NEW YORK – A multi-omic study of advanced renal cell carcinoma (RCC) has found that molecular subsets of the disease are associated with differential clinical outcomes to angiogenesis blockade alone or with a checkpoint inhibitor.
In a study published on Thursday in Cancer Cell, researchers in the US and Europe described their evaluations and analyses of 823 tumors from advanced RCC patients. An unsupervised transcriptomic analysis identified seven molecular subsets of the disease, with distinct angiogenesis, immune, cell-cycle, metabolism, and stromal programs.
While sunitinib (Pfizer's Sutent) and atezolizumab (Genentech's Tecentriq) combined with bevacizumab (Genentech's Avastin) were effective for treating subsets with high angiogenesis, the researchers found that atezolizumab combined with bevacizumab improved clinical benefit in tumors with high T-effector and/or cell-cycle transcription. Meanwhile, somatic mutations in PBRM1 and KDM5C were associated with high angiogenesis and AMPK/fatty acid oxidation gene expression, while CDKN2A/B and TP53 alterations were associated with increased cell-cycle and anabolic metabolism.
The researchers also found that sarcomatoid tumors exhibited lower prevalence of PBRM1 mutations and angiogenesis markers, frequent CDKN2A/B alterations, and increased PD-L1 expression. They noted that the overall findings could be used to molecularly stratify patients, explain improved outcomes of sarcomatoid tumors to checkpoint blockade versus antiangiogenics alone, and to develop personalized therapies for RCC and other indications.
The researchers performed integrated RNA sequencing and targeted somatic variant analysis on tumor samples from the Phase III IMmotion151 clinical trial, before treatment, and evaluated biomarkers associated with objective response and progression-free survival.
The investigators had previously reported associations between angiogenesis and T-effector gene expression signatures and clinical outcome to treatment with atezolizumab plus bevacizumab or sunitinib in the Phase II IMmotion150 trial. They evaluated the association of these signatures with clinical outcomes in IMmotion151 by pre-determining transcriptional cutoffs for both signatures in IMmotion150 and retrospectively applying them to define high- and low-expression patient subsets.
They reported that high expression of the angiogenesis signature was associated with improved progression-free survival in the sunitinib treatment arm, and that no difference in survival was observed in the angiogenesis-high or T-effector-low tumors when compared across treatment arms. Further, atezolizumab plus bevacizumab improved progression-free survival versus sunitinib in T-effector-high and in angiogenesis-low tumors, underscoring the relevance of immune and angiogenesis biology as biomarkers of differential clinical outcomes to checkpoint and angiogenesis blockade in independent advanced RCC cohorts.
The investigators next used this data to further identify and refine transcriptionally defined subgroups of RCC, identifying seven clusters of patients based on the top 10 percent most variable genes in the IMmotion151 cohort. To understand the main biological features driving these clusters, they compared them individually to each other, and found that clusters 1 and 2 were primarily characterized as highly angiogenic, with enrichment of vascular and VEGF pathway-related genes as well as inferred endothelial cell presence.
Tumors in cluster 3 were characterized by relatively lower expression of both angiogenesis and immune genes and moderate expression of cell-cycle genes. Those in clusters 4, 5, and 6 were characterized by enrichment of cell-cycle transcriptional programs and lower expression of angiogenesis-related genes. The researchers termed cluster 4 as T-effector/Proliferative, cluster 5 as Proliferative, and cluster 6 as Stromal/Proliferative. And finally, cluster 7 was characterized by enrichment of expression of small nucleolar RNAs (snoRNAs), especially C/D box snoRNAs (SNORDs), which have been implicated in alterations of epigenetic and translation programs and have been linked to carcinogenesis.
The researchers subsequently evaluated clinical outcomes to atezolizumab plus bevacizumab and sunitinib treatment in each cluster. Patients in clusters 1 and 2 demonstrated longer progression-free survival in both treatment arms, suggesting better outcome regardless of treatment, while those in cluster 5 had relatively shorter survival, suggesting poor prognostic association of proliferative/stromal biology with clinical outcomes.
When evaluated across treatment arms, there was no apparent difference in clinical outcomes between atezolizumab plus bevacizumab and sunitinib arms in clusters 1, 2, and 3, the researchers added. Atezolizumab plus bevacizumab demonstrated improved overall response versus sunitinib in cluster 4, confirming the contribution of pre-existing intratumoral adaptive immune presence in determining benefit to regimens containing immunotherapy. Further, atezolizumab plus bevacizumab showed an improved overall response rate and progression-free survival in cluster 5, implicating the relevance of PD-L1 blockade in low angiogenesis but high proliferative tumors, the researchers said.
Atezolizumab plus bevacizumab also showed improved progression-free survival in cluster 7, but the biological basis of that effect was unclear.
A subsequent evaluation of clinical outcomes in somatic alteration subgroups showed that PBRM1 mutations conferred overall better prognosis, regardless of treatment arm, the investigators added. Sunitinib-treated patients whose tumors harbored PBRM1 mutations showed longer survival compared with those with wild-type PBRM1. This trend was also observed in atezolizumab plus bevacizumab-treated patients, but did not reach statistical significance.
"Overall, findings from this randomized phase III study expand our understanding of RCC biology and provide a molecular basis for differential clinical outcomes and resistance mechanisms associated with angiogenesis blockade, checkpoint inhibition, and their combinations in patients with untreated advanced RCC," the authors concluded. "Given that these combinations are under clinical evaluation and have shown promising activity in additional indications, such as hepatocellular carcinoma, non-small cell lung cancer, and endometrial cancer, the findings from this study may be applicable in interpreting clinical outcomes and developing personalized therapies across many cancers."