The Role of Lymphadenectomy for Testicular Cancer: Indications, Controversies, and Complications

Retroperitoneal lymphadenectomy remains an integral component in the multidisciplinary management of testicular cancer. Through the evolution of the procedure, continued refinements have led to diagnostic and therapeutic benefits as well as improvements in perioperative and long-term morbidity. This article discusses the surgical management of testicular cancer, specifically addressing the indications, controversies, and complications of retroperitoneal lymph node dissection for both low-stage and high-stage disease.

Germ cell tumors of the testis

Testicular germ cell tumors (GCTs) are the most common solid malignancy in young men ages 20 to 35. The incidence of testis tumors is rising, with 8480 new diagnoses in 2010, resulting in 350 deaths. With survival rates exceeding 90%, the multidisciplinary management of testicular cancer serves as a model for the treatment of other solid tumors. Surgery, cisplatin-based chemotherapy, and radiotherapy play integral roles in the treatment of this disease and have contributed to these excellent cure rates. This article discusses the surgical management of testicular cancer, focusing on retroperitoneal lymphadenectomy (RPLND) for nonseminomatous GCT treatment (NSGCT), specifically addressing the indications, controversies, and complications of lymphadenectomy for both low-stage and high-stage disease.

Development of the contemporary RPLND

Retroperitoneal lymph node dissection was first performed in the 1950s using a surgical template based on the initial lymphatic drainage studies described by Jamieson and Dobson in 1910. Further mapping studies performed by Donohue and colleagues, Weissbach and colleagues, and Ray and colleagues, established specific anatomic regions within the retroperitoneum described as “primary landing zones” for right-sided and left-sided tumors. Right-sided testicular tumors most commonly metastasize to the interaortocaval lymph nodes followed by precaval and paracaval lymph nodes, whereas left-sided testicular tumors most commonly spread to the para-aortic and preaortic lymph nodes. Contralateral involvement occurs more commonly with right-sided primary tumors, especially with higher-stage tumors.

The role of RPLND has been well established in the management of NSGCT; however, surgical templates have become the focus of considerable debate. In an era where clinical staging was limited and effective chemotherapy was not available, emphasis for RPLND was placed on extensive dissection and removal of all lymph nodes in the retroperitoneum. Initial templates, as described by Donohue and colleages, included bilateral suprahilar dissection as well as removal of all nodal tissue between both ureters down to the bifurcation of the common iliac arteries. Several investigators confirmed therapeutic effectiveness with this extended template RPLND; however, suprahilar dissection was associated with increased pancreatic injury, lymphatic injury, and renovascular complications. Given the morbidity of performing suprahilar dissection, Donohue and colleagues investigated relapse rates for patients with low-stage NSGCT and found that relapse was independent of whether or not the patient underwent a suprahilar node dissection. Thus, bilateral infrahilar dissection, with boundaries set superiorly by the renal hilum, laterally by the ureters, and inferiorly by the bifurcation of the common iliac vessels, decreased the morbidity of RPLND while maintaining oncologic efficacy and has been established as the standard of care.

With the modification of the RPLND template to an infrahilar dissection, the most common long-term complication was infertility secondary to retrograde ejaculation from injury to the sympathetic nerve fibers. Normal antegrade ejaculation requires sympathetic innervation to trigger the coordinated processes of closure of the bladder neck, seminal emission, and expulsion of semen. Efferent impulses originate in the preganglionic fibers at the T12 to L3 thoracolumbar spinal cord. After leaving the lumbar sympathetic trunks, these fibers converge near the inferior mesenteric artery just above the aortic bifurcation and form the hypogastric plexus. From the hypogastric plexus, these fibers innervate the seminal vesicles, prostate, vas deferens, and bladder neck through the pelvic plexus. Ejaculation is triggered by nerves originating at the lumbosacral spinal cord levels, primarily S2 to S4. The coordinated event of antegrade ejaculation requires sympathetic innervation to close the bladder neck and sacral somatic innervation to cause external urethral sphincter relaxation and rhythmic contractions of the bulbourethral and perineal muscles.

