Phase 1B Study to assess safety and efficacy of Neo-Adjuvant Bladder Urothelial Carcinoma COmbination-immunotherapy (NABUCCO)

Although muscle-invasive urothelial cancer can be cured by surgery, recurrence
rates are high and 5-year survival is only 45-55% for pT3N0 tumors, and even
worse for pT4aN0 (35-45%) or pTxN+ (10-35%) patients. Neo-adjuvant
cisplatin-based chemotherapy has a high rate of response. Despite impressive
responses, including a pathological complete response (pCR) rate of 22-40% [1],
the absolute benefit in terms of overall survival is only 5%[2], at the cost of
substantial toxicity. Given the high rate of distant recurrences, systemic
treatment needs to be improved to change the abysmal prognosis of high-risk
bladder cancer. Due to the time period between diagnosis and surgery for
resectable UCC ( often 6-8 weeks), there is an opportunity to test drugs while
patients are waiting for surgery. Promising activity of anti-PD-L1 and
anti-PD-1 antibodies has been observed in patients with urothelial cancer [3].
As responses to immunotherapy appear to be durable, the introduction of
immunotherapy in the peri-operative setting may improve prognosis, as has been
shown for melanoma [4]. Judging from the metastatic setting however, response
rates for monotherapy may not be high enough to translate into sufficient
benefit peri-operatively. Ideally, response rates should approach response
rates from neo-adjuvant chemotherapy. Reasons for failure to respond to
anti-PD-(L)1 therapy could include insufficient priming of the immune system by
cancer antigens and/or negative regulation of other steps in the *cancer
immunity cycle* [5]. The addition of anti-CTLA4 treatment could prevent
negative regulation of T-cell priming and thereby broaden and intensify the
immune response to cancer antigens. Although combination therapy with
anti-CTLA4 and anti-PD1 has not been formerly compared to anti-PD1 alone,
response rates appear to be higher. In addition, higher response rates are seen
when treating in first-line rather than second line in the metastatic setting
and when treating patients with only lymph node metastases vs patients with
visceral disease. We therefore hypothesize that treating patients with
combination immunotherapy in the neo-adjuvant setting, where visceral
metastases are absent, will bring response rates within the range of those seen
with platinum-based chemotherapy.
There is currently no clinical data to inform whether immunotherapy will work
better pre- or post-operatively. In contrast to chemotherapy, immunotherapeutic
approaches depend on sufficient activation of the immune system. Expression of
CTLA-4 and PD-1 on T-cells are regulated by TCR triggering. Thus, blockade of
these two pathways will be of greatest value when a sufficient TCR trigger is
present. Importantly, the amount of antigen that can provide this TCR trigger
will correlate with tumor load, and because of this we postulate that
immunotherapy given pre-operatively will work more efficiently. As cystectomy
is a procedure with a high risk of morbidity and even mortality, toxicity
should be manageable to make sure it will not preclude resection in a timely
manner. A study with neo-adjuvant ipilimumab showed that this drug can safely
be administered prior to cystectomy and does not endanger resection [6, 7].
In cohort 1 of this study, we used an attenuated schedule of ipilimumab and
nivolumab. This cohort has now fully enrolled; all patients had resection; 23
(96%) had resection <12 weeks from 1st cycle. Grade 3/4 irAEs occurred in 54%
of patients; 42% when excluding clinically insignificant lab deviations. A
total of 11/24 patients (46%) achieved a pCR. 3 additional pts (13%) had
noninvasive cancer at resection (2 ypTis, 1 ypTa), resulting in an overall path
downstaging (<=ypT1N0) rate of 59% (13/22). In the current extension (cohort 2),
we aim to further improve the balance between efficacy and tolerability. As we
will collect pre- and post-treatment tissue, this study will provide a unique
opportunity to study effects of treatment on the anti-tumor immune response. We
will study the tumor micro-environment with specific emphasis on
immune-inhibiting processes that can potentially be targeted by therapy.
Furthermore, optional baseline and on-treatment MRI assessment will be
performed to extract tumor radiomic features to develop predictive models for
immunotherapy response.

