After gathering proper approval from the University of Iowa Human Subjects Office Institutional Review Board-1 (IRB-1) under Genitourinary Retrospective Umbrella Projects Institutional Review Board (IRB) (Approval number 201404766), the University of Iowa Hospitals and Clinics (UIHC) Electronic Medical Records (EMR) from the years 1998–2023 were queried for GU sarcomas in patients aged 18 and above, involving the prostate, bladder, kidneys, and paratesticular region. Gynecological and retroperitoneal sarcomas were excluded from the search.

All sarcomas were reviewed by specialized soft-tissue pathologists at the University of Iowa. Sarcomas were categorized into different subtypes based on the predominant tissue type, microscopic features of cellular and nuclear pleomorphism, and immunohistochemistry stains. Grading of the tumor was done based on the Fédération Nationale des Centres de Lutte Contre le Cancer (FNCLCC) grading into a low grade (G1), intermediate grade (G2), or high grade (G3). The FNCLCC system is based on 3 factors: degree of differentiation, mitotic count, and tumor necrosis, with each factor being assigned a score of 1 to 3. The sarcomas were staged in accordance with the American Joint Committee on Cancer (AJCC) Tumor Node Metastasis (TNM) system for soft tissue sarcomas. Detailed tumor characteristics are provided in Table 1.

Survival probabilities were estimated and plotted using the Kaplan-Meier (K-M) method. Estimates along with 95% pointwise confidence intervals (CIs) were reported. Recurrence-free survival (RFS) was defined as the time from when the patient was disease-free to recurrence. Otherwise, patients were censored at the last disease evaluation. Patients still alive were censored at the last known point of being alive, non-cancer-related deaths were also counted as a censoring event. Using this data, descriptive tables (Tables 2 and 3) were generated, and univariate analysis of different risk factors was performed. Due to the small sample size, further subgroup analysis on RFS and OS could not be done. Neither was it possible to do a multivariate analysis of the risk factors affecting prognosis. All statistical tests were performed using SAS v9.4 (SAS Institute, Cary, NC, USA) software.

There were 4 types of sarcomas in this patient cohort with the commonest being LPS and LMS, other subtypes include RMS and angiomyosarcoma. There were 12 reported cases of LMS, 15 cases of LPS, 5 cases of RMS (with 3 alveolar subtypes), and 1 angiosarcoma. Based on the site of origin, the most frequent location of origin/organ system involvement, was paratesticular (n = 23), bladder (n = 5), prostate (n = 2) and kidney (n = 2), prostatic urethra (n = 1). Among the LPS 7 were well differentiated and 8 were de-differentiated.

Median age for diagnosis was 51 years (range: 21–79) for LMS, 60 years (range: 36–89) for LPS, 24 years (range: 20–71) for RMS. The median tumor size for LPS was 8.7 cm (range: 3.0–17.3 cm) in the greatest dimension compared to 5 cm (range: 1.2–15.0 cm) for LMS, and 5.0 cm (range: 4.2–10.3 cm) for RMS.

LMS tended to have a higher grade on diagnosis (intermediate and high grade: n = 10, 83.33% of total vs. n = 8/15, 53.33% for LPS). Two patients with LPS developed recurrent disease after resection of a presumed lipoma.

Immunohistochemistry and testing for different markers and chromosomal abnormalities were performed on most, but not all, GU sarcomas. The staining details are provided in Table 1. Most LPS were assessed using fluorescent in situ hybridization (FISH) for the presence of MDM2 and other amplifications. Among the 16 total cases, 8 were positive for MDM2, 1 was equivocal, 2 were negative and in cases, FISH was not done. Other amplifications detected included CDK4 in 4 cases. The single case of alveolar RMS tested negative for fusion transcript. Most LMS tested positive for the presence of smooth muscle actin and myosin and negative for other markers including CK5/6, CD45, CD34, and CD 117 (details in Table 1).