Preservation of the paravertebral sympathetic ganglia, postganglionic sympathetic fibers from T2 to L4, and the hypogastric plexus during RPLND are critical to prevent ejaculatory dysfunction. Through an improved understanding of neuroanatomy, unilateral metastatic distribution patterns, and surgical mapping studies, surgeons began developing strategies to minimize the risk of retrograde ejaculation and subsequent infertility. Thus, two strategies were developed to preserve these structures, unilateral modified template RPLND and nerve-sparing RPLND.

Modified template RPLND was initially introduced for carefully selected patients with low-stage disease and eliminated contralateral dissection below the inferior mesenteric artery. Several different modified templates have been used and have reported excellent functional outcomes, ranging from 51% to 88% preserved antegrade ejaculation. Despite varying surgical boundaries, unilateral modified templates share the common goals of (1) dissection of the interaortocaval and ipsilateral lymph nodes between the renal hilum and the bifurcation of the common iliac artery ipsilateral to the primary tumor, and (2) minimizing contralateral dissection, especially below the level of the inferior mesenteric artery, thereby preserving the contralateral sympathetic trunk, postganglionic sympathetic fibers, and hypogastric plexus.

Further understanding of the sympathetic innervation and anatomy has led to the development of the modern nerve-sparing bilateral template RPLND. Prospective identification and meticulous preservation of the sympathetic trunks, postganglionic sympathetic fibers, and hypogastric plexus have resulted in antegrade ejaculation in more than 95% of patients undergoing primary RPLND in high-volume centers. Using a similar technique in select patients undergoing a more challenging postchemotherapy (PC)-RPLND, Pettus at Memorial Sloan-Kettering Cancer Center (MSKCC) reported a 79% success rate for antegrade ejaculation in patients with involved lymph nodes. Although bilateral nerve-sparing RPLND is feasible in experienced hands, oncologic outcome should never be compromised to preserve antegrade ejaculation and fertility. The issue of decreasing template size and potential risks of extratemplate disease is discussed later.

With further modifications to RPLND templates, lymph node count, as with other malignancies, has been evaluated in an effort to establish benchmarks in surgery. Thompson and colleagues evaluated the lymph node counts of 255 patients who underwent primary RPLND at MSKCC and reported a median node count of 48 lymph nodes. These investigators reported that increasing lymph node count increased the odds of identifying a positive lymph node; and on multivariate analysis, a lymph node count greater than 40 retained a significant association with identifying positive lymph nodes, even after adjusting for year of surgery, clinical stage (CS), and surgeon (odds ratio 2.0; 95% CI, 1.1–3.7; P = .026). In the postchemotherapy setting, Carver and colleagues evaluated the number of lymph nodes obtained in 628 patients who underwent PC-RPLND and were found to have fibrosis or teratoma. The investigators reported decreasing lymph node count as a significant predictor of disease recurrence ( P = .04) and a 2-year relapse-free survival of 97% when greater than 50 lymph nodes were removed. These data suggest that lymph node counts during RPLND may be an independent predictor of disease recurrence and, thus, may have important implications in determining adequacy of surgical quality and pathologic assessment.

Development of the contemporary RPLND

Retroperitoneal lymph node dissection was first performed in the 1950s using a surgical template based on the initial lymphatic drainage studies described by Jamieson and Dobson in 1910. Further mapping studies performed by Donohue and colleagues, Weissbach and colleagues, and Ray and colleagues, established specific anatomic regions within the retroperitoneum described as “primary landing zones” for right-sided and left-sided tumors. Right-sided testicular tumors most commonly metastasize to the interaortocaval lymph nodes followed by precaval and paracaval lymph nodes, whereas left-sided testicular tumors most commonly spread to the para-aortic and preaortic lymph nodes. Contralateral involvement occurs more commonly with right-sided primary tumors, especially with higher-stage tumors.

