BERKELEY, CA (UroToday.com) - Pelvic lymph node dissection (PLND) is most accurate method for detecting lymph node metastases.[1] However, the extent and indication of PLND is controversial, and most of the published literature is based on retrospective analyses. Without randomized prospective design, biases can occur in the analysis of oncologic outcomes. In a non-randomized observational study, investigators have no control over the selection of treatment (extended PLND vs. standard PLND) and the selection of patients to perform extended PLND is based on the cancer features. However, extended PLND provides more accurate staging information, affecting pathological nodal staging. For example, patients with lymph node metastases, diagnosed with extended PLND, may be classified as not having lymph node metastases should standard PLND be performed on them. This phenomenon is the well-known statistical artifact "Will Rogers phenomenon."[2] Thus, it will undoubtedly have a negative effect on the prognosis of the group with standard PLND, specifically if analyses stratified by the presence of lymph node metastases were performed. Multivariate analysis, adjusting lymph node staging, also would not be free of this artifact.
Propensity score methods are increasingly used in non-randomized study to reduce the impact of confounding which occurs when characteristics of patients influence treatment selection.[3] Propensity score is defined as an individual’s probability of receiving a specific treatment based on observed pretreatment variables, and investigators can balance the observed covariates by adjustments using the propensity score. Several adjustment methods using propensity score have been proposed, including regression adjustment, weighting, and matching. In this study, we aimed to compare the pathological and biochemical outcomes between extended PLND and standard PLND in patients who underwent robot-assisted radical prostatectomy for intermediate- or high-risk prostate cancer using propensity core matching. Propensity score matching was performed using preoperative covariates such as age, body mass index, prostate-specific antigen, clinical stage, biopsy, Gleason score, total number of biopsied cores, number of positive cores, and prostate volumes. This study included 170 patients who underwent robotic extended PLND and 294 patients who underwent robotic standard PLND, and propensity score matching resulted in 141 patients in each group. While patients subjected to extended PLND had unfavorable preoperative characteristics, there were no preoperative between-group differences in the matched cohort. In the matched cohort, the rate of lymph node metastases was significantly higher in the extended PLND group (12.1% vs. 5.0%, p=0.033). Although biochemical outcomes were not significantly different in the matched cohort, the trend of the hazard ratio was inverted in favor of extended PLND (3-year biochemical free survival rates, 77.8% and 73.5%, hazard ratio 0.85, p=0.497). A significant proportion of patients in our study had unfavorable pathologic features such as extracapsular extension, seminal vesicle invasion, or positive surgical margin. We speculated that they are still at increased risk of biochemical recurrence even though micrometastatic nodal disease was eliminated by PLND. However, the median follow up period was only 36 months, which is too short to draw meaningful conclusion regarding oncologic outcomes, and we all know that biochemical recurrence does not necessarily translate into clinical metastasis or cancer death.
Recently, the first prospective randomized trial which examined the impact of extent of PLND was published from China.[4] Median follow up was 74 months, and extended PLND was associated with improved biochemical outcomes in patients with intermediate- and high-risk prostate cancer.[5] Also, phase 3 randomized trials comparing extended vs limited PLND are ongoing, and these trials will evaluate the role of PLND on long-term outcomes with cancer-specific and overall survival.[6] In light of this, long-term follow up of our cohort would be also valuable in the robotic era; all of the patients in our cohort underwent PLND using robotic system, and we believe robotic PLND can be performed with comparable staging accuracy and oncologic outcomes to open technique.
References:
- Briganti A, Blute ML, Eastham JH et al. Pelvic lymph node dissection in prostate cancer. Eur Urol 2009; 55: 1251.
- Gofrit ON, Zorn KC, Steinberg GD et al. The Will Rogers phenomenon in urological oncology. J Urol 2008; 179: 28.
- Kurth T, Walker AM, Glynn RJ et al. Results of multivariable logistic regression, propensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect. Am J Epidemiol 2006; 163: 262.
- Ji J, Yuan H, Wang L et al. Is the impact of the extent of lymphadenectomy in radical prostatectomy related to the disease risk? A single center prospective study. J Surg Res 2012; 178: 779.
- Prospective study comparing extended with limited pelvic lymphadenectomy in intermediate and high risk prostate cancer patients undergoing radical prostatectomy (LFD). NCT01812902. www.clinicaltrials.gov.
- Prospective study to compare a limited versus extended pelvic lymphadenectomy during prostatectomy. NCT01555086. www.clinicaltrials.gov.
Written by:
Kwang Hyun Kim, MD and Koon Ho Rha, MD, PhD as part of Beyond the Abstract on UroToday.com. This initiative offers a method of publishing for the professional urology community. Authors are given an opportunity to expand on the circumstances, limitations etc... of their research by referencing the published abstract.
Department of Urology, Urological Science Institute, Yonsei University College of Medicine, Seoul, Korea
More Information about Beyond the Abstract