Paratesticular rhabdomyosarcoma (PTRMS) is the most common primary solid tumor arising from the mesenchymal tissue of the testis. Traditionally, retroperitoneal lymph node dissection is not recommended for children aged <10 years because of the morbidity of the procedure and low risk of retroperitoneal lymph node involvement. In the current study, the authors analyzed the patient and tumor characteristics of PTRMS as well as survival outcomes associated with lymph node dissection status.
A total of 255 cases of PTRMS were identified from the patient data reported by the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Institute from 1973 through 2009.
Among 173 patients aged ≥10 years, lymph node dissection was found to improve the 5-year overall survival (OS) rate from 64% to 86% (P < 0.01). Conversely, patients aged <10 years fared extremely well regardless of lymph node dissection status; the 5-year OS rate was 100% and 97%, respectively, for patients who did versus those who did not undergo lymph node dissection (P = .37). The yield of positive lymph nodes was approximately ≥ 20% when < 11 lymph nodes were removed. The incidence of lymph node involvement was also higher in older patients compared with younger patients (40% vs 8%). Radiotherapy improved the OS rate in patients with lymph node involvement (5-year OS rate: 90% with vs 36% without radiation; P < .0001).
Paratesticular rhabdomyosarcoma (PTRMS) is a solid tumor arising from the mesenchymal tissue of the testis, spermatic cord, epididymis, and tunica vaginalis. It is the most common primary paratesticular malignant neoplasm diagnosed in patients aged 7 years to 36 years, with a mean age of 10 years.[1-3] PTRMS comprises 7% of all RMS cases. The Intergroup Rhabdomyosarcoma Study (IRS)-I and IRS-II report an excellent overall prognosis for patients with PTRMS, with 3-year recurrence-free survival rates of 93% and 90%, respectively, reported for IRS-I and IRS-II.
However, controversy has persisted regarding lymph node management in patients with PTRMS. In a recent research protocol from the Children's Oncology Group, staging ipsilateral retroperitoneal lymph node dissection (SIRPLND) is recommended for all boys aged ≥10 years. However, in patients aged < 10 years, SIRPLND is reserved for those with an enlarged lymph node suspicious for metastatic disease on computed tomography (CT). In a study published by the Italian and German Cooperative Group, among 72 patients with a negative retroperitoneal CT scan, only 1 patient was found to have lymph node involvement, thereby demonstrating the high sensitivity of CT for retroperitoneal lymph node involvement. Conversely, lymph node involvement was confirmed after surgery in 65% of the 23 patients found to have suspicious lymph nodes on CT scan.
In the current study, we analyzed patient and tumor characteristics of PTRMS as well as survival data associated with lymph node evaluation in different age groups. We also attempted to evaluate the optimal number of dissected lymph nodes. In addition, we examined outcomes for patients treated with radiotherapy (RT) with or without lymph node involvement.
MATERIALS AND METHODS
The Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Institute is a population-based cancer registry that routinely collects high-quality patient data in 17 registries and represents an estimated 26% of the US population. The SEER database was queried for the diagnosis of RMS from 1973 to 2009. To identify the histology associated with RMS, the third edition of the International Classification of Diseases for Oncology (IDC-0-3) codes 8900 to 8904, 8910, 8912, 8920 to 8921, and 8991 were used. Paratesticular anatomical sites were then identified using the primary site codes of C62.0, C62.1, C62.9, C63.0, C63.1, and C63.2, which coded for undescended testis, descended testis, testis, epididymis, spermatic cord, and scrotum, respectively. Data regarding the number of positive lymph nodes were extracted from the variable “regional nodes positive (1988+).” No RPLND was performed if the patient was coded as “no nodes were examined” in the variable “regional nodes examined (1988+)”; “no regional lymph node removed” in the variable “scope of regional lymph node surgery (1998-2002)”; or “no regional lymph nodes removed or aspirated, diagnosed at autopsy,” “biopsy or aspiration of regional lymph node, [not otherwise specified] NOS,” or “sentinel lymph node biopsy” in the “treatment summary-scope of regional surgery (2003+)” variable. The same 3 variables were used if > 1 lymph node was examined without the above coding mentioned.
