EURURO Vol. 72 No. 6

973 1030

poor-risk patients benefit the least . No significant
differences appear across subgroups divided by age. These
data call into question the prevailing dogma highlighted
earlier. Rather than reserving cabozantinib for ‘‘aggressive’’
disease (synonymous with poor risk), it might be reasonable
to use the drug in populations with good and intermediate
risk as well. Similarly, subset analyses in patients divided by
age suggest a similar benefit in older adults.
So where do these subset analyses ultimately position
nivolumab? The data presented here suggest that it may be
best used in patients with poor-risk disease. Paradoxically,
however, nivolumab appears to be associated with a much
higher rate of primary progressive disease (PD) . PD as a
best response in Checkmate 025 was observed in 35% of
patients receiving nivolumab. With the caveat of cross-trial
comparisons, PD as a best response was only observed in
14% of patients receiving cabozantinib in METEOR . Thus,
in a poor-risk patient, one might make the argument that
cabozantinib offers the best opportunity for clinical benefit.
One would be remiss to omit toxicity froma discussion of
nivolumab versus cabozantinib. Nivolumab has been
purported to offer a toxicity profile far superior to targeted
therapies, but again no comparative data exist to support
this statement. In the first publication of data from
Checkmate 025, grade 3/4 adverse events were noted in
19% of patients treated with nivolumab, compared to 37% of
patients treated with everolimus At first glance, this
appears to compare favorably to the 68% rate of grade 3/4
adverse events in patients receiving cabozantinib in
METEOR However, the astute investigator will note a
subtle but important difference in adverse event reporting:
all-cause adverse events were reported for cabozantinib in
the METEOR trial, while only treatment-related adverse
events were reported for nivolumab in Checkmate 025. The
package insert for nivolumab suggests an all-cause grade 3/
4 adverse event rate of 56%, which is more balanced with
the toxicities observed for other targeted agents
An ideal scenario would be a biomarker that could predict
differential benefit with cabozantinib, nivolumab, and
lenvatinib/everolimus. There is ongoing work to confirm
observations of the mutational load associated with lung
cancer and bladder cancer with PD-1 inhibition, with some
supportive evidence in mRCC . Furthermore, retro-
spective studies correlating TSC1, TSC2, and MTOR alterations
to mTOR-inhibitor response and MET alteration with MET-
inhibitor response suggest that these biomarkers could be
associated with lenvatinib/everolimus and cabozantinib
activity, respectively . Tremendous investment
would be required to validate these findings prospectively.
Apart from biomarkers, specific clinical scenarios may also
add to clinical decision-making. One example is bone
metastasis, for which there is compelling evidence from
subgroup analysis of the METEOR trial that cabozantinib has
increased activity in this population . Until then, the
standard for second-line treatment should remain cabozan-
tinib. Both nivolumab and cabozantinib offer benefits in
response rate and OS, and distinct toxicity considerations
exist for each. The benefit in PFS and the lower rate of PD as
best response should sway clinicians towards cabozantinib.
Conflicts of interest: Sumanta K. Pal is a consultant for Exelixis, BMS,
GSK, Novartis, Pfizer, Astellas, and Genentech, and receives honoraria
from Genentech. Neeraj Agarwal has been a consultant for Pfizer,
Novartis, Exelixis, Merck, Argos, EMD Serono, and Eisai, and has received
research funding from Pfizer, EMD Serono, Novartis, Takeda, and Bayer.
Manuel C. Maia and Nazli Dizman have nothing to disclose.
References
[1]
.
[2]
[3]
.
[4]
.
[5]
.
[6]
.
[7]
.
[8]
.
[9]
.
[10] Opdivo package insert.
[11]
.
[12]
.
[13]
[14]
.
[15]
.
[16]
.
[17]
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 9 7 2 – 9 7 3 973