SARCOMATOID/RHABDOID RENAL TUMORS RESPND TO IMMUNOTHERAPY?

• Sarcomatoid and rhabdoid (S/R) renal cell carcinoma (RCC) tumors are associated with poor prognosis and resistance to targeted therapies.
• Recent data suggest S/R RCC is especially responsive to immune checkpoint inhibitor (ICI)–based therapies.
• Further studies are needed to pinpoint the origins of the responsiveness of S/R RCC to ICIs and to improve outcomes for patients with these aggressive forms of cancer.

Renal cell carcinoma (RCC) accounts for more than 175,000 deaths worldwide, with 400,000 new cases diagnosed annually.¹ Although most deaths from RCC occur in patients with metastatic disease, the majority of patients are initially diagnosed with localized or locoregional disease.² Sarcomatoid and rhabdoid (S/R) RCC are the most common forms of de-differentiation in RCC, can occur over background histologies of clear cell or non–clear cell RCC, are associated with a poor prognosis, and are over-represented in metastatic disease—accounting for up to 15% of patients with metastatic RCC.^3,4

Although therapies targeting the VEGF and mTOR pathways have been shown to significantly improve survival in patients with clear cell RCC, S/R RCC is largely refractory to these agents.⁵ Cytotoxic chemotherapy can yield responses in S/R RCC, although many are not durable, and toxicity can be significant.⁶ However, recent data have suggested that immune checkpoint inhibitor (ICI)–based therapies may be particularly effective for these tumors.^7-12 There are currently little data evaluating the molecular characteristics underlying the aggressive clinical phenotype and responses to ICIs of these tumors. Herein, we summarize the clinical data evaluating the efficacy of targeted therapies and ICIs in S/R RCC and evaluate whether novel biologic and clinically relevant insights can be gained from studying the biology of these tumors

Poor Prognosis, Resistance to Targeted Therapies, and Promising ICI Data

S/R components, which can arise in the same tumors or in distinct RCC tumors, often form only a fraction of the overall volume of RCC tumors. The proportion of cells within RCC tumors harboring these features ranges from 1% to 100%, with a median of approximately 25%.^{4, 13-14} Pure S/R RCC tumors are rare, and generally an epithelioid background histology (clear cell or non–clear cell) can be identified with sufficient tumor sampling.¹⁴Although the volumetric proportion of the S/R RCC components can vary significantly, the presence of any percentage of S/R RCC components is considered to constitute grade 4 histology¹⁴ and is associated with a poor prognosis.⁵

Some reports have previously evaluated the outcomes of patients with sarcomatoid and/or rhabdoid features treated by targeted therapies. Golshayan et al. first reported in 2009 that VEGF pathway inhibitors had some efficacy in patients with sarcomatoid RCC.¹⁵ Although no patients had complete responses (CR), the overall response rate (ORR) was around half that of nonsarcomatoid RCC tumors (25% vs. 50%), and patients with sarcomatoid RCC had generally poor outcomes. Similar subsequent reports confirmed the dismal outcomes of these tumors on VEGF pathway–targeted therapies, with similar outcomes on mTOR pathway–targeted therapies, chemotherapy, and cytokine therapy.^5,16,17However, it is notable that a subsequent phase II trial of a combination of sunitinib and gemcitabine (VEGF pathway inhibitor and chemotherapy, respectively) led to a relatively improved ORR of 26% in patients with sarcomatoid RCC. On the other hand, data for RCC tumors with rhabdoid de-differentiation have been sparser, with only a few case reports published on the efficacy of targeted therapies.^18,19

Recently, there has been a surge of interest in S/R RCC after multiple real-world studies and subgroup analyses of clinical trials reported that ICIs significantly improved outcomes in these tumors (Table). The interest in these tumors was perhaps best encapsulated by the title of a dedicated poster discussion session during the 2019 ASCO Annual Meeting: “Sarcomatoid RCC: Defining a New Treatment Paradigm?” This interest was based on the readout of the subgroup analyses of three phase III clinical trials (CheckMate 214, KEYNOTE-426, and IMmotion 151),^10-12 as well as two large real-world cohorts⁹ evaluating the role of ICI-based therapies in S/R RCC. Each of these studies showed that patients with sarcomatoid and/or rhabdoid RCC had response and survival outcomes that were significantly improved by ICI-based therapies compared with targeted therapies. Strikingly, the CR rate ranged from 10% to 18% in patients with sarcomatoid RCC treated by ICI-based regimens in the three clinical trials, versus just 0% to 3% in patients who had been randomly assigned to the sunitinib (control) arms of these studies.^10-12View larger

