Clinicopathological Characteristics and Prediction of Postoperative Mortality Risk in Patients with Non-metastatic Sarcomatoid Renal Cell Carcinoma

Abstract

Introduction

Sarcomatoid renal cell carcinoma (sRCC) is rare but highly aggressive and is associated with poor prognosis and limited treatment responsiveness. Despite several studies investigating its clinicopathological features, existing research is often limited by small sample sizes and short follow-up periods, and currently, no prognostic risk model is specific to patients with non-metastatic sRCC. This study aimed to investigate the clinicopathological characteristics of patients with non-metastatic sRCC and develop a predictive model for postoperative mortality risk.

Methods

In this retrospective study, we analyzed the clinical data of 45 patients diagnosed with non-metastatic sRCC who underwent surgical treatment at our institution's Department of Urology, between January 2008 and June 2024. These patients were compared with 527 patients with non-sarcomatoid renal cell carcinoma (non-sRCC). The primary endpoint was death, and the exact cause of death was recorded. Routine postoperative examinations and treatment details were documented through outpatient and inpatient electronic medical record systems.

Results

The results indicated significant differences in body mass index, hypertension, surgical approach, nephrectomy type, surgical duration, maximum tumor diameter, tumor necrosis, T stage, and Ki-67 expression between patients with sRCC and those with non-sRCC (P < 0.05). Survival analysis revealed that the cancer-specific survival (CSS) for patients with sRCC was significantly lower than that for patients with non-sRCC (P < 0.001). Cox univariate and multivariate analyses identified maximum pathological tumor diameter, T stage, and high Ki-67 expression as independent risk factors. Based on these factors, we developed a postoperative mortality risk prediction model for patients with sRCC, with the calibration curves demonstrating a good fit of the model.

Conclusions

The proposed model is designed for patients with non-metastatic sRCC. It has potential clinical application value, aiding in the identification of high-risk patients and providing guidance for individualized treatment and close follow-up.

Keywords

sarcomatoid renal cell carcinoma clinical pathological features mortality risk prediction survival analysis

Introduction

Research Background

Renal cell carcinoma (RCC) is the most common malignant tumor of the kidney and one of the most prevalent cancers of the urinary system. According to statistics, the incidence of RCC ranks after those of prostate and bladder cancers, accounting for approximately 90% of all malignant kidney tumors.¹ Due to population aging, the incidence of RCC is anticipated to show a marked upward trend in the coming decades.² Based on pathological characteristics, RCC is generally classified into several subtypes, with the most common types including clear cell carcinoma, papillary carcinoma, and chromophobe carcinoma.³ In addition to these common subtypes, sarcomatoid renal cell carcinoma (sRCC) represents a rare and highly aggressive form of RCC, accounting for 4%–5% of all RCC cases, with the majority of patients diagnosed between the ages of 54 and 63 years.^4–5 sRCC is not an independent histological subtype but rather a high-grade malignant transformation that can occur in any RCC subtype.^4,6–9 It was first described by Farrow et al in 1968 and is characterized by epithelial-to-mesenchymal transition within renal epithelial tissue, leading to the formation of tumors containing sarcomatoid components.¹⁰ sRCC is typically associated with aggressive invasiveness, rapid progression, high metastatic potential, and poor prognosis.^{4–5,8–9,11–12} Compared with non-sarcomatoid RCC (non-sRCC), sRCC exhibits significant differences in clinical presentation and treatment response. Although sRCC has a low incidence, its highly malignant nature poses significant clinical challenges, making it a subject of critical research interest.

Research Objectives

The highly malignant nature of sRCC presents a major challenge in the clinical management of urological cancers. Despite marked progress in research on certain targeted drugs and natural products for RCC,¹³ surgical intervention remains the primary treatment approach for patients with non-metastatic sRCC and has been shown to significantly improve overall survival (OS).^4,9,11,14 However, because of the aggressive behavior and high metastatic potential of sRCC, the risk of postoperative mortality remains high, making an effective assessment of postoperative mortality risk essential in RCC research. Several studies have explored the clinicopathological characteristics and prognostic factors of sRCC.^{4,9,12,15–18} However, patients with advanced-stage sRCC account for a relatively high proportion in existing research cohorts,¹¹ leaving the localized, non-metastatic population demands understudied. This group warrants urgent scientific attention, as timely surgical intervention and adjuvant therapy can often lead to a significant improvement in survival outcomes. Although a few sRCC prognostic risk models have been developed,^19,20 their clinical application remains limited, and currently, no prognostic risk model is specific to patients with non-metastatic sRCC. Therefore, this study aimed to retrospectively analyze the clinicopathological data of patients with non-metastatic sRCC to explore the clinical and pathological differences between sRCC and common types of renal cancer, identify independent risk factors for postoperative mortality, and develop a predictive model based on these factors. This model will facilitate the early identification of high-risk patients through the evaluation of clinicopathological parameters and support the implementation of personalized treatment strategies, thereby providing a scientific basis for the clinical management of non-metastatic sRCC and promoting precise treatment and postoperative follow-up.

Materials and Methods

Study Population and Data Collection

This study was a retrospective analysis of the clinical data of 45 patients who underwent surgical treatment at our institution's Department of Urology, between January 2008 and June 2024, and were pathologically diagnosed with sRCC postoperatively. The collected clinical data included basic patient information (age and sex), medical history (smoking, hypertension, and diabetes), physical examination findings (body mass index [BMI] and tumor laterality), surgical details (surgical approach, nephrectomy type, and surgical duration), preoperative symptoms (lumbar pain, hematuria, abdominal mass, weight loss, and fever), and pathological findings (pathological subtype, maximum tumor diameter, cystic tumor component, tumor necrosis, T stage, and immunohistochemical expression of Ki-67). Among the 45 patients with sRCC, 36 were classified as having sarcomatoid clear cell carcinoma, whereas 9 had sarcomatoid non-clear cell carcinoma (4 cases of pRCC, 2 cases of chRCC and 3 cases of other subtypes). As of December 2024, 31 patients died from renal tumors, whereas 14 patients remained alive. Data were collected from 527 patients who underwent surgery during the same period and were pathologically diagnosed with non-sRCC, composed of 457 cases of clear cell carcinoma and 70 cases of non-clear cell carcinoma (36 cases of pRCC, 29 cases of chRCC and 5 cases of other subtypes), who were included in the control group. Among them, 54 patients died, and renal tumor was the cause of death in 45 patients. Additionally, within this cohort of 527 patients with non-sRCC, 27 were classified as having World Health Organization/International Society of Urological Pathology (WHO/ISUP) grade 4 RCC; among them, 10 died during follow-up, with all deaths attributed to renal tumors. This study adhered to the ethical principles outlined in the Declaration of Helsinki (revised in 2013) and was approved by the Ethics Committee of our institution, and written informed consent was obtained from all participants. The reporting of this study conforms to TRIPOD guidelines.²¹