JMP 10 software (SAS Institute Inc, Cary, NC) was used for statistical analysis. Estimates of overall survival (OS) were performed using the Kaplan-Meier method. Estimates of differences between survival curves among different host, tumor, and treatment characteristics were also performed. For comparison and analysis, we also divided patients into a younger age group (aged < 10 years) and an older age group (aged ≥10 years) based on previous North American and European studies demonstrating significance for survival with this age cutoff.[4, 5]
A total of 255 males with PTRMS were identified in the SEER database from 1973 through 2009. Among these patients, 193 were white (76%), 45 were black (18%), 13 were Asian/Pacific Islander (5%), 2 were American Indian/Alaska Native (1%), and 2 were of unknown race/ethnicity (1%) The median age at diagnosis was 15 years (range, <1 year to 79 years). Approximately 1 of 3 of patients were aged <10 years whereas 2 of 3 of patients were aged ≥10 years (Table 1).
Table 1. Patient, Tumor, and Surgical Treatment Characteristics
Total No. of Available Cases (%)
Age, y (median, 15 y)
American Indian/Alaska Native
Tumor size, cm
2 to 5
No. of positive lymph nodes
Lymph node dissection
No lymph node dissection
Lymph node dissection performed
Type of surgery for primary tumor
Excision of testicle without cord
Excision of testicle with cord/or cord not mentioned
Spermatic cord, scrotum, and epididymis
Local tumor destruction
Simple/partial surgical removal of primary site
Total surgical removal of primary site
PTRMS involved the testis in 178 cases (70%), the spermatic cord in 33 cases (13%), the scrotum in 30 cases (12%), and the epididymis in 14 cases (5%).It is interesting to note that 2 cases involved the undescended testis. Embryonal RMS was the most common histologic type, comprising 60% of cases, followed by alveolar (6%), spindle cell (5%), mixed (5%), pleomorphic (4%), ganglionic differentiation (< 1%), and unknown histology (19%).
Surgery of the primary tumor was performed in all but 1 patient. However, data regarding surgical procedures were only available for 241 patients. Among 167 patients with tumor involving the testis, the surgical procedure included excision of the testicle without the cord (16%), excision of the testicle with the cord/or cord not mentioned (57%), orchiectomy (6%), and nonspecific surgery (21%). The remaining 74 patients for whom surgical data were available had tumor involving the spermatic cord, scrotum, and epididymis; these patients were treated with various procedures including tumor debulking only (4%), local tumor destruction (3%), simple/partial surgical removal of the primary tumor site (10%), total surgical removal of the primary tumor site (35%), radical surgery (20%), and nonspecific surgery (28%).
Lymph Node Evaluation
Data detailing the lymph node evaluation were coded in the SEER data set only after 1998. Thus, they were available for only 196 patients. Of those patients, 137 did not undergo surgical lymph node evaluation. These patients were distributed throughout the treatment era examined. The remaining 59 patients underwent lymph node evaluation with ipsilateral (3 patients) and bilateral (2 patients) RPLND; in 54 patients, the degree of dissection (ipsilateral or bilateral) was not specified. The median number of lymph nodes examined was 10 (range, 1 lymph node-49 lymph nodes). Data detailing the number of positive lymph nodes were available for 57 patients. Among these were 34 patients (60%) with no lymph node involvement (N0 disease), 10 patients (18%) with 1 positive lymph node, and 13 patients (22%) with ≥2 positive lymph nodes. Among patients with documented lymph node involvement, the mean number of positive lymph nodes was 1.5 (range, 1 lymph node-7 lymph nodes).
We next analyzed the yield of lymph node dissection, which was defined as the percentage of positive lymph nodes of the total number of lymph nodes removed for all patients. Figure 1 shows that the yield of lymph node positivity remained at approximately ≥ 20% when <11 lymph nodes were dissected. However, the yield of positivity dropped to 0% to 20% for patients in whom 11 to 23 lymph nodes were dissected and 0% to 5% for those who had ≥24 lymph nodes removed.
To analyze the subgroup of patients without lymph node involvement, we identified 61 patients with a known number of lymph nodes examined. Eleven patients (18%) had ≥10 lymph nodes examined whereas 50 patients (82%) had <10 lymph nodes examined. There was no difference in survival noted between the 2 groups. The 5-year and 10-year OS rates were 90% and 86%, respectively, for those with <10 lymph nodes evaluated and 100% and 77%, respectively, for those with ≥10 lymph nodes examined (P = .62).
