Immunotherapy in Urothelial Carcinoma- A New Era
Dr. Eli Rosenbaum
8.3.2018
TNM Staging SystemMuscle-invasive Urothelial Cancer
Tx paradigm in non metastatic muscle invasive disease
surgery or chemoradiation
Two large randomized trials and a meta-analysis support the administration of
cisplatin based neoadjuvant chemotherapy (DD MVAC, GC) for T2-4 disease
The data regarding adjuvant chemotherapy is limited (trials were small,
terminated early, used old protocols). However, a metaanalyses suggests a
survival benefit for T3-4 / N+ disease
For cisplatin non fit patients (Galsky, JCO 2011; ECOG PS ≥ 2, GFR <60, neuropathy/hearing loss grade
≥2, CHG NYHA grade ≥ 3), may consider a carboplatin based regimen; Benefit not shown
Slide 5
Presented By Alison Birtle at 2018 Genitourinary Cancers Symposium: Translating Evidence to Multidisciplinary Care
Slide 17
Presented By Alison Birtle at 2018 Genitourinary Cancers Symposium: Translating Evidence to Multidisciplinary Care
Urothelial Carcinoma
Treatment of Metastatic Disease
Gemcitabine & Cisplatinversus MVAC
MVAC
Gem/Cis
Proportion
surviving
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
Pats at risk
202
203
6
161
167
12
124
120
18
54
52
24
18
18
30
4
1
36
0
0
months
MVAC
Gem/Cis
GC M-VAC
Neut. Sepsis 1% 12% < .001
Admissions F-N 9 49
G- or GMCSF 6% 21%
Mucositis Gr 3/4 1% 22% < .001
Toxic Death 1% 3%
CR + PR 50% 46%
CR 12% 12%
EORTC 30986: Gemcitabine/Carboplatin (GC) and Methotrexate/Carboplatin/Vinblastine (M-CAVI) in Cisplatin-
Unfit Patients
mOS ~9 months
De Santis et al. JCO 2012
Standards in Advanced UC
mOSORRRegimenSetting
12-15 months40-50%MVAC1
Gemcitabine + Cisplatin2
PGC3
CisplatinEligible1st
Line
7-9 months36-56%Gemcitabine + Carboplatin4-6
CisplatinIneligible
5-8 months~10%Single agent
chemotherapy7-92nd Line
1. Loehrer et al., JCO 19922. Von der Maase et al., JCO 2000 3. Bellmunt et al. JCO 20124. De Santis et al., JCO 2012 5. Linardou et al., Urology 2004 6. Nogué-Aliguer et al., Cancer 20037. Bellmunt et al., NEJM 20178. Bellmunt et al., JCO 20099. Petrylak et al., JCO 2015
Treatment after platinum failure
• Advanced UC is uniformly fatal after failure of platinum chemotherapy– Responses are transient– High toxicity– Frail patients
• No global consensus for treatment following platinum-based chemotherapy exists– Taxanes are typically used in
the US – Vinflunine (not approved in
the US) is often used in Europe
mOSVinflunine + BSC: 6.9 moBSC: 4.6 mo
2L Regimena ORR mPFS mOS
Paclitaxel (n = 31) 10% 2.2 mo 7.2 mo
Docetaxel (n = 30) 13% — 9.0 mo
Vinflunine (n = 51) 18% 3.0 mo 6.6 mo
Vinflunine (n = 253) 9% 3.0 mo 6.9 mo
Metastatic UBC: New Drug Approvals
1995-2012
Bellmunt J, et al. Ann Oncol. 2013;24:1466-1472.
UBC
RCC
Prostate
Slide credit: clinicaloptions.com
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Yr
14
12
10
8
6
4
2
0
Nu
mb
er
of
Dru
gs
Ap
pro
ved
by F
DA
2017
Pembrolizumab
Nivolumab
Avelumab
Immunotherapy in Bladder Cancer
Cystoscopy showing
bladder tumor
TURBT Pathology: High Grade TCC extensively invading the muscle layer
TURBT
Proposed Model of BCG Therapy in Bladder Cancer
T cells
NeutrophilsMonocytes
Bladder urothelium
BCG Tumor cells
1. Attachment and invasion of the urothelium
4. Cytolysis of bladder tumor cells by CD8+ T cells
3. Attraction and activation of T cells2. Attraction of innate immune cells and release of cytokines/chemokines
BCG Versus Mitomycin-C(SWOG 8795)
Time To Recurrence
Perc
ent
Recurr
ence
363024181260
100
90
80
70
60
50
40
30
20
10
0
BCG
MMC
190
187
44
64
Not
Reached
20
At Risk FailMedian
in Months
Lamm DL
Urol Oncol
1:119-126, 1995
High Somatic Mutation Frequencies
Observed in Bladder Cancer[1]
▪ High mutational complexity rates similar to tobacco/environmental carcinogen exposure[2-4]
▪ Similar rates observed between pts exposed to tobacco or other environmental carcinogens[3]
▪ Potential for many neoantigens to be seen as foreign by host immune system[4]
