Anda di halaman 1dari 9

Review Article

Immunotherapy for the Treatment of Urothelial Carcinoma


Nicholas M. Donin, Andrew T. Lenis, Stuart Holden, Alexandra Drakaki,
Allan Pantuck, Arie Belldegrun and Karim Chamie*
From the Department of Urology, Institute of Urologic Oncology (NMD, SH, AD, AP, AB, KC) and Department of Medicine,
Division of Hematology and Oncology (AD), David Geffen School of Medicine at UCLA and Jonsson Comprehensive Cancer
Center (AD, AP, AB, KC), University of California, Los Angeles, California

Purpose: We review the biological mechanisms of action, clinical safety and ef-
Abbreviations
ficacy of immunotherapies for urothelial carcinoma. We also describe current
and Acronyms
areas of research in immunotherapy, and highlight ongoing trials and promising
AE adverse event and novel investigational agents.
APC antigen presenting cell Materials and Methods: Data were obtained by a search of PubMed, Clin-
BCG bacillus Calmette-Guerin icalTrials.gov and Cochrane databases for English language articles published
CAR chimeric antigen receptor through February 2016. Applicable abstracts from recent Society of Urologic
CIS carcinoma in situ Oncology, European Association of Urology, American Urological Association
CTA cancer testis antigen and ASCO meetings were used.
CTLA-4 cytotoxic T-lymphocyte Results: Bacillus Calmette-Guerin is one of the most successful immunother-
antigen 4 apies in cancer treatment and remains the gold standard of care for patients with
FDA U.S. Food and Drug
high risk, nonmuscle invasive bladder cancer, with initial response rates of
Administration approximately 70%. However, with the exception of valrubicin and standard
chemotherapeutics there is a paucity of available treatment options for patients
HLA human leukocyte antigen
with recurrence or progression to more advanced disease. Recently there has
IFN interferon been significant interest in novel immunotherapeutic agents in the management
IHC immunohistochemistry of cases where bacillus Calmette-Guerin fails, as well as cases of more advanced
MCNA Mycobacterium phlei cancer. These investigational therapies can generally be classified into several
cell wall nucleic acid complex broad categories, including recombinant bacillus Calmette-Guerin and cell wall
MHC major histocompatibility derived therapies, cytokines, gene therapy, cancer vaccines, immune checkpoint
complex inhibitors, oncolytic viruses, adoptive immunotherapies and immune agonists, as
NK natural killer well as several additional immunomodulatory agents. The majority of these
NMIBC nonmuscle invasive agents are currently under investigation in phase I or II clinical trials. Recently
bladder cancer investigators reported evidence that inhibition of the PD-1/PD-L1 pathway has
NSCLC nonsmall cell lung
clinical activity in patients with advanced bladder cancer. These findings, along
cancer with successful phase III trials and U.S. Food and Drug Administration ap-
provals of other checkpoint inhibitors in melanoma, nonsmall cell lung cancer
RB retinoblastoma
and renal cell carcinoma, ultimately led to Food and Drug Administration
RCC renal cell carcinoma approval of atezolizumab for advanced disease, the first new treatment approved
TCR T-cell receptor for advanced urothelial carcinoma in 20 years.
TLR Toll-like receptor
TUR transurethral resection
UC urothelial cancer
Accepted for publication February 20, 2016.
No direct or indirect commercial incentive associated with publishing this article.
The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional
review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics
committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with gua-
rantees of confidentiality; IRB approved protocol number; animal approved project number.
* Correspondence: Department of Urology, David Geffen School of Medicine, University of California, Los Angeles, 300 Stein Plaza, 3rd Floor,
Los Angeles, CA 90095. e-mail: kchamie@mednet.ucla.edu

0022-5347/17/1971-0014/0 http://dx.doi.org/10.1016/j.juro.2016.02.3005

14 j www.jurology.com
THE JOURNAL OF UROLOGY
2017 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
Vol. 197, 14-22, January 2017
Printed in U.S.A.
IMMUNOTHERAPY FOR UROTHELIAL CANCER 15

Conclusions: While bacillus Calmette-Gu erin has demonstrated significant clinical efficacy in the treatment
of patients with bladder cancer, additional therapies are needed for those in whom bacillus Calmette-Gu erin
fails, as well as for those with advanced disease. Immunotherapy for urothelial carcinoma remains a prom-
ising and active area of research, and numerous agents, particularly the monoclonal antibodies targeting
checkpoint inhibition pathways, are showing encouraging signs of clinical activity.

Key Words: BCG vaccine, cancer vaccines, cell cycle checkpoints, immunotherapy, urologic neoplasms

