Non–Germ Cell Intrascrotal Tumors in Adults: A Retrospective Cohort of Localized and Metastatic Disease From a Tertiary Cancer Center

Abstract

Purpose: 

Given their rarity, little is known about outcomes and optimal treatment strategies for non–germ cell intrascrotal tumors (NGCITs). Here, we report a retrospective cohort of NGCITs and describe clinicopathological features and treatment patterns and outcomes.

Materials and Methods: 

Patients 18 years or older with a histological diagnosis of NGCIT were eligible. Electronic health records were reviewed to assess clinical features, disease characteristics, and outcomes. Radiological responses were determined by RECIST 1.1. Time-to-event outcomes were estimated using the Kaplan-Meier method and compared through the log-rank test and Cox univariate regression.

Results: 

From January 2009 to April 2022, 896 patients were reviewed for eligibility, with 36 patients with NGCIT identified. At diagnosis, 29 were nonmetastatic (Leydig tumors, n = 9; sarcomas, n = 8; Sertoli tumors, n = 3; others, n = 9) and 7 were metastatic (sarcomas, n = 5; others, n = 2). Five patients (sarcomas, n = 2; Leydig tumors, n = 2; mesothelioma, n = 1) developed metastatic disease after a median follow-up of 18.6 months. Systemic treatment (n = 10) was given according to primary histology, with partial responses observed in 2 patients with metastatic rhabdomyosarcoma treated with VAC regimen (vincristin, adriamycin, and cyclophosphamide). Overall survival varied according to disease classification (benign vs malignant, hazard ratio [HR] 4.5, confidence interval [CI] 95% 1.0-21.2) or presentation (all patients: metastatic vs localized at diagnosis, HR 7.1, CI 95% 2.0-24.5; patients with malignant disease: metastatic vs localized at diagnosis, HR 5.0, CI 95% 1.1-24.4).

Conclusions: 

NGCITs comprise a heterogeneous group of diseases, with a relatively better prognosis at early stages. Most metastatic cases occurred within patients with paratesticular sarcomas and mesotheliomas.

Introduction

Approximately 5% of intrascrotal cancers have a non–germ cell origin, comprising a heterogeneous group of distinct neoplasms. Sex cord stromal tumors, mesotheliomas, lymphomas, adenocarcinoma of the rete testis, paratesticular sarcomas, and metastatic involvement from other primary sites all represent potential sources of intrascrotal malignancies.¹ The literature is based mostly on case reports and series, with few populational cross-sectional studies, scattered along different populations in different periods, and focusing on different tumor types.^2-6

When localized, intrascrotal non–germ cell tumors are treated with radical orchiectomy or hemiscrotectomy, with or without adjuvant chemotherapy and/or radiotherapy, depending on the primary histology. When metastatic, intrascrotal non–germ cell tumors may be treated according to the underlying histology. However, given the paucity of medical literature, there is a lack of knowledge to support optimal clinical evaluation, prognostic assessment, and treatment decisions for this rare group of diseases.^2-6

Here, we report on a retrospective cohort of patients with non–germ cell intrascrotal tumors (NGCITs) from a tertiary specialized cancer center. Patient and disease characteristics, treatment patterns, and outcomes are described.

Methods

Study Design and Patient Eligibility

This was a retrospective, noninterventional study that included patients with non–germ cell solid intrascrotal tumors admitted at Instituto do Cancer do Estado de Sao Paulo (ICESP). Patients aged 18 years or older with scrotal or testicular masses were reviewed for eligibility. Search terms included (1) International Classification of Diseases 10th Edition C62, (2) testicle, (3), spermatic cord, (4) rete testis, (5) scrotum/scrotal, (6) tunica vaginalis, (7) mesothelioma, (8) granulosa, (9) sarcoma, (10) liposarcoma, (11) rhabdomyosarcoma, (12) leiomyosarcoma, (13) myxoid, (14) spindle cell, (15) undifferentiated, (16) Leydig, and (17) Sertoli. Patients with a diagnosis of any non–germ cell intrascrotal neoplasms were included, and those with infectious diseases, hematological tumors, or germ cell tumors were excluded from analysis.

