Ability to detect LNI relates to extensiveness of PLND

September 1, 2006

Atlanta-The ability to detect lymph node invasion (LNI) in men undergoing radical prostatectomy relates directly to the extensiveness of pelvic lymph node dissection (PLND), a review of a large clinical series suggests.

Atlanta-The ability to detect lymph node invasion (LNI) in men undergoing radical prostatectomy relates directly to the extensiveness of pelvic lymph node dissection (PLND), a review of a large clinical series suggests.

The rate of LNI detection increases to about 90% with a PLND involving 28 nodes. More extensive node sampling does not increase the LNI yield. In contrast, a limited PLND of 10 or fewer nodes has a low yield for detection of LNI and probably can be eliminated from use in clinical practice.

"The yield of PLND, when less than 10 nodes were removed, is less than 10%," Alberto Briganti, MD, a urologist at the University of Milan, Italy, told Urology Times. "There is a marked increase in the rate of lymph node invasion when 10 to 30 nodes are removed. A plateau is reached when over 28 nodes are removed, and the probability of detecting lymph node invasion exceeds 90%."

The ideal number of nodes that should be removed to achieve optimal cancer staging at radical prostatectomy has not been determined. In an attempt to define that ideal, investigators analyzed data from 858 prostate cancer patients who predominantly underwent extensive PLND prior to radical prostatectomy. Using logistic regression models, they assessed the prognostic value of four clinical variables for detection of LNI: total PSA at diagnosis, clinical stage, biopsy Gleason sum, and number of lymph nodes removed and examined.

The PLNDs revealed positive nodes in 88 cases (10.3%) and negative nodes in 770 (89.7%). Patients with LNI were significantly more likely to have stage T3 disease (17.1% vs. 2.5%) than T1c disease (38.6% vs. 57.1%, p<.001). LNI was also significantly more likely to be associated with a biopsy Gleason sum of 7+ (70.5% vs. 29.0%, p<.001). Mean baseline PSA was 8.8 ng/mL in men without LNI versus 12.1 ng/mL in those with LNI (p<.001). Number of nodes evaluated averaged 15 in men without LNI compared with 18 in men with LNI (p<.001), who averaged three positive nodes.

Detection of 90%

Sensitivity analysis based on receiver operator characteristics curve showed that examination of 10 nodes was associated with an LNI detection rate of 12%, increasing to about 90% when 28 nodes were examined. Sensitivity for detection of LNI did not significantly improve further when as many as 40 or more nodes were sampled. Virtually identical findings emerged when the analysis was restricted to men who had a biopsy Gleason sum of 7+.

The results also provided validation for a nomogram to predict LNI in patients who undergo extensive PLND, said Dr. Briganti. The number of nodes removed correlated significantly with the likelihood of LNI, as did age, clinical stage, biopsy Gleason sum, and PSA distribution. Previous risk assessment tools had all been validated in patients with limited PLND.

Despite the findings, the clinical significance of the link between extent of PLND and risk of LNI remains unresolved.

"What we know is that the percentage of positive nodes removed is highly associated with cancer-specific survival and biochemical recurrence," Dr. Briganti said. "Unfortunatley, there is no study showing that extent of PLND in patients with positive nodes is associated with improved survival."

The investigators also found a disparity in the risk of LNI between patients treated in Europe and North America. European patients had a six-fold greater risk of LNI after controlling for other variables.

"The variables we used could not account for this difference, which means we need to look at other variables," Dr. Briganti said.