Pathological Diagnosis and Staging Criteria

In the present study, sRCC was diagnosed based on the presence of sarcomatoid features within the tumor. According to existing literature, the minimum threshold for the proportion of sarcomatoid components required for diagnosis has not been universally established. Most studies have considered the presence of sarcomatoid tissue identifiable in low-power fields or clearly demarcated regions sufficient for diagnosis.^6,18,22 Tumor staging in this study was performed according to the eighth edition of the Cancer Staging Manual published by the American Joint Committee on Cancer. As a unique pathological variant of RCC, sRCC follows the standard RCC staging criteria.

Inclusion and Exclusion Criteria

Inclusion Criteria:

Cases with medical treatment conducted at our hospital between January 2008 and June 2024 and individualized surgical treatment performed based on diagnostic evaluation by an experienced team of urological specialists.

Presence of a renal tumor without regional lymph node invasion or distant metastasis (N0M0).

Surgical pathology specimens reviewed and verified by at least two senior pathologists.

Postoperative regular follow-ups, including thoracoabdominal imaging conducted every 6 months for the first 3 years, followed by annual imaging thereafter. High-risk or recurrent disease requiring treatment according to the surgeon's clinical experience.

Exclusion Criteria:

Presence of a primary malignant tumor originating from another site.

Evidence of undergoing renal biopsy only, having incomplete tumor resection during surgery, or presenting with positive surgical margins postoperatively.

Regional lymph node involvement or distant metastasis (N1 or M1).

Incomplete clinical data or patient's refusal to participate

Follow-Up and Outcome Assessment

In this study, survival outcomes and survival times for all patients, including 45 patients with sRCC and 527 patients with non-sRCC, were determined through telephone follow-up. The primary endpoint was death, and the exact cause of death was recorded. Survival time was calculated from the date of surgery to the date of death or last follow-up date (December, 2024). Routine postoperative examination and treatment details were documented through outpatient and inpatient electronic medical record systems, which served as the basis for follow-up. The cause of death was confirmed through medical records and telephone follow-ups. For surviving patients, the disease status was updated based on the most recent imaging findings.

Statistical Analysis

All statistical analyses were performed using R software (version 4.4.1). For continuous variables, data were presented as mean ± standard deviation (x̄ ± s) if they followed a normal distribution, and comparisons between the groups were conducted using the t-test. If the data did not follow a normal distribution, they were expressed as medians and interquartile ranges (M [P25, P75]), and intergroup comparisons were performed using the Wilcoxon rank-sum test. For categorical variables, if the expected frequency in all cells was greater than 5 and the sample size was equal or greater than 40, the chi-squared test was used. If the expected frequency ranged between 1 and 5 and the sample size was equal or greater than 40, a continuity-corrected chi-square test was applied. If the simple size was less than 40, Fisher's exact test was applied. Survival analysis was conducted using the Kaplan–Meier method, and survival differences between the groups were compared using the log-rank test. To identify independent prognostic factors, Cox univariate and multivariate regression analyses were performed. Based on the identified factors, a nomogram was developed to predict the risk of postoperative mortality, and a calibration curve was plotted to assess the model's goodness of fit. A significance level of P < 0.05 was considered statistically significant.

Through these statistical analyses, this study aimed to gain a deeper understanding of the clinicopathological characteristics of non-metastatic sRCC, identify independent prognostic risk factors, and construct a clinically valuable mortality risk prediction model to provide guidance for optimizing patient management.

Results

Comparison of Clinical and Pathological Characteristics and Survival Analysis between Patients with Sarcomatoid Renal Cell Carcinoma and Those with Non-Sarcomatoid Renal Cell Carcinoma

During the follow-up period, all baseline and follow-up data were collected as of December 2024. The follow-up results showed that among the 45 patients with sRCC, 31 died due to renal tumors, whereas 14 remained alive. In contrast, among the 527 patients with non-sRCC, 54 died, of whom 45 had a confirmed cause of death due to renal tumors, and 473 remained alive. A comparison of the baseline characteristics between the sRCC (n = 45) and non-sRCC groups (n = 527) revealed no significant differences in sex, age, smoking history, history of diabetes, tumor laterality, or pathological subtype (P > 0.05). However, significant differences in BMI, hypertension, surgical approach, nephrectomy type (radical vs partial), operation duration, presenting symptoms at diagnosis (lumbar pain, hematuria, abdominal mass, weight loss, and fever), maximum pathological tumor diameter, cystic tumor components, tumor necrosis, T stage, Ki-67 expression levels, median postoperative survival, and causes of death were observed between the two groups (P < 0.05) (Table 1).

Table 1.

Comparison of Clinicopathological Characteristics Between Patients with non-sRCC and Those with sRCC.