During the median follow-up period of 19.9 months (range: 0.1–103.1) for LMS, 3 patients developed the metastatic disease (GU-05, 06, and 10: metastasis to pleura, lungs, and liver; Table 1), compared to 2 cases of LPS recurring (GU-18: recurring locally; GU-32: recurring as disseminated peritoneal disease) over a follow-up period of 32.1 months (range: 2.9–119.8) and 1 recurrent case of RMS (GU-14: metastasis to lung) over a median follow-up period of 93.4 months (range: 25.8–172.0). Using K-M analysis, the RFS at 12, 24, and 36 months was 91% (95% CI: 70%–98%), 83% (95% CI: 60%–93%) and 63% (95% CI: 39%–79%), respectively (Figure 1).

Among patients with no metastatic disease or lymph node spread at presentation, 6 patients (n = 6 out of 26, 23.07%) died. Four had prior LMS and 1 each had RMS and LPS. Among patients who went into remission, i.e., no residual disease radiologically post-treatment, 27.58% died (n = 8/29). By KM analysis, the OS at 12, 24, and 36 months was 100%, 91% (95% CI: 69%–98%), and 81% (95% CI: 57%–92%), respectively (Figure 2).

4 patients had metastatic disease and 3 had lymph node involvement on presentation. Disease did not go into remission, i.e., macroscopic disease is still present or distant metastasis present in 4 patients (GU-09, 13, 17, 33). Among these patients, 3 patients died and 1 was in hospice (GU-09) 2.5 years after treatment on the last follow-up.

Paratesticular soft tissue sarcomas can arise from the epididymis, and mesenchymal layers surrounding the testis and the spermatic cord.⁵ LPS are the most common tumors (64%) of the paratesticular region,² often presenting as a nonreducible painless inguinal lump.⁷ In line with our series, other studies have also reported unplanned hernia surgery in 20.1% of cases (compared to n = 17/33, 50% cases) for spermatic cord sarcomas necessitating additional oncologic surgery.⁷ Local recurrence is a common occurrence for paratesticular tumors as circumferential negative resection margins are difficult to achieve, with one study reporting an actuarial local recurrence rate of 30% at 10 years and 42% at 15 years when surgery is the sole treatment modality.⁸ There exist two approaches to resection, simple tumorectomy vs. high inguinal resection. High inguinal resection offers better outcomes in terms of needing re-resection in the subcutaneous variant.⁹ Thus, it has been adopted as the standard of care. For the one local recurrence of dedifferentiated intermediate grade LPS, pembrolizumab, salvage radiotherapy, and repeat surgery were performed. Although the inked margins were positive after repeat resection, CT shows no evidence of local recurrence or metastasis at 36 months follow-up.

While the presence of MDM2 and CDK4 amplifications suggests a more aggressive clinical course due to their role in uncontrolled cell cycle progression and inhibition of tumor suppressor pathways, they also present potential pathways for targeted therapies. Currently, palbociclib a CK4/6 inhibitor shows the most promising results with a phase 2 trial demonstrating 12 weeks Progression Free Survival (PFS) of 66% in advanced dedifferentiated LPS.¹⁰ Another notable clinical trial NCT02343172 is currently assessing the efficacy of an MDM2 inhibitor siremadlin in combination with a CDK4 inhibitor ribociclib in patients with advanced dedifferentiated liposarcoma (DDLPS).¹¹

Although RMS do tend to present in the pediatric age group, they can also present in adult patients accounting for 2%–5% of adult soft tissue tumors. Among the subtypes of RMS, the embryonal subtype is the more commonly found accounting for 60% of all RMS, with a favorable prognosis.¹² This contrasts with the less common alveolar variant (accounting for 20%–30% of all RMS cases), which may harbor chromosomal translocations involving the FOXO1 gene, such as the PAX3-FOXO1 fusion (t [2;13]) or the PAX7-FOXO1 fusion (t [1;13]) which have a more aggressive clinical course and a worse prognosis.¹³ The presence of the PAX3/FOX01 gene fusion in alveolar RMS is associated with a worse prognosis, having a 5-year OS of 7% in fusion-positive vs. 69% in fusion negative.^12,14 In a SEER database study, the 1- and 5-year OS for prostate RMS was 76.2% (51.9%–89.3%) and 33.0% (12.8%–55.0%), respectively.¹⁵ Prognostic factors for RMS include older age (>21 years) or the presence of distal metastasis at presentation. Interestingly, the site of tumor origin is not associated with disease-specific overall survival.¹⁶ Treatment of adult patients with non-metastatic RMS with prospective RMS protocols has been shown to improve 3-year, 4-year, and 5-year survival, with survival rates approaching that of pediatric RMS.¹⁷