1. Bellmunt J, et al. Ann Oncol. 2013;24:1466-1472.
2. Cancer Genome Atlas Research Network. Nature. 2014;507:315-322.
3. Lawrence MS, et al. Nature. 2013;499:214-218.
4. Kandoth C, et al. Nature. 2013;502:333-339. Slide credit: clinicaloptions.com
New US FDA approved treatments for UC (2016-2017)
Agent Mechanism Schedule Post
Platinum
Frontline
Cis-
ineligibleAtezolizumab Anti-PD-L1 Q3Weeks accelerated* accelerated
Nivolumab Anti-PD-1 Q2Weeks accelerated -
Durvalumab Anti-PD-L1 Q2Weeks accelerated -
Avelumab Anti-PD-L1 Q2Weeks accelerated -
Pembrolizumab Anti-PD-1 Q3Weeks Level 1 accelerated
*
Phase 1 evaluation of PD-L1 inhibitors shows robust
activity in refractory disease
IC0IC1
IC2IC3
Atezolizumab: PD-L1
Avelumab: PD-L1
Durvalumab: PD-L1
**
*
*******
****
**
**
*
PD-L1 <1% PD-L1 indeterminate/not
evaluable/missing
–100
–75
–50
–25
0
25
50
75
100
Be
st
Re
du
cti
on
Fro
m B
as
eli
ne
in T
arg
et
Le
sio
n (
%)
PD-L1 ≥1%
Nivolumab: PD1
Pembrolizumab: PD1
PD1 inhibitors in bladder cancer
Avelumab: PD-L1 Durvalumab: PD-L1
Atezolizumab: PD-L1Nivolumab: PD1Pembrolizumab: PD1
Anti PD-1/Pd-L1 Abs Show Promising Antitumor Activity
ORR= 21.1%CR= 9.3%
ORR= 17.8%CR= 3.7%
ORR= 17.4%CR= 6.2%
ORR= 19.6%CR= 2.0%
ORR= 16.0%CR= 6.0%
Change in sum of longest diameters over time (IRF-assessed per RECIST v1.1)
Rosenberg J, et. al, Lancet 2016;387:1909–20
EfficacyStable disease
EfficacyAtezolizumab demonstrates clinical benefit beyond progression
Robert Dreicer et. al, Abstract 4515, ASCO 2016
In patients treated beyond PD:• 19% had SLD reductions ≥30% in
target lesions• 28% had disease stabilization (-
30% to +20% SLD change)• 12-mo OS was 50% in all
patients treated beyond progression
Atezolizumab in patients with metastatic urothelial carcinoma (mUC): A 2-year clinical update from a phase Ia study
Post Platinum Phase III Trials
Median Follow Up- 27.7 months
Key Eligibility Criteriaa
• mUC with progression during or
following platinum-based chemotherapy
– ≤ 2 prior lines of therapy
• Measurable disease per RECIST v1.1
• ECOG PS 0-1
• Evaluable sample for PD-L1 testing
• TCC histology as primary component
(N = 931)
▪ Primary endpoint
– OS, tested hierarchically
in pre-specified populations
41 Powles T, et al. EAS 2017, IMvigor211.
DOR, duration of response; ECOG, Eastern Cooperative Oncology Group; EORTC, European Organisation for Research
and Treatment of Cancer; PRO, patient-reported outcome; q3w, every three weeks; RECIST, Response Evaluation
Criteria In Solid Tumors; TCC, transitional cell carcinoma. a ClinicalTrials.gov, NCT02302807. b Defined by time from prior
chemotherapy < 3 mo, ECOG performance status > 0 and hemoglobin < 10 g/dL. c Confirmed response was not required
for secondary efficacy endpoints. This analysis reports exploratory confirmed responses.
