Molecular subtypes of clear-cell renal cell carcinoma (ccRCC) were prognostic for outcomes after metastasectomy, with significant differences seen in overall survival (OS) and disease-free survival (DFS) between groupings, according to findings presented at the 2017 European Multidisciplinary Meeting on Urological Cancers (EMUC).1
“Subtype may aid in patient selection after clinical selection. The differences between relapse rate and survival were striking,” said Annelies Verbiest, a PhD candidate at VIB-KU Leuven Center for Cancer Biology in Leuven, Belgium. “Clear-cell renal cell carcinoma subtypes have intrinsic differences that are correlated with survival, response to systemic therapy, and response to metastasis-directed therapies. These are not just prognostic but also predictive.”
Overall, 4 molecular subtypes (ccRCC 1 to 4) have been identified in metastatic ccRCC based on clustering of whole-genome expression data,2 Verbiest said during a presentation at EMUC, which was held November 16-19 in Barcelona, Spain. For the analysis, investigators used nCounter, a 35-gene classifier, developed by NanoString Technologies. Prior research suggests that 2 of these subtypes, ccRCC 2 and 3, have a good prognosis while subtypes ccRCC 1 and 4 have a poor prognosis. Specifically, Beuselinck et al2 demonstrated that ccRCC molecular subtypes are predictive of response to sunitinib in patients with metastatic disease, and could be used for personalized metastatic renal cell carcinoma treatment with tyrosine kinase inhibitors, demethylating, or immunomodulatory drugs.
The researchers performed global transcriptome analyses on 53 primary resected ccRCC tumors from patients who developed metastatic disease and were treated with first-line sunitinib (Sutent). They also determined chromosome copy-number aberrations, methylation status, and mutations in VHL, which are associated with von Hippel-Lindau syndrome, and PBRM1. Molecular data were analyzed in relation to response rate, progression-free survival, and OS.
“Patients with subtypes 2 and 3 who are treated with sunitinib or pazopanib [Votrient] usually respond well to therapy,” Verbiest said in an interview with OncologyLive®. “Because comorbidity is associated with surgery, it might be best to avoid it, especially in these patients.”
Differences in Categories
The study presented at EMUC involved 32 patients: 22 with ccRCC subtypes 2 and 3 and 10 with ccRCC subtypes 1 and 4. Study objectives included correlating the molecular subtype with DFS, time to systemic therapy, and OS after complete metastasectomy. The researchers selected patients who had undergone a first metastasectomy with curative intent without prior systemic treatment. Other inclusion criteria included the availability of fresh frozen primary tumor tissue for genomic analysis, and having had at least 1 follow-up CT scan after metastasectomy.
In the overall study population, at a median follow up of 62 months (range, 20-143), the median DFS was 14.5 months (5-year DFS rate, 24%), the time to systemic therapy was 57 months, and the median OS was 133 months (5-year OS rate, 60%).
When the subtypes were categorized, there was a marked improvement in outcomes for those classified as having a good versus a poor prognosis (Table1 ). In subtypes 2 and 3, the median DFS was 21 month (5-year DFS rate, 33%) compared with a median of 9 months (5-year DFS rate, 0%) for those in the ccRCC 1 and 4 groups (HR, 2.58; P
= .009). For time to systemic therapy, the median was 98 months for the ccRCC 2 and 3 groups versus 28 months for ccRCC 1 and 4 (HR, 363; P = .003).
Table. RCC Outcomes by Molecular Subtype1
The median OS in the ccRCC 2 and 3 groups was 133 months (5-year OS rate, 85%) compared with a median OS of 50 months (5-year OS rate, 19%) for the ccRCC 1 and 4 groups. This represented a significant reduction in the risk of death for those in the good-prognosis arm (HR, 5.47; P
“The survival outcomes after metastasectomy are much better for the good-prognosis subtypes compared with the poor-prognosis subtypes,” Verbiest said. “The interesting part is that in other research, conducted by us and other research groups, we have seen that the poor-prognosis subtypes are correlated with more prognostic features.”