While the majority of renal cell carcinomas (RCCs) are clear cell RCC (ccRCC), non-ccRCCs comprise up to 25% of RCCs. Non-ccRCCs are further divided into histologic subtypes, including papillary (10%), chromophobe (5%), collecting duct (1%), and translocation RCC (<1%) (1). Sarcomatoid differentiation can be found in any subtype (2). A systematic review of 49 studies in 2015 found that systemic therapies for primarily metastatic RCC conferred poorer progression-free survival and overall survival (OS) in patients with non-ccRCC compared to those in patients with ccRCC (3). However, robust data regarding oncologic outcomes in patients with non-ccRCC, particularly with respect to neoadjuvant and adjuvant therapy, are more limited.

Currently, the standard treatment algorithm for localized or locoregional non-ccRCC is surgery followed by surveillance, mirroring management for ccRCC (4). While oncologic outcomes and the role of adjuvant therapy have not been well studied for localized non-ccRCC, there have been trials comparing systemic therapies in advanced non-ccRCC. Notably, the ESPN, ASPEN, and RECORD-3 trials compared sunitinib and everolimus (5–7). A meta-analysis found that sunitinib may be more effective than everolimus, but the difference in outcomes was not statistically significant (8). However, these trials were limited by not differentiating between histologic subtypes.

Recent molecular profiling has increasingly shown non-ccRCC to be a heterogenous disease (9). MET mutations are frequently present in papillary RCC, whereas mitochondrial alterations are frequently found in the chromophobe subtype (10, 11). The collecting duct subtype appears to be immunogenic, with upregulation of genes associated with T-cell activity along with lymphocytic infiltration on biopsy (12). Sarcomatoid RCCs may have increased frequency of mutations in TP53, CKDN2A, and NF2 (13). These molecular differences suggest that different systemic therapies may be required for the various subtypes of non-ccRCC.

In light of the rarity and lack of robust data for localized non-ccRCC, we utilized a large hospital-based dataset to describe oncologic outcomes of papillary, chromophobe, collecting duct, and sarcomatoid subtypes of non-ccRCC in comparison to ccRCC.

Overall, 220,170 RCC patients were included in our analysis. A total of 178,066 patients were found to have non-metastatic ccRCC and 42,104 with non-ccRCC. The majority of the non-ccRCC cohort was composed of papillary and chromophobe subtypes, 27,510 (65.3%) and 12,760 (30.3%), respectively. There were notable differences in the distribution of demographic characteristics and tumor-specific information ( Table 1 ). Patients with papillary histology were more commonly male (72.6% male vs. 27.4% female), and chromophobe histology was more likely to present in patients below age 65 years compared to those above 65 years of age (65.2% vs. 34.8%).

At the time of diagnosis or treatment, there were 73.8% collecting duct and 87.9% sarcomatoid RCC patients with poorly or undifferentiated histology. Conversely, only 37.2%, 30.1%, and 32.4% patients with chromophobe, clear cell, and papillary RCC had poorly or undifferentiated histology, respectively. Similarly, 49.1% and 55.2% of patients with collecting duct and sarcomatoid histology had stage III disease. While 12.2%, 16.0%, and 9.4% patients with chromophobe, clear cell, and papillary RCC had stage III disease, respectively.

A higher proportion of patients with collecting duct (63.0%) and sarcomatoid (71.7%) histology underwent radical surgery compared to patients with chromophobe, clear cell, and papillary histology. Similarly, regional lymph node surgery was performed in 28.9% and 35.1% patients with collecting duct and sarcomatoid histology. Regional lymph node surgery was less common in the remaining subtypes. In terms of radiation therapy, most patients with RCC did not receive radiation treatment. A higher proportion of patients with collecting duct (12.1%) and sarcomatoid (10.1%) RCC received chemotherapy compared to other histologies.

We analyzed a large hospital-based dataset to demonstrate several key findings regarding non-ccRCC. Our results demonstrate that 1) patients with collecting duct and sarcomatoid RCC present with higher stage and have significantly inferior oncologic outcomes relative to other subtypes and 2) patients who received radiation or chemotherapy had worse OS, although they are likely some of the sickest patients with other unmeasured risk factors (genetic mutations) for poor outcomes for any non-ccRCC subtype. These findings highlight the unmet need for more effective therapies for non-ccRCC.

Our findings are consistent with the existing literature regarding presenting characteristics of non-ccRCC. As demonstrated in prior studies, patients with the collecting duct and sarcomatoid subtypes presented more commonly with clinical T3+ and metastatic than regional disease (14, 15). Additionally, case volume was associated with outcomes, as patients treated at facilities that treat <15 cases per year experienced worse survival in our analysis. Black race was associated with both worse presentation and worse outcomes relative to others. Nearly a quarter of collecting duct RCC was observed in the Black race and was found to have higher grade and stage. Also, carrying no health insurance or public insurance was associated with worse survival, and earning <$48,000 per residential zip code was associated with worse outcomes.