Adjuvant radiation has been shown to be beneficial for the treatment of alveolar RMS with microscopic positive margins. For paratesticular tumors, ipsilateral retroperitoneal lymph node dissection (RPLND) is recommended, and Lymph node (LN)-positive disease should be followed up by radiotherapy to the para-aortic chain.¹⁸ For RMS of prostate/bladder, while cystoprostatectomy may achieve local control, the high rate of rectal and urinary dysfunction makes neoadjuvant radiotherapy an attractive choice in unresectable disease, which if followed by resection can result in acceptable urinary and bowel function.¹⁹

Predictors for decreased cancer-specific survival (CSS) of LMS include increased patient age (with patients >60 years having a dramatic decrease in median survival²⁰), size (>5 cm as an independent prognostic factor for disease),²¹ absence of cancer-directed surgery, and the presence of distant metastasis. While standard treatment for large LMS of the bladder is radical cystectomy, smaller tumors can be managed with partial cystectomy or even transurethral resection for bladder tumor (TURBT).²⁰ There are few case reports in the literature that mention smaller LMS (<4 cm) managed with partial cystectomy with no adverse oncologic outcomes.²² While there may be local recurrence (lower Disease Specific Survival (DSS)) when tumors are managed with TURBT, there are no long-term differences in survival compared to cystectomy.²⁰ For advanced or metastatic LMS and LPS 1st line treatment is chemotherapy with doxorubicin alone or combination doxorubicin-ifosfamide therapy, doxorubicin is the standard first-line therapy with no other combinations yielding superior results.^23,24 In patients previously treated with anthracyclines, a combination of trabectedin with doxorubicin is recommended as 2nd line therapy. LMS-04, a phase 3 randomized controlled trial (RCT), evaluated the efficacy of current 1st line doxorubicin against a combination of doxorubicin and trabectedin for metastatic LMS with the highest reported PFS (12.2 months, 95% CI: 10.1–15.6) in adult Soft Tissue Sarcoma (STS) till date.²⁵ Immunotherapy has yielded limited and conflicting results in the treatment of metastatic LMS with an Overall Response Rate (ORR) of 0.10 (95% CI: 0.06–0.17) and LPS (ORR: 0.11, 95% CI: 0.07–0.17) in a meta-analysis, with current recommendations placing them at 2nd or 3rd line therapy in certain situations.²⁶ Other newer therapies include pazopanib, a multi-kinase inhibitor of vascular endothelial growth factor receptor (VEGFR), platelet-derived growth factor inhibitor (PDGFi), and Cluster of Differentiation CD 117 which has been shown to improve PFS in STS including LMS in a Phase III trial (NCT00753688)²⁷ (see Table 4 for the role of chemotherapy regimens in LMS, LPS, and RMS).

