Pelvic lymph node dissection for prostate cancer

Introduction

Lymph node status is a primary factor in determining a patient’s prognosis and future therapy for prostate cancer. Lymph node involvement (LNI) is associated with a significant increase in recurrence and mortality. ^, Pelvic lymph node dissection (PLND) remains the most accurate means of determining lymph node status. Conventional cross-sectional imaging (computed tomography [CT] and magnetic resonance imaging [MRI]) has significant limitations in detecting LNI, with newer molecular imaging modalities such as prostate-specific membrane antigen (PSMA) and fluciclovine (Axumin) showing added accuracy, but seeming to lack specificity. ^,

While some men with limited LNI may have prolonged recurrence free survival following surgery with PLND, most will require additional therapy, and the therapeutic benefit of PLND for prostate cancer remains controversial. ^, Most retrospective analyses of this are limited by stage migration and the “Will Rogers phenomenon.”

Following our description of the technique for performance of PLND, including tips for minimization of morbidity, this chapter will review the guidelines and evidence around PLND in prostate cancer.

Guidelines and pelvic lymph node dissection

Contemporary guidelines from the European Association of Urology (EAU), National Comprehensive Cancer Network (NCCN), and American Urological Association (AUA) suggest that a PLND be performed in higher risk cases for the purposes of staging and prognosis. The EAU recommends an extended PLND (ePLND) for all patients with a risk for nodal metastasis greater than 5% based on preoperative risk nomograms, including the Briganti/Gandaglia nomogram, Memorial Sloan Kettering Cancer Center (MSKCC) or Partin nomograms, or the Roach formula. These nomograms utilize several preoperative variables—age, pretreatment prostate-specific antigen (PSA), clinical tumor stage—as well as biopsy data—Gleason grade and percentage of positive cores—to predict risk of LNI. They also have been extensively validated by several studies with no significant differences in predictive ability.

The rationale for the 5% risk cutoff focuses on maximizing diagnostic accuracy and potentially the therapeutic benefit, while minimizing missed diagnoses and additional morbidity associated with PLND. As an example, the negative predictive value of a 5% cutoff in the Briganti nomogram is 98.4%, sparing ePLND in 65% of patients while only missing 1.5% of nodal involvement cases. As these stratification systems are all based on preoperative data as well as systematic template prostate biopsy, a new Briganti nomogram was created to allow for MRI-targeted biopsy information. This novel nomogram utilizes a cutoff point of 7% risk, which would avoid unnecessary PLNDs in 70% of men at the cost of missing only 1.5% of cases with nodal involvement. The 2021 NCCN guidelines also support the use of ePLND as the optimal staging modality while also providing the potential for cure for certain individuals with limited micrometastatic disease. The NCCN has a lower cutoff of 2% for risk of nodal metastasis based on preoperative nomograms. AUA guidelines have a slightly different approach, though based on the same principles of targeting higher risk patients. These state that PLND can be considered for any localized prostate cancer patients undergoing radical prostatectomy and is recommended for those with unfavorable intermediate-risk or high-risk disease. While it is not explicitly mandated, these guidelines also suggest the benefits of a more extensive PLND in sampling more nodes, resulting in improved detection of nodal metastases.

Extent of pelvic lymph node dissection

While it is widely accepted that PLND improves staging of prostate cancer through detection of lymph node metastases, the question remains as to the extent of the lymph node dissection. Understanding the lymphatic drainage of the prostate and prostate cancer is crucial to maximizing the benefits of PLND. Mattei et al. studied this in 34 patients with organ-confined prostate cancer, utilizing single-photon emission computed tomography (SPECT)/CT and SPECT/MRI after intraprostatic injection of technetium (Tc-99m) nanocolloid to map the lymphatic landing sites. The accuracy of this technique was validated by intraoperative use of a gamma probe and ultimately by a super extended PLND (sePLND). Thirty-eight percent of the lymph nodes were located in the region of the external iliac vein and obturator fossa. An additional 25% of the nodes were within the region medial and lateral to the internal iliac artery, indicating that 63% of the positive nodes were in the field of what many describe as an extended template PLND. The other lymphatic regions for the prostate include the common iliac (16%), paraaortic/paracaval (12%), and presacral (8%). Another study with a similar approach—intraprostatic injection of technetium (Tc-99m) nanocolloid followed by planar scintigraphy, SPECT/CT, and sePLND—investigated the sites of lymph node metastases for localized prostate cancer. This revealed the following distribution: internal iliac region (35%), external iliac region (26%), obturator region (25%), presacral region (9%), and common iliac region (3%). It is clear that the lymphatic drainage is quite variable and may be affected by tumor location within the prostate gland as well as malignant obstruction of the predicted channels of lymphatic flow. As a result, PLND, especially if more extensive, is beneficial for staging because it encompasses multiple channels and regions. This is supported by many studies, showing that more extensive PLND improves nodal yield and detection of nodal metastasis. ^,