IMvigor211 Study Design
41
Atezolizumab 1200 mg q3w
R
1:1
No crossover permitted
per protocol
Survival
follow-up
Loss of
clinical benefit
RECIST v1.1
progression
Stratification Factors
• No. of risk factorsb (0 vs. 1/2/3)
• Liver metastases (yes vs. no)
• PD-L1 status (0/1 vs. 2/3)
• Chemotherapy (vinflunine vs. taxanes)
▪ Additional endpoints
– Efficacy: RECIST v1.1 ORR, PFS and DORc
– Safety
– PROs: EORTC QLQ-C30
Chemotherapy
(investigator’s choice)
• Vinflunine q3w
• Docetaxel q3w
• Paclitaxel q3w
Key Eligibility Criteriaa
• mUC with progression during or
following platinum-based chemotherapy
– ≤ 2 prior lines of therapy
• Measurable disease per RECIST v1.1
• ECOG PS 0-1
• Evaluable sample for PD-L1 testing
• TCC histology as primary component
(N = 931)
▪ Primary endpoint
– OS, tested hierarchically
in pre-specified populations
42 Powles T, et al. EAS 2017, IMvigor211.
DOR, duration of response; ECOG, Eastern Cooperative Oncology Group; EORTC, European Organisation for Research
and Treatment of Cancer; PRO, patient-reported outcome; q3w, every three weeks; RECIST, Response Evaluation
Criteria In Solid Tumors; TCC, transitional cell carcinoma. a ClinicalTrials.gov, NCT02302807. b Defined by time from prior
chemotherapy < 3 mo, ECOG performance status > 0 and hemoglobin < 10 g/dL. c Confirmed response was not required
for secondary efficacy endpoints. This analysis reports exploratory confirmed responses.
IMvigor211 Study Design
42
Atezolizumab 1200 mg q3w
R
1:1
No crossover permitted
per protocol
Survival
follow-up
Loss of
clinical benefit
RECIST v1.1
progression
Stratification Factors
• No. of risk factorsb (0 vs. 1/2/3)
• Liver metastases (yes vs. no)
• PD-L1 status (0/1 vs. 2/3)
• Chemotherapy (vinflunine vs. taxanes)
▪ Additional endpoints
– Efficacy: RECIST v1.1 ORR, PFS and DORc
– Safety
– PROs: EORTC QLQ-C30
Chemotherapy
(investigator’s choice)
• Vinflunine q3w
• Docetaxel q3w
• Paclitaxel q3w
Key secondary endpoints:
ORR, then PFS
Primary endpoint:
OS
OS: IC2/3
OS: IC1/2/3
OS: ITT
PFS: IC2/3
PFS: IC1/2/3
PFS: ITT
ORR: IC2/3
ORR: IC1/2/3
ORR: ITT
2-sided = 0.05
43 Powles T, et al. EAS 2017, IMvigor211.HR, hazard ratio.
OS Analysis: IC2/3 Population
HR = 0.87 (95% CI: 0.63, 1.21)
P = 0.41
Events/
Patient
sMedian OS
(95% CI)
12-mo OS
Rate(95% CI)
Atezolizumab 72/116 11.1 mo (8.6, 15.5) 46% (37, 56)
Chemotherap
y
88/11810.6 mo (8.4, 12.2) 41% (32, 50)
No. at Risk
Atezolizumab 116 100 85 77 71 58 51 39 27 19 11 6 0
Chemotherapy 118 100 91 82 71 61 47 32 24 15 9 5 1
80
60
0
10 12 14 16 18 202 4 6 80 2422
20
40
Overa
ll S
urv
ival
100
Months
▪ Median follow-up duration in ITT population: 17.3 mo (range, 0 to 24.5 mo)
45 Powles T, et al. EAS 2017, IMvigor211.
OS Analysis: ITT Population
Events/
Patient
sMedian OS
(95% CI)
12-mo OS
Rate(95% CI)
Atezolizumab 324/467 8.6 mo (7.8, 9.6) 39% (35, 44)
Chemotherap
y
350/4648.0 mo (7.2, 8.6) 32% (28, 37)
80
60
0
10 12 14 16 18 202 4 6 80 2422
20
40
Overa
ll S
urv
ival
100
Months
80
60
0
10 12 14 16 18 202 4 6 80 2422
20
40
Overa
ll S
urv
ival
100
Months
HR = 0.85 (95% CI: 0.73, 0.99)
P = 0.038
No. at Risk
Atezolizumab 467 405 327 280 245 201 177 138 90 59 34 13 1
Chemotherapy 464 397 330 268 219 175 140 99 60 42 17 7 1
OS by PD-L1 IC Status
Unstratified HRs are displayed. Reprinted in part from The Lancet, Powles T, et al. 2017 Dec 18. [Epub], © 2017, with permission requested from Elsevier.