Characteristics	Description	Non-sRCC (N = 527)	sRCC (N = 45)	χ²/T/Z	P
Sex	Male	355 (67.4%)	33 (73.3%)	0.431	0.511
	Female	172 (32.6%)	12 (26.7%)
Age	Median (IQR)	60.0 (51.0-66.5)	63.0 (53.0-69.0)	−2.282	0.19
BMI	Median (IQR)	24.1 (22.3-26.0)	22.5 (20.8-23.5)	3.491	< .001
Smoking	Yes	70 (13.3%)	7 (15.6%)	0.041	0.84
	No	457 (86.7%)	38 (84.4%)
Hypertension	Yes	247 (46.9%)	13 (28.9%)	4.705	0.03
	No	280 (53.1%)	32 (71.1%)
Diabetes	Yes	87 (16.5%)	7 (15.6%)	0.000	1
	No	440 (83.5%)	38 (84.4%)
Tumor laterality	Left	252 (47.8%)	25 (55.6%)	0.708	0.4
	Right	275 (52.2%)	20 (44.4%)
Surgical approach	Open	33 (6.3%)	16 (35.6%)	41.759	< .001
	Laparoscopic	494 (93.7%)	29 (64.4%)
Nephrectomy type	Radical	261 (49.5%)	41 (91.1%)	27.125	< .001
	Partial	266 (50.5%)	4 (8.9%)
Surgical duration	Median (IQR)	120.0 (93.5-160.5)	155.0 (105.0-191.0)	−3.709	0.006
Symptoms at diagnosis	Asymptomatic	401 (76.1%)	22 (48.9%)	14.545	< .001
	Symptomatic	126 (23.9%)	23 (51.1%)
Pathological subtype	Clear cell RCC	457 (86.7%)	36 (80%)	1.058	0.304
	Non-clear cell RCC	70 (13.3%)	9 (20%)
Maximum tumor diameter	Median (IQR)	4.0 (3.0 to 5.5)	6.5 (5.0 to 9.0)	−6.843	< .001
Cystic tumor component	Yes	84 (15.9%)	1 (2.2%)	5.129	0.024
	No	443 (84.1%)	44 (97.8%)
Tumor necrosis	Yes	70 (13.3%)	27 (60%)	60.981	< .001
	No	457 (86.7%)	18 (40%)
T Stage	T1/T2	447 (84.8%)	20 (44.4%)	42.443	< .001
	T3/T4	80 (15.2%)	25 (55.6%)
WHO/ISUP grade	Grade I	97 (18.4%)	0 (0%)	339.184	< .001
	Grade II	302 (57.3%)	0 (0%)
	Grade III	64 (12.1%)	0 (0%)
	Grade IV	27 (5.1%)	45 (100%)
	Not applicable	37 (7%)	0 (0%)
Ki-67 expression	Low	472 (89.6%)	21 (46.7%)	60.538	< .001
	High	55 (10.4%)	24 (53.3%)
Median postoperative survival	Median	31.0	-	186.049	< .001
Follow-up outcome	Death	54 (10.2%)	31 (68.9%)	108.104	< .001
	Survival	473 (89.8%)	14 (31.1%)
Cause of death	Renal tumor	45 (83.3%)	31 (100%)	4.152	0.042
	Other causes	9 (16.7%)	0 (0%)

Abbreviations: BMI, body mass index; IQR, interquartile range; RCC, renal cell carcinoma; sRCC, sarcomatoid renal cell carcinoma; WHO/ISUP, World Health Organization/International Society of Urological Pathology. For numerical variables: T-test or One-way ANOVA used for normally distributed data with equal variances; Welch versions if variances are unequal; Wilcoxon or Kruskal-Wallis tests if not normally distributed. For categorical variables: Chi-squared test used if expected counts > 5 and sample size >= 40; apply Yates’ correction if counts are between 1 and 5; Fisher's exact test if counts are < 40.

Kaplan–Meier survival curve analysis demonstrated that CSS was significantly lower the sRCC group than in the non-sRCC group (P < 0.001). This result indicates that patients with sRCC have a markedly poorer survival prognosis, as shown in Figure 1.

Figure 1.

Kaplan-Meier Curves of CSS in the Patients with Non-sarcomatoid and those with Sarcomatoid Renal Cell Carcinoma. Abbreviations: RCC, Renal Cell Carcinoma; sRCC, Sarcomatoid Renal Cell Carcinoma. Kaplan-Meier Curve Plotted using R Packages ‘Survival [version 3.3.1]’, ‘Survminer [version 0.4.9]’, and ‘ggplot2 [Version 3.4.4]’.

Comparison of Clinical and Pathological Characteristics and Survival Analysis between Sarcomatoid Renal Cell Carcinoma Patients and WHO/ISUP Grade 4 Renal Cell Carcinoma Patients

The WHO/ISUP grading system is widely used for prognostic evaluation of patients with RCC.^23–24 This grading system classifies tumors based on nucleolar prominence and nuclear morphology, with higher grades indicating worse prognoses. Notably, all patients with sRCC are categorized as having WHO/ISUP grade 4 RCC.^7,24 In this study, 27 patients with non-sarcomatoid RCC classified as WHO/ISUP grade 4 RCC were included in the control group for survival comparison. The median survival time for patients with sRCC was 31 months, whereas that of patients with WHO/ISUP grade 4 non-sRCC was 112 months. Kaplan–Meier survival analysis and log-rank tests revealed that the difference in survival between the two groups was statistically significant (P = 0.037). However, baseline comparisons between patients with sRCC and those with WHO/ISUP grade 4 non-sRCC showed no significant differences in clinicopathological characteristics (P > 0.05) (Table 2, Figure 2).

Figure 2.

Kaplan-Meier Curves of CSS in the Patients with wHO/ISUP Grade 4 RCC and Those with sRCC. Abbreviations: wHO/ISUP, World Health Organization/International Society of Urological Pathology; RCC, Renal Cell Carcinoma; sRCC, Sarcomatoid Renal Cell Carcinoma. Kaplan-Meier Curve Plotted Using R Packages ‘Survival [Version 3.3.1]’, ‘Survminer [Version 0.4.9]’, and ‘ggplot2 [Version 3.4.4]’.

Table 2.

Comparison of Clinicopathological Characteristics Between Patients with wHO/ISUP Grade 4 RCC and Those with sRCC.