THE use of bacillus Calmette-Gu erin in the treat- IMMUNOBIOLOGY


ment of urothelial carcinoma remains one of the The current understanding of antitumor cellular
most successful stories in the history of cancer immunity outlines several key events that occur
immunotherapy, and currently represents one of during immune destruction of tumor cells.4 First,
the most active and exciting areas of ongoing uro- tumor associated antigen must be sampled, pro-
logical research. As early as 1891, the interaction cessed and presented by APCs in the context of
between the immune system and neoplasia was MHC molecules. These cells must receive a suffi-
recognized and exploited as a potential target for cient secondary activation signal (ie B7-1 or B7-2) to
cancer therapeutics.1 However, not until the semi- achieve an immunogenic phenotype. Once acti-
nal article of Morales et al in 1976 revealing the vated, APCs travel to lymphoid tissue, where they
efficacy of bacillus Calmette-Gu erin was immuno- interact with T-cells. If these APCs are sufficiently
therapy proved effective for bladder cancer.2 Some immunogenic, they can induce the production of
40 years later bacillus Calmette-Gu erin remains the CD8 cytotoxic and NK effector cells. Finally, these
standard of care for patients with high risk non- effector cells travel to the tumor to infiltrate the
muscle invasive bladder cancer, and serves as proof microenvironment and carry out an antitumor
of principle that immunotherapy for urothelial response.
carcinoma can be effective. These processes are subject to complex modula-
Immunotherapy represents a technically feasible tion, and if sufficient negative regulation occurs, an
treatment strategy for bladder cancer for multiple antitumor effect will not be achieved. Tumors
reasons. Bladder tumors possess one of the highest themselves evade this process through various
rates of mutation of all human tumors, and are thus mechanisms working at multiple points along the
likely to be highly antigenic.3 Other advantages pathway. Secretion of paracrine mediators such as
include the ability to achieve direct contact between adenosine, prostaglandin E2, TGF-b and VEGF-A
therapeutic agents and the tumor via intravesical suppress activation of APCs and prevent T-cell
administration, the ability to deliver high concen- infiltration of the tumor bed. Cell surface ligands on
trations of agents with theoretically limited systemic tumor cells, such as PD-L1 and PD-L2, may be up-
exposure and the ability to monitor for treatment regulated, interacting with tumor infiltrating lym-
response via cystoscopic surveillance, biopsy, and phocytes and suppressing their cytotoxic activity.
noninvasive monitoring of cells and urinary bio- Tumors may down-regulate the normal antigen
markers. Nonetheless, few effective therapeutic processing and presentation on MHC-I molecules,
agents exist for the treatment of UC. A large pro- thus avoiding TCR and MHC interaction and sub-
portion of agents ultimately fail to achieve cure, sequent cytotoxicity. Finally, tumors are known to
including BCG and platinum based systemic thera- recruit populations of immunomodulatory cells,
pies. Given the significant burden of morbidity and such as regulatory T-cells, tumor associated mac-
mortality associated with the disease, alternative rophages and myeloid derived suppressor cells,
therapies are needed. As such, multiple immuno- which can blunt the antitumor response. The pres-
therapeutic agents are being investigated, several of ence of these immunosuppressive cell types within
which have demonstrated preliminary promise. tumor tissues has been shown to correlate with poor
In this analysis we discuss basic cancer immu- prognosis and tumor stage.5
nology as it pertains to the mechanism of action of
currently approved and investigational agents. We
review the discovery of BCG and the data support- 
BACILLUS CALMETTE-GUERIN
ing its use, and we discuss cytokines, adoptive cell BCG is a unique strain of mycobacterium bovis
therapies, checkpoint inhibition and additional developed by Albert Calmette and Camille Gu erin
agents currently being investigated in patients with at the Pasteur Institute in the early 1920s as an
UC. Ongoing clinical trials and completed seminal antituberculosis vaccine.6 Concurrent with its
trials are summarized in the supplementary table development was a growing appreciation of the
and supplementary Appendix (http://jurology.com/). connection between the immune system and
16 IMMUNOTHERAPY FOR UROTHELIAL CANCER

