EURURO Vol. 72 No. 3

374 476

The concept of treatment beyond progression is not new
and there is a precedent for success with this approach with
various therapies for multiple cancers . In one
example, time to second progression was estimated at
7.3 mo with sunitinib treatment for untreated metastatic
RCC
The current analysis of a larger patient population also
characterized disease features and outcomes of patients
TBP. Though the key differences in disease characteristics at
first progression in patients TBP versus NTBP with
nivolumab were better KPS, lower incidence of new bone
lesions, higher ORR, shorter time to response, and improved
QoL, the decision to continue treatment beyond progression
was made based on tolerability and investigator-assessed
clinical benefit. These favorable disease characteristics may
represent the benefit that influenced the investigator’s
decision to treat the patient beyond progression and,
independent of continued treatment, may impact survival.
In an exploratory descriptive analysis that compared
demographic and baseline disease characteristics of
patients who responded to treatment beyond first progres-
sion versus those who did not, characteristics such as
favorable Memorial Sloan Kettering Cancer Center risk were
identified in patients TBP who benefited, not surprising
given that this is a feature of better prognosis. The patients
TBP who benefited in terms of tumor burden reduction also
had better QoL scores compared with the patients who did
not have significant tumor burden reduction from the new
baseline. However, the sample size was small and the
results should be interpreted with caution. Whether tumor
burden reduction or length of treatment from the first
progression/OS, is a better variable to describe the outcome
of patients TBP still remains to be elucidated.
These results support OS as a primary endpoint for future
trials, through demonstration that even with progression,
some patients may still benefit from nivolumab. Longer
follow-up in patients TBP, further research in intrinsic
immune system mechanisms of action, and prospective
studies designed to evaluate how long patients should be
TBP, including a qualitative assessment of physicians’
rationale for treating patients beyond progression, may
help to elucidate these findings.
Patients TBP and NTBP had a similar frequency of
treatment-related AEs from initial treatment to first
progression, suggesting that toxicity—or lower degree of
toxicity—did not necessarily guide the decision to stop or
continue treatment. Incidence of any-grade treatment-
related AEs was lower after progression in patients TBP with
nivolumab, suggesting that most AEs occurred earlier in the
course of treatment.
Although CheckMate 025 randomized patients to
nivolumab and everolimus, the rationale for continuing
treatment beyond progression was based on the immune-
modulating mechanism of action of nivolumab potentially
leading to tumor flare, so an assessment in patients treated
with everolimus beyond progression was not included in
the main report. Additionally, comparisons of nivolumab
with everolimus would not be meaningful and should not
be made because of the nonrandomized, and post-hoc
nature of this analysis.
Limitations include the lack of direct assessment of
physician rationale for treating patients beyond progres-
sion. Additionally, patients TBP generally had favorable
disease characteristics, which may have overestimated the
TBP benefit. Subsequent therapy was received by the
majority of patients NTBP, minimizing potential bias
toward inferior results in patients NTBP should they not
have been subsequently treated and, consequently, pro-
gressed. Future prospective trials should be designed to
reduce bias and may include a second randomization of
patients at progression.
Definitions of treatment beyond progression vary
depending on the treatment and radiographic assessment
schedule of a given drug . In our analysis,
treatment beyond progression was defined as treatment
for 4 wk after first progression and in the published
analysis from the same study was defined as receipt of last
dose after progression . The revised definition is more
conservative and aimed to capture patients who were
intentionally rather than unintentionally TBP. The
18 patients who were treated <4 wk beyond progression
were included with those NTBP. Results were similar when
analyses were conducted that excluded these 18 patients.
Another definition, used by the US Food and Drug
Administration (FDA) analysis from this study, defined
treatment beyond progression as patients who received
more than one dose of treatment after progressive disease
and who had radiographic progressive disease . The
findings from the FDA differed from the current analysis
the FDA analysis excluded patients who responded to
treatment and then progressed and excluded patients with
nontarget progression even if there was a reduction in
target lesions (personal communication with author B.
Escudier). A revision of the guidelines to determine
progression is warranted for immunotherapies. Immune-
related RECIST criteria were developed to address differ-
ences in mechanism of action of newer immunotherapies.
However, these criteria also have limitations, namely the
inability to fully address all patterns of clinical activity
Functional imaging may add information to RECIST
criteria to guide treatment decisions
5. Conclusions
In conclusion, tumor shrinkage after RECIST[9_TD$DIFF]-defined
progression and improved survival was observed in some
patients TBP, and an acceptable safety profile was main-
tained. This observation supports the use of nivolumab
beyond progression in patients with symptomatic improve-
ment despite progressive disease. However, further inves-
tigations are warranted to better determine which patients
may benefit from treatment beyond progression to avoid
excessive duration of therapy. Presented as a poster
(number 132) at the 52nd Annual American Society of
Clinical Oncology Meeting, Chicago, IL, USA, June 3–7, 2016.
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 3 6 8 – 3 7 6 374