One newer area of interest to minimize the morbidity of PLND is sentinel lymph node biopsy. The concept behind this approach is to identify the exact pathway of lymphatic drainage and remove the first echelon node as a means of staging the disease. Therein, an ePLND may possibly be avoided if the sentinel node does not harbor malignancy. The two most common tracers are indocyanine green, which is combined with intraoperative fluorescence, and technetium (Tc-99m) nanocolloid, which is detected with the intraoperative gamma probe. Both can be injected into the prostate to define the lymphatic channels. Some studies show promising results, with one review concluding that sentinel node biopsy (SNB) has considerable diagnostic ability with a negative predictive value of 98% and sensitivity of 95%. However, several other articles have demonstrated quite variable results, with sensitivity as low as 50%. Overall, there is currently not enough evidence to support SNB as the sole staging modality, though it may provide additional information when utilized simultaneously with ePLND.

Is pelvic lymph node dissection therapeutic?

An area of great controversy is the therapeutic utility of PLND. Theoretically, removing micrometastatic disease that has solely spread to the lymph nodes could improve cancer control by decreasing recurrence, metastasis, and even potentially mortality. At this time, however, there is a dearth of evidence to support this notion. The majority of articles on this topic are retrospective studies with inconsistent definitions of PLND and variable degrees of follow-up. This makes it challenging to draw any definitive conclusions about the oncological benefits of PLND, especially when the effects of recurrent and metastatic prostate cancer may take several years to manifest in measurable outcomes. Thus, longer follow-up of at least 10 to 15 years is needed to provide a more accurate assessment of any therapeutic potential with PLND.

Earlier studies had revealed a positive impact of PLND on biochemical recurrence (BCR), particularly in patients with low volume node-positive disease. Bader et al. showed this by performing an ePLND, including external iliac, internal iliac, and obturator regions, on patients with clinically organ-confined prostate cancer. At a median follow-up of 45 months, 39% of patients with only one positive lymph node remained free of BCR. BCR-free survival was only 14% in those with more than two positive nodes. A study by Allaf et al. also described the potential benefit of ePLND in low volume node-positive prostate cancer. For patients with less than 15% positive lymph nodes, there was a significant improvement between standard PLND (sPLND)—external iliac and obturator regions—and ePLND—which also included the hypogastric region—with respect to 5-year PSA progression-free survival. There was also a nonstatistically significant trend toward improved survival in all patients who underwent ePLND ( P = .07). A follow-up study of this same cohort of node-positive men followed for at least 10 years revealed improved metastasis-free survival with ePLND. Importantly, patients were excluded from the aforementioned studies if they received adjuvant treatment with radiation or hormone deprivation, allowing for measurement of the potential merits of surgery alone. An additional retrospective study also found that removal of a higher number of lymph nodes in pathologic node-positive patients is associated with improved 10-year cancer-specific survival.

More recent studies have failed to find an oncological benefit with PLND for prostate cancer. Fossati et al. tackled this question in a 2017 comprehensive review. First, 21 retrospective comparative studies were used to compare oncological outcomes with any form of PLND versus no PLND. There was no difference in BCR, metastasis-free survival, or cancer-specific or overall mortality in patients who underwent any form of PLND, although a significant number of these studies included limited PLND (lPLND) only (or the extent of PLND was unknown). The same authors evaluated eight articles comparing limited/sPLND and sePLND and found no significant differences in BCR.