Data Collection

Electronic health records were reviewed to collect epidemiological, clinical, pathological, and laboratory data. All data were encrypted to protect patients' privacy according to the best practices in human research. The protocol was approved by our center's institutional review board (protocol no.: 1778/20). All proceedings were conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines.

End Points

Overall survival (OS) was measured as the time from histopathological diagnosis to death or last contact. Progression-free survival (PFS) was measured as the time from date of local treatment or chemotherapy start to radiological progression per RECIST 1.1, death, or last contact. Relapse-free survival (RFS) was measured as the time from orchiectomy or hemiscrotectomy to radiological relapse detection, death, or last contact.

Statistical Analysis

Descriptive statistics were used to describe patients and disease characteristics. For continuous variables, measures of central tendency (median) and measures of dispersion (IQR) were calculated. Time-to-event variables were estimated by using the Kaplan-Meier method. Comparison of survival between groups was performed with log-rank test. All programing for data analysis was done in Python version 3.6. Hazard ratios (HRs) and 95% confidence intervals (CIs 95%) were calculated in a univariate manner using the lifelines library, version 0.27.0.⁷P values <0.05 were considered statistically significant.

Results

Cohort Characteristics

From January 1, 2009, to June 30, 2020, 896 patients were assessed for eligibility (Figure 1). Of those patients, 36 had a diagnosis of non–germ cell solid intrascrotal tumors that were confirmed by an experienced pathologist for this analysis. The median follow-up was 36.5 months (IQR 11.6-82.0 months). The median age at diagnosis was 45.4 years (IQR 39.0-53.4 years).

Figure 1.: Flow diagram of all patients assessed for eligibility. TTAGS, testicular tumor of the adrenogenital syndrome.

Twenty-nine (80.5%) patients had localized disease at diagnosis. Of those patients, 12 (41.3%) had malignant disease (sarcomas, n = 8; mesothelioma, n = 2; Leydig tumor with malignant features, n = 1; adenocarcinoma of the rete testis, n = 1). Seven (19.5%) patients had metastatic disease at diagnosis (sarcomas, n = 5; mesothelioma, n = 1; metastatic adult-type granulosa cell tumor, n = 1). All patients underwent primary disease excision (radical orchiectomy, n = 31; hemiscrotectomies, n = 5). All hemiscrotectomies were performed in patients with malignant tumors. Two patients underwent primary retroperitoneal lymph node dissection (adenocarcinoma of the rete testis, n = 1; sarcoma, n = 1). The median OS for all patients was 19.8 years (CI 95% 3.1-19.8).

Of the 10 metastatic patients who received systemic chemotherapy, 2 achieved partial response (2 patients with rhabdomyosarcoma treated with vincristine, adriamycin, and cyclophosphamide) and 3 had stable disease (1 patient with rhabdomyosarcoma treated with vincristine, adriamycin, and cyclophosphamide; 1 patient with myxoid liposarcoma treated with doxorubicin; and 1 patient with metastatic adult-type granulosa tumor treated with bleomycin, etoposide, and cisplatin) while 5 had progressive disease as best response (2 patients with mesothelioma treated with cisplatin and pemetrexed; 1 patient with liposarcoma treated with ifosfamide and doxorubicin; 1 patient with liposarcoma treated with doxorubicin; and 1 patient with rhabdomyosarcoma who received vincristine, adriamycin, and cyclophosphamide).

Benign vs Malignant Disease

Benign tumors were found in 17 of 36 patients (Leydig tumors, n = 8; Sertoli tumors, n = 3; adult-type granulosa tumor, n = 1; adenomatoid leiomyoma, n = 1; benign adenomatoid paratesticular tumor, n = 1; benign fibrous pseudotumor, n = 1; paratesticular genital angioleiomyoma, n = 1; testicular tumor of the adrenogenital syndrome, n = 1). Patients with benign tumors trended toward an improved OS (Figure 2) compared with those with malignant disease (median not reached vs 4.4 years; HR 4.5, CI 95% 1.0-21.2, log-rank p = .03).

Figure 2.: Kaplan-Meier overall survival plots of all patients (blue) with 95% confidence interval (light blue area), patients with benign tumors (green), and patients with malignant tumors (red).

