Tertiary Gleason Pattern 5 and a High Genomic Classifier Score Better Predicts Lymph Node Invasion at Radical Prostatectomy in 20,441 Intermediate-Risk Disease Patients

Abstract

<h3>Purpose/Objective(s)</h3> The POP-RT and SSPORT trial showed that pelvic lymph node radiation therapy (PLNRT) can improve prostate or bed-only oncologic outcomes. Grade groups 2 (GG2) or 3 (GG3) and Tertiary Gleason pattern 5 (TP5) are heterogeneous patient groups with varied prognoses. Genomic classifiers (GC) may provide additional information essential to identifying optimal pelvic radiotherapy candidates. We hypothesized that TP5, and a GC, could improve prognostication of lymph node invasion (LNI) after radical prostatectomy (RP) over traditional clinicopathologic risk factors alone. <h3>Materials/Methods</h3> The Decipher GRID database (Veracyte, San Diego, CA) contains de-identified GC testing and pathology results for patients tested between November 2015 and March 2020. Intermediate risk patients with GG2 or GG3 were included in the analysis of continuous GC scores (range 0-1) and GC risk groups (low, intermediate, and high). Multivariable logistic regression (MVA) was used to quantify the relationship between LNI with GC, TP5, and other clinicopathologic features (age, race, extraprostatic extension, seminal vesicle invasion, bladder neck invasion). <h3>Results</h3> 20,441 patients were included (11,934 with GG2 (58.4%) and 8,507 (41.6%) with GG3 disease). Of those with GG2 and GG3 disease, 1,451 (12.2%) and 2,911 (34.2%) had TP5, respectively. Patients with TP5 had a significantly higher GC score (median [interquartile range (IQR)]: 0.39 [0.24-0.61] vs 0.53 [0.32-0.75], p<0.001 for GG2 and median [IQR]: 0.58 [0.35-0.81] vs 0.66 [0.43-0.86, p<0.001 for GG3. In MVA adjusted for clinicopathologic features, TP5 and higher GC score were independently associated with the presence of LNI in both GG2 (TP5: odds ratio [OR] 1.60, 95% CI 1.07-2.38, p=0.02 and GC: OR 1.20 per 0.1 increase, 95% CI 1.11-1.29, p<0.001) and GG3 (TP5: OR 1.47, 95% CI 1.16-1.87, p<0.01 and GC: OR 1.13 per 0.1 increase, 95% CI 1.07-1.19, p<0.001) patients. TP5 patients with GG2 or GG3 and high GC have a 3.8- and 1.8-fold risk of LNI over those with a low GC, respectively. Patients with GC high and GG2 or GG3 have a 2.7- and 1.8-fold risk of LNI over those with or without TP5, respectively (Table 1). <h3>Conclusion</h3> TP5 was associated with higher GC scores and adverse clinicopathologic features. TP5 and higher GC are independently associated with LNI while controlling for other clinicopathologic risk factors, demonstrating their importance in future studies of pelvic RT treatment intensification or de-intensification for these patients.