Is There an Oncological Benefit of Performing Bilateral Pelvic Lymph Node Dissection in Patients with Penile Cancer and Inguinal Lymph Node Metastasis?

We aim to review the literature for studies investigating the oncological outcomes of patients with penile cancer (PC) undergoing bilateral pelvic lymph node dissection (PLND) in the presence of inguinal lymph node metastasis (LNM) who are at risk of harboring pelvic metastasis. A search of English language literature was performed using the PubMed-MEDLINE database up to 3 December 2020 to identify articles addressing bilateral PLND in PC patients. Eight articles investigating bilateral PLND met our inclusion criteria. Patients with pelvic LNM have a dismal prognosis and, therefore, PLND has an important role in both the staging and treatment of PC patients. Ipsilateral PLND is recommended in the presence of ≥2 positive inguinal nodes and/or extranodal extension (ENE). Significant survival improvements were observed with a higher pelvic lymph node yield, in patients with pN2 disease, and in men treated with bilateral PLND as opposed to ipsilateral PLND. Nevertheless, the role of bilateral PLND for unilateral inguinal LNM remains unclear. Although the EAU guidelines state that pelvic nodal disease does not occur without ipsilateral inguinal LNM, metastatic spread from one inguinal side to the contralateral pelvic side has been reported in a number of studies. Further studies are needed to clarify the disseminative pattern of LNM, in order to establish PLND templates according to patients’ risk profiles and to investigate the benefit of performing bilateral PLND for unilateral inguinal disease.


Introduction
The development of lymph node metastases (LNM) in penile cancer (PC) follows the anatomical loco-regional drainage route and is characterized by a stepwise disseminative pattern [1]. The presence and extent of LNM is the single most important prognostic factor in determining long-term survival in men with invasive PC [2]. The European Association of Urology (EAU) and the National Comprehensive Cancer Network (NCCN) guidelines recommend performing an inguinal lymph node dissection (ILND) in patients with a Union for International Cancer Control TNM stage ≥ T1bG2 [3,4]. The number, diameter, and extracapsular nodal extension (ENE) of inguinal metastases have been established as independent risk factors for pelvic lymph node involvement, which carries a poor prognosis. Furthermore, the proportion of pelvic LNM was shown to increase from 0% to 57.1% in cases with no risk factors and when all three are present, respectively [5]. Therefore, ipsilateral pelvic lymph node dissection (PLND) is recommended as an adjunct to ILND in patients with ≥2 inguinal LNM on one side and/or ENE for proper lymphatic assessment and staging [3].
Nevertheless, controversy exists regarding the therapeutic role of PLND since the proportion of patients with pelvic LNM among those with inguinal LNM is highly variable, and it has been reported that only a fraction of patients with pelvic LNM may benefit from PLND alone [6]. Other studies suggest that PLND has a curative role only in patients with a single microscopic focus in the pelvic specimen [7]. A further polemic is whether PLND should be performed ipsilaterally only or bilaterally in patients with unilateral inguinal LNM since crossover from inguinal to contralateral pelvic nodes has not been well studied [8]. The rarity of PC and the lack of randomized studies preclude high-level evidence recommendations in the management of pelvic lymph nodes, and its benefit in low-risk patients remains to be confirmed.
This study aimed to retrieve articles addressing the oncological outcomes of patients with penile cancer (PC) undergoing bilateral PLND for inguinal LNM at risk of harboring pelvic LNM.
The titles and abstracts of identified articles were retrieved for evaluation. The full text of potentially eligible articles was independently screened against the study selection criteria by two independent reviewers (M.B. and R.S.). Disagreement was resolved by discussion; if no agreement was reached, a third independent party acted as an arbiter (A.M). Figure 1 shows the Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) flow diagram detailing the search strategy and identification of studies used in the evidence synthesis.

