Plasma levels of hepatocyte growth factor (HGF), BAP1 and PBRM1 gene mutation status, and angiogenesis gene expression are promising predictive biomarkers for survival outcomes among patients undergoing different treatment regimens for renal cell carcinoma (RCC), according to research (abstracts 4522 and 4523) presented at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting, held June 2–6 in Chicago.
Low baseline plasma HGF levels among 310 patients with advanced clear-cell RCC correlated with overall survival (OS) in the patients treated with interferon-α (IFN-α) with or without bevacizumab—and plasma HGF levels remained prognostic at 4 weeks in multivariate analyses, reported the lead study author of the first study (abstract 4522), Daniel J. George, MD, of the Duke Cancer Institute in Durham, North Carolina.
Elevated HGF levels at 4 weeks (exceeding the median of 161.4 pg/mL) were associated with poorer OS (14 vs 27 months; hazard ratio [HR], 1.69; 95% CI, 1.31–2.19; P < .0001). However, compared with patients with persistently high HGF levels, those experiencing a decline in HGF levels by week 4 to the median or lower than the median of 161.4 pg/mL also experienced improved survival times (OS, 19 vs 13 months; HR, 1.46; 95% CI, 1.04–2.04; P = .029), George said.
The findings bolster previous work tying low baseline HGF levels to improved OS in patients with RCC.
“Low HGF levels that remain low suggest that some patients may benefit from more selective VEGF-only targeted agents,” George noted. “High HGF levels at baseline or over time suggest that some patients may need agents that also target the HGF-MET axis.”
However, HGF’s causal role in the association remains to be established, cautioned James Brugarolas, MD, PhD, director of the kidney cancer program at the University of Texas Southwestern Medical Center in Dallas.
Another report on RCC biomarker development (abstract 4523) involved an analysis of the correlation of PBRM1 and BAP1 mutations with angiogenesis expression signatures in patients on first-line tyrosine kinase inhibitor (TKI) therapy, by Martin H. Voss, MD, and colleagues at Memorial Sloan Kettering Cancer Center in New York. The team examined sunitinib or pazopanib treatment outcomes and 352 patients’ mutation status for BAP1 and PBRM1 and RNA-based angiogenesis pathway gene expression. BAP1 and PBRM1 mutations are “largely mutually exclusive,” Voss noted. Progression-free survival (PFS) was only marginally significantly associated with BAP1 mutation status (P = .058), but OS was significantly worse among patients whose tumors harbored BAP1 mutations (BAP1 wild-type OS vs mutation OS: HR, 1.57; 95% CI, 1.09–2.27; P = .0116). In contrast, PBRM1 mutations were associated with improved PFS and OS (OS, 35.5 vs 26.1 months; HR, 0.64; 95% CI, 0.48–0.87; P = .004).
Likewise, the team’s 43-gene angiogenic gene expression RNA score was prognostic in univariate and multivariate analyses; the angiogenic signature score also correlated with PBRM1 mutation status.
“Loss of PBRM1 enhances the pro-angiogenic microenvironment of RCC with favorable effects on response to TKI,” explained Voss. “BAP1 loss associates with decreased angiogenic signaling and adverse outcome to TKI.”
BAP1 and PBRM1 mutations should be considered together rather than independently, commented Brugarolas. Only 30% of study participants had PBRM1 mutations, he cautioned, questioning whether such a small sample of patients is representative of the RCC patient population. Patients with low angiogenesis signature scores and BAP1 mutations might benefit from combination therapy approaches.