malignancy, and research gradually began on the reported rates of progression of CIS in the pre-BCG
use of BCG as a cancer therapy. Early investigators era were upward of 50%, BCG has been shown in
found that inoculation of mice with BCG exerted a multiple trials to result in initial complete response
protective effect against subsequently implanted rates of approximately 70%.12,13 Complete response
tumors.7 Clinical findings in leukemia, melanoma, rates appear to increase between the 3 and 6-month
and cancers of the head and neck had already been evaluations and, as such, a proportion of patients
described when in 1975 deKernion et al published with residual CIS at the 3-month point will go on to
their report of a melanoma lesion metastatic to the experience a complete response, even without
bladder that was successfully eradicated with an maintenance therapy.13 The benefit of maintenance
endoscopic intralesional injection of BCG.8 This case therapy remains questioned by some. While early
was followed soon after by the seminal report of smaller studies of varying schedules failed to illus-
Morales et al on use of intravesical BCG in the trate a benefit of BCG maintenance, the totality of
treatment of NMIBC.2 The initial findings were the evidence, including that from large prospective
subsequently reproduced in prospective randomized trials and meta-analyses,12e14 reveals that the
trials.9 BCG was approved for use intravesically in maximal benefits of BCG require the use of main-
1990 and currently remains the standard of care in tenance therapy.
patients with high risk NMIBC. Several groups have attempted to modify or
While a complete understanding of the mecha- create derivatives of BCG that might prove more
nisms underlying the therapeutic effect of BCG re- effective with fewer side effects. Recombinant BCG
mains the subject of investigation, a broad strains expressing IL-2, IL-12, IL-18, IFN-a and
understanding of the events underlying its activity IFN-g have been developed and tested in preclinical
has emerged.10 It is clear that BCG works via acti- models in the hope of enhancing the response.
vation of the immune system and induction of an However, no clinical trials have evaluated these
inflammatory response. This effect begins with agents in humans. A preparation of MCNA, which
attachment of BCG to urothelial cells, followed by has immunostimulatory and directly cytotoxic ac-
internalization via macropinocytosis. These cells tivity, was evaluated in a single arm trial of 129
then up-regulate MHC-II molecules and secrete patients with BCG refractory or relapsing
cytokines, resulting in recruitment of immune cells, NMIBC.15 With a 1-year disease-free survival rate
including lymphocytes, to the tumor environment. of 25% the study failed to meet its primary end
These lymphocytes induce what is believed to be a point. In addition, metastatic disease developed in
predominantly Th1 cytokine milieu, which ulti- 11% of patients during the trial, prompting concerns
mately leads to immune mediated cytotoxicity via regarding safety.
CD8 lymphocytes, NK cells and granulocytes.
In the original study of Morales et al patients
with NMIBC who had frequent recurrence and pa- CYTOKINE AND GENE THERAPY
tients with incompletely resected tumors were Interferons are a class of cytokines produced by
treated with 120 mg/50 cc BCG administered immune cells and tumor cells in response to the
intravesically plus intradermal injections for presence of pathogens, and cause a myriad of
6 weeks.2 In the period before BCG treatment the immunomodulatory effects. Initial studies evalu-
cohort of 9 patients experienced 22 recurrences ated IFN-a2B as a potential intravesical mono-
during 77 patient-months. Following BCG treat- therapy for NMIBC. While a small but detectable
ments these same 9 patients experienced 1 recur- antitumor effect has been demonstrated when given
rence during 41 patient-months. These findings in higher doses (80 million units or greater),16 IFN-
were followed by a randomized trial of 37 patients in a2B monotherapy has consistently been observed to
which those treated with BCG for 6 weeks exhibited be inferior to mitomycin C and BCG.17,18
a 22% recurrence rate at 1 year, compared to 42% IFN-a2B has also been investigated in combina-
recurrence in untreated controls. A Cochrane Group tion with BCG, given that the 2 agents are
meta-analysis published in 2000 of 6 randomized biocompatible and may act synergistically. A large
trials including 585 patients revealed that BCG prospective study of combination BCG plus IFN-
decreased the odds of recurrence at 1 year by 70% a2B in 670 BCG nave patients with NMIBC failed
compared to TUR alone (odds ratio 0.30, CI 0.21 to to find an improvement in disease-free rates
0.43).11 BCG has since been compared with multiple compared to BCG alone.19 Therefore, combination
chemotherapeutic agents and has demonstrated therapy is not recommended for use in BCG nave
persistent superiority for the prevention of patients. IFN-a2B has also been evaluated for use in
recurrence.12 the setting of BCG failures, and in this context
BCG is widely accepted as the standard of care several smaller series have shown disease-free rates
for the treatment of CIS of the bladder. While of approximately 50% at 1 year.20,21 The largest
IMMUNOTHERAPY FOR UROTHELIAL CANCER 17

study of combination therapy included 1,007 in- costimulatory molecules B7.1, ICAM-1 and LFA-3.26
dividuals, including BCG nave patients (53%) and The virus is given as a series of subcutaneous in-
BCG failures (46%).22 In this study a recurrence- jections and is subsequently processed by APCs,
free survival rate of 45% was noted in the BCG with the goal of eliciting a T-cell antigen specific
failure group at a median of 24 months. As such, response. PANVAC-VF was studied in advanced
BCG plus IFN-a2B is a viable alternative for pa- pancreatic cancer and failed to display a survival
tients in whom BCG monotherapy has failed. advantage. It is currently being investigated in a
One explanation for the limited efficacy of intra- phase II study in patients with NMIBC in whom
vesical IFN-a2B may be its transient presence BCG has failed (NCT02015104).
within the bladder, and thus prolonged delivery of HS-410 is a cellular vaccine based on a human
the cytokine within the urothelium may improve its cancer cell line that has been transfected to express
efficacy. An intravesical gene therapy using a re- HLA-A1 and secrete gp96-Ig, an endoplasmic retic-
combinant IFN replication deficient adenovirus ulum chaperone protein known to act as an immune
system has been developed and tested in early stimulant, particularly to CD8 T-cells. Hypotheses
phase clinical trials. In a phase I multicenter study regarding the potential mechanism of action of HS-
17 patients with NMIBC in whom at least 2 cycles of 410 include activation of dendritic cells, NK cells
BCG had failed were treated with the investiga- and cytotoxic lymphocytes, and stimulation of anti-
tional agent, which includes Syn3, an excipient that gen cross-presentation via CD19 and Toll-like re-
enhances transduction of the urothelium.23 Pro- ceptors. A phase I/II study is currently recruiting,
longed and increased levels of IFN-a were detected and will include patients who have undergone TUR
in urine. Of 14 patients treated at higher doses 6 and are candidates for BCG (NCT02010203). The
(43%) were complete responders at 3 months and 2 FDA has granted fast track designation for HS-410.
(14%) were disease-free at 39 months. A phase II Cancer testis antigens are a class of proteins
study of 40 patients is ongoing (NCT01687244). whose expression in normal tissues is limited to the
testis but that are aberrantly expressed in various
tumor types. As a result of this limited expression in
ONCOLYTIC VIRUSES normal tissues, CTAs are an attractive potential
Oncolytic viruses work by selective replication target for cancer vaccines. UC has high levels of
within tumor cells, causing tumor cell lysis and expression of various CTAs,27 and early phase vac-
release of additional oncolytic virions as well as cine trials targeting CTAs are ongoing for UC. A
tumor antigens. One such agent being evaluated is dendritic cell based vaccine, in which patient
CG0070, a serotype 5 oncolytic adenovirus that autologous dendritic cells pulsed with the CTA
selectively replicates in and destroys RB pathway MAGE-A3 were injected into patients with meta-
deficient cells, and carries the gene for the immu- static bladder cancer, exhibited some efficacy in a
nomodulatory cytokine GM-CSF. This virus is a small pilot study.28 A phase I trial of recombinant
logical agent in patients with UC, given the sub- MAGE-A3 plus BCG given to patients following
stantial proportion of tumors that harbor RB gene TUR has completed accrual and the results are
mutations or inactivation.24 In a phase I study 35 pending (NCT01498172). In a phase II trial evalu-
subjects underwent intravesical administration ating the efficacy of the recombinant MAGE-A3
using varying doses and treatment schedules.25 A vaccine after cystectomy the tumor tissue of pa-
49% complete response at a median of 10.4 months tients will be tested for MAGE-A3 expression and, if
was noted, and responses were greater than 80% in present, a course of 13 injections will be adminis-
the subset of patients with borderline or high RB tered spanning 27 months (NCT01435356).
status and 77% in those receiving 6 weekly doses. NY-ESO-1 is another CTA that has garnered in-
An open-label phase II/III study is ongoing terest based on its high degree of immunogenicity
(NCT01438112). and high frequency of expression in several cancer
types, including UC. In a phase I study 6 patients
who underwent cystectomy for NY-ESO-1 tumors
VACCINE THERAPY were vaccinated postoperatively with recombinant
Therapeutic cancer vaccines eliminate tumors by NY-ESO-1, GM-CSF and BCG.29 CD4 T-cell re-
activation of acquired cellular immunity and come sponses were found in all 6 patients, NY-ESO-1
in various forms, including synthetic peptides, re- specific antibody responses were induced in 5 and
combinant proteins and viral vectors. PANVAC-VF CD8 T-cell responses were seen in 1.
is a cancer vaccine using recombinant vaccinia and HER2 is a cell surface receptor tyrosine kinase
fowlpox viruses engineered to contain genes for best known for its association with aggressive forms
human CEA and MUC-1 (both known to be over- of breast cancer but is overexpressed or amplified in
expressed in urothelial carcinoma), along with the some urothelial tumors. Overexpression appears to
18 IMMUNOTHERAPY FOR UROTHELIAL CANCER