The subsequent presentation of the subgroup analysis of the JAVELIN Renal 101 trial during the 2019 ESMO Congress served as additional validation for the improved results of sarcomatoid RCC with ICI therapies.²⁰ These results were further corroborated by two large real-world cohorts (from the Harvard group and International Metastatic RCC Database Consortium), in which survival and response outcomes were shown to be significantly increased in patients with S/R RCC on ICI-based therapies compared with non-ICI therapies.⁹ Moreover, the benefit derived from ICIs for S/R RCC seemed to hold in both patients with clear cell and non–clear cell background histologies.^21-23 However, although the extent of benefit from ICI therapies seems to be greater in patients with RCC tumors harboring S/R components, the survival outcomes of these patients remain inferior to those without S/R RCC, further attesting to the aggressive nature of these tumors.¹⁰

S/R RCC: A Model for Immune-Responsive Aggressive Tumors?

Despite the recent increased interest in S/R RCC tumors and the reports of responsiveness to ICI-based therapies, our understanding of the molecular biology of these tumors and the drivers of their aggressivity and responsiveness to ICIs is fairly limited. A few small studies have evaluated the molecular features of these tumors, and their conclusions have been somewhat discordant. Paired analyses of the sarcomatoid and epithelioid components of these tumors, through either DNA sequencing or evaluation of X chromosome inactivation patterns, have shown that both the sarcomatoid and epithelioid components originate from the same clonal origin.^24,25 Some studies have further suggested that there are few differences in genomic alterations between the sarcomatoid and epithelioid components,²⁶ while others have suggested that the sarcomatoid component harbors a higher overall mutational load in addition to more frequent mutations in BAP1, TP53, and ARID1A.²⁴ Other studies still have not corroborated these findings of differences between the sarcomatoid and epithelioid components and have instead suggested that mutational differences are found to be more pronounced when comparing sarcomatoid- with nonsarcomatoid-containing RCC tumors.²⁷

Despite the recent increased interest in S/R RCC tumors and the reports of responsiveness to ICI-based therapies, our understanding of the molecular biology of these tumors and the drivers of their aggressivity and responsiveness to ICIs is fairly limited.

In particular, a study by Wang et al. suggested that sarcomatoid-containing RCC tumors harbor more frequent mutations in PTEN, TP53, and RELN along with less frequent two-copy losses in VHL and PBRM1.²⁷ Moreover, although this study further suggested that the inactivation of RELN leads to upregulation of TGFβ signaling in these tumors (which is associated with ICI resistance and immune exclusion),²⁸ these results seem difficult to reconcile with other reports, suggesting that sarcomatoid RCC tumors harbor an immune-infiltrated phenotype characterized by PD-L1 expression, increased CD8⁺ T-cell infiltration, and increased responsiveness to ICIs.^11,20,29,30 Two recent studies have further suggested that genomic amplification at the CD274 (or PD-L1) gene locus, 9p24.1, may be partly driving the increased PD-L1 expression and responses to immunotherapy. However, the proportion of S/R RCC tumors that harbor these amplifications (5%-6%) is insufficient to alone account for the high rates of PD-L1 expression and response to ICI therapies of these tumors.^30,31,32

Overall, although these studies evaluating the molecular features of S/R RCC provide interesting insights into the biology underlying these tumors, the small sample sizes included in these studies and their somewhat contradictory findings reinforce the need for further studies to better elucidate the drivers of aggressivity and responsiveness to ICIs in these tumors. The importance of such studies is 2-fold; first, although the outcomes of patients with S/R RCC have significantly improved with ICI-based therapies, their outcomes remain inferior to those without S/R RCC, which reflects the inherent aggressivity of these tumors. Therefore, better exploring the genomic drivers of these tumors could allow for more potent targeting of these tumors to improve patients’ poor outcomes. Second, elucidating the molecular underpinnings of the responses of S/R RCC tumors to ICI therapies may allow better selection of subsets of these tumors for ICI therapies and more broadly inform us about the molecular features that drive ICI response in all RCC tumors as well as aggressive solid tumors in general.

In sum, S/R RCC tumors are relatively frequent forms of RCC that represent forms of subclonal de-differentiation, are over-represented in metastatic tumors, and are associated with poor prognosis and resistance to targeted therapies. Recent data have suggested that these tumors are especially responsive to ICI-based combinations, but the drivers of aggressivity and responsiveness to ICIs in these tumors remain largely elusive. Further studies are needed to build on recent improvements in the treatment of patients with these tumors and to more generally inform the molecular mechanisms of response to ICIs in aggressive forms of cancer.