The role of RPLND has been well established in the management of NSGCT; however, surgical templates have become the focus of considerable debate. In an era where clinical staging was limited and effective chemotherapy was not available, emphasis for RPLND was placed on extensive dissection and removal of all lymph nodes in the retroperitoneum. Initial templates, as described by Donohue and colleages, included bilateral suprahilar dissection as well as removal of all nodal tissue between both ureters down to the bifurcation of the common iliac arteries. Several investigators confirmed therapeutic effectiveness with this extended template RPLND; however, suprahilar dissection was associated with increased pancreatic injury, lymphatic injury, and renovascular complications. Given the morbidity of performing suprahilar dissection, Donohue and colleagues investigated relapse rates for patients with low-stage NSGCT and found that relapse was independent of whether or not the patient underwent a suprahilar node dissection. Thus, bilateral infrahilar dissection, with boundaries set superiorly by the renal hilum, laterally by the ureters, and inferiorly by the bifurcation of the common iliac vessels, decreased the morbidity of RPLND while maintaining oncologic efficacy and has been established as the standard of care.

With the modification of the RPLND template to an infrahilar dissection, the most common long-term complication was infertility secondary to retrograde ejaculation from injury to the sympathetic nerve fibers. Normal antegrade ejaculation requires sympathetic innervation to trigger the coordinated processes of closure of the bladder neck, seminal emission, and expulsion of semen. Efferent impulses originate in the preganglionic fibers at the T12 to L3 thoracolumbar spinal cord. After leaving the lumbar sympathetic trunks, these fibers converge near the inferior mesenteric artery just above the aortic bifurcation and form the hypogastric plexus. From the hypogastric plexus, these fibers innervate the seminal vesicles, prostate, vas deferens, and bladder neck through the pelvic plexus. Ejaculation is triggered by nerves originating at the lumbosacral spinal cord levels, primarily S2 to S4. The coordinated event of antegrade ejaculation requires sympathetic innervation to close the bladder neck and sacral somatic innervation to cause external urethral sphincter relaxation and rhythmic contractions of the bulbourethral and perineal muscles.

Preservation of the paravertebral sympathetic ganglia, postganglionic sympathetic fibers from T2 to L4, and the hypogastric plexus during RPLND are critical to prevent ejaculatory dysfunction. Through an improved understanding of neuroanatomy, unilateral metastatic distribution patterns, and surgical mapping studies, surgeons began developing strategies to minimize the risk of retrograde ejaculation and subsequent infertility. Thus, two strategies were developed to preserve these structures, unilateral modified template RPLND and nerve-sparing RPLND.

Modified template RPLND was initially introduced for carefully selected patients with low-stage disease and eliminated contralateral dissection below the inferior mesenteric artery. Several different modified templates have been used and have reported excellent functional outcomes, ranging from 51% to 88% preserved antegrade ejaculation. Despite varying surgical boundaries, unilateral modified templates share the common goals of (1) dissection of the interaortocaval and ipsilateral lymph nodes between the renal hilum and the bifurcation of the common iliac artery ipsilateral to the primary tumor, and (2) minimizing contralateral dissection, especially below the level of the inferior mesenteric artery, thereby preserving the contralateral sympathetic trunk, postganglionic sympathetic fibers, and hypogastric plexus.

Further understanding of the sympathetic innervation and anatomy has led to the development of the modern nerve-sparing bilateral template RPLND. Prospective identification and meticulous preservation of the sympathetic trunks, postganglionic sympathetic fibers, and hypogastric plexus have resulted in antegrade ejaculation in more than 95% of patients undergoing primary RPLND in high-volume centers. Using a similar technique in select patients undergoing a more challenging postchemotherapy (PC)-RPLND, Pettus at Memorial Sloan-Kettering Cancer Center (MSKCC) reported a 79% success rate for antegrade ejaculation in patients with involved lymph nodes. Although bilateral nerve-sparing RPLND is feasible in experienced hands, oncologic outcome should never be compromised to preserve antegrade ejaculation and fertility. The issue of decreasing template size and potential risks of extratemplate disease is discussed later.