In a study conducted at MD Anderson Cancer Center by Sexton et al., comprising largely of LMS and RMS, it was concluded that multimodal therapy with pre-operative chemotherapy and radiotherapy resulted in appreciable tumor necrosis and appreciable downsizing, respectively.³² Hence, this approach may be considered for locally advanced LMS with a high chance of positive surgical margin, as positive margins are associated with a 0% 5-year survival rate.³² Two of our cases recurred within 2 years of achieving disease-free status as metastasis to the lungs (commonest site of metastasis)³³ and liver, and were treated with pembrolizumab. In a retrospective analysis conducted at four Midwest Sarcoma Trials Partnership institutions, immunotherapy (pembrolizumab, nivolumab, combined pembrolizumab, nivolumab, and other combinations) resulted in a radiographic partial response in 45% of metastatic LMS cases.³⁴ One patient developed progressive disease with hepatic metastasis and bony pelvis involvement. Tissue sampling showed extensive PD-L1 expression (Table 5), hence, pembrolizumab and adjuvant radiotherapy to the pelvis was prescribed. However, in spite of treatment with pembrolizumab, death occurred at 4 months. One of the limitations of this study is the small sample size (n = 3) of patients treated with pembrolizumab, hence it is difficult to draw conclusions from our outcomes. One patient with high-grade (FNCLCC grade 3) spindle cell variant LMS developed a solitary metastasis to the pleura after achieving disease-free status (remission), which was treated with palliative radiotherapy due to advanced age and poor performance status.

One patient with paratesticular LMS developed recurrence in the form of distal metastasis to the pleura that was treated with radiotherapy. Our patient demonstrated a mitotic count of >19/High Powered Field (HPF) with tumor necrosis (<50%). A second patient with paratesticular epithelioid LMS was found to have metastatic disease to the liver, posterior fifth rib, and lung. This patient was observed initially and subsequently treated with doxorubicin upon progression of the disease after about one year of surveillance. The patient subsequently succumbed to the sequela of brain metastasis and further progressive disease at about 40 months after diagnosis.

The kidneys are a frequent site of origin of sarcomas within the genitourinary tract, ranking just behind the paratestis.^3,35 Renal sarcomas also have worse survival outcomes, possibly due to the high proportion of high-grade tumors and anatomic location, with DSS worse in the kidney compared to other GU sites.² In the largest study to date on renal sarcomas based on the SEER database, the CSS of renal sarcomas is bleak, with a 5-year OS and CSS of 46% and 58% for nonmetastatic disease.³⁶ These renal sarcomas have worse DSS in comparison to GU sarcomas from other sites.² LMS are the most commonly found subgroup of sarcomas found in the kidney,³⁶ having an OS of 28 months and a DSS time of 34 months.² In our series, we had only one patient with renal angiosarcoma (GU-09) who had undergone cytoreductive surgical resection. There is some documented benefit to this as, cytoreductive surgery combined with chemotherapy remains a reasonable approach for metastatic disease, with surgery improving survival benefit for renal sarcomas from 10% to 40%.³⁶ However, the outlook is only marginally improved, as in our case (angiosarcoma arising within angiomyolipoma); the disease progressed and the patient was on palliative treatment upon the last follow-up 2 years later. Primary renal angiosarcomas, in general, have a poor outlook with one-third of the cases being metastatic at presentation and 2/3rd of non-metastatic patients developing metastasis after surgical resection, with the commonest sites being lung, liver, and bones.³⁷ Although previous studies have demonstrated a very poor prognosis (DFS: 6.0 months, PFS: 2.0 months, and OS: 5.0 months) with patients with metastatic disease having a threefold increase in the risk of death (HR = 3.27, p = 0.004), our case exhibited an exceptionally good survival rate at the most recent follow-up.³⁷

While our study comprised an acceptable cohort, due to the relative heterogeneity and variety in the subtypes of GU sarcomas further subgroup analyses could not be performed. There is a wide variation in the survival outcome based on the initial stage, grade, and subtype of the sarcoma. Hence, the calculated OS may not be evenly applicable for all the sarcomas. As a limitation, we acknowledge that while we considered a Cox proportional hazards model, the small sample size precluded its reliable application, limiting our ability to perform a multivariate survival analysis. Our study did yield a similar 3-year RFS (66%) reported in other studies (63%) with a similar sample size⁵ and higher 3-year OS (81%) compared to studies with a similar sample size.³ Furthermore, each sarcoma responds differently to different treatment modalities. For instance, while surgery alone may be adequate for a well-differentiated LPS, RMS may require pretreatment chemotherapy and radiotherapy. Also, as previously noted, stage, grade, and presence of metastasis are independent predictors of mortality for GU sarcomas. Hence, OS and median survival will differ based on the cancer staging.