46Presented by: Powles T, IMvigor211.
• In our study design, we
hypothesized that efficacy
was associated with
PD-L1 status
• Unexpectedly, PD-L1
overexpression resulted in
favorable outcomes in
both arms
• Subsequent biomarker
analyses focused on the
ITT population
10.6 mo 11.1 mo 8.2 mo7.3 mo
MonthsMonths
HR = 0.81 (95% CI: 0.59, 1.10) HR = 0.84 (95% CI: 0.71, 1.00)
OS by TMB and PD-L1 Status
47
Unstratified HRs are displayed. Reprinted in part from The Lancet, Powles T, et al. 2017 Dec 18. [Epub], © 2017, with permission requested from Elsevier.
• Improved OS benefit was observed in patients with high TMB as well as high PD-L1 IC scores
Presented by: Powles T, IMvigor211.
HR = 0.81 (95% CI: 0.59, 1.10)HR = 0.68 (95% CI: 0.51, 0.90) HR = 0.50 (95% CI: 0.29, 0.86)
48 Powles T, et al. EAS 2017, IMvigor211.NE, not estimable. Unstratified hazard ratios are plotted. Dashed line refers to HR for ITT population.
OS in Clinical and Treatment Subgroups
1.0
Atezolizumab better Chemotherapy better
SubgroupPD-L1 status IC2/3
IC1/2/3
ITT
Tobacco use history Current
Previous
Never
ECOG PS 0
1
Primary tumor site Bladder
Urethra
Renal pelvis
Ureter
Liver metastases Yes
No
Lymph node–only metastases Yes
No
Chemotherapy stratification Taxane
Vinflunine
Atezolizumab mOS
Chemotherapy
mOS11.1 mo 10.6 mo
8.9 mo 8.2 mo
8.6 mo 8.0 mo
9.2 mo 6.7 mo
8.4 mo 8.2 mo
10.4 mo 8.1 mo
12.0 mo 10.1 mo
6.1 mo 6.4 mo
8.9 mo 7.7 mo
NE 12.5 mo
5.9 mo 8.5 mo
8.9 mo 8.1 mo
4.0 mo 5.2 mo
10.1 mo 9.7 mo
17.4 mo 12.2 mo
8.1 mo 7.2 mo
8.3 mo 7.5 mo
9.2 mo 8.3 mo
12 14
Months
Confirmed ORRa
IC2/3 IC1/2/3 ITT
Atezo (n = 113)
Chemo (n = 116)
Atezo (n = 312)
Chemo (n = 306)
Atezo (n = 462)
Chemo (n = 461)
Responders, n (%) 26 (23%) 25 (22%) 44 (14%) 45 (15%) 62 (13%) 62 (13%) 95% CI, % 16, 32 15, 30 10, 19 11, 19 11, 17 11, 17
CR, n (%) 8 (7%) 8 (7%) 11 (4%) 13 (4%) 16 (3%) 16 (3%)
▪ Objective response was
similar between arms
▪ Responses to
atezolizumab were durable
regardless of PD-L1 status
– 63% of patients in the
atezolizumab arm and
21% in the chemotherapy
arm had ongoing responses
at data cutoff
49 Powles T, et al. EAS 2017, IMvigor211.•mDOR, median DOR. a Confirmed RECIST v1.1 responses were assessed as an exploratory endpoint.
Response by PD-L1 Subgroup
DOR in ITT Population
No. at Risk
Atezolizumab 62 61 56 50 42 35 23 14 9 5 2 0
Chemotherapy 62 62 59 40 28 23 16 8 5 4 0 0
Events/
Patients mDOR (95% CI)Atezolizumab 23/62 21.7 mo (13.0, 21.7)
Chemotherap
y
49/627.4 mo (6.1, 10.3)
80
60
0
10 12 14 16 18 202 4 6 80 22
20
40
Ob
jective
Re
sp
on
se
100
Months
▪ The safety profile for atezolizumab was consistent with Phase I-II data1,2
50 Powles T, et al. EAS 2017, IMvigor211.1. Powles Nature 2014. 2. Rosenberg Lancet 2016.
Treatment-Related AEs
Treatment-Related AEs in ≥ 10% (All Grade) or ≥ 4% (Grades 3-4) for Either Arm
ChemotherapyAtezolizumab
Proportion of Patients (%)
30% 30%10%10% 0%40% 40%20%20%
■■ All Grade
■■ Grade 3-4
First Line Immunotherapy
Frontline Therapy for UC: Cis-Ineligible
54
Results: mOS (primary end-point) of 15.9m
NE, not estimable. Patients at risk of an event are displayed at indicated time points below plot.