Inclusion and Exclusion Criteria
According to the PRISMA statements, the PICOS: Population (P), Intervention (I), Comparator (C), Outcomes (O), and Study design (S), approach was used to specify the eligibility criteria. Therefore, studies were considered eligible if PC patients (P) were managed by bilateral PLND for inguinal LNM (I) as a single-arm or compared to patients in whom unilateral PLND or no PLND was performed (C) and if their oncological outcomes (O) were assessed in retrospective or prospective studies (S). After article selection and according to the eligibility criteria, the following types of studies were excluded: articles not describing bilateral PLND, articles not written in English, review articles, conference abstracts, and case reports.

Inclusion and Exclusion Criteria
According to the PRISMA statements, the PICOS: Population (P), Intervention (I), Comparator (C), Outcomes (O), and Study design (S), approach was used to specify the eligibility criteria. Therefore, studies were considered eligible if PC patients (P) were managed by bilateral PLND for inguinal LNM (I) as a single-arm or compared to patients in whom unilateral PLND or no PLND was performed (C) and if their oncological outcomes (O) were assessed in retrospective or prospective studies (S). After article selection and according to the eligibility criteria, the following types of studies were excluded: articles not describing bilateral PLND, articles not written in English, review articles, conference abstracts, and case reports.

Data Extraction and Analysis
A data extraction form was developed to collect information on author and journal, study design and period, purpose of investigation, number of patients included, patient age, type of intervention, TNM stage, tumor differentiation and lymphovascular invasion, ENE, PLND template, complications, local, regional, or systemic recurrences, neo/adjuvant treatment, follow-up period, lymph node yield and positivity, and survival analysis.

Data Extraction and Analysis
A data extraction form was developed to collect information on author and journal, study design and period, purpose of investigation, number of patients included, patient age, type of intervention, TNM stage, tumor differentiation and lymphovascular invasion, ENE, PLND template, complications, local, regional, or systemic recurrences, neo/adjuvant treatment, follow-up period, lymph node yield and positivity, and survival analysis. Two reviewers independently extracted data for further assessment of qualitative and quantitative evidence synthesis. Descriptive statistics were used for baseline data and a narrative synthesis of the evidence was performed. Table 1 summarizes the demographic, baseline characteristics, and perioperative data of studies which included patients treated with bilateral PLND for PC. Eight studies were included that reported on the oncological outcomes of bilateral PLND in 619 patients with PC who underwent PLND, of which 420 were bilateral. All eight studies reported the anatomical boundaries of the PLND templates, of which three described the surgical technique for PLND in detail [7,9,10]. Seven studies were of a retrospective nature, while one study performed ilioinguinal lymph node dissections prospectively [10]. Table 2 summarizes the inguinal and pelvic lymph node status and survival analysis of patients treated for PC.     Optimal PLND may extend to the common iliac artery, including common iliac, external iliac, internal iliac, and obturator LNs.

Prophylactic PLND
Djajadiningrat et al. prophylactically treated 79 chemonaïve PC patients with PLND, of which 23 (29%) were bilateral, when ≥2 positive inguinal LN or ENE were found. Tumor-positive pelvic nodes were found in 19 (24%) patients, and both inguinal ENE and ≥2 positive inguinal LN were identified as predictors of pelvic nodal involvement. A plausible explanation for ENE predicting pelvic LNM may lie in tumors' aggressiveness and ability to penetrate the lymph node capsule. Moreover, patients with positive pelvic LNM had worse 5-year DSS compared to patients without pelvic LNM (17% vs. 62%). There was no comparison of outcomes between ipsilateral and bilateral PLND [7].

Bilateral PLND vs. No-PLND
In a multicenter study by Li et al., 69 patients underwent bilateral PLND and were compared to 121 patients in whom PLND was spared. Among the PLND group, 16 patients had unilateral and five patients had bilateral pelvic LNM, and no patient had crossover metastatic spread from one inguinal side to the contralateral pelvic side. The PLND group did not demonstrate higher 1-and 3-year DSS rates than the no-PLND group (65.7% and 39.0% vs. 65.4% and 39.6%, p = 0.796). However, propensity score matching among pN2 patients showed higher DSS in patients who underwent PLND compared to those who did not (83.3% and 83.3% vs. 69.0% and 50.2%, p = 0.030). Based on their analyses, a significant benefit was found for a subset of pN2 cases, while pN3 patients may not benefit from bilateral PLND [12].

