Page 78 - Read Online

Serzan et al. J Cancer Metastasis Treat 2021;7:39 https://dx.doi.org/10.20517/2394-4722.2021.76 Page 13 of 19

two studies may be due to less robust antitumor activity of the anti-PD-1 backbone compared to
nivolumab/ipilimumab and the increasing ability for patients with less aggressive tumors who progress on
sunitinib to receive salvage anti-PD1 based immunotherapy. In contrast to the TFS period observed with
nivolumab/ipilimumab, there is currently no evidence that anti-PD-1/VEGFR TKI combination regimens
can produce continued response after cessation of the VEGFR TKI component. On the other hand, despite
the fact that anti-PD-1/VEGFR TKI regimens have high rates of grade 3-4 adverse effects and frequent dose
reductions, the Checkmate 9ER trial demonstrated significant improvement in health-related quality of life
[66]
and burden of symptoms compared with sunitinib . As data from several pivotal trials of anti-PD-
1/VEGFR TKI therapy continue to mature, their ability to produce improvements in late endpoints such as
landmark PFS/OS and treatment free survival relative to nivolumab/ipilimumab will be essential to
understanding their role as a front line therapy for patients with advanced ccRCC.

Predictive biomarkers
Although the MSKCC and IMDC models reliably predicted overall survival in patients receiving TKI
therapy, there are currently no validated biomarkers of disease prognosis or prediction of response to ICI
therapy or combination TKI/ICI therapy. Tumor PD-L1 expression was initially thought to be a promising
biomarker given the observations that it is commonly overexpressed in 23%-56% of ccRCC tumors
(depending on the assay) and is associated with poor outcomes [67,68] . However, the ability for PD-L1
expression to reliably predict response to ICI has been inconsistent across studies. In Checkmate 214,
patients with PD-L1 positive tumors treated with nivolumab/ipilimumab had improved PFS and OS
compared to those with PD-L1 negative tumors. In the Keynote 426 study, patients responded to
axitinib/pembrolizumab regardless of tumor PD-L1 status. Lastly, in the IMmotion 151 study patients with
PD-L1 negative disease appeared to have the most benefit from atezolizumab/bevacizumab relative to
sunitinib. Taken together, PD-L1 positivity appears to predict better ORR on ICI therapy in patients,
especially those aggressive disease; however, it remains limited in its ability to predict long-term PFS or OS
benefits. PD-L1 expression appears to also predict lack of benefit from VEGFR TKI therapy alone,
confounding its use as a biomarker in anti-PD1/VEGFR TKI combinations.

Acknowledging that most patients with ccRCC have PD-L1 negative tumors, it is important to recognize
that the majority of responses to ICI will occur in the PD-L1 negative population. Although its role as a
prognostic and predictive biomarker remains unclear, PD-L1 expression remains an important risk
stratification tool for clinical trials. Clearly additional biomarkers predictive of benefit for a particular
regimen are needed to help with treatment choices in the current crowded first line therapeutic space.

Perhaps the most promising biomarkers have emerged from extensive genetic and gene expression profiling
[69]
on tumors from the IMmotion 151 trial. Motzer et al. conducted integrated multi-omics evaluation of 823
tumors from patients with ccRCC and identified 7 distinct tumor molecular subsets (1) angiogenic/stromal;
(2) angiogenic; (3) complement/oxidation; (4) T-effector/proliferative; (5) proliferative; (6)
stromal/proliferative; and (7) SnoRNA . The angiogenic groups 1 and 2 appeared to show benefit to both
[69]
sunitinib and atezolizumab/bevacizumab. The proliferative groups 4-5 as well as group 7 appeared to show
greater benefit to atezolizumab/bevacizumab compared to sunitinib alone. Groups 3 and 6 did not appear to
benefit from either treatment approach. Lastly, somatic mutations in PBRM1 and KDM5C associate with
high angiogenesis and AMPK fatty acid oxidation suggesting benefits from angiogenesis blockade .
[70]
Whereas CDKN2A, BAP1, and TP53 appear to associate with increase cell cycle and anabolic metabolism
[71]
suggesting benefit from ICI therapies . Taken together these discoveries, if validated using contemporary
FDA approved regimens, could be used to categorize patients based on genetic and gene expression profiles
and thus determine the optimal treatment approach (e.g., combination ICI, combination VEGFR TKI/ICI,
or novel agents) while also mitigating drug and financial toxicities . Importantly, such validation studies
[72]