Figure 2 represents a Kaplan-Meier survival plot for patients treated with and without lymph node resection. The 5-year and 10-year OS rates were 92% and 87% versus 76% and 74% (P = .028) for those who did and those who did not undergo LND, respectively. Among patients aged ≥10 years, LND improved the 5-year OS rate (86% vs 64%; P < .009) (Fig. 3A). However, a benefit for LND was not observed in younger patients. As expected, patients aged <10 years fared extremely well overall. At 5 years, there was no survival benefit found to be associated with LND (100% and 97%; P = .37) (Fig. 3B). Younger patients aged <10 years also were found to have significantly lower rates of lymph node involvement compared with older patients (8% vs 40%; P < .0001). On subset analysis, when compared with older patients, children aged <10 years demonstrated better 5-year OS among those with lymph node-negative disease (100% vs 81%; P = .002) but similar survival in those with lymph node-positive disease (75% vs 70%) (P = .64).
The addition of RT improved the 5-year OS rate in patients with positive lymph node disease (90% vs 36%; P < .0001). RT was not found to be associated with a better OS in patients with N0 disease (74% vs 90%; P = .08). We analyzed the subset of patients aged <10 years with positive lymph nodes to determine whether RT was beneficial in this favorable group. We identified only 1 patient treated with both LND and RT and 1 patient treated with LND but not RT. Both patients were alive at 96 months and 67 months of follow-up, respectively. We also analyzed patients aged <10 years who were not treated with either LND or RT; all 44 patients were alive at the time of last follow-up.
RPLND in patients with PTRMS is controversial. IRS-III required all patients with PTRMS to undergo RPLND. However, because of the morbidity associated with the procedure, IRS-IV replaced RPLND with CT scans of the abdomen and pelvis to evaluate lymph node involvement. The use of CT for staging led to a significant change in the distribution of patients with groups I and II disease in these trials. The result was a higher percentage of patients enrolled in IRS-IV staged as group I compared with those in IRS-III (81% vs 68%) because CT failed to detect lymphatic involvement in some patients. Subsequently, group I patients in IRS-IV were found to have inferior 3-year failure-free survival compared with those in IRS-III (81% vs 96%). Based on these results from IRS-IV, Wiener et al concluded that the use of CT scans in place of RPLND decreased the detection of retroperitoneal lymph node involvement in patients with true lymph node disease, which resulted in a higher rate of regional disease recurrence. Similarly, the Italian and German Cooperative Group found a significant difference in the rate of lymph node involvement in children aged <10 years compared with older patients (5% vs 21%; P < .01). The difference in the rate of lymph node involvement between the younger patients (8%) and older patients (40%) was also observed in the current study (P < .0001). Thus, staging RPLND has been generally recommended for all patients aged >10 years, but only in younger patients if there is evidence of lymph node enlargement on CT scan. Results from the current study are consistent with the above recommendations. For patients aged ≥10 years, RPLND resulted in improved 5-year OS (86% vs 64%; P = .019). Alternatively, younger patients have excellent 5-year OS regardless of lymph node dissection status (100% and 97%, respectively; P = .37). The results of the current study are similar to the Malignant Mesenchymal Tumors studies (MMT 84 and MMT 89) of the International Society of Pediatric Oncology with regard to the outcome of children aged <10 years with nonmetastatic PTRMS who did not undergo LND. The study reported a 5-year OS rate of 96% and 80%, respectively, for patients aged <10 years and those aged ≥10 years, compared with 97% and 64%, respectively, in the current study. In the group of patients aged >10 years with tumors measuring >5 cm, the 5-year OS rate was only 62%. The better survival noted in the older group of patients with PTRMS in the MMT 84 and 89 studies may possibly be because of the relatively younger patients in that group (range, 10 years-18 years) compared with the age range of 10 years to 79 years in the current study.
