RECORD-3 Molecular Analysis Characterizes Poor-Risk Patients With mRCC

To best treat patients with clear cell renal cell carcinoma, incorporating precision biomarkers into the current nomogram is essential.

James J. Hsieh, MD

To best treat patients with clear cell renal cell carcinoma (ccRCC), incorporating precision biomarkers into the current nomogram is essential, according to James J. Hsieh, MD.

Currently, “We see 12 drugs approved and more to come, and I think we have entered what I call the ‘golden age’ of kidney cancer therapeutics,” said Hsieh, who gave a presentation during the 2016 International Kidney Cancer Symposium, adding that treatments for ccRCC have emerged over recent years. “The question now is, ‘how do we now learn to sequence or use combination therapy to benefit our patients?’”

The RECORD-3 study was a large (n = 471) randomized trial in which treatment-naïve patients received either everolimus (Afinitor) followed by sunitinib (Sutent) or sunitinib followed by everolimus as first-line treatment.1 Results showed that first-line progression free survival (PFS) was 10.7 months for sunitinib and 7.9 months for everolimus, affirming the primary role of sunitinib in the frontline setting.

To evaluate the relationship between somatic mutations and treatment response, researchers performed next-generation sequencing (NGS) on 41 genes for 220 patients with ccRCC.2 Of the genes sequenced, 6 were mutated in >10% of patients: VHL, PBRM1, SETD2, BAP1, KDM5C, and PTEN. The mutation frequency of these genes in the RECORD-3 cohort was compared with the frequencies reported in The Cancer Genome Atlas (TCGA) and 2013 studies.3,4

In the RECORD-3 and previously published cohorts,VHL is mutated in 75% and 82% of patients, respectively. The VHL mutation frequency in the TCGA and earlier cohorts is lower (52% and 49%, respectively), but this difference is likely due to the technical difficulty associated with sequencing the GC-rich exon 1 of VHL. As VHL is the most frequently mutated gene in all 4 cohorts, it is likely the primary genetic event associated with pathogenesis of ccRCC.

Importantly, the RECORD-3 cohort consists of metastatic precancerous metastatic disease, while other studies of mutation frequency consist primarily of stage I to IV disease, a difference that may drive the variability in mutation frequency between these studies. The comparatively high frequency of SETD2, BAP1, KDM5C, and PTEN mutation in the RECORD-3 study suggests a role for these genes in the metastatic progression of ccRCC.

Evaluation of the relationship between gene mutation status and PFS revealed that 3 genes—PBMR1, BAP1, and KDM5C—have distinct correlations with treatment selection. Patients with PBMR1 mutations do equally well when treated with sunitinib or everolimus, whereas patients with BAP1 mutations do poorly on both therapies, especially everolimus. Patients with a KDM5C mutation fare better when treated with sunitinib.

To evaluate the relationship between mutation status and OS, researchers considered the mutation status of the patients treated with either sunitinib followed by everolimus or everolimus followed by sunitinib. In the sunitinib-everolimus arm, patients performed well irrespective of genotype, reflecting the finding for the overall RECORD-3 cohort.

However, in the everolimus-sunitinib arm, patients with PBRM1 mutations (39.6; 95% CI, 34.0—not estimable) fared better than those with the wild-type allele (16.2; 95% CI, 10.7-26.9) or BAP1 mutations (9.8; 95% CI, 8.2-22.4). Patients with KDM5C mutations did exceptionally well (40.3; 95% CI, 5.1-not estimable).

Further analysis divided the cohort into groups: group 1, BAP1-mutant, PBRM1 wild-type/mutant (19% of cohort); group 2, BAP1 wild-type, PBRM1 mutant (43%); and group 3, BAP1 wild-type, PBRM1 wild-type (38%). BAP1 and PBRM1 mutations are mutually exclusive.

Within the sunitinib-everolimus arm, there was not a significant difference in OS between groups. Within the everolimus-sunitinib arm, however, there is a significant difference in median OS between groups: OS was shortest in group 1 (9.8; 95% CI, 7.8—20.0), longest in group 2 (39.6; 95% CI, 31.7–not estimable), and intermediate in group 3 (18.1; 95% CI, 13.7–30.0). When comparing OS for different treatment sequences within each of the molecular groups, there was not a statistically significant difference. However, a trend suggested differences for everolimus-sunitinib versus sunitinib-everolimus in group 1 (HR, 1.5), group 2 (HR, 0.8), and group 3 (HR, 1.2). During the study, investigators uncovered that BAP1 and KDM5C mutations are also mutually exclusive. Patients with KDM5C mutations do better in both arms than patients with the wild-type allele.

The results of this study suggest that specific tumor genotype may represent distinct molecular subtypes of ccRCC with potentially predictive vales. Further, tumor genotype may inform treatment sequence to optimize survival outcome, laying the foundation for prospective studies to address the hypothesis generated by this work with independent data sets.

References

  1. Motzer RJ, Barrios CH, Kim TM, et al. Phase II randomized trial comparing sequential first-line everolimus and second-line sunitinib versus first-line sunitinib and second-line everolimus in patients with metastatic renal cell carcinoma. J Clin Oncol. 2013;33 (suppl; abstr 4504).
  2. Hsieh J, Chen D, Wang P, et al. Differential overall survival (OS) results in RECORD-3 study based on three distinct mRCC molecular subgroups classified by BAP1 and/or PBRM1 mutations. J Clin Oncol. 2016;34 (suppl; abstr 561).
  3. Sato Y, Yoshizato T, Shiraishi Y, Maekawa S, Okuno Y, et al. Integrated molecular analysis of clear-cell renal cell carcinoma. Nat Genet. 2013;45:860—867.
  4. Hakimi AA, Ostrovnaya I, Reva B, Schultz N, Chen YB, et al. Adverse outcomes in clear cell renal cell carcinoma with mutations of 3p21 epigenetic regulators BAP1 and SETD2: a report by MSKCC and the KIRC TCGA research network. Clin Cancer Res. 2013;19:3259—3267.

In an effort to address this question, Hsieh’s presentation focused on differential overall survival (OS) results of the RECORD-3 study based on 3 distinct mRCC molecular subgroups, classified by patients who harbored BAP1 and/or PBRM1 mutations.