Localized vs Metastatic Disease

Considering the whole cohort, patients who were metastatic at diagnosis (n = 7) had worse OS (median 2.5 years vs 19.7 years; HR 7.1, CI 95% 2.0-24.5, log-rank p < .005) when compared with the ones with localized disease (n = 29). Within the 12 patients with localized malignant disease, 5 developed metastases after a median follow-up of 3.11 years (malignant Leydig cell tumor, n = 1; mesothelioma, n = 1; myxoid sarcoma, n = 1; liposarcoma, n = 1; rhabdomyosarcoma, n = 1). The median RFS was 1.5 years (CI 95% 0.43-16.92).

At the end of the follow-up, 24 patients had localized disease only while 12 developed metastases at any time (de novo metastatic disease, n = 7). Overall survival was better among patients with localized disease at any time (median 2.1 years vs not reached; HR 12.0, CI 95% 2.5-56.0, log-rank p < .005).

Patients with malignant tumors who had metastatic disease at diagnosis also showed shorter OS compared with patients with localized disease (median 2.5 vs 19.7 years, HR 5.0, CI 95% 1.1-24.4). Figure 3 shows Kaplan-Meier OS estimates of patients with malignant disease who had metastatic disease at any time, stratified by time of metastasis (ie, all patients, patients with de novo metastatic disease or relapse after localized disease).

Figure 3.: Kaplan-Meier overall survival plots of all patients with malignant disease (blue) with 95% confidence interval (light blue area), patients with localized malignant disease at diagnosis (green), and patients with de novo metastatic disease (red).

For PFS, the median value for all tumors of patients who had metastasis and received any therapies (local or systemic) was 6.7 months (CI 95% 0.3-14.2). Patients with initially localized disease had a median PFS of 6.7 months (CI 95% 0.06-34.46) while patients with de novo metastatic disease had a median PFS of 13.3 months (CI 95% 0.33-not applicable [NA]), with no significant difference between groups (P value 0.46). Figure 4 shows the Kaplan-Meier PFS estimates of patients who had metastatic disease at any time, stratified by time of metastasis.

Figure 4.: Kaplan-Meier progression-free plots of all patients with metastatic disease at any time (blue) with 95% confidence interval (light blue area), patients with localized malignant disease at diagnosis (green), and patients with de novo metastatic disease (red).

Accounting for only malignant tumors (n = 19; with or without eventual development of metastatic disease), the median OS was 4.4 years (CI 95% 1.0-19.8). Localized disease at diagnosis conferred a median OS of 19.8 years (CI 95% 1.0-19.8) vs 2.2 years (CI 95% 0.3-NA) among patients with de novo metastatic disease, with a significant difference in the log-rank test (P value 0.03).

Description by Tumor Type

Sex Cord Stromal Tumors

All patients had localized disease at diagnosis (Leydig tumors, n = 9; Sertoli tumors, n = 3). The median age at diagnosis was 45.1 years (IQR 39.2-55.2 years). All patients underwent orchiectomy, and none underwent retroperitoneal lymph node dissection. One patient with localized Leydig cell tumor received malignant disease diagnosis having an RFS of 16.9 years, with the eventual discovery of a lung metastasis and treated with a lobectomy. This patient died after 19.8 years of follow-up due to an infection not related to cancer.

Mesotheliomas

Three patients had mesotheliomas, and no previous exposure to asbestos was reported. Of those patients, 1 was of epithelioid histology and 2 were biphasic. The median age at diagnosis was 62.34 years (IQR 44.76-91.88 years). Two patients had localized disease at diagnosis. All patients underwent hemiscrotectomy. None underwent retroperitoneal lymph node dissection. One of the patients with localized disease had a relapse, with an RFS of 158 days, with lymph node involvement. The patient with de novo metastatic disease had lymph node and lung involvement. Both patients received cisplatin and vinorelbine as first-line treatment and had a PFS of 21 and 38 days and OS of 1.0 and 0.4 years, respectively. The patient with localized disease who experienced no relapse was aged 91 years and died 84 days after hemiscrotectomy due to health care–associated pneumonia.