Characteristics	Description	Grade 4 non-sRCC (N = 27)	sRCC (N = 45)	χ²/T/Z	P
Sex	Male	20 (74.1%)	33 (73.3%)	0.000	1
	Female	7 (25.9%)	12 (26.7%)
Age	Median (IQR)	61.0 (44.5-68.0)	63.0 (53.0-69.0)	−12.981	0.349
BMI	Median (IQR)	22.8 (21.0-24.6)	22.5 (20.8-23.5)	−11.248	0.432
Smoking	Yes	5 (18.5%)	7 (15.6%)	0.000	1
	No	22 (81.5%)	38 (84.4%)
Hypertension	Yes	9 (33.3%)	13 (28.9%)	0.017	0.895
	No	18 (66.7%)	32 (71.1%)
Diabetes	Yes	6 (22.2%)	7 (15.6%)	0.156	0.692
	No	21 (77.8%)	38 (84.4%)
Tumor laterality	Left	12 (44.4%)	25 (55.6%)	0.448	0.503
	Right	15 (55.6%)	20 (44.4%)
Surgical approach	Open	5 (18.5%)	16 (35.6%)	1.618	0.203
	Laparoscopic	22 (81.5%)	29 (64.4%)
Nephrectomy type	Radical	22 (81.5%)	41 (91.1%)	0.686	0.408
	Partial	5 (18.5%)	4 (8.9%)
Surgical duration	Median (IQR)	146.0 (117.5-185.0)	155.0 (105.0-191.0)	−12.359	0.753
Symptoms at diagnosis	Asymptomatic	16 (59.3%)	22 (48.9%)	0.372	0.542
	Symptomatic	11 (40.7%)	23 (51.1%)
Pathological subtype	clear cell RCC	26 (96.3%)	36 (80%)	2.508	0.113
	Non-clear cell RCC	1 (3.7%)	9 (20%)
Maximum tumor diameter	Median (IQR)	7.1 ± 2.7	6.7 ± 2.6	0.717	0.474
Cystic tumor component	Yes	0 (0%)	1 (2.2%)	0.000	1
	No	27 (100%)	44 (97.8%)
Tumor necrosis	Yes	17 (63%)	27 (60%)	0.000	1
	No	10 (37%)	18 (40%)
T stage	T1/T2	15 (55.6%)	20 (44.4%)	0.448	0.503
	T3/T4	12 (44.4%)	25 (55.6%)
WHO/ISUP grade	Grade I	0 (0%)	0 (0%)	-	1
	Grade II	0 (0%)	0 (0%)
	Grade III	0 (0%)	0 (0%)
	Grade IV	27 (100%)	45 (100%)
	Not Applicable	0 (0%)	0 (0%)
Ki-67 expression	Low	17 (63%)	21 (46.7%)	1.204	0.273
	High	10 (37%)	24 (53.3%)
Median postoperative survival	Median	112.0	31.0	4.369	0.037
Follow-up outcome	Death	10 (37%)	31 (68.9%)	5.744	0.017
	Survival	17 (63%)	14 (31.1%)

Comparison of Clinical and Pathological Characteristics and Survival Analysis between Patients with Sarcomatoid Clear Cell Carcinoma and Those with Sarcomatoid Non-Clear Cell Carcinoma

Among the 45 patients with sRCC, 36 were classified as having sarcomatoid clear cell RCC and 9 were classified as having sarcomatoid non-clear cell RCC. A comparison of baseline clinicopathological characteristics revealed no significant differences between the sarcomatoid clear cell RCC and sarcomatoid non-clear cell RCC groups (P > 0.05). The Kaplan–Meier survival curve analysis further demonstrated no statistically significant difference in CSS between the two groups (P = 0.52) (Table 3, Figure 3).

Figure 3.

Kaplan-Meier Curves of CSS in the Patients with Sarcomatoid Clear Cell RCC and Those with Sarcomatoid Non-clear cell RCC. Abbreviations: RCC, Renal Cell Carcinoma; sRCC, Sarcomatoid Renal Cell Carcinoma. Kaplan-Meier Curve Plotted using R Packages ‘Survival [Version 3.3.1]’, ‘Survminer [version 0.4.9]’, and ‘ggplot2 [Version 3.4.4]’.

Table 3.

Comparison of Clinicopathological Characteristics Between Patient with Sarcomatoid Clear Cell RCC and Those with Sarcomatoid non-Clear Cell RCC.

Characteristics	Description	Clear cell RCC (N = 36)	Non-clear cell RCC (N = 9)	χ²/T/Z	P
Sex	Male	28 (77.8%)	5 (55.6%)	0.859	0.354
	Female	8 (22.2%)	4 (44.4%)
Age	Mean ± SD	60.0 ± 13.6	60.8 ± 13.5	−0.154	0.878
BMI	Mean ± SD	22.4 ± 2.1	22.5 ± 1.7	−0.218	0.849
Smoking	No	29 (80.6%)	9 (100%)	0.856	0.355
	Yes	7 (19.4%)	0 (0%)
Hypertension	No	25 (69.4%)	7 (77.8%)	0.007	0.934
	Yes	11 (30.6%)	2 (22.2%)
Diabetes	No	31 (86.1%)	7 (77.8%)	0.011	0.918
	Yes	5 (13.9%)	2 (22.2%)
Tumor laterality	Left	18 (50%)	7 (77.8%)	1.266	0.261
	Right	18 (50%)	2 (22.2%)
Surgical approach	Open	12 (33.3%)	4 (44.4%)	0.055	0.815
	Laparoscopic	24 (66.7%)	5 (55.6%)
Nephrectomy type	Radical	33 (91.7%)	8 (88.9%)	0.000	1
	Partial	3 (8.3%)	1 (11.1%)
Surgical duration	Median (IQR)	155.0 (104.0-212.5)	150.0 (120.0-180.0)	−18.472	0.68
Symptoms at diagnosis	Asymptomatic	18 (50%)	4 (44.4%)	0.000	1
	Symptomatic	18 (50%)	5 (55.6%)
Maximum tumor diameter	Mean ± SD	6.5 ± 2.6	7.3 ± 2.8	−0.800	0.406
Cystic tumor component	No	35 (97.2%)	9 (100%)	0.000	1
	Yes	1 (2.8%)	0 (0%)
Tumor necrosis	No	15 (41.7%)	3 (33.3%)	0.006	0.939
	Yes	21 (58.3%)	6 (66.7%)
T stage	T1/T2	15 (41.7%)	5 (55.6%)	0.141	0.708
	T3/T4	21 (58.3%)	4 (44.4%)
WHO/ISUP grade	Grade I	0 (0%)	0 (0%)	-	1
	Grade II	0 (0%)	0 (0%)
	Grade III	0 (0%)	0 (0%)
	Grade IV	36 (100%)	9 (100%)
	Not applicable	0 (0%)	0 (0%)
Ki-67 expression	Low	18 (50%)	3 (33.3%)	0.273	0.601
	High	18 (50%)	6 (66.7%)
Median postoperative survival	Median	37.0	28.0	0.422	0.516
Follow-up outcome	Death	24(66.7%)	7(77.8%)	0.058	0.809
	Survival	12(33.3%)	2(22.2%)