be associated with aggressive tumors and poor clin- The CTLA-4 receptor is similar in structure to
ical outcomes.30 DN24-02 is an autologous cellular the CD28 costimulatory receptor on the surface of
immunotherapy that uses the same antigen engi- T-cells and is known to be a critical element in the
neering technology as sipuleucel-T and targets T-cell activation process (fig. 1). CTLA-4 expression
HER2. Patients undergo leukapheresis in which im- on T-cells is induced with T-cell activation and has
mune cells are harvested and activated ex vivo with a been observed to compete with CD28 for interaction
recombinant fusion protein that has components of with the costimulatory B7-1 and B7-2 molecules on
the extracellular and intracellular domains of HER2 APCs. However, rather than promote T-cell activa-
linked to GM-CSF. NeuACT (NCT01353222) is an tion and effector functions, the B7:CTLA-4 interac-
open-label, randomized, phase II trial comparing tion inhibits T-cell activation primarily in lymphoid
adjuvant DN24-02 with surveillance in patients with tissues. Ipilimumab, a monoclonal anti-CTLA-4
HER2 UC at high risk for relapse following cys- antibody, blocks the B7:CTLA-4 interaction, thus
tectomy or nephroureterectomy. shifting T-cell equilibrium toward activation and
effector function, with subsequent antitumor ef-
fects. Successful trials of ipilimumab in metastatic
CHECKPOINT INHIBITION melanoma led to its FDA approval in 2011.33 Ipili-
Immune checkpoint inhibition is one of the most mumab was evaluated in a small study of 12 pa-
exciting and promising areas of research in cancer tients with bladder cancer to investigate its
therapeutics. Positive phase III clinical trials in biological effects on bladder cancer tissue.34 In this
advanced melanoma, nonsmall cell lung cancer and study patients scheduled for cystectomy for cT1-
RCC have led to FDA approval of 3 checkpoint in- T2N0M0 disease were given 2 doses of the drug
hibitors. A large number of clinical studies evalu- beginning approximately 7 weeks before cys-
ating these agents in various malignancies are now tectomy. The majority of AEs were grade 1 or 2,
ongoing, including multiple trials in UC. Check- although 2 patients had a delay in surgery due to
point inhibition involves targeting T-cell regulatory immune related AEs. A phase II ongoing clinical
pathways to reduce inhibitory signaling and pro- trial evaluating ipilimumab in combination with
mote T-cell activation and enhanced antitumor ac- gemcitabine and cisplatin in patients with advanced
tivity. It is now recognized that in addition to TCR disease was presented at the ASCO 2016 Genito-
activation, costimulatory and inhibitory molecules urinary Cancers Symposium and failed to meet its
on the surface of T-cells exist that further regulate primary end point (NCT01524991).
T-cell behavior. The discovery that blockade of The programmed cell death 1 receptor (PD-1) and
inhibitory signaling pathways resulted in tumor its ligands, PD-L1 and PD-L2, are an exciting set of
rejection in in vivo models ultimately led to an up- receptor ligands that are vital in T-cell immuno-
surge in investigation and discovery in this modulation and tolerance (fig. 2). Like CTLA-4, the
area.31,32 PD-1 receptor becomes expressed on activated