With respect to oncologic outcomes, our findings also confirm that collecting duct and sarcomatoid histologies are associated with worse survival compared to other types of RCC, with over two times increased risk of death compared to patients with ccRCC. Conversely, patients with chromophobe were at 36% decreased risk of death compared to the clear cell, which is consistent with prior literature (15). OS for papillary RCC was not found to be significantly different from RCC ( Figure 1 ). Other findings may have been influenced by the small sample size and uncontrolled selection bias: 1) regional lymphadenectomy was associated with worse outcomes, 2) adjuvant radiation was not associated with improved survival, and 3) adjuvant chemotherapy was not associated with improved survival. Vascular endothelial growth factor (VEGF), mammalian target of rapamycin (mTOR), and immune checkpoint inhibitors have shown significant clinical benefit in metastatic RCC (16). These agents are being evaluated in the adjuvant setting for non-metastatic RCC. For patients with high-risk stage III ccRCC, adjuvant sunitinib was approved by the Food and Drug Administration (FDA) in 2017 based on the S-TRAC trial, which showed improved disease-free survival (DFS) relative to placebo (17). However, due to lack of OS benefit and concerns with toxicity, adjuvant sunitinib is infrequently administered in practice. In the ASSURE clinical trial, which assessed patients with resected RCC, of whom 21% had non-ccRCC, adjuvant sunitinib or sorafenib did not show DFS benefit in comparison to placebo (18). Due to these studies, adjuvant sunitinib for stage I–III non-ccRCC is not recommended.

Currently, the role of adjuvant immune checkpoint inhibitors is unknown. In the metastatic setting, the combination of nivolumab and ipilimumab has shown better progression-free survival and OS in comparison to sunitinib in patients with metastatic sarcomatoid RCC (19). Currently, there are multiple ongoing phase III clinical trials evaluating the role of immune checkpoint inhibitors in the adjuvant setting of RCC with or without sarcomatoid features (20). In the adjuvant study evaluating the efficacy of nivolumab, any RCC histologic subtype is allowed.

Our study has several limitations. Due to the rarity of collecting duct and sarcomatoid subtypes, despite our inclusion of more than 220,000 patients with stage I–III RCC, there were only 273 patients with collecting duct and 1,561 patients for sarcomatoid histology, and the number of patients for both subtypes decreased even further with multivariate analysis, as patients without matching variables were excluded from the multivariate analysis. Furthermore, the NCDB dataset is limited by additional unknown confounding variables and missing data. NCDB does not collect specific granular treatment-related information regarding systemic therapy such as chemotherapy, VEGF, mTOR inhibitors, or immunotherapy types, and hence, it was not feasible to derive more specific treatment-associated outcomes. Limited economic factors included in the NCDB are also adjusted for in the multivariable model. While the patient’s income is not collected by NCDB, the median income for the ZIP code the patient lives in is included. Additionally, primary payor, area, and distance to hospital can be considered proxies for access to care. After adjusting for these available metrics, black race was still significantly associated with increased risk of death relative to white race. However, the statistically significant result may not have practical implications given the smaller in magnitude HR evidenced (HR = 1.11). Lastly, renal medullary and translocation RCC were not available as distinct histologies on NCDB; therefore, they were not included in this study.

Collecting duct and sarcomatoid histologies were uncommon but were more likely to present with higher stage and have worse survival when compared to other RCC subtypes. Consistent with the current standard of care, surgical treatment for stage I–III RCC showed the greatest survival benefit compared to those without surgery or ablation procedures only. Further studies are needed to evaluate effective adjuvant treatment for non-ccRCC, particularly to improve outcomes for the collecting duct and sarcomatoid subtypes due to significantly worse survival compared to other subtypes.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

JA: Conception and design, collection and assembly of data, data analysis and interpretation, article writing, final approval of article. VP: Conception and design, collection and assembly of data, data analysis and interpretation, article writing, final approval of article. AC: Conception and design, article writing, final approval of article. JR: Data analysis and interpretation, article writing, final approval of article. JJ: Data analysis and interpretation, article writing, final approval of article. SM: Collection and assembly of data, data analysis and interpretation, article writing, final approval of article. RG: Conception and design, collection and assembly of data, data analysis and interpretation, article writing, final approval of article. All authors (a) have made substantial contributions to the work including conception, performance, or interpretation of data and (b) writing the article, (c) approved the final version to be published, and (d) agreed to be accountable for the accuracy and integrity of the work.

Department funding by the Division of Hematology and Oncology at the University of Iowa Hospitals and Clinics.

RG received research funding from Endocyte/Advanced Accelerator Applications and Pfizer. VP has done consulting for Cold Genesys and Guidepoint.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.