Certainly, there is a need for randomized prospective data and a few studies have tried to address this. Lestingi et al. measured early oncological outcomes in 300 patients with intermediate- or high-risk clinically localized prostate cancer, randomized to lPLND (obturator nodes) or ePLND (obturator, external iliac, internal iliac, common iliac, and presacral nodes). The BCR-free, metastasis free, and cancer specific survival was similar between the groups, though an exploratory subgroup analysis revealed a significant difference in BCR free survival for patients with grade group 3 to 5 prostate cancer undergoing ePLND. Another randomized prospective trial evaluated lPLND versus ePLND in 1440 patients with clinically localized prostate cancer and reported similar findings—there was no difference in BCR. A notable limitation in many of these studies is the short follow-up, which in the two aforementioned articles was 5 and 3 years, respectively.

The Will Rogers phenomenon should be factored into consideration in the interpretation of all of these studies. More extensive PLND for higher risk patients may transfer the more aggressive tumors with micrometastatic disease from a better prognostic group (N0) to a worse prognostic group (N1). Consequently, the prognostics and outcomes of both groups improve. For example, pN0 patients undergoing ePLND are more likely to truly have organ-confined prostate cancer. As a result, their oncological outcomes may appear better than pN0 patients who underwent no or only lPLND because some of those patients may, in fact, have had pN0 disease and had a truly worse prognosis. The difference in outcomes may not be related to the PLND, but rather to the improved diagnostics and accurate staging (stage migration) associated with removal of a high number of lymph nodes.

An important consideration is not just how the surgical removal of more nodes affects disease control, but also how the diagnosis of more advanced node-positive disease affects the utilization of additional therapies, as this could indirectly impact oncological outcomes. Selecting patients that would benefit from adjuvant hormone deprivation and/or radiation is dependent on accurate staging, and ePLND serves that purpose. Studies have examined this question—the benefit of adjuvant therapy in node-positive patients—and found an oncological benefit, particularly in patients with high-risk features. Touijer et al. retrospectively evaluated 1338 patients with lymph node metastases after prostatectomy and PLND who were managed with either observation, lifelong androgen deprivation therapy (ADT), or adjuvant ADT and radiotherapy (RT). Treatment with ADT and RT was associated with better overall survival (OS) and cancer-specific survival (CSS) than ADT alone or observation, and this difference was most pronounced in high-risk men. Because of the selection bias and the stratification of higher risk patients to the treatment arms, one might expect the opposite results, which provide even greater support for the use of adjuvant therapy. Another report found that adjuvant radiation in node-positive patients improves oncological outcomes more than salvage RT or observation, regardless of tumor characteristics. These findings were corroborated by a systematic review by Marra et al. Management of LNI can be tailored to the patient and tumor characteristics and the severity of the disease, and several strategies exist to improve oncological outcomes. In support of this, the EAU offers three management options for pN1 patients after prostatectomy and ePLND: observation, adjuvant ADT, or ADT and radiation.

Finally, any benefits of PLND must be weighed against the risks of the procedure itself. PLND is time-consuming and can add costs for both the patient and the facility. It is technically challenging and has the potential for intraoperative complications—vascular injury, nerve injury, and ureteral injury. There is also a significant risk of postoperative morbidity in the form of lymphocele or deep venous thrombosis (DVT). The likelihood of any of these complications during radical prostatectomy increases with the inclusion of a PLND in the procedure as well as the extent of PLND.

Anatomical extent of pelvic lymph node dissection

The extent of a PLND varies widely in the literature and among surgeons. A lPLND generally includes only the obturator nodes. A sPLND includes the obturator and external iliac nodes. An ePLND encompasses the obturator, external, and internal iliac nodes. And finally, a sePLND includes the ePLND template plus the common iliac and presacral nodes. The EUA defines an ePLND as the following: the nodes overlying the external iliac artery and vein, the nodes within the obturator fossa, and the nodes medial and lateral to the internal iliac artery. According to the NCCN, the boundaries of ePLND are the external iliac vein anteriorly, pelvic sidewall laterally, bladder wall medially, floor of the pelvis posteriorly, Cooper’s ligament distally, and internal iliac artery proximally.

(Refer to Table 16.1 for the anatomical landmarks/boundaries for each type of PLND. Figs. 16.1–16.8 demonstrate intraoperative images of the progression of steps in an ePLND. Please see Table 16.2 for key steps and Table 16.3 for special equipment required.)