T-CELL APC INTERACTION

CTLA-4 ANTIBODY BLOCKS CTLA-4 SIGNALING

CTLA-4
Lymph nodes
B7-1, Activated
B7-2 T-cell

28
CD

Bladder
carcinoma
Antigen TCR
Presenting MHC
Cell IL-2 production
Tumor T-cell proliferation
antigen
Cytotoxic function
CTLA-4 expression

Figure 1. T-celleAPC interaction. Medical Illustration by JustinKleinCMI.com 2016.


IMMUNOTHERAPY FOR UROTHELIAL CANCER 19

T-CELL TUMOR CELL INTERACTION


PD-L1 INHIBITORS
AND PD-1 INHIBITORS
INHIBIT THE DOWN REGULATION
OF CYTOTOXIC FUNCTION

Bladder
carcinoma
PD-1 T-cell
cytotoxic
PD-L1,
function
PD-L2

Tumor
TCR
cell
Tumor antigen
MHC
Tumor cells

Oncogenic activation, induction


from immune activity in
tumor microenvironment

Figure 2. T-celletumor cell interaction. Medical Illustration by JustinKleinCMI.com 2016.

T-cells, and PD-1/ligand interaction results in inhi- responses, and in 4 of 35 patients with lymphocyte
bition of TCR mediated functions and suppression of IHC 0 to 1 tumors (11%). Interestingly responses to
the T-cell effector function. While CTLA-4 activation atezolizumab were associated with the tumor-
is generally thought to down-regulate T-cell activa- infiltrating immune cell IHC scores but not with
tion in lymphoid tissues, PD-1 activity is believed to tumor cell IHC scores. There were no grade 4 or 5
act primarily in the tumor microenvironment, treatment related AEs, and only 4% of patients
where it restrains T-cell mediated tumor destruc- experienced a grade 3 event (NCT01375842).32
tion. The observation of up-regulation of PD-L1 on More recently investigators reported results from
tumor cells indicated activation of the PD-1 pathway IMvigor 210, a phase II study of atezolizumab in
as a mechanism of immune evasion.35 Immunohis- patients with metastatic or locally advanced UC
tochemical studies have revealed that increased PD- who experienced disease progression during or after
L1 expression is associated with increasing bladder platinum based therapy.37 In 310 evaluable patients
tumor stage and grade, suggesting that blockade the overall response rate was 15%, with 38 of 45
may be efficacious.36 This hypothesis was ultimately (84%) of responses ongoing at the time of data cut-
borne out in successful phase III trials in melanoma, off. Response rate correlated with IHC status (ORR
squamous NSCLC and RCC. 26% in IHC in 2 to 3 tumors, 18% in IHC 1 to 3
In what is arguably the most exciting develop- tumors). Grade 3 or 4 treatment related AEs
ment in bladder cancer therapeutics in decades, in occurred in 5 of 310 patients (16%).
May 2016 the FDA approved atezolizumab Another emerging PD-1 inhibiting antibody,
(MPDL3280A), a monoclonal IgG1 antibody that pembrolizumab, is demonstrating antitumor activ-
binds PD-L1, for the treatment of advanced stage ity in early phase clinical trials. In data presented at
bladder cancer in patients who experienced disease ASCO 2015 a total of 33 patients with advanced UC
progression on platinum based therapy.31,37 This were treated with pembrolizumab every 2 weeks
approval was based on data from seminal phase I until complete response, progression or unaccept-
and phase II studies in which pathological speci- able toxicity. A complete response was noted in 3 of
mens from patients with advanced bladder cancer 28 patients (11%) and a partial response in 4 (14%)
were graded for the presence of PD-L1 before at a median of 13 months. Grade 3 to 4 drug related
treatment. In the phase I study 67 patients with AEs occurred in 15% of patients (NCT01848834).38
metastatic bladder cancer had the pathological Combination therapy with PD-1/PD-L1 and
specimen graded for the presence of PD-L1 before CTLA-4 inhibition is being investigated in a number
treatment. An objective response was seen in 17 of malignancies, including UC (NCT01928394,
patients overall (25%), with 15 partial responses NCT02496208). Several other T-cell surface re-
and 2 complete responses. Objective responses ceptors that regulate cell activation and efficacy
occurred in 13 of 30 patients with lymphocyte IHC 2 have been identified as potential therapeutic tar-
to 3 tumors (43%), of which 2 were complete gets, such as B7-H3 and OX40, and antibodies
20 IMMUNOTHERAPY FOR UROTHELIAL CANCER