Paratesticular Sarcomas

Thirteen patients were diagnosed with paratesticular sarcomas. The median age at diagnosis was 44.8 years (IQR 18.5-51.9 years). Eight cases had localized disease at diagnosis (leiomyosarcoma, n = 2; spindle cell sarcoma, n = 3; myxoid liposarcoma, n = 1; liposarcoma, n = 1; rhabdomyosarcoma, n = 1). All patients with initially localized disease underwent local treatment, being the patient with rhabdomyosarcoma submitted to hemiscrotectomy. Three patients had disease relapse (liposarcoma, n = 1; rhabdomyosarcoma, n = 1; myxoid liposarcoma, n = 1) with a median RFS of 1.5 years. All patients had bone metastasis; however, the patient with liposarcoma also developed skin, lung, and adrenal gland as secondary disease sites while the other two had concurrent lymph node involvement. These patients received, respectively, ifosfamide and doxorubicin; vincristine, doxorubicine, and cisplatin; and isolated doxorubicine as first-line treatment and had a median PFS of 6.7 months. The median OS of initially nonmetastatic patients with posterior disease relapse was 3.1 years.

Five patients had metastatic disease at diagnosis (liposarcoma, n = 1; rhabdomyosarcoma, n = 4). One patient underwent hemiscrotectomy and retroperitoneal lymph node dissection. The median age at diagnosis was 19 years (IQR 18-65 years) for those patients. All patients had lymph node involvement; however, the patient with liposarcoma had concurrent lung, adrenal gland, and peritoneal involvement, and 1 patient with rhabdomyosarcoma had concurrent bone marrow and meningeal involvement. Although the patient with liposarcoma received doxorubicin as first-line treatment, patients with rhabdomyosarcoma received vincristin, adriamycin, and cyclophosphamide (VAC). The median OS of de novo metastatic patients was 2.2 years, and the median PFS was 1.2 years. The patient who underwent lymph node resection and posterior VAC regimen remains alive, with an OS of 10.2 years.

Other Tumors

Eight patients had testicular tumors not listed in other categories (adult-type granulosa cell tumor, n = 2; adenocarcinoma of the rete testis, n = 1; paratesticular adenomatoid tumor, n = 1; adenomatoid leiomyoma of the epididymis, n = 1; peritesticular fibrous pseudotumor, n = 1; paratesticular genital angioleiomyoma, n = 1; testicular tumor of the adrenogenital syndrome, n = 1). After a median follow-up time of 2.77 years, no deaths were observed. The patient with a metastatic adult-type granulosa cell tumor was diagnosed at age 26 years and had peritoneal and lung metastases. He received bleomycin, etoposide, and cisplatin for 1 cycle, complicated by a myocardial infarction 2 days after treatment completion. The patient remains in good condition after 7.2 months of follow-up since diagnosis and 4.4 months of PFS since day 1 of bleomycin, etoposide, and cisplatin.

Discussion

Here, we retrospectively analyzed a single-center cohort of patients with non–germ cell intrascrotal solid tumors to assess patients and disease characteristics, treatment patterns, and outcomes. Furthermore, we provided a more detailed description of characteristics and outcomes according to specific tumor types. NGCITs are rare, occurring in only approximately 1% to 5% of all orchiectomies.¹ Medical literature is scarce given the paucity of cases. Patient workup and treatment decisions are generally carried out according to tumor-specific histology, but optimal management remains elusive.^2-6 To our knowledge, this is one of the largest contemporary single-centered cohorts of patients with unusual NGCITs.

Patients who developed metastatic disease had poor prognosis, with a median OS close to 2 years. The histological types that most frequently present as de novo metastatic or metastatic relapse after local treatment were sarcomas and mesotheliomas. Consequently, in our study, most death events occurred in patients with those diagnoses. The outcomes observed in our study are in line with previous retrospective findings showing poor outcomes in patients with mesotheliomas⁸ and paratesticular sarcomas.^9,10

Radiological responses were achieved only in patients with rhabdomyosarcoma, whereas stable disease was seen among patients with liposarcoma, rhabdomyosarcoma, and metastatic adult-type granulosa tumor. Although data regarding response rate are limited, the populational study by Nazemi and Daneshmand⁹ showed that these sarcoma subtypes portray relatively better prognosis when compared with other paratesticular sarcoma histologies.