Primary objectives:
1. Establish whether a sequenced pre-operative schedule of ipilimumab and
nivolumab is safe in urothelial cancer patients (cohort 1)
2. Study objective for cohort 2: to evaluate which treatment regimen is most
effective. Effectiveness expressed in pathological complete response

Secondary objectives:
2. Dissect the tumor micro-environment of urothelial cancers before and after
combination immunotherapy using multiplexed immunohistochemistry methods and
transcriptional profiles
3. Test efficacy of pre-operative ipilimumab - nivolumab in cohort 1 (sequenced
ipilimumab/nivolumab), cohort 2a (ipi 3 mg/kg and nivo 1 mg/kg) and cohort 2b
(ipi 1 mg/kg and nivo 3 mg/kg). Efficacy is defined as pCR rate at resection.
4. Provide an estimate of >=grade 3 immune-related toxicity in the ipi3/nivo1
and ipi1/nivo3 cohorts
5. Explore how the (dys)functional state of the tumor-specific T cells is
altered by sequenced combination therapy.

This is an open-label phase Ib trial to evaluate the effects of short-term
preoperative ipilimumab and nivolumab in patients with high-risk resectable
urothelial cancer (upper urinary tract allowed), defined as:
Stage III UCC according to American Joint Committee on Cancer (AJCC) guidelines:
- cT3-4aN0M0 OR
- T1 -aN1-3M0
The primary endpoint of this trial is safety. We will determine the number of
patients that have surgical resection <12 weeks from first infusion in cohort
1. In cohort 1, patients will be receiving:
- Day 1: Ipilimumab 3 mg/kg (wk1)
- Days 22: Ipilimumab 3 mg/kg + Nivolumab 1 mg/kg (wk4)
- Day 43: Nivolumab 3 mg/kg (wk 7)
- Day 56-84: Radical cystectomy or nefro/ureterectomy with appropriate
lymph node dissection (wk9-11)
After 8 patients have been followed until resection, a report of observed
toxicity in cohort 1 will be submitted to the DSMB. Enrollment will continue
until 12 patients have been enrolled. At this point, an update of toxicity will
be sent to the DSMB. If toxicity is judged by the DSMB as manageable, enrolment
will continue to a total of 24.
CT scans will be required at baseline and week 7- 9 to evaluate response to
immunotherapy. An optional MRI assessment for translational purposes will be
scheduled prior to first infusion and week 7-9. After surgery, patients attend
study visits at day 8 and day 29 . Their final study visit for physical
examination and laboratory testing is at day 57 (+/- 7 days), which is
scheduled to anticipate late-onset adverse events (particularly endocrine).
After this final visit, patients will be followed according to standard
clinical guidelines. Tumor biopsies/material preservation is required at
baseline and during surgery.
The study will be expanded after the first cohort to include 2 more cohorts of
15 patients and additional centers to test 2 different treatment schedules.
Patients in cohort 2 will be randomized between cohort 2a and 2b:
Cohort 2a
-Ipilimumab 3 mg/kg, days 1 and 22
-Nivolumab 1 mg/kg, days 1 and 22, followed by Nivolumab 3 mg/kg at day 43
-Radical cystectomy or nefro/ureterectomy with appropriate lymph node
dissection, day 56-84
Cohort 2b
-Ipilimumab 1 mg/kg, days 1 and 22
-Nivolumab 3 mg/kg, days 1 and 22, followed by Nivolumab 3 mg/kg at day 43
-Radical cystectomy or nefro/ureterectomy with appropriate lymph node
dissection, day 56-84. In cohort 2, the DSMB will be consulted after 12
patients have been randomized. Enrollment can be continued pending DSMB
assessment

Cohort 1:
- Day 1: Ipilimumab 3 mg/kg
- Days 22: Ipilimumab 3 mg/kg + Nivolumab 1 mg/kg
- Day 43: Nivolumab 3 mg/kg
- Day 57-71: Radical cystectomy or nefro/ureterectomy with appropriate lymph
node dissection

The study will be expanded after the first cohort to include 2 more cohorts of
15 patients and additional centers to test 2 different treatment schedules.
Patients in cohort 2 will be randomized between cohort 2a and 2b:
Cohort 2a
- Ipilimumab 3 mg/kg, days 1 and 22
- Nivolumab 1 mg/kg, days 1 and 22, followed by Nivolumab 3 mg/kg at day 43
- Radical cystectomy or nefro/ureterectomy with appropriate lymph node
dissection, day 56-84
Cohort 2b
- Ipilimumab 1 mg/kg, days 1 and 22
- Nivolumab 3 mg/kg, days 1 and 22, followed by Nivolumab 3 mg/kg at day 43
- Radical cystectomy or nefro/ureterectomy with appropriate lymph node
dissection, day 56-84. In cohort 2, the DSMB will be consulted after 12
patients have been randomized. Enrollment can be continued pending DSMB
assessment

Metastatic UC is an aggressive disease; patients have a median survival of
approximately 15 months when treated with modern chemotherapy regimens.
Therefore, therapeutic approaches in the perioperative setting could
potentially lower progression to metastatic disease and thereby improve
prognosis.