With further modifications to RPLND templates, lymph node count, as with other malignancies, has been evaluated in an effort to establish benchmarks in surgery. Thompson and colleagues evaluated the lymph node counts of 255 patients who underwent primary RPLND at MSKCC and reported a median node count of 48 lymph nodes. These investigators reported that increasing lymph node count increased the odds of identifying a positive lymph node; and on multivariate analysis, a lymph node count greater than 40 retained a significant association with identifying positive lymph nodes, even after adjusting for year of surgery, clinical stage (CS), and surgeon (odds ratio 2.0; 95% CI, 1.1–3.7; P = .026). In the postchemotherapy setting, Carver and colleagues evaluated the number of lymph nodes obtained in 628 patients who underwent PC-RPLND and were found to have fibrosis or teratoma. The investigators reported decreasing lymph node count as a significant predictor of disease recurrence ( P = .04) and a 2-year relapse-free survival of 97% when greater than 50 lymph nodes were removed. These data suggest that lymph node counts during RPLND may be an independent predictor of disease recurrence and, thus, may have important implications in determining adequacy of surgical quality and pathologic assessment.

Indications for retroperitoneal lymph node dissection

Clinical Stage I NSGCT

Approximately one-third of patients with testicular cancer present with CS I disease, defined as normal postorchiectomy tumor markers (β-human chorionic gonadotropin, α-fetoprotein, and lactate dehydrogenase) and absence of metastatic disease on imaging studies of the chest, abdomen, and pelvis. Treatment options for patients with stage I NSGCT include surveillance, 2 cycles of cisplatin-based adjuvant chemotherapy, and retroperitoneal lymph node dissection. The management of stage I NSGCT has been risk-adapted based on pathologic features (histology and presence or absence of vascular invasion) of the primary tumor with a trend toward surveillance for compliant patients with low-risk features and chemotherapy for patients with high-risk features. Recent data questioning the long-term morbidities of imaging-related ionizing radiation exposure, however, as well as the long-term morbidities of chemotherapy, have re-emphasized the importance of surgery. Some of the advantages provided by RPLND include (1) definitive pathologic nodal staging, (2) removal of chemoresistant teratoma, and (3) therapeutic removal of viable GCT.

Despite advances in modern radiographic imaging techniques, the accurate staging of these patients remains challenging. Approximately 30% of patients who present with CS I disease harbor occult metastasis in the retroperitoneum after orchiectomy, and up to 35% of patients presenting with CS IIa disease have negative nodes at the time of RPLND. This inaccuracy in staging results in both the undertreatment and the overtreatment of these patients and increases their long-term morbidity. RPLND allows for definitive nodal pathologic staging of patients with CS I or CS IIa NSGCT and provides critical information to guide further treatment and minimize treatment-related long-term morbidity.

In addition to improved staging, RPLND removes chemoresistant teratomatous elements. Failure to control the retroperitoneum during the initial treatment phase may compromise overall patient curability. Additionally, retroperitoneal relapse is rare (2%) after a properly performed RPLND, thus eliminating the need for routine postoperative CT scanning, aside from a baseline scan. Although histologically described as benign, teratoma is biologically unpredictable and, when left unresected, possesses the potential to invade into adjacent organs (growing teratoma syndrome), undergo malignant transformation, and increase the risk of late relapse. Late relapse is defined as disease recurrence more than 2 years after primary therapy and occurs in 2% to 3% of all patients with testicular cancer. Late relapse is associated with relatively poor survival, with approximately 50% of patients remaining free of disease.

The therapeutic benefit of RPLND has been well established, providing a cure rate of 99% for patients with CS I NSGCT. The retroperitoneum is the first site of metastatic spread in more than 90% of patients and metastatic testicular cancer to the retroperitoneum can be cured with surgery alone. In a review of the literature, Stephenson and Sheinfeld found RPLND, without subsequent chemotherapy, to be curative in 60% to 92% of patients with pathologic N1 disease. In patients with pathologic N2 disease, cure rates with surgery alone decreased to 50%; however, administration of 2 cycles of cisplatin-based adjuvant chemotherapy improved relapse-free survival rates to 98%.