modulating their activity are also being evaluated receptor can bind directly to the cell surface to become
in early phase clinical trials (NCT01391143, activated, as opposed to the engineered TCRs, which
NCT02221960, NCT02219724). must bind specific allelic variants of MHC molecules.
Thus, CAR therapies are not limited to a particular
HLA subtype and circumvent the problem of MHC
ADOPTIVE IMMUNOTHERAPY down-regulation. However, they are limited in
Adoptive immunotherapy involves administering epitope repertoire to cell surface proteins and cannot
immune cells with direct anticancer activity to a bind intracytoplasmic derived peptides.
cancer bearing host. Lymphocytes are first obtained In CAR therapy T-cells from a patient are ob-
from the patient via biopsy of the primary or met- tained via leukapheresis, activated, transfected
astatic tumor site, or via leukapheresis. These cells with a gene coding for the fusion complex, expanded
are cultured in vitro to expand the lymphocyte to the desired dosage and infused back into the pa-
population and ultimately transferred back to the tient. CAR therapy has displayed notable efficacy in
patient, typically following a nonmyeloablative patients with leukemia and lymphoma. The anti-
conditioning regimen. Studies of melanoma and CD19 CAR T-cell agent CTL019 was granted
lymphoma have shown complete and durable re- breakthrough therapy designation in July 2014.
sponses.39 In UC adoptive immunotherapy remains Clinical results in UC are not yet available.
in the early stages of clinical investigation. In a
feasibility trial involving 12 patients with metasta-
tic bladder cancer lymphocytes were extracted from OTHER EMERGING THERAPIES
metastatic lymph nodes, and tumor specific T helper ALT-801 is a unique molecule in which an IL-2
cells were expanded in vitro and subsequently molecule has been fused to a single-chain TCR
infused into patients. The process was safe and no that binds a peptide derived from p53 when dis-
major AEs were noted. However, the process was played in the context of HLA-A*0201. The purpose
completed in only 50% of the patients due to tech- of this fusion protein is to direct IL-2 activation of
nical limitations.40 T-cells to cells with aberrant p53. In a phase I trial
Adoptive immunotherapy is somewhat limited by of 26 patients with p53 metastatic malignancies
the need to obtain tumor tissue for the harvest of increases in IFN-g were detected in the serum after
tumor-infiltrating lymphocytes. TCR T-cell therapy treatment, although in contrast to treatment with
is an emerging technology that circumvents this IL-2, no TNF-a was detected.42 Of the patients 10
problem by taking peripherally obtained lympho- had stable disease at 11 weeks. Given that a
cytes and transfecting them with a gene for a re- significant proportion of UCs have mutated p53,
combinant TCR specifically engineered to bind a ALT-801 is currently being investigated for UC
particular CTA-MHC complex. Advantages include in the muscle invasive and metastatic settings
access to peripheral lymphocytes, as well as the in combination with gemcitabine and cisplatin
ability to engineer a TCR that is optimized for (NCT01326871), as well as in the setting of BCG
maximal affinity for the CTA-MHC complex. Early failures for NMIBC (NCT01625260).
reports show success with this therapy in melanoma Another promising agent under investigation is a
and synovial sarcoma, including 1 study evaluating fusion protein consisting of an IL-15 super agonist
a TCR engineered to recognize NY-ESO-1.41 Given bound to a soluble domain of the IL-15 receptor pro-
that a proportion of urothelial tumors are known to tein. This fusion complex, known as ALT-803, has
express NY-ESO-1, TCR T-cell therapy is being displayed 25 times the in vivo biological activity of
investigated for use in bladder cancer in early phase native IL-15. Because IL-15 is a critical factor in the
trials (NCT02457650). Some limitations of TCR development, proliferation and activation of effector
T-cell therapy are the known propensity of tumors NK and CD8T cells, it is speculated to be a prom-
to down-regulate MHC and, thus, evade TCR ising agent. In a carcinogen induced rat NMIBC
mediated autoimmunity, as well as the specificity of model tumor burden was reduced by 35% and 15%
a given TCR to only certain HLA alleles. with ALT-803 and BCG, respectively.43 When the
CAR T-cell therapy represents a similar emerging agents were given in combination, tumor burden was
adoptive T-cell technology that potentially circum- decreased by 46%. This agent is being evaluated in
vents some of the aforementioned limitations of TCR clinical trials in several tumor types, including mel-
therapy. CARs are engineered fusion complexes anoma, lymphoma and NSCLC, and in combination
consisting of a single-chain variable region of an with BCG for NMIBC (NCT02138734).
antibody domain that binds a specific CTA, a trans- Toll-like receptors are transmembrane proteins
membrane domain, a costimulatory domain and the that function in innate and adaptive immunity.
CD3-zeta domain, which serves as the intracellular TLRs are expressed in normal urothelium and in
signaling domain. T-cells expressing this chimeric NMIBC cells. Imiquimod, a small molecule of the
IMMUNOTHERAPY FOR UROTHELIAL CANCER 21

imidazoquinoline family, is known to activate TLR7 represents one of the first immuno-oncologic ap-
and TLR8, a pathway that is also believed to be provals in solid tumors and shows that UC is an
activated by BCG, and has exhibited antitumor ef- immunosensitive tumor amenable to immune medi-
fect in basal cell carcinoma and penile CIS when ated treatment strategies. Progressive understand-
used topically. TMX-101 is a formulation of imiqui- ing of the complex regulatory mechanisms that
mod optimized for intravesical administration, and govern cellular immunity has led to the development
has been evaluated in several early phase studies of of a number of novel agents and strategies, most
low grade disease44 and CIS (unpublished data), notably checkpoint inhibitors, which have resulted in
totaling 33 patients. While the majority of patients overall survival benefits in patients with advanced
reported AEs, virtually all were grade 1 or 2, and all melanoma, NSCLC and RCC. The recent FDA
patients received all 6 planned instillations. approval of atezolizumab heralds a new era in the
treatment of UC, and this agent will likely assume a
central role in the management of many cases of
CONCLUSIONS advanced stage disease. There is ample reason to
Urothelial carcinoma remains a common malignancy believe a number of novel immunotherapeutics
with few treatment advances in the last 20 years. currently under investigation will soon emerge as
However, the 1990 approval of BCG for NMIBC important treatment options for patients with UC.