Abbreviations: BMI, body mass index; IQR, interquartile range; RCC, renal cell carcinoma; SD, standard deviation; WHO/ISUP, World Health Organization/International Society of Urological Pathology. For numerical variables: T-test or One-way ANOVA used for normally distributed data with equal variances; Welch versions if variances are unequal; Wilcoxon or Kruskal-Wallis tests if not normally distributed. For categorical variables: Chi-squared test used if expected counts > 5 and sample size >= 40; apply Yates’ correction if counts are between 1 and 5; Fisher's exact test if counts are < 40.

Cox Univariate and Multivariate Analyses for Patients with Non-Metastatic sRCC

To evaluate independent prognostic risk factors, we performed a Cox univariate analysis, which revealed that presenting symptoms at diagnosis, maximum pathological tumor diameter, T stage, and immunohistochemical Ki-67 expression to be significantly associated with patient prognosis (P < 0.05) (Table 4). Furthermore, the Cox multivariate analysis confirmed that maximum pathological tumor diameter, T stage, and Ki-67 expression were independent prognostic factors influencing postoperative outcomes in patients with non-metastatic sRCC (P < 0.05). (Table 5)

Table 4.

Cox Univariate Analysis of Clinicopathological Characteristics in Patients with non-Metastatic sRCC.

Variable	Description	Statistics	Univariable analysis HR (95%CI, p value)
Sex	Male	33 (73.3%)
	Female	12 (26.7%)	0.96 (0.44-2.10, p = .920)
Age	Median (IQR)	60.2 ± 13.4	1.02 (0.99-1.04, p = .261)
BMI	Median (IQR)	22.4 ± 2.0	1.06 (0.90-1.25, p = .459)
Smoking	No	38 (84.4%)
	Yes	7 (15.6%)	0.74 (0.26-2.13, p = .579)
Hypertension	No	32 (71.1%)
	Yes	13 (28.9%)	0.92 (0.41-2.07, p = .838)
Diabetes	No	38 (84.4%)
	Yes	7 (15.6%)	1.17 (0.44-3.09, p = .748)
Tumor laterality	Left	25 (55.6%)
	Right	20 (44.4%)	0.69 (0.33-1.42, p = .313)
Surgical approach	Open	16 (35.6%)
	Laparoscopic	29 (64.4%)	0.94 (0.46-1.96, p = .878)
Nephrectomy type	Radical	41 (91.1%)
	Partial	4 (8.9%)	1.11 (0.33-3.66, p = .867)
Surgical duration	Mean ± SD	168.8 ± 87.8	1.00 (1.00-1.00, p = .864)
Symptoms at diagnosis	Asymptomatic	22 (48.9%)
	Symptomatic	23 (51.1%)	2.14 (1.03-4.42, p = .041)
Pathological subtype	Clear cell RCC	36 (80.0%)
	Non-clear cell RCC	9 (20.0%)	1.32 (0.57-3.08, p = .514)
Maximum tumor diameter	Mean ± SD	6.7 ± 2.6	1.17 (1.01-1.35, p = .032)
Cystic tumor component	No	44 (97.8%)
	Yes	1 (2.2%)	-*
Tumor necrosis	No	18 (40.0%)
	Yes	27 (60.0%)	1.43 (0.68-2.99, p = .344)
T Stage	T1/T2	20 (44.4%)
	T3/T4	25 (55.6%)	2.47 (1.16-5.26, p = .019)
Ki-67 expression	Low	21 (46.7%)
	High	24 (53.3%)	3.88 (1.69-8.93, p = .001)

* Coefficients in Cox regression analysis failed to converge. Abbreviations: BMI, body mass index; IQR, interquartile range; RCC, renal cell carcinoma; sRCC, sarcomatoid renal cell carcinoma; SD, standard deviation. Cox univariate analysis conducted using R packages ‘survival [version 3.3.1]’ and ‘rms [version 6.3-0]’.

Table 5.

Cox Univariate and Multivariate Analyses of Clinicopathological Characteristics in Patients with non-Metastatic sRCC.

Variable	Description	Statistics	Univariable analysis HR (95%CI, p value)	Multivariable analysis HR (95%CI, p value)
Symptoms at diagnosis	Asymptomatic	22 (48.9%)
	Symptomatic	23 (51.1%)	2.14 (1.03-4.42, p = .041)	1.85 (0.88-3.90, P = 0.107)
Maximum pathological tumor diameter	Mean ± SD	6.7 ± 2.6	1.17 (1.01-1.35, p = .032)	1.20 (1.02-1.41, P = 0.028)
T stage	T1/T2	20 (44.4%)
	T3/T4	25 (55.6%)	2.47 (1.16-5.26, p = .019)	2.44 (1.11-5.39, P = 0.027)
Ki-67 expression	Low	21 (46.7%)
	High	24 (53.3%)	3.88 (1.69-8.93, p = .001)	2.91 (1.25-6.77, P = 0.013)

Cox multivariate analysis conducted using R packages ‘survival [version 3.3.1]’ and ‘rms [version 6.3-0]’.

Development of a Postoperative Mortality Risk Prediction Model and Calibration Curve

Based on the results of Cox multivariate analysis, we developed a postoperative mortality risk prediction model incorporating the maximum pathological tumor diameter, T stage, and immunohistochemical Ki-67 expression as independent risk factors. A calibration curve was generated to evaluate the predictive performance of the model. The calibration curve demonstrated a good fit, with a mean value of 0.069 and a 0.9 quantile value of 0.151, indicating that the model had satisfactory predictive capability for actual clinical outcomes (Figure 4).

