Phase II Trial and Correlative Genomic Analysis of Everolimus Plus Bevacizumab in Advanced Non-Clear Cell Renal Cell Carcinoma

Abstract

Purpose The decreased effectiveness of single-agent targeted therapies in advanced non-clear cell renal cell carcinoma (ncRCC) compared with clear cell renal cell carcinoma (RCC) supports the study of combination regimens. We evaluated the efficacy of everolimus plus bevacizumab in patients with metastatic ncRCC. Patients and Methods In this single-center phase II trial, treatment-naive patients received everolimus 10 mg oral once per day plus bevacizumab 10 mg/kg intravenously every 2 weeks. The primary end point was progression-free survival (PFS) at 6 months. Correlative analyses explored candidate tissue biomarkers through next-generation sequencing. Results Thirty-five patients were enrolled with the following histologic subtypes: chromophobe (n = 5), papillary (n = 5), and medullary (n = 2) RCC and unclassified RCC (uRCC, n = 23). The majority of patients had papillary growth as a major component (n = 14). For 34 evaluable patients, median PFS, overall survival, and objective response rate (ORR) were 11.0 months, 18.5 months, and 29%, respectively. PFS varied by histology ( P < .001), and ORR was higher in patients with significant papillary (seven of 18) or chromophobe (two of five) elements than for others (one of 11). Presence of papillary features were associated with benefit, including uRCC, where it correlated with ORR (43% v 11%), median PFS (12.9 v 1.9 months), and overall survival (28.2 v 9.3 months; P < .001). Several genetic alterations seemed to segregate by histology. In particular, somatic mutations in ARID1A were seen in five of 14 patients with papillary features but not in other RCC variants. All five patients achieved treatment benefit. Conclusion The study suggests efficacy for this combination in patients with ncRCC characterized by papillary features. Distinct mutational profiles among ncRCCs vary according to specific histology.

Conflict of interest statement

Authors’ disclosures of potential conflicts of interest are found in the article online at www.jco.org. Author contributions are found at the end of this article.

Figures

Fig 1.

Efficacy assessment by subject (colors encode renal cell carcinoma [RCC] variants). (A) Swimmer plot depicts individual patients as lines. Arrows indicate patients who remained on therapy. The primary end point for the trial was progression-free survival at 6 months (vertical line). (B) The best radiographic response for 33 evaluable patients treated. Best response was assessed per Response Evaluation Criteria in Solid Tumors 1.1. OR, objective response. (*) Discontinuation due to toxicity. (†) Voluntary withdrawal from study.

Fig 2.

Kaplan-Meier curves that summarize outcomes for patients with unclassified renal cell carcinoma (uRCC). Separate curves depict the comparison between subjects with uRCC with (n = 14; red line) and without (n = 9; black line) significant papillary features. (A) Median progression-free survival differed significantly between groups (12.9 v 1.9 months; log-rank P = .01). (B) Median overall survival was significantly longer for uRCC with papillary features than for other uRCC variants (28.2 v 9.3 months; log-rank P < .001).

Fig 3.

Oncogenomic changes detected by deep-sequence analysis from archival tumor specimens across 28 subjects. Columns represent individual subjects; rows represent selected genes of interest examined for each sample. Subjects who reached the primary efficacy goal of 6 months progression-free survival (PFS) are marked blue; those who did not are marked gold. Subjects censored within their first 6 months of the trial are marked gray. Colored squares mark the presence of somatic alterations detected by sequence analysis. ARID1A mutations were seen in five of 14 tumors with major papillary features; no ARID1A mutations were seen in all other renal cell carcinoma (RCC) variants. For patients with ARID1A mutations, PFS was > 6 months in five of five, and three of the five achieved a partial response.

Fig A1.

Analysis of archival tissue by immunohistochemistry (IHC) failed to demonstrate correlates of treatment benefit. (A) Box plot of median and range of 4E-BP1 IHC scores in patients who experienced disease progression or died within v 57.5; Wilcoxon rank sum P = .96). (B) Box plot of median and range of ERG-positive cells/mm2 in patients who experienced disease progression or died within < 6 months of starting study treatment (failure) versus those alive and free from disease progression (success; median score, 176 v 144; Wilcoxon rank sum P = .38).