Clinical Stage IIa and IIb NSGCT

The management strategy for patients with CS IIa or CS IIb is dependent on many factors, including extent of disease, serum tumor marker levels, and presence or absence of tumor-related back pain. Candidates best suited for RPLND include those patients with a single focus of retroperitoneal disease measuring less than or equal to 3 cm at the primary landing zone, patients with normal postorchiectomy markers, and patients without signs of tumor-related back pain. Elevated tumor markers or tumor-related back pain may indicate metastatic systemic disease or unresectable disease and these patients should be considered for induction chemotherapy.

Stephenson and colleagues reviewed the experience at MKSCC and identified important selection factors that have an impact on the risk of recurrence for patients undergoing primary RPLND. Elevated postorchiectomy markers ( P <.001), CS ( P = .0002), and pre-1999 RPLND ( P = .05) were identified as independent pretreatment predictors of progression. Excluding patients with elevated tumor markers and patients with CS IIb disease improved 4-year relapse-free survival rates from 83% to 96% (95% CI, 79%–88% and 91%–100%; P = .005), increased the percentage of pN1 disease among patients who were pathologic stage II (64% vs 40%, P = .01), and did not affect the rate of retroperitoneal teratoma (21 vs 22%, P = .89). Patients with elevated tumor markers or CS IIb should be referred for induction chemotherapy.

After primary RPLND, accurate pathologic assessment is critical to assess prognosis and determine the need for adjuvant therapy. A select group of patients with pathologic stage II disease may benefit from receiving 2 cycles of adjuvant chemotherapy. The Testicular Cancer Intergroup Study randomized 195 patients who underwent RPLND with complete resection and pathologic stage II disease to receive either 2 cycles of adjuvant chemotherapy or observation. Their study showed a significant reduction in the risk of relapse after receiving adjuvant chemotherapy (6% vs 49%) but did not show any survival difference. The investigators were not able to show any difference between pathologic N1 versus N2 disease and attribute this to the small sample size. The majority of other investigators, however, have demonstrated that the risk of relapse is dependent on the size and/or number of lymph node involvement. In general, observation is recommended for compliant patients with completely resected pathologic N1 disease and 2 cycles of adjuvant chemotherapy for patients with pathologic N2 or N3 disease. Chemotherapy cannot compensate for poor quality surgery, however. Two cycles of adjuvant chemotherapy should only be recommended for patients with complete resection of their retroperitoneal disease. Patients with incomplete resection should be referred for induction chemotherapy.

Clinical Stage IIc and III NSGCT

Patients with CS IIc or CS III should receive induction cisplatin-based chemotherapy, based on IGCCG risk criteria. Elevated tumor markers after primary induction chemotherapy often represent viable metastasis and second-line chemotherapy should be administered to these patients. The criteria for use of PC-RPLND have undergone continued refinement; however, most experts agree that PC-RPLND is indicated in the setting of normalized tumor markers with radiographic evidence of a residual retroperitoneal mass. Recent studies have demonstrated a histologic stage migration due to the use of more effective cisplatin-based chemotherapy regimens. The histologic outcomes of patients who underwent induction chemotherapy followed by PC-RPLND have been reported as 45% fibrosis, 40% teratoma, and 15% viable GCT. After second-line chemotherapy, histologic evaluation of PC-RPLND specimens revealed approximately 50% malignant GCT, 40% teratoma, and 10% fibrosis. Several investigators have proposed predictive models to determine the retroperitoneal pathology and avoid the morbidity of PC-RPLND in patients with necrosis or fibrosis. The accuracy of these models is limited and has ranged from 70% to 90% and, thus, bilateral PC-RPLND (with nerve sparing if feasible) remains the treatment of choice for these patients. Similar to RPLND in the primary setting, PC-RPLND provides several advantages, including improved pathologic staging and removal of chemoresistant elements.

Clinicians have had difficulty differentiating between necrosis and teratoma or viable malignant GCT on postchemotherapy imaging. Thus, the radiographic criteria defining the “normal” retroperitoneum have varied widely among institutions. The controversy regarding management of small residual retroperitoneal masses is discussed later. Nonetheless, PC-RPLND provides definitive nodal staging to guide further treatment. Modern imaging cannot accurately stage the postchemotherapy retroperitoneum; thus, surgical resection of any radiographic abnormality is required to evaluate for the presence of malignant GCT or teratoma.