For melanoma, safety and efficacy have been demonstrated with both single agent
nivolumab and the combination of ipilimumab and nivolumab. Same day sequential
administration of nivolumab (1 mg/kg) followed by ipilimumab (3 mg/kg) was
initially evaluated in advanced/metastatic melanoma in 2 studies, the Phase 2
study CA209069 and the Phase 3 study CA209067, and demonstrated statistically
significant and clinically meaningful improvements in progression-free survival
(PFS) and objective response rate (ORR) compared to nivolumab or ipilimumab
monotherapy and served as the basis for an application to extend the indication
of nivolumab to include the use of nivolumab and ipilimumab in combination for
the treatment of advanced melanoma (United States Packaging Insert [USPI] for
nivolumab and Summary of Product Characteristics [SmPC] for nivolumab).
Subsequently, nivolumab plus ipilimumab has also demonstrated clinical activity
in several tumor types, including renal cell cancer (RCC), NSCLC, small cell
lung cancer (SCLC), and gastric cancer (Nivolumab IB V15; June 2016).
Early data suggests that the combination of dual check-point blockade with both
of these agents administered together has an efficacy advantage over single
agent anti-PD(L)1 in metastatic UC. In the Checkmate 032 trial, two schedules
of ipilimumab and nivolumab were tested: 4 courses of either ipilimumab 1
mg/kg/nivolumab 3 mg/kg (n=104) or ipilimumab 3 mg/kg/nivolumab 1 mg/kg
(n=26), followed by nivolumab monotherapy 3 mg/kg every 2 weeks. Response rates
were 26% and 38.5%, respectively (Sharma et al, SITC 2016), comparing favorably
to the nivolumab monotherapy response rate of 19.6% in the Checkmate 275
trial[17].
Although responses in the platinum-refractory metastatic setting are
encouraging, many patients do not respond. Earlier treatment with atezolizumab
or pembrolizumab in the metastatic setting, in cisplatin-ineligible patients
who did not receive chemotherapy yet, showed increased response rates ([20] and
O*Donnell et al, ASCO 2017). As response rates are higher in patients without
visceral metastases, we expect response rates in the pre-operative setting to
be even higher, reaching response rates high enough to compete with neoadjuvant
cisplatin-based chemotherapy.
In melanoma, a neoadjuvant combination immunotherapy study was conducted at our
institute (OpACIN, CA209-278). Although toxicity was commonly present, this
trial showed that neo-adjuvant combination immunotherapy had remarkable
response rates. 8 out of 10 patients showed profound regression of their
macrometastases after 2 courses of neo-adjuvant combination immunotherapy.
Preliminary pathology assessment showed 3/9 pCRs, 3/9 near pCR (remaining
micrometastasis), and 1/9 pPR (micrometastasis 0.5mm). An earlier study with
neo-adjuvant ipilimumab in resectable bladder cancer showed that this drug can
safely be administered prior to cystectomy and does not endanger resection [6,
7] In cohort 1 of this study, we used an attenuated schedule of ipilimumab and
nivolumab. This cohort has now fully enrolled; all patients had resection; 23
(96%) had resection <12 weeks from 1st cycle. Grade 3/4 irAEs occurred in 54%
of patients; 42% when excluding clinically insignificant lab deviations. A
total of 11/24 patients (46%) achieved a pCR, and another 8/24 patients showed
a significant reduction in tumor burden.
As cystectomy is a procedure with a high risk of morbidity and even mortality,
toxicity should be manageable to make sure it will not preclude resection in a
timely manner. For this reason we use a sequenced approach with an attenuated
dosing scheme. The design of this sequenced dosing schedule appeared to have
similar response rates with lower toxicity based on preliminary data from
metastatic melanoma patients treated with alternative combination schemes of
ipilimumab and pembrolizumab ( KEYNOTE-029, G. Long, ASCO 2016 suppl; abstract
9506).
In conclusion, although toxicity may be an issue in this setting, we believe
risk is mitigated by our sequenced treatment schedule. Data on efficacy in
metastatic setting suggest treating early in the course of disease and
combining anti-PD-1 with anti-CTLA4 may both augment response rate, supporting
our premise that this strategy will lead to response rates that are sufficient
to compete with cisplatin-based therapy.