with a systematic biopsy (SBx). We compared TBx alone versus TBx + SBx and evaluated whether the presence of MLs increases the risk of PCa and csPCa using a multivariable logistic regression model (MVA), while accounting for confounders.

Key findings and limitations

Overall, PCa was detected in 57.6% and 51.2% of IL-TBx for the SL and ML groups, respectively (p ≤ 0.01). The rates of detection of csPCa with IL-TBx were 46.2% for the SL group and 36.1% for the ML group (p < 0.01). When combining all TBx and SBx procedures, PCa was detected in 63.8% for the SL group and 67.0% for the ML group (p = 0.2), while csPCa was detected in 54.1% for the SL group and 48.1% for the ML group (p = 0.01). SBx yielded PCa detection rates of 58.5% for the SL group and 56.7% for the ML group (p = 0.5) and csPCa detection rates of 42.1% for the SL group and 38.8% for the ML group (p = 0.2). ISUP upgrading targeting the 2° and 3° lesions was found in 14 (12.2%) and six (5.2%) cases, respectively. In the MVA, the presence of MLs was identified as an independent predictor of PCa (odds ratio [OR]: 0.6, 95% confidence interval [CI]: 0.5–0.8, p < 0.01) and csPCa (OR: 0.5, 95% CI: 0.4–0.6, p < 0.01) in TBx and combined TBx + SBx (PCa: OR: 0.6, 95% CI: 0.5–0.9, p = 0.02, and (csPCa: OR: 0.4, 95% CI: 0.3–0.7, p < 0.01), respectively.

Conclusions and clinical implications

Patients with MLs have a lower risk of PCa and csPCa. In case of MLs, a TBx of the IL + a concomitant SBx allows for the diagnosis of the vast majority of PCa and csPCa cases. The added value of a second TBx on the non-IL is modest, including only a 5.6% increase in the diagnosis of csPCa.

Patient summary

We studied whether targeting multiple suspicious areas on prostate magnetic resonance imaging increases cancer detection. We found that patients with multiple lesions had a lower risk of prostate cancer than those with a single lesion. Targeting the largest suspicious area along with sampling the entire prostate led to the detection of most cancer cases, with little added benefit from targeting the other smaller areas.