REFERENCES
1. Coley WB: The treatment of malignant tumors by bladder cancer. Cochrane Database Syst Rev 19. Nepple KG, Lightfoot AJ, Rosevear HM et al:
repeated inoculations of erysipelas. With a 2000; 4: CD001986. Bacillus Calmette-Guerin with or without inter-
report of ten original cases. 1893. Clin Orthop feron a-2b and megadose versus recommended
12. Sylvester RJ, van der Meijden AP, Witjes JA
Relat Res 1991; 262: 3. daily allowance vitamins during induction and
et al: Bacillus Calmette-Guerin versus chemo- maintenance intravesical treatment of non-
2. Morales A, Eidinger D and Bruce AW: Intra- therapy for the intravesical treatment of patients muscle invasive bladder cancer. J Urol 2010;
cavitary bacillus Calmette-Guerin in the treat- with carcinoma in situ of the bladder: a meta-
184: 1915.
ment of superficial bladder tumors. J Urol 1976; analysis of the published results of randomized
116: 180. clinical trials. J Urol 2005; 174: 86.
20. ODonnell MA, Krohn J and DeWolf WC: Salvage
3. Lawrence MS, Stojanov P, Polak P et al: Mutational 13. Lamm DL, Blumenstein BA, Crissman JD et al: intravesical therapy with interferon-alpha 2b
heterogeneity in cancer and the search for new Maintenance bacillus Calmette-Guerin immuno- plus low dose bacillus Calmette-Guerin is
cancer-associated genes. Nature 2013; 499: 214. therapy for recurrent Ta, T1 and carcinoma in situ effective in patients with superficial bladder
transitional cell carcinoma of the bladder: a cancer in whom bacillus Calmette-Guerin alone
4. Mellman I, Coukos G and Dranoff G: Cancer randomized Southwest Oncology Group study. J previously failed. J Urol 2001; 166: 1300.
immunotherapy comes of age. Nature 2011; Urol 2000; 163: 1124.
480: 480. 21. Lam JS, Benson MC, ODonnell MA et al: Ba-
14. Oddens J, Brausi M, Sylvester R et al: Final re- cillus Calmette-Guerin plus interferon-alpha2B
5. Curiel TJ, Coukos G, Zou L et al: Specific sults of an EORTC-GU Cancers Group randomized intravesical therapy maintains an extended
recruitment of regulatory T cells in ovarian car- study of maintenance bacillus Calmette-Guerin treatment plan for superficial bladder cancer
cinoma fosters immune privilege and predicts in intermediate- and high-risk Ta, T1 papillary with minimal toxicity. Urol Oncol 2003; 21: 354.
reduced survival. Nat Med 2004; 10: 942. carcinoma of the urinary bladder: one-third dose
6. Calmette A, Guerin C, Boquet A et al: La versus full dose and 1 year versus 3 years of 22. Joudi FN, Smith BJ and ODonnell MA: Final
Vaccination Preventive contre la Tuberculose par maintenance. Eur Urol 2013; 63: 462. results from a national multicenter phase II
le BCG. Paris: Masson et Cie 1927. 15. Morales A, Herr H, Steinberg G et al: Efficacy and trial of combination bacillus Calmette-Guerin
safety of MCNA in patients with nonmuscle plus interferon alpha-2B for reducing recur-
7. Old LJ, Clarke DA and Benacerraf B: Effect of rence of superficial bladder cancer. Urol Oncol
bacillus Calmette-Guerin infection on trans- invasive bladder cancer at high risk for recurrence
and progression after failed treatment with ba- 2006; 24: 344.
planted tumours in the mouse. Nature, suppl.,
1959; 184: 291. cillus Calmette-Guerin. J Urol 2015; 193: 1135.
23. Dinney CP, Fisher MB, Navai N et al: Phase I trial
8. deKernion JB, Golub SH, Gupta RK et al: Suc- 16. Glashan RW: A randomized controlled study of of intravesical recombinant adenovirus mediated
cessful transurethral intralesional BCG therapy intravesical alpha-2b-interferon in carcinoma in interferon-a2b formulated in Syn3 for bacillus
of a bladder melanoma. Cancer 1975; 36: 1662. situ of the bladder. J Urol 1990; 144: 658. Calmette-Guerin failures in nonmuscle invasive
bladder cancer. J Urol 2013; 190: 850.
9. Lamm DL, Thor DE, Harris SC et al: Bacillus 17. Malmstrom PU: A randomized comparative dose-
Calmette-Guerin immunotherapy of superficial ranging study of interferon-alpha and mitomycin-
24. Mitra AP, Birkhahn M and Cote RJ: p53 and
bladder cancer. J Urol 1980; 124: 38. C as an internal control in primary or recurrent
retinoblastoma pathways in bladder cancer.
superficial transitional cell carcinoma of the
World J Urol 2007; 25: 563.
10. Redelman-Sidi G, Glickman MS and Bochner BH: bladder. BJU Int 2002; 89: 681.
The mechanism of action of BCG therapy for
bladder cancerda current perspective. Nat Rev 18. Jimenez-Cruz JF, Vera-Donoso CD, Leiva O et al: 25. Burke JM, Lamm DL, Meng MV et al: A first in
Urol 2014; 11: 153. Intravesical immunoprophylaxis in recurrent su- human phase 1 study of CG0070, a GM-CSF
perficial bladder cancer (stage T1): multicenter expressing oncolytic adenovirus, for the treat-
11. Shelley MD, Court JB, Kynaston H et al: Intra- trial comparing bacille Calmette-Guerin and ment of nonmuscle invasive bladder cancer. J
vesical bacillus Calmette-Guerin in Ta and T1 interferon-alpha. Urology 1997; 50: 529. Urol 2012; 188: 2391.
22 IMMUNOTHERAPY FOR UROTHELIAL CANCER