Despite the potential benefit of RPLND in the staging and treatment of older patients, the procedure involves the risk of significant morbidity including bowel obstruction, loss of normal ejaculatory function, lymphedema, and the development of hydrocele as reported in IRS-I and IRS-II. Minimally invasive alternatives to open RPLND with laparoscopic approaches have been reported in the literature.[9, 10] The data from the current study demonstrated no survival benefit associated with the examination of >10 lymph nodes. The mean number of positive lymph nodes among those with >10 lymph nodes and those with <10 lymph nodes examined were 0.64 and 1, respectively. The yield of lymph node positivity was approximately ≥20% when <11 lymph nodes were dissected. However, it dropped to <20% for patients in whom ≥11 lymph nodes were dissected. Thus, we suggest that the optimal number of lymph nodes removed during staging RPLND is 10 lymph nodes. However, this recommendation should be taken with caution because of small patient numbers in the current study. It is also possible that on frozen section, a positive lymph node was found and further dissection was aborted. Further investigation with a larger number of patients for the optimal number of lymph nodes dissected is needed for a definitive conclusion to be reached.
The result of RPLND can be used to determine adjuvant treatment. In general, RT is recommended in patients with PTRMS with lymph node involvement. Our results confirmed the survival benefit of RT in patients with PTRMS with lymph node-positive disease. At 5 years, the OS rate was 90% and 36%, respectively, for patients with lymph node-positive disease treated with and without RT (P < .0001). It is unclear whether patients aged <10 years who have a very favorable prognosis require RT if they have lymph node-positive disease because there was only 1 patient in the current study who received RT and 1 who did not; both were alive at the time of last follow-up. However, the benefit of RT in patients with N0 disease is uncertain. In actuality, the results of the current study demonstrated a trend toward inferior survival outcomes in patients with N0 disease who received RT versus patients with N0 disease who did not (5-year OS rates of 90% vs 85%, respectively; P = .08). This result suggests that the potential morbidity associated with RT may outweigh its benefit in patients with N0 disease. Another possibility is that some patients with lymph node-negative disease received RT to a metastatic site and not the regional lymph node chain. Although the SEER database includes information regarding whether the patient received RT, it does not specify the details of RT such as treatment site, dose, and fractionation. It is interesting to note that the 5-year survival rate of patients with PTRMS with lymph node-positive disease who received RT was identical to that of patients with lymph node-negative disease who did not receive RT (90%). Therefore, the results of the current study indicate that it is appropriate to withhold lymph node RT for patients with lymph node-negative disease. However, these data support the use of RT in patients with lymph node-positive disease.
A recent study from the St. Jude Children's Research Hospital compared the use of positron emission tomography (PET) and CT versus conventional imaging for staging patients with RMS. The overall accuracy for detecting lymph node disease was 95% for PET-CT and 49% for conventional imaging. Because the SEER data analyzed were primarily from the era in which PET-CT was not performed, the use of PET-CT may change the eventual management of PTRMS in children aged ≥10 years. It is possible that in the future, LND might not be used if PET-CT can identify patients with positive lymph nodes. In this particular instance, only those patients with positive lymph nodes on PET-CT will undergo RPLND. Further studies are needed to determine whether this is possible.
As with all retrospective analyses, there are limitations to the current study. First, patient data were collected over a 36-year period, during which there have been alterations and improvements in surgical, chemotherapeutic, and RT techniques. Thus, there is significant heterogeneity of treatment techniques in this study. Second, information regarding chemotherapy was not coded in SEER data. Third, RT was recorded as beam radiation, radioactive implant, or isotope. However, the dose, fractionation, and irradiated area were not specified. Fourth, the coding for lymph node evaluation for RMS changed during the 3 periods including 1983 through 1987, 1988 through 2003, and 2004 and beyond. Lastly, information regarding the number of positive lymph nodes and the number of lymph nodes evaluated was available only for patients diagnosed after 1997. Thus, the number of cases included in sections of the analysis is limited. Despite these limitations, the SEER data provide large numbers of cases for rare cancers such as PTRMS, which may not be available in single-institution settings.
RPLND is recommended in patients aged ≥10 years. In younger patients, the risk of lymph node involvement is low, with an excellent prognosis. Thus, staging RPLND in younger patients may not be necessary if the CT scan does not reveal enlarged lymph nodes. RT improves the survival outcome in patients with lymph node involvement but not in those with lymph node-negative disease.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
Dr. Paulino is employed by The Methodist Hospital and has received royalties from Elsevier for his previously published textbook on positron emission tomography-computed tomography in radiotherapy.