Points of technique

Patient positioning with robotic PLND is unchanged from robotic prostatectomy. Port placement may be shifted cephalad, especially in the case of ePLND or sePLND in order to access nodes at or proximal to the iliac bifurcation. ^, Instrument choice is at the discretion of the surgeon. Control of the lymphatic drainage may be accomplished with the use of robotic or laparoscopic Weck Hem-o-lok polymer clips (Teleflex), laparoscopic titanium clips, or bipolar electrocautery with the Maryland bipolar forceps (Intuitive), fenestrated bipolar forceps (Intuitive), or a similar variant. In the only study to examine these modalities, prospectively randomized patients underwent robotic prostatectomy and ePLND to receive either titanium clips or bipolar coagulation to control lymphatic channels at the level of the femoral canal, and there was no difference in lymphocele rate between the two modalities.

When assessing the technical aspects of lymphatic control, it is important to consider the pathophysiology of lymphocele formation. The mechanism involves leakage from unsealed lymphatic channels. This most likely originates from either the proximal or distal end of the PLND, though there is incredible variability in the lymphatic network and the drainage may not follow the expected path perfectly from the femoral canal and node of Cloquet distally toward the iliac vessels proximally. Thus, it is recommended to employ a systematic approach to lymphatic control.

Several approaches have been adapted in attempts to prevent lymphatic leaks during PLND. The utility of intraoperative placement of a pelvic drain was evaluated in a prospective randomized trial and no difference was found in the rate of symptomatic lymphoceles at 90 days. This result has been validated by multiple groups. Another approach to control lymphatic leakage is the use of absorbable hemostatic agents. The data are mixed about the benefit of these products in terms of lymphocele formation. Waldert et al. demonstrated that the FloSeal hemostatic matrix (Baxter) may reduce the rate of symptomatic lymphocele from 14.5% to 3.1%. This was shown to be cost effective even when the product cost was taken into account. Studies exploring other hemostatic agents such as the TachoSil hemostatic patch (Nycomed-Takeda) and autologous fibrin glue (Vivostat) have revealed a lower incidence of radiographic lymphocele, but have failed to find a significant impact on the rate of symptomatic lymphoceles. ^, Though these findings are intriguing, it is difficult to interpret these studies and apply their findings broadly given the variability in extent of the PLND, technical application of the hemostatic agent, and postoperative follow-up.

A recent development in the approach to PLND is the attempt to maintain continuity of the lymphadenectomy bed with the peritoneal cavity. The peritoneal lining enables absorption of lymphatic fluid. In the case of extraperitoneal prostatectomy and PLND, there is no peritoneal lining within the area of dissection predisposing to lymphocele formation. In transperitoneal robotic prostatectomy and PLND, the region of the PLND should be in continuity with the peritoneum, hypothetically mitigating the chance of lymphocele(s). However, lymphoceles still form in a significant number of these cases. One theory to explain this is that the bladder folds over the PLND bed and isolates a cavity of lymphatic fluid, preventing continuity with the peritoneal lining. Several studies have evaluated the technique of a peritoneal interposition flap to prevent this from happening. In its initial description, this technique was applied in 78 patients and compared to 77 patients who did not receive it. In a follow-up of over 1 year, the interposition flap lowered the clinically significant lymphocele rate from 11.6% to 0%. Variations of this technique have had similar improvements in the rate of symptomatic lymphoceles without any adverse effects on nonlymphocele complications or postoperative urinary function. ^,

Finally, the timing of PLND during the operation can vary. PLND can be performed early or late in the procedure depending on surgeon preference. The difference is in relation to dissection of the space of Retzius and performing the bladder drop. No studies have investigated the timing of PLND in relation to other steps of the operation, though it is believed that this aspect likely does not have a significant impact on the PLND or complications. One potential advantage of performing ePLND or sePLND prior to the bladder drop is that the surgeon can take advantage of natural bladder retraction anteriorly via its attachments to the abdominal wall. This may achieve easier exposure of the common iliac region and even more proximal tissue, especially in the absence of an experienced assistant to aid in optimal retraction. Ultimately, though, the timing of the PLND should be left to the discretion and comfort of the surgeon.