Censored values are indicated with a plus (+) symbol.
Landmark Analysis of OS by PD-L1
Balar, et al. Lancet 2017
Presented by O’Donnell et al., ASCO 2017
Presented by de Wit et al., #LBA37, ESMO 2017
Presented by O’Donnell et al., ASCO 2017
KN-052: ORR by Age and ECOG PS
Combinations in Immunotherapy of UC
Study Design & Patient Population
PLATINUM/GEM +
PLACEBO
Key Eligibility N= 1200• Platinum-eligible• ECOG PS 0-2• TCC histology• Locally Adv/Metastatic
Stratification Factors :• PD-L1 IHC Status0 vs 1 vs. 2/3 • Investigator chemo
choice• Bajorin risk factor
R
1:1:1 ATEZOLIZUMAB
PLATINUM/GEM +
ATEZOLIZUMAB A
B
C
Primary endpoints: PFS and OS as Co-primary for Arm A, OS testing for Arm B only.
Design Details:
• N = 1200 (no minimum required for any IHC IC subgroup)
• Secondary endpoints: (mod)ORR, Safety, PRO as part of totality of evidence (HRQoL, Physical
function, symptoms measured with EORTC QLQ-C30), Euro QoL 5 Dimensions 5-Level (EQ-5D-5L)
• Cross-over not allowed
KEYNOTE 361- 1st Line Therapy in mUC
Presented by Powles et al., ASCO 2017
Immune checkpoint blockade comes to bladder
cancer
Drake CG. Nat Rev Immunol. 2010;10(8):580-593.
PD-L1
PD1
CTLA4
Modified from Drake, Nat Rev Immunol 2010
P Sharma: SITC 2016
Ipilimumab + Nivolumab:CheckMate 032
P Sharma: SITC 2016
Ipilimumab + Nivolumab:Antitumor Activity
P Sharma: SITC 2016
Ipilimumab + Nivolumab: Tumor Change
from Baseline in Target Lesion
P Sharma: SITC 2016
Ipilimumab + Nivolumab: AdverseEvents
DANUBE Trial Design
69
aStage IV T4b, any N; or any T, N2-N3; or M1. bCisplatin +
gemcitabine or carboplatin + gemcitabine, depending on cisplatin
eligibility. IV = intravenous; PD-L1 = programmed cell death-ligand 1;
q4w = every 4 weeks.
Phase III, randomized, open-label,
multicenter, global study1,2
1. Powles T, et al. Poster presented at ASCO Annual Meeting;
June 3-7 2016; Chicago, IL; Poster TPS4574.
2. Study NCT02516241. ClinicalTrials.gov website. Accessed
October 30, 2017.
Arm 1:
Durvalumab IV 1500 mg q4w +
Tremelimumab IV 75 mg q4w
Arm 2:
Durvalumab IV 1500 mg q4w
Arm 3:
Standard of Careb
Randomizat
ion
1:1:1
Patients with
Unresectable Stage
IVa Urothelial
Carcinoma as 1st
Line Therapy
Estimated
enrollment =
1005 patients
IDO1 Enzyme and Epacadostat
Presented By David Smith at 2017 ASCO Annual Meeting
Best Objective Response by RECIST v1.1<br />Epacadostat Plus Pembrolizumab<br />Phase 1/2 Advanced Urothelial Carcinoma
Presented By David Smith at 2017 ASCO Annual Meeting
Percentage Change From Baseline in Target Lesions<br />Epacadostat Plus Pembrolizumab<br />Phase 1/2 Advanced Urothelial Carcinoma by Number of Prior Lines of Treatment
Presented By David Smith at 2017 ASCO Annual Meeting
PD-1 Blockade Based Combinations in mRCC:<br />Are they Additive or Synergistic?
Presented By David McDermott at 2018 Genitourinary Cancers Symposium: Translating Evidence to Multidisciplinary Care
Thank You!