Our study is limited by its retrospective nature, the heterogeneity of histological types, the relatively few numbers of patients included, and the survival events observed, which taken altogether preclude further definitive conclusions. Taking into consideration the extreme rarity of cases and the consequent scarcity of available data, larger cohorts are mostly possible in regional databases. In Table 1, we summarize the available data on NGCITs.

Table 1. - Retrospective Studies on Non–Germ Cell Instrascrotal Tumors

Study	Histology of interest	Number of patients assessed for eligibility	N	Data origin
Zidar et al 11	Intrascrotal mesothelioma	26 (malignant mesotheliomas)	1	Institutional
An et al 12	Intrascrotal mesothelioma	7 (case series report)	7	Institutional
Marinaccio et al 13	Intrascrotal mesothelioma	9166 (malignant mesotheliomas)	31	Italian National Mesothelioma Register—ReNaM
Nazemi et al 8	Intrascrotal mesothelioma	Not mentioned	113	National Cancer Institute's SEER 18 Database
Farkas et al 14	Non–germ cell testicular tumors	504 (testicular tumors)	10	Institutional
Paffenholz et al 15	Non–germ cell testicular tumors	522 (testis sparing surgery–benign testicular masses)	11	Institutional
Conkey et al 16	Non–germ cell testicular tumors	18 (case series report)	18	Institutional
Scott et al 17	Non–germ cell testicular tumors	1526 (testicular tumors)	56	Thames Valley Germ Cell Database
Gatto et al 18	Paratesticular sarcomas	274 (sarcomas)	4	Institutional
Korkes et al 10	Paratesticular sarcomas	179 orchiectomies from 1993 to 2006	12	Institutional
Murray et al 19	Paratesticular sarcomas	72 (case series report)	72	Institutional
Nazemi et al 9	Paratesticular sarcomas	3007 (genitourinary sarcomas)	1087	National Cancer Institute's SEER Database

SEER, Surveillance, Epidemiology, and End Results Program.

Conclusion

In conclusion, NGCITs are rare entities with relatively good prognosis if detected and treated at early stages. Patients with metastatic disease had an overall poorer prognosis when compared with patients with localized disease. Sarcomas and mesotheliomas are probably more prone to relapse and develop metastasis, and radiological responses or disease stabilization may be achieved with systemic chemotherapy.

Author Contributions

MTC: Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing—original draft, reviewing and editing. CAC: Data curation; Formal analysis; Investigation; Visualization; Writing—reviewing and editing. JAS: Formal analysis; Investigation; Visualization; Writing—reviewing and editing. GFF: Formal analysis; Investigation; Visualization; Writing—reviewing and editing. DQM: Formal analysis; Investigation; Visualization; Writing—reviewing and editing. GBG: Formal analysis; Investigation; Visualization; Writing—reviewing and editing. MDC: Formal analysis; Investigation; Visualization; Writing—reviewing and editing. LC: Data curation; Formal analysis; Investigation; Methodology; Visualization; Writing—reviewing and editing. WCN: Formal analysis; Investigation; Methodology; Visualization; Resources; Supervision; Writing—reviewing and editing. JMM: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Visualization; Writing—original draft, reviewing and editing.

Financial Support

None to disclose.

Additional Disclosures

The results of this study were partially presented at the 2022 Genitourinary Cancers Symposium of the American Society of Clinical Oncology and the abstract was published in the supplement of Journal of Clinical Oncology (DOI: 10.1200/JCO.2022.40.6_suppl.415).

Conflicts of Interest

MTC and CAC: no conflicts of interest to disclose. JAS: speaker's bureau (Janssen, MSD), research funding (Astellas). GFF: honoraria (Roche). DQM: advisory board (Janssen), speaker's bureau (Janssen, Pfizer, BMS, Ferring), research funding (Pfizer). GBG, MDC, LC, WCN: no conflicts of interest to disclose. JMM: participation as a compensated member of advisory boards and speaker for Janssen and Astellas and speaker for Technopharma, Bristol-Myers-Squibb, Bayer, Ipsen, Pfizer, Amgen, and Merck Sharp & Dohme.