26. Petrulio CA and Kaufman HL: Development of the 33. Hodi FS, ODay SJ, McDermott DF et al: patients with metastatic melanoma using T-cell
PANVAC-VF vaccine for pancreatic cancer. Expert Improved survival with ipilimumab in patients transfer immunotherapy. Clin Cancer Res 2011;
Rev Vaccines 2006; 5: 9. with metastatic melanoma. N Engl J Med 2010; 17: 4550.
363: 711.
27. Sharma P, Shen Y, Wen S et al: Cancer-testis 40. Sherif A, Hasan MN, Marits P et al: Feasibility of
antigens: expression and correlation with sur- 34. Carthon BC, Wolchok JD, Yuan J et al: Preop-
T-cell-based adoptive immunotherapy in the first
vival in human urothelial carcinoma. Clin Cancer erative CTLA-4 blockade: tolerability and immune
12 patients with advanced urothelial urinary
Res 2006; 12: 5442. monitoring in the setting of a presurgical clinical
bladder cancer. Preliminary data on a new
trial. Clin Cancer Res 2010; 16: 2861.
immunologic treatment based on the sentinel
28. Nishiyama T, Tachibana M, Horiguchi Y et al:
35. Dong H, Strome SE, Salomao DR et al: Tumor- node concept. Eur Urol 2010; 58: 105.
Immunotherapy of bladder cancer using autolo-
gous dendritic cells pulsed with human lympho- associated B7-H1 promotes T-cell apoptosis: a
cyte antigen-A24-specific MAGE-3 peptide. Clin potential mechanism of immune evasion. Nat 41. Robbins PF, Morgan RA, Feldman SA et al: Tumor
Cancer Res 2001; 7: 23. Med 2002; 8: 793. regression in patients with metastatic synovial
cell sarcoma and melanoma using genetically
36. Inman BA, Sebo TJ, Frigola X et al: PD-L1 (B7-H1)
29. Sharma P, Bajorin DF, Jungbluth AA et al: Im- engineered lymphocytes reactive with NY-ESO-1.
expression by urothelial carcinoma of the
mune responses detected in urothelial carci- J Clin Oncol 2011; 29: 917.
bladder and BCG-induced granulomata: associa-
noma patients after vaccination with NY-ESO-1
tions with localized stage progression. Cancer
protein plus BCG and GM-CSF. J Immunother 42. Fishman MN, Hadjenberg J, Kuzel T et al:
2007; 109: 1499.
2008; 31: 849. Phase I/II clinical trial of ALT-801, a T-cell
37. Rosenberg JE, Hoffman-Censits J, Powles T receptor/IL-2 fusion protein, plus gemcitabine
30. Zhao J, Xu W, Zhang Z et al: Prognostic role of et al: Atezolizumab in patients with locally and cisplatin in urothelial cancer. J Clin Oncol,
HER2 expression in bladder cancer: a systematic advanced and metastatic urothelial carcinoma suppl., 2013; 31: 271.
review and meta-analysis. Int Urol Nephrol 2015; who have progressed following treatment with
47: 87. platinum-based chemotherapy: a single-arm, 43. Gomes-Giacoia E, Miyake M, Goodison S et al:
multicentre, phase 2 trial. Lancet 2016; 387: Intravesical ALT-803 and BCG treatment reduces
31. Powles T, Eder JP, Fine GD et al: MPDL3280A
1909. tumor burden in a carcinogen induced bladder
(anti-PD-L1) treatment leads to clinical activity
in metastatic bladder cancer. Nature 2014; cancer rat model; a role for cytokine production
38. Plimack ER, Bellmunt J, Gupta S et al: Pem-
515: 558. and NK cell expansion. PLoS One 2014; 9:
brolizumab (MK-3475) for advanced urothelial
e96705.
cancer: updated results and biomarker analysis
32. Petrylak DP, Powles T, Bellmunt J et al: A from KEYNOTE-012. J Clin Oncol, suppl., 2015;
phase Ia study of MPDL3280A (anti-PDL1): 33: 4502. 44. Falke J, Lammers RJ, Arentsen HC et al: Results
updated response and survival data in urothe- of a phase 1 dose escalation study of intra-
lial bladder cancer (UBC). J Clin Oncol, suppl., 39. Rosenberg SA, Yang JC, Sherry RM et al: Du- vesical TMX-101 in patients with nonmuscle
2015; 33: 4501. rable complete responses in heavily pretreated invasive bladder cancer. J Urol 2013; 189: 2077.