Bilateral vs. Unilateral PLND
In another multicentric study, Zargar-Shoshtari et al. investigated criteria predicting bilateral pelvic LNM in patients with pathologically confirmed inguinal nodes. In a cohort of 140 patients, 83 had both bilateral inguinal and pelvic LNM, and bilateral PLND was performed in 64 (77%) patients. Of note, 27 (32.5%) patients received neoadjuvant chemotherapy (NAC). The presence of ≥4 positive inguinal nodes had a sensitivity of 95% and an area under the curve (AUC) of 0.76 for predicting bilateral pelvic LNM, which was present in only one patient with <4 inguinal LNM. In a separate analysis which excluded patients who received NAC, ≥4 inguinal LNM was still the strongest predictor for bilateral disease; however, it did not reach statistical significance. The rate of regional failure was higher for patients who only had ipsilateral PLND compared to bilateral PLND (47% at 5.53 vs. 20% at 12.8 months, p = 0.02). Additionally, there was a trend for improved OS in men treated with bilateral PLND compared to unilateral (11.8 vs. 10.9 mo., p = 0.10), and OS was significantly lower with ≥4 positive inguinal LNM (8.7 vs. 15.4 months, p = 0.04) [13].
In a subsequent multicentric study, the same authors assessed the oncological outcomes of patients treated with bilateral PLND in the presence of unilateral pelvic LNM. In 51 patients with unilateral inguinal LNM and positive pelvic nodes, 38 (75%) had ipsilateral and 13 (25%) bilateral PLND. In the 13 patients with bilateral PLND, metastases were only found in ipsilateral pelvic nodes. It is important to note that 86% of the cohort received preoperative and postoperative chemotherapy, radiotherapy, or both, and a higher proportion of adjuvant radiotherapy was utilized in the unilateral PLND patients (50% vs. 8%, p = 0.01). Despite this, the median OS was significantly longer in bilateral PLND patients compared to unilateral (21.7 vs. 13.1 mo., p = 0.05) at a median 13.3 months follow-up. Cancer-specific survival (CSS) was statistically similar in bilateral and ipsilateral PLND, but regression analysis showed that bilateral PLND, adjusted for additional therapies and presence of multiple pelvic LNM, improved CSS [14].

Lymph Node Mapping
Recently, Yao et al. provided an accurate dissemination map of LNM and aimed to determine the extent of PLND. One hundred and twenty-eight chemo-radio-naive patients received bilateral ILND, 111 received bilateral PLND, and 17 received unilateral PLND. Pelvic LNM was present in 57 (44.5%) patients and bilateral LNM was found in 13 patients. The median number of pelvic LNs removed was 18 (IQR: 10-30). All 57 patients with PLNM had inguinal LNM. As opposed to other studies where crossover was not found [9,10,12], two patients with unilateral inguinal LNM had bilateral PLNM, suggesting crossover from inguinal nodes to contralateral pelvic nodes. Lopes et al. similarly documented metastasis to the pelvic nodes with no inguinal involvement [15].
OS was significantly longer in patients that underwent bilateral PLND than those with unilateral (30 vs. 18 mo., p = 0.004). Among the 57 patients with pelvic LNM, the estimated median OS was higher in the bilateral than in the unilateral group (16 vs. 11 mo., p = 0.07). Moreover, the prevalence of pelvic LNM was similar in patients with ENE and without ENE (42.9% vs. 45.8%, p = 0.74), which questions the validity of ENE for predicting pelvic LNM, as has been reported by a multitude of studies [8].