Figure 4.

Nomogram for the Postoperative Mortality Risk Prediction Model and Calibration Curve for Patients with non-Metastatic Sarcomatoid Renal Cell Carcinoma. Nomogram and Calibration Curves Developed Using R Packages ‘Survival [Version 3.3.1]’ and ‘rms [Version 6.3-0]’.

Discussion

In this study, we examined sarcomatoid renal cell carcinoma (sRCC), a rare and highly aggressive type of renal cell carcinoma (RCC), accounting for only 4%–5% of all RCCs, with a typical diagnosis between the ages of 54 and 63 years.⁵ The incidence is higher in men than women, as shown by data from Wang et al,⁴ using the National Cancer Database (NCDB), noted an average patient age of 62 years with a male prevalence of 68%. Our study included 45 patients with non-metastatic sRCC, presenting an average age of 60.2 years and a male-to-female ratio consistent with prior large-scale studies.^4,7,14,16 Other baseline data, such as tumor size and major subtypes, were similar to those of previously reported studies. Despite the limited number of participants due to sRCC's rarity, our study spanned a broad timeframe, permitting comprehensive follow-up and data collection. Findings underline that sRCC contrasts markedly with non-sarcomatoid RCC (non-sRCC) in terms of clinical and pathological features. Critical prognostic factors were identified, culminating in a mortality risk prediction model.

The clinical progression of sRCC is notably rapid and invasive. Unlike non-sRCC cases, which often present incidentally,³ the majority of sRCC cases manifest with clinical symptoms, including weight loss and cachexia, contributing to a lower average BMI compared to non-sRCC patients.¹¹ Tumors in sRCC are typically larger,^4–5,7,11 with 90% of cases exhibiting coagulative necrosis, a contrast supported by de Peralta-Venturina et al's findings of 75.2% necrosis rates.²⁵ In our cohort, the median maximum tumor diameter was 6.5 cm versus 4.0 cm in non-sRCC. Furthermore, based on T staging, if the tumor invades surrounding tissues such as the renal capsule, veins or ipsilateral adrenal gland, it is classified as T3 or T4.²⁶ A higher percentage of sRCC cases were staged as T3 or T4 compared to non-sRCCs, as corroborated by studies from Wang et al⁴ and Tully et al.¹⁴ These characteristics complicate surgical management due to extrarenal extension and required open surgical approaches, often extending the duration of surgery.

From a molecular perspective, high Ki-67 expression ties closely with poor prognosis and high malignancy, consistent with previous research acknowledging Ki-67 as a proliferation marker linked to worse survival.^27,28 Studies have shown that patients with Ki-67 expression ≥10% have significantly lower OS, CSS, and disease-free survival compared to those with Ki-67 expression < 10%,^22,29 based on this, we divided Ki-67 into high-expression and low-expression groups. Our results indicate that sRCC patients have a markedly higher rate of high Ki-67 expression than those with non-sRCC, suggesting elevated proliferative activity and aggressiveness, which ties into poorer survival outcomes.

The prognosis of grade 4 RCC is significantly worse than that of lower-grade RCC (grades 1-3).³⁰ Notably, a significant proportion of patients with RCC are diagnosed with lower-grade tumors, which generally have a better prognosis. This explains the relatively better survival outcomes observed in patients with non-sarcomatoid RCC than in those with sRCC, as lower-grade tumors were more prevalent in the former patient group. Our findings align with those of previous studies, such as those by Wang et al⁴ and Ullah et al,¹² which confirmed that patients with sRCC exhibit notably worse survival outcomes compared to patients with non-sRCC. However, in our study, we found that patients with sRCC had an even poorer prognosis than those with grade 4 RCC. These findings highlights the aggressive nature of sRCC and the limitations of the current WHO/ISUP grading system in accurately reflecting the extreme malignancy of sRCC, since sRCC is categorized as grade 4 RCC. Beyond these survival differences, we observed that the median survival time for patients with sRCC in our study was 31 months, which exceeds prior publications, the stringent inclusion criteria only focusing on non-metastatic cases might partly elucidate this deviation. In addition, advancements in imaging technology and heightened public awareness have likely extended patient survival by facilitating earlier-stage interventions.^4,12

In-depth prognostic factor assessment further clarifies this study's contribution. Tumor size remains critical, with larger tumors linked to unfavorable prognosis,^4,12,17,31 our study finding is consistent with those reported previously, as the maximum tumor diameter was confirmed as an independent risk factor for postoperative mortality. T stage continues to hold prognostic value, as confirmed in studies by Tully et al¹⁴ and empirically validated in our study's multivariate analysis. Several studies have demonstrated that Ki-67 overexpression is significantly associated with poor prognosis in patients with sRCC,^27–28,32 our result align with those reported by them. Despite clinical symptom presentation correlating with higher mortality rates, it didn’t retain statistical significance as an independent predictor post-covariate adjustment. This may be attributed to subjective variability in tumor invasiveness, tumor location, individual sensitivity, and timing of medical consultation. Previous studies have suggested that the pathological subtype of sRCC may influence prognosis.^7,33 However, our histological subtype analysis within sRCC revealed slight prognostic disparities that did not reach statistical significance, further studies, particularly large-scale ones like those by Blum et al,⁷ are requisite for defining correlations.

To address these insights, we derived an innovative mortality risk prediction model integrating maximum tumor diameter, T stage, and Ki-67 expression — determinants verified through calibration curve analysis for predictive validity. This is the first postoperative mortality risk prediction model for non-metastatic sRCC. The nomogram represents a proactive step in tailoring patient prognosis toward personalizing clinical strategy deployment.

