Clinical Risk Based Associations of Lymph Node Dissection and Detection Yield Among Men Treated with Radical Prostatectomy for Prostate Cancer
Alejandro Abello, MD, Patrick A. Kenney, MD, Preston Sprenkle, MD, Michael Leapman, MD
Yale School of Medicine, New Haven, CT
Clinical Risk Based Associations of Lymph Node Dissection and Detection Yield Among Men Treated with Radical Prostatectomy for Prostate Cancer
Introduction and objectivesPelvic lymph node dissection (PLND) is recommended for men at risk for lymph node involvement at the time of radical prostatectomy (RP) yet is frequently omitted. We aimed to examine the probability of PLND based on clinical risk status, and to evaluate the impact of increasing lymph node yield on cancer detection rate across risk strata, with particular interest at the extremes of risk.
Methods
We queried the National Cancer Database to identify patients with clinically localized PCa who underwent RP as their primary treatment from 2004 to 2014. We extracted clinical and sociodemographic variables. Risk status was assessed using the Cancer of the Prostate Risk Assessment (CAPRA) score. We fit conditional logistic regression models to estimate likelihood of PLND and incremental value of increasing lymph node count by risk strata. As a secondary measure, we evaluated the association of PLND and increasing lymph node count with 30-day readmission.
Results
We identified 698,728 men with PCa treated with RP including 380.201 (54.41%) with PLND. Mean age at diagnosis was 62.6. PLND was omitted (Nx) in 56.1%, 31.4%, and 24.7% of patients with low, intermediate and high CAPRA-risk disease, respectively. Adjusting for clinical and pathologic factors, treatment in a community versus academic hospital (OR=1.62, 95% CI 1.59-1.66; P <0.001) and black race (OR=1.13, 95% CI 1.09-1.17, P: 0.01) was associated with pNx status. Increasing lymph node count was independently associated with greater likelihood of detection of lymph node metastasis in all risk strata. In patients at the lowest spectrum of risk (CAPRA-0), greater reported lymph node yield remained associated with detection of metastasis (relative to 0-10, 11-20 nodes: OR: 3.28 , 95% CI 3.06-3.53, P<0.001; 20-30 nodes: OR: 5.77 , 95% CI 5.16-6.45, P<0.001; >30 nodes OR: 7.90, 95% CI 6.56-9.51, P<0.001). In a multivariable model adjusted for clinical, pathologic, and demographic variables, reported lymph node counts greater than 30 were independently associated with higher odds of 30-day readmission (adjusted OR: 1.52, 95% CI: 1.02-2.25; P 0.03)
Conclusion
Among men treated in the United States at Commission on Cancer accredited hospitals, PLND continues to be omitted in a substantial proportion of intermediate and high risk patients. Increasing lymph node yield was associated with greater odds of detecting lymph node metastasis in all groups of patients, including those at the lowest level of risk by clinical criteria.
TABLE 1. Multiple variable model including predictors for N1 disease after RP + PLND
| Variable | Odds ratio | 95% CI | P value |
| Age at diagnosis | 0.99 | 0.99-1.00 | 0.160 |
| Median Income | 0.96 | 0.92-1.00 | 0.105 |
| Education | 0.968 | 0.927-1.012 | 0.158 |
| Distance to hospital | 1.00 | 0.99-1.00 | 0.212 |
| Race | 1.12 | 0.86-1.46 | 0.370 |
| Type of Hospital | 0.95 | 0.88-1.02 | 0.199 |
| Government Insurance | 1.13 | 1.04-1.22 | 0.003 |
| Facility Region Location | 0.986 | 0.97-0.998 | 0.028 |
| Capra Score 0 1 2 3 4 5 6 7 8 9 10 | Omitted 0.18 0.41 Ref. 1.59 3.63 5.26 9.01 10.19 11.65 19.05 | Omitted 0.12-0.27 0.33-0.51 Ref. 1.33-1.89 3.17-4.16 4.59-6.03 7.83-10.3 8.60-12.09 8.86-13.77 13.80-26.30 | Omitted <0.001 <0.001 Ref. <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 |
| LN count 1-10 11-20 21-30 >30 | Ref. 3.13 5.07 6.58 | Ref. 2.90-3.37 4.50-5.73 5.38-8.05 | Ref. <0.001 <0.001 <0.001 |
Back to 2019 Abstracts