Kidney Cancer Journal 115
therapy, with a hazard ratio of 0.61 (95% CI). In the
overall population, a slightly smaller benefit was observed
in PFS, with axitinib and avelumab achieving 13.8
months versus 8.4 months with sunitinib with a hazard
ratio of 0.69 (95% CI). There was little discrepancy
between this and investigator assessment of PFS. Response
rates were higher with the combination of axitinib with
avelumab, ranging between 50% and 55% depending on
PD-L1 status. This nearly doubled the response rate
observed with sunitinib therapy.
One of the biggest questions and challenges that
emerge is related to OS. At the time of this interim
analysis, without sufficient events for final assessment,
there was no significant difference in OS. This lack of
maturity in OS data is one of the conundrums associated
with the study. To be considered as a more compelling
study, JAVELIN 101 needs to (1) demonstrate positive OS
data, and (2) demonstrate that there is a higher rate of
durable, complete responses. Still to be addressed is the
provocative question of whether adding an antiangiogenic
agent to an IO agent is additive or synergistic.
IMotion 151 Confirms Different Gene Expression
Profiles for Atezolizumab + Bevacizumab vs Sunitinib
The IMmotion151 (NCT02420821) phase 3 study previously
demonstrated improved PFS with combined
atezolizumab/ bevacizumab vs sunitinib in treatmentnaive
patients with metastatic renal cell carcinoma
(mRCC). The presentation at ESMO correlated molecular
gene expression signatures with clinical outcomes, prognostic
risk groups, and tumor histology, extending data
from the IMotion150 study.
The investigators, led by Brian Rini, MD, noted that
biomarker analyses performed in a phase 2 study
(IMmotion150) suggested that T effector (Teff) and interferon
gamma, as well as angiogenesis gene expression
signatures (GEs) were associated with differential outcomes
to atezolizumab plus bevacizumab and suni-tinib.
They subsequently conducted prespecified genomic
analyses to validate these GEs with clinical outcomes
from 823 patients in IMmotion151 and also evaluated
association of GEs with Memorial Sloan Kettering Cancer
Centre (MSKCC) risk groups, and sarcomatoid histology.
The IMmotion151 study met its co-primary endpoint
of improved PFS with atezolizumab plus bevacizumab
over sunitinib in PD-L1-positive patients across all
MSKCC risk groups, hazard ratio (HR) 0.74; 95%
confidence interval (CI), 0.57 - 0.96 (P = 0.02). Improved
PFS was also observed in patients with sarcomatoid
histology, HR 0.56; 95% CI, 0.38 - 0.83. Tumor molecular
analyses showed that high Teff GE was associated with PDL1
expression, as evaluated by immunohistochemistry.
High Teff GE also associated with longer PFS for atezolizumab
plus bevacizumab compared to sunitinib, HR
0.76; 95% CI, 0.59 - 0.99.
In patients receiving sunitinib, high angiogenesis GE
was associated with improved PFS, HR 0.59, 95% CI, 0.47
- 0.75; however high angiogenesis GE did not differentiate
between atezolizumab plus bevacizumab versus suniti-
nib clinical activity, HR, 0.95; 95% CI, 0.75 - 1.19.
Atezolizumab plus bevacizumab improved PFS vs sunitinib
in the subset of patients with low angiogenesis GE,
HR 0.68; 95% CI, 0.52 - 0.89. Angiogenesis GE was found
to be higher in favorable versus intermediate-to-poor
MSKCC risk groups (P = 4.28e-06).
Prospective evaluation of biomarkers in IMmotion 151
added further evidence molecular signatures that could
identify subpopulations of patients more likely to benefit
from ther combination immunotherapy plus antiangiogenic
therapy versus antiangiogenic monotherapy.
These data also demonstrated that sarcomatoid histology
is possesses an immune signature that signals a unique
sensitivity to immunotherapy. Aside from providing
further confirmation of the IMotion150 data, this study
is one of the first steps toward resolving an unmet need:
development of predictive biomarkers based on an
understanding of tumor biology. Ultimately it can inform
clinicians as to which treatment should be used for which
patient and at what time.
Future Directions
Ironclad evidence is still lacking that the IO-TKI combination
is the new standard of care. Information from
Merck indicates that the combination of pem-brolizumab
and axitinib met its primary endpoint of OS as well as
PFS. Perhaps that could serve as an example of an IO-TKI
combination that will soon gain ascendancy. The next 12
months will help determine how various data sets evolve
and hopefully will provide a clear signal on how to
change the treatment algorithm. This remains one of the
overarching challenges as the next round of international
meetings present data with which to revisit this issue. KCJ