However, certain limitations should be acknowledged. First, this study was conducted at a single center with a relatively small sample size, which may have introduced selection bias. The limited cohort size prevented us from performing external validation. Future studies should incorporate larger cohorts and multicenter collaborations to enhance the generalizability of the model. Second, owing to its retrospective design, this study may have been subject to information bias and confounding factors. Therefore, a prospective study is required to validate the effectiveness and robustness of this predictive model. Third, this study excluded the proportion of sarcomatoid component proportion from the analysis, even though previous studies indicate that the extent of sarcomatoid differentiation within tumors may impact prognosis.^9,22,31–32 Future studies should integrate quantitative pathological assessment methods to explore the prognostic significance of the proportions of sarcomatoid components.

Projected research enhancements center on broader, collaborative investigations and incorporating machine learning capabilities to couple multimodal data — spanning genetic, pathological, and radiomic dimensions — into enriched, actionable predictive frameworks. The detailed role of sarcomatoid components warrants integration to dissect their mechanistic impact on progression and treatment responsiveness. Future research directives should also devote efforts to optimizing therapeutic interventions, assessing the relevance and effectiveness of emerging targeted or immunotherapeutic modalities. Such refinements could invigorate management principles and improving the collective quality-of-life metrics and survival prospects for sRCC patients.

In summary, this study confirmed that patients with sRCC exhibit significant clinicopathological differences from patients with non-sRCC and revealed independent prognostic risk factors affecting survival outcomes. We found that even non-metastatic patients exhibited dismal outcomes, indicating tumor aggressiveness and malignant potential exceeding initial expectations. We have developed the first postoperative mortality risk prediction model for this population that can be utilized for the early identification of high-risk patients, optimization of follow-up strategies, and personalized treatment guidance, ultimately aiming to improve survival outcomes in patients with sRCC.

Conclusion

Our study demonstrated that patients with sRCC exhibit significant clinicopathological differences compared to patients with non-sRCC, highlighting the highly malignant and aggressive nature of sRCC. Notably, patients with sRCC have a worse prognosis than those with WHO/ISUP grade 4 non-sRCC. Using Cox univariate and multivariate analyses, we identified maximum pathological tumor diameter, T stage, and Ki-67 expression as independent prognostic factors for postoperative mortality in patients with non-metastatic sRCC. Based on these findings, we developed a mortality risk prediction model to identify high-risk patients after surgery, thereby enabling closer follow-up and more proactive treatment interventions. This is the first predictive model designed specifically for patients with non-metastatic sRCC. It holds significant clinical value for early recurrence detection, timely intervention, and improvement in patient survival outcomes.

Footnotes

Acknowledgments

We would like to thank SAGE Author Services for editing and reviewing this manuscript for English language.

ORCID iDs

Lian Fang

Zhiyu Zhang

Xuefeng Zhang

Ethical

The authors assume full responsibility for ensuring the accuracy and integrity of all aspects of the study. Any concerns regarding the precision or honesty of any part of the work were thoroughly investigated and addressed. The study was conducted in accordance with the Declaration of Helsinki (revised in 2013). The study was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (No. 344, 2023), and written informed consent was obtained from all participants.

Author Contributions

(I) Conception and design: Shicheng Sun, Nan Wang and Lian Fang; (II) Methodology and investigation: Nan Wang, Lian Fang and Zhiyu Zhang; (III) Software and formal analysis: Lian Fang, Ouyang Song, Yufeng Hou; (IV) Data curation: Xuefeng Zhang, Lian Fang, Hujuan Yang; (V)Supervision and funding acquisition: Shicheng Sun, Xuefeng Zhang and Jun Ouyang; (VI) Manuscript writing: Lian Fang, Zhiyu Zhang, Ouyang Song; (VII) Final approval of the manuscript: All authors.

Funding

The authors disclose receipt of the following financial support for the research, authorship, and/or publication of this article. This work was supported by the Suzhou Science and Technology Project (grant numbers SLJ201906).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The datasets generated during and analyzed during the current study are available from the corresponding author on reasonable request.

References

Powles

Albiges

Bex

, et al. Renal cell carcinoma: ESMO clinical practice guideline for diagnosis, treatment and follow-up. Ann Oncol. 2024;35(8):692–706. doi:10.1016/j.annonc.2024.05.537

Feng

, et al. Global burden and cross-country inequalities in urinary tumors from 1990 to 2021 and predicted incidence changes to 2046. Mil Med Res. 2025;12(1):12. doi:10.1186/s40779-025-00599-y

Campbell

Clark

Chang

Karam

Souter

Uzzo

. Renal mass and localized renal cancer: Evaluation, management, and follow-up: AUA guideline: Part I: AUA guideline. J Urol. 2021;206(2):199–208. doi:10.1097/JU.0000000000001911

Wang

Puri

Saitta

, et al. Trends and outcomes in sarcomatoid renal cell carcinoma: Analysis of the national cancer data base. Eur Urol Open Sci. 2024;71:96–105. doi:10.1016/j.euros.2024.10.002

Blum

Gupta

Tickoo

, et al. Sarcomatoid renal cell carcinoma: Biology, natural history and management. Nat Rev Urol. 2020;17(12):659–678. doi:10.1038/s41585-020-00382-9

Moch

Cubilla

Humphrey

Reuter

Ulbright

. The 2016 WHO classification of tumours of the urinary system and male genital organs-part A: Renal, penile, and testicular tumours. Eur Urol. 2016;70(1):93–105. doi:10.1016/j.eururo.2016.02.029

Blum

Silagy

Knezevic

, et al. Localised non-metastatic sarcomatoid renal cell carcinoma: A 31-year externally verified study. BJU Int. 2024;133(2):169–178. doi:10.1111/bju.16125

Ciccarese

Büttner

Cerbone

, et al. Clinical features and response to immune combinations in patients with renal cell carcinoma and sarcomatoid de-differentiation (ARON-1 study). Int J Cancer. 2024;155(11):2036–2046. doi:10.1002/ijc.35141

Shan

Shao

, et al. Clinicopathologic factors linked to oncologic outcomes for renal cell carcinoma with sarcomatoid dedifferentiation: A PRISMA-compliant systematic review and meta-analysis. Front Surg. 2022;9:922150. doi:10.3389/fsurg.2022.922150

10.