The removal of teratoma in the resected PC-RPLND specimen confers a survival benefit to the patient because of the unpredictable biology of teratoma. As discussed previously, there are several key benefits to removing residual teratoma in the retroperitoneum. First, teratoma may grow, obstruct, or invade into adjacent organs (growing teratoma syndrome) and become unresectable. Secondly, teratoma with malignant transformation is reported in 3% to 6% of men undergoing PC-RPLND. The incidence of malignant transformation increases to 12% to 18% in patients undergoing reoperative PC-RPLND for late relapse. In this scenario, complete surgical resection remains the treatment of choice. Lastly, unresected residual teratoma may be associated with late relapse which, as discussed previously, is often chemoresistant, requiring surgical resection, and is often associated with poor outcomes.

High-Risk PC-RPLND for NSGCT

Combined modality treatment for advanced GCTs has yielded improved outcomes; however, Donohue and associates have identified surgical scenarios where patients are at higher risk of relapse and experience lower survival rates: (1) PC-RPLND after salvage or second-line chemotherapy, (2) redo RPLND, (3) desperation RPLND, and (4) RPLND with viable cancer in resected specimen.

The first subset of patients includes patients who require salvage or second-line chemotherapy with subsequent normalization of tumor markers. PC-RPLND after salvage chemotherapy is technically difficult, due to desmoplastic reactions, and thus is associated with lower rates of complete resection and higher histologic proportions of viable cancer. Furthermore, additional standard-dose chemotherapy does not seem to benefit the patients in this setting.

The second subset of high-risk patients includes patients who undergo redo RPLND. McKiernan and colleagues reviewed the experience at MSKCC and evaluated 56 patients who underwent redo operations for NSGCT, 22 after primary RPLND and 34 after PC-RPLND. Left-sided recurrence occurred more commonly and the sites of recurrence most commonly identified were at the left renal hilum and para-aortic regions, likely secondary to the technical demands necessary to gain adequate dissection in these regions. The overall 5-year survival rate for patients who underwent redo PC RPLND was 56%, thus underscoring the importance of initial complete resection. Data from MSKCC and Indiana University clearly demonstrate that patients who undergo incomplete initial resection and require redo RPLND are at a severe disadvantage, regardless of other risk factors.

The third subset of high-risk patients includes patients who undergo desperation RPLND in the setting of persistently elevated serum tumor markers after chemotherapy. Elevated tumor markers after cisplatin-based chemotherapy are conventionally considered a contraindication to surgical therapy due to persistent presence of viable GCT. In a select subset of patients, however, RPLND may be curative in the setting of elevated tumor markers. Historic and contemporary series report a 20% to 55% 5-year survival rate after desperation RPLND, thus conferring some therapeutic benefit despite the elevated tumor markers. Careful patient selection is critical, however, in determining which patients benefit from desperation RPLND. General indications include (1) stable or declining serum tumor markers, (2) radiographic resectable disease in 1 or 2 sites, and (3) resectable disease after exhausting salvage chemotherapy regimens.

The last subset of high-risk patients includes patients with viable cancer in a resected specimen after RPLND. Patients who undergo first-line chemotherapy followed by complete resection during PC-RPLND with viable GCT currently receive 2 additional cycles of chemotherapy and achieve disease-free survival rates of up to 70%. Fizazi and colleagues recently challenged the role of adjuvant chemotherapy after completely resected GCT during PC-RPLND in a retrospective analysis of 146 patients. They identified 3 independent prognostic variables for survival: (1) complete resection, (2) good risk IGCCCGT classification, and (3) less than 10% viable malignant cells. They concluded that additional chemotherapy benefited patients with 1 risk factor but not those with 2 or more risk factors. After salvage or second-line therapy, Fox and colleagues reported that 2 additional cycles of chemotherapy, in the setting of viable GCT in the PC-RPLND specimen, did not have any additional therapeutic benefit. For patients with residual teratoma, complete surgical resection remains the treatment of choice.

Only gold members can continue reading. Log In or Register to continue