European Urology

Pelvic lymph node dissection (PLND) represents one of the most intriguing topics in the management of prostate cancer (PCa). For years, the decision to perform PLND and to what extent has been left to the clinical judgment of each treating physician. As indeed often happens in medicine, the lack of a clear benefit of any treatment (either pharmacologic or surgical) opens the doors to a wide variety of experience-based approaches.

Some urologists did not consider PLND useful at all, since radical prostatectomy (RP) was reserved only for patients with favourable PCa and, thus, at low risk of nodal metastases. Patients with high-risk disease were not even treated with surgery due to the presumed high probability of being affected by systemic disease. All the recent series reporting excellent outcomes after RP in men with high-risk disease have clearly demonstrated the inadequacy of such approach [1]. For those few patients receiving surgery for more aggressive disease, PLND was often not extended beyond the obturator fossa. Again, this was due to the wrong assumption that lymph nodes located in this area represented the only primary lymphatic drainage of the prostate. In these cases, PLND was given only a (small) staging significance.

Conversely, other urologists did consider it useful to extend PLND to several pelvic lymphatic landing sites. This was done whenever RP was indicated and performed. These clinicians well knew the complex lymphatic drainage of the prostate. However, the main question was whether this extensive approach could improve disease control. In the view of these urologists, removal of micrometastatic disease could only be beneficial for patients, even in the absence of solid, prospective data. The scepticism of opponents was met most frequently with the question, “If this is true for the vast majority of cancers, why should it be different for prostate?”

Finally, other urologists did perform extended PLND (ePLND) in highly selected patients treated with RP. This was done to spare morbidity of ePLND to a certain proportion of patients with less aggressive disease. Why expose patients to higher rates of complications in the absence of a real proof of benefit of ePLND on cancer control? Morbidity has always been the Trojan horse for those reluctant to deliver ePLND. However, the equation more extensive PLND equals higher rates of complications has not always been confirmed [2] and [3]. Moreover, no serious concerns about complications were raised (at least not very often) when addressing the role of ePLND in bladder cancer, given the fact that a meticulous PLND can save lives in that setting. The risk in that case, urologists said, is justified.

This heterogeneity in personal views created heterogeneous patient series that precluded any objective well-performed evaluation of the role of ePLND on PCa control. Inadequate templates of PLND given on the basis of different indications by different surgeons were often merged in the same retrospective series. Not surprisingly, no association between PLND and PCa outcomes was found [4]. Similarly, none of the series in which meticulous PLND was invariably performed in all patients was able to assess the impact of ePLND for one main reason: the lack of a corresponding comparable group not receiving PLND.

So, how do we solve this problem? The answer to this question is coming from recent high-quality reports, such as the study by Joniau et al. [5]. These works have contributed to meeting two needs in this field: (1) standardisation of the anatomic extent of PLND and (2) indication for ePLND itself. Data coming from such studies have provided the basis for prospective evaluation of the benefit of PLND on cancer control.

1. Anatomic extent of and indications for extended pelvic lymph node dissection

First, we would like to state the premise that the extent of PLND should never be based on the number of lymph nodes removed. This is not a reliable proxy of the extent of PLND. For example, patients with bilateral PLND limited to the obturator fossa may have fairly high number of lymph nodes removed. However, they would remained significantly understaged. Conversely, some patients may have limited representation of lymphatic tissue within the pelvis. In this case, even an adequately performed ePLND would result in a relatively limited number of lymph nodes retrieved. Instead, the extent of PLND should be anatomically defined and include at least bilateral removal of all lymphatic tissue in the obturator fossa, along (above, medially, and laterally) the hypogastric and the external iliac vessels [6] and [7]. The reason for this extensive approach resides in the anatomic complexity of lymphatic drainage of the prostatic gland. Some authors have proposed to change the term extended to standard, since the extended approach should be the only one contemplated in PCa [8]. While we do not believe that this change in terms is strictly needed, all urologists should be aware that limited PLND is the equivalent of no PLND at all. This, therefore, would leave only two possible options: ePLND or no PLND.

Having said this, when ePLND is indicated in men with PCa, should it be further extended to other lymphatic landing sites in addition to those previously cited? Should it resemble what is currently done in bladder cancer, where presacral and common iliac lymph nodes are usually included in the ePLND template? In their study, Joniau et al, using planar scintigraphy and single-photon emission tomography (SPECT) after intraprostatic technetium 99m nanocolloid injection combined with a back-up ePLND performed by a single surgeon, proposed to extend ePLND to the presacral area [5]. This landing site was invaded by PCa in 9% of the cases (alone or in combination with other anatomic sites). Standard ePLND plus dissection of the presacral region would correctly stage 97% of patients and would remove all nodal metastases in 88% of them. Therefore, the authors suggest also including the presacral area in the ePLND template. This would improve staging without significantly increasing either complication rates or surgical time.