Farrow

Harrison

Utz

ReMine

. Sarcomas and sarcomatoid and mixed malignant tumors of the kidney in adults. I. Cancer. 1968;22(3):545–550. doi:10.1002/1097-0142(196809)22:3<545::aid-cncr2820220308>3.0.co;2-4

11.

Adeniran

Shuch

Humphrey

. Sarcomatoid and rhabdoid renal cell carcinoma: Clinical, pathologic, and molecular genetic features. Am J Surg Pathol. 2024;48(7):e65–e88. doi:10.1097/PAS.0000000000002233

12.

Ullah

Yasinzai

AQK

Sakhalkar

, et al. Demographic patterns and clinicopathological analysis of sarcomatoid renal cell carcinoma in US population. Clin Genitourin Cancer. 2024;22(1):38–46. doi:10.1016/j.clgc.2023.07.010

13.

Wang

Tuo

, et al. Natural products and derivatives in renal, urothelial and testicular cancers: Targeting signaling pathways and therapeutic potential. Phytomedicine. 2024;127:155503. doi:10.1016/j.phymed.2024.155503

14.

Tully

Berg

Paciotti

, et al. The natural history of renal-cell carcinoma with sarcomatoid differentiation, a stage-by-stage analysis. Clin Genitourin Cancer. 2023;21(1):63–68. doi:10.1016/j.clgc.2022.11.015

15.

Alevizakos

Gaitanidis

Nasioudis

Msaouel

Appleman

. Sarcomatoid renal cell carcinoma: Population-based study of 879 patients. Clin Genitourin Cancer. 2019;17(3):e447–e453. doi:10.1016/j.clgc.2019.01.005

16.

Zhao

Chen

Xie

Zhang

Hou

Jin

. Clinicopathologic features and prognostic factors in patients with renal cell carcinoma with sarcomatoid differentiation. APMIS. 2020;128(5):378–386. doi:10.1111/apm.13035

17.

Brookman-May

May

Shariat

, et al. Prognostic effect of sarcomatoid dedifferentiation in patients with surgically treated renal cell carcinoma: A matched-pair analysis. Clin Genitourin Cancer. 2013;11(4):465–470. doi:10.1016/j.clgc.2013.04.026

18.

Hahn

Lebenthal

Genovese

Sircar

Tannir

Msaouel

. The significance of sarcomatoid and rhabdoid dedifferentiation in renal cell carcinoma. Cancer Treat Res Commun. 2022;33:100640. doi:10.1016/j.ctarc.2022.100640

19.

Hou

Zheng

, et al. Construction and validation of a novel prognostic nomogram for patients with sarcomatoid renal cell carcinoma: A SEER-based study. Int J Clin Oncol. 2020;25(7):1356–1363. doi:10.1007/s10147-020-01681-2

20.

, et al. Prognostic value of preoperative inflammatory response biomarkers in patients with sarcomatoid renal cell carcinoma and the establishment of a nomogram. Sci Rep. 2016;6:23846. doi:10.1038/srep23846

21.

Collins

Reitsma

Altman

Moons

. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement. Br Med J. 2015;350:g7594–g7594. doi:10.1136/bmj.g7594

22.

Zhang

Thompson

Lohse

, et al. A novel prognostic model for patients with sarcomatoid renal cell carcinoma. BJU Int. 2015;115(3):405–411. doi:10.1111/bju.12781

23.

Galtung

Lauritzen

Baco

Berg

Naas

Rud

. Predictive performance of prospectively applied ISUP and Fuhrman grade in nonmetastatic renal cell carcinoma. Anticancer Res. 2022;42(6):2967–2975. doi:10.21873/anticanres.15780

24.

Xiao

Guan

, et al. Validation of the world health organization/international society of urological pathology grading for Chinese patients with clear cell renal cell carcinoma. Transl Androl Urol. 2020;9(6):2665–2674. doi:10.21037/tau-20-799

25.

de Peralta-Venturina

Moch

Amin

, et al. Sarcomatoid differentiation in renal cell carcinoma: A study of 101 cases. Am J Surg Pathol. 2001;25(3):275–284. doi:10.1097/00000478-200103000-00001

26.

Young

Jackson-Spence

Beltran

, et al. Renal cell carcinoma. Lancet. 2024;404(10451):476–491. doi:10.1016/S0140-6736(24)00917-6

27.

Zeng

Zhou

Wen

Zhu

Luo

Wang

. The relationship between the expression of Ki-67 and the prognosis of osteosarcoma. BMC Cancer. 2021;21(1):210. doi:10.1186/s12885-021-07880-y

28.

Zheng

Zhu

Tan

Yang

Zhang

. Retrospective analysis of a large patient sample to determine p53 and Ki67 expressions in renal cell carcinoma. J BUON. 2014;19(2):512–516.

29.

Xie

Chen

, et al. Prognostic and clinicopathological role of high Ki-67 expression in patients with renal cell carcinoma: A systematic review and meta-analysis. Sci Rep. 2017;7:44281. doi:10.1038/srep44281

30.

Khor

L-Y

Dhakal

Jia

, et al. Tumor necrosis adds prognostically significant information to grade in clear cell renal cell carcinoma. A study of 842 consecutive cases from a single institution. Am J Surg Pathol. 2016;40(9):1224–1231. doi:10.1097/PAS.0000000000000690

31.

Adibi

Thomas

Borregales

, et al. Percentage of sarcomatoid component as a prognostic indicator for survival in renal cell carcinoma with sarcomatoid dedifferentiation. Urol Oncol. 2015;33(10):427.e17. doi:10.1016/j.urolonc.2015.04.011

32.

Wang

Zeng

Chen

. Ki-67 index and percentage of sarcomatoid differentiation were two independent prognostic predictors in sarcomatoid renal cell carcinoma. Cancer Manag Res. 2018;10:5339–5347. doi:10.2147/CMAR.S176242

33.

Chahoud

Msaouel

Ross

, et al. Outcomes of patients with metastatic renal cell carcinoma with sarcomatoid dedifferentiation to immune checkpoint inhibitors. Urol Oncol. 2021;39(2):134.e9–134.e16. doi:10.1016/j.urolonc.2020.10.019