Conclusions
There is a paucity of data on the disseminative pattern of pelvic LNM in patients with PC. Moreover, there are scarce data comparing the outcomes of patients treated with bilateral vs. unilateral PLND in the presence of unilateral inguinal disease. Although the EAU guidelines state that crossover metastatic spread from one groin to the contralateral pelvis has never been reported [3], a number of studies have found metastases to pelvic nodes without ipsilateral inguinal involvement [8,12,15]. Interestingly, despite the rarely found crossover, data gathered in the current review suggest that bilateral PLND is beneficial, but the ideal candidate is yet to be determined.
Among the justifications for performing bilateral PLND is the improved survival observed compared to patients treated unilaterally, and particularly in the pN2 subset of patients, both of which may correlate with the treatment of occult micrometastatic disease. Moreover, a possible explanation for the observation of minimal crossover metastatic spread to contralateral pelvic LNs in patients with unilateral inguinal LNM may be disease underestimation even when invasive nodal staging is performed.
Despite patients with invasive or high-grade PC seemingly benefiting from undergoing extensive lymph node dissection, both ILND and PLND are characteristically associated with considerable physical and psychological morbidity. However, only a single study reported data on postoperative complications in patients that underwent bilateral PLND, where 14 patients (18%) had some type of complication, 9 of these wound-related and 5 non-wound-related such as pneumonia, delirium, and ileus [7]. Modern series report complication rates of 42-57% for all patients undergoing ILND, the most common being wound necrosis and dehiscence, lymphocele, lymphoedema, and venous thromboembolism [16]. Several strategies have been recommended to prevent adverse effects of ILND such as prophylactic antibiotics until drain removal, early ambulation, debridement of non-viable skin, meticulous control of lymphatics, preservation of the saphenous vein, and antiembolism stockings, none of which have been studied in prospective trials [17]. Contemporary minimally invasive management strategies for LN involvement aim to minimize the morbidity associated with traditional radical inguinal and pelvic lymphadenectomy through appropriate risk stratification whilst optimizing oncological outcomes.
Due to inherent methodological limitations, false-negative rates of inguinal and pelvic lymphadenectomy remain unknown. The implementation of new imaging modalities such as 18 FDG-PET/CT into clinical workflows may improve the diagnostic accuracy for the detection of LNM [18][19][20]. A recent Danish study by Jakobsen et al. found 18 FDG-PET/CT to have a groin false-negative rate of 2.1% and a sensitivity of 85.4% per patient for assessing inguinal lymph node metastasis and distant metastasis compared to 47.5% for the conventional CT scan [21]. Graafland et al. reported a sensitivity of 91% and a specificity of 100% with PET/CT for pelvic staging [6]. In contrast, for dynamic sentinelnode biopsy, false-negative rates of up to 12-15% have been reported [22]. Consequently, these approaches might help to address challenges in adequately identifying patients that might benefit from bilateral PLND and assist in treatment decision making.
However, specific bilateral PLND indications and patient selection need to be better defined. Furthermore, studies are needed addressing the value of the LN yield as a surrogate of survival. Since retrospective studies cannot establish causality properly, ongoing prospective randomized trials, such as the International Penile Advanced Cancer Trial (InPACT), may help clarify the role of PLND and its integration with multimodal therapies. Studies are warranted to determine the benefit of bilateral PLND in patients with unilateral inguinal LNM suspected of harboring contralateral metastatic disease. This implies studies that help clarify the disseminative pattern of metastatic spread and the limits of dissection in a precise and personalized manner.
The clinical heterogeneity within studies and the limitations of the study need to be acknowledged. Firstly, the search strategy was not systematically performed, and therefore the acquisition of all bilateral PLND studies cannot be confirmed. Secondly, there was not enough information to propose a meta-analysis given the little convergence within studies, which would incur in a high risk of outcome measurement bias. Not all patients that underwent bilateral PLND had the same tumor stage, nor did they receive the same treatment regimen or LND templates, which precludes the comparison of outcomes. The different PLND templates, aims/outcomes of studies, the low patient number, interference of neo/adjuvant treatments, and the retrospective nature of studies contribute to the heterogeneity of data.
Although pelvic and inguinal LNM dissemination patterns are predominantly reported following the anatomical staged pathway, exceptions exist and further efforts and larger cohorts are needed to deepen the understanding of LNM spread, as they may provide insights for directing LND strategies. Moreover, data suggest that preoperative imaging and staging may play a role in identifying pelvic LN involvement. Therefore, further research in this field may contribute to better patient selection for either unilateral or bilateral PLND, while weighing the oncological benefits, perioperative complications, and patients' quality of life.