What about common iliac lymph nodes? Interestingly, although 19% of all lymph nodes potentially harbouring nodal metastases were preoperatively detected by SPECT in this area, dissecting the common iliac region as well did not increase the number of correctly staged N+ patients [5]. Moreover, positive lymph node density in the common iliac area was very low (1%), suggesting a low volume of nodal invasion.

These data are extremely interesting. Indeed, although the common iliac area has been suggested as a possible primary prostate lymphatic landing site on the basis of the simultaneous appearance at SPECT of these nodes with those in the regions close to the prostate [9], no skip lesion has ever been identified between the common iliac and lower pelvic areas [5] and [10]. In other words, no patient with positive common iliac lymph nodes had negative lymph nodes in the lower pelvic lymphatic landing sites. Based on these data, it is possible to hypothesise that common iliac lymph nodes represent the natural ascending pathway of spread of nodal metastases from lower to upper landing sites. Therefore, dissection of this area might be reserved for those patients with more aggressive cancer features (ie, one or several nodal metastases found in the lower pelvic areas). Frozen section might be helpful in deciding whether to further increase the extent of PLND or not. However, we still do not know whether extending nodal dissection up to this level really would be beneficial for patients due to their risk of having other concomitant pelvic and/or retroperitoneal nodal metastases [5], [9], and [10].

Despite a similar methodology, the study by Joniau et al. differs from the landmark study by Mattei et al. [9] with regards to the anatomic definition of the external iliac area. Indeed, the group from Bern defined the lateral limit of the external iliac dissection as the medial border of the external iliac artery [9]. According to these authors, lymph nodes located laterally to the external iliac vessels should be left untouched due to an imbalanced risk–benefit ratio in favour of complications. Conversely, Joniau et al. defined the external iliac area as an area limited laterally by the genitofemoral nerve and psoas muscle and medially by the medial border of the external iliac artery. Interestingly, 26% of nodal metastases were located in this area [5]. Although there might be a certain degree of overlapping between the two definitions depending on the case, this is an example of the heterogeneity in nodal dissections among surgeons even when an anatomic definition for ePLND is used.

Taken together, these data seem to suggest that the surgical approach used for ePLND in PCa would only slightly differ from that commonly used for bladder cancer. However, in contrast to bladder cancer, it is certainly true presently that not all PCa patients can be considered candidates for such an extensive approach. Patients at low risk of nodal invasion should not receive ePLND. However, it is likely that these patients do not even need RP at all [11]. Certainly, they do not need limited PLND.

Having said this, the characteristics of patients at a low risk for lymph node metastases are still to be clearly defined. Since risk grouping might be less accurate than an individualised risk-based approach, we recently proposed the use of a nomogram developed on a single-institution ePLND series based on routinely available preoperative data [12]. A 5% risk of positive lymph nodes has been suggested as the cut-off to decide to perform ePLND. While we are aware that any cut-off might be associated with a risk of missing a certain number of events, this approach also demonstrated excellent performance characteristics when externally validated [13] and it was incorporated in the updated European Association of Urology guidelines [7].

2. Prospective assessment of extended pelvic lymph node dissection on prostate cancer outcomes

One might argue anyway that all previous considerations would be of limited value in the absence of a solid proof supporting the role of ePLND in cancer control. Is such prospective randomised data available? Yes, they are now. Ji et al. recently published the first prospective assessment of the role of ePLND on PCa control [14]. The authors randomised 360 consecutive patients to either ePLND (including removal of hypogastric, obturator, and external iliac lymph nodes) or limited PLND (dissection limited to the obturator fossa and to the tissue along the external iliac vein) at a single institution. No patient received any form of neoadjuvant or adjuvant therapy. At a median follow-up of 74 mo, ePLND did show a significant benefit on biochemical recurrence (BCR)-free survival in patients with intermediate- and high-risk disease (12% and 20% higher BCR-free survival rates in favour of ePLND over limited PLND, respectively; all p≤ 0.04). Conversely, ePLND did not have any impact in men with low-risk disease. Although several limitations may apply to this study, we cannot ignore the first Level 1 evidence supporting the role of ePLND in intermediate- and high-risk disease in cancer control. Certainly, data focusing on patient survival as an end point instead of on BCR will have to come. However, given the low rates of cancer mortality after radical treatment, even in the high-risk setting, it is unlikely that such an end point will ever be reached.

Taken together, these data contribute to bring light into the shadows where PLND remained for years. We hope that these advances in the knowledge of PCa treatment will not be cancelled by the introduction of minimally invasive procedures. Use of such approaches has been reported as a major determinant not to performing ePLND, despite indications for it based on PCa features [15]. This, we believe, represents a dangerous step back in the treatment of the disease. We are confident that excellent studies like that reported by Joniau et al. [5] will contribute to changing this attitude in the near future.

Conflicts of interest

The authors have nothing to disclose.