1 EPA Staff Assessment Report on application APP203750 To release genetically modified live Chimaeric Antigen Receptor T-cells for use in a Phase 1 dose escalation clinical trial to examine safety and efficacy in patients with relapsed and refractory B-cell lymphomas. 20 September 2019 Application number: APP203750 Purpose: To release genetically modified live Chimaeric Antigen Receptor T-cells for use in a Phase 1 dose escalation clinical trial to examine safety and efficacy in patients with relapsed and refractory B-cell lymphomas. Applicant: The Malaghan Institute for Medical Research Application Lead:
34
Embed
EPA Staff Assessment Report on application APP203750€¦ · self’) by so-called antigen presenting cells (APCs; Fig. 1). The antigen is recognised by T-cell receptors (TCRs) on
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
EPA Staff Assessment Report on application APP203750
To release genetically modified live Chimaeric Antigen
Receptor T-cells for use in a Phase 1 dose escalation clinical
trial to examine safety and efficacy in patients with relapsed and
refractory B-cell lymphomas.
20 September 2019
Application number: APP203750
Purpose:
To release genetically modified live Chimaeric Antigen
Receptor T-cells for use in a Phase 1 dose escalation
clinical trial to examine safety and efficacy in patients
with relapsed and refractory B-cell lymphomas.
Applicant: The Malaghan Institute for Medical Research
Application Lead:
2
EPA staff assessment report for application APP203750
September 2019
ADVICE TO THE DECISION MAKER
Executive summary and recommendation
On 16 September 2019, the Environmental Protection Authority (EPA) formally received an application
from the Malaghan Institute for Medical Research (the applicant) to release genetically modified
organisms from containment. The organisms, live Chimaeric Antigen Receptor T-cells (CAR T-cells)
are intended for the treatment of patients with relapsed and refractory B-cell lymphomas in a Phase 1
dose escalation clinical trial at Wellington Hospital. The application to release the CAR T-cells for the
trial and for sterility testing in the Wellington SCL clinical laboratories at Wellington Hospital was
lodged pursuant to section 34 of the Hazardous Substances and New Organisms (HSNO) Act 1996
(the “HSNO Act”).
Section 38I of the HSNO Act provides for a rapid assessment of applications received under section
34, if the application seeks the release of a qualifying organism. A qualifying organism is, in part, a
new organism (including a genetically modified organism) that is a medicine or is contained in a
medicine.
Based on the information in the application and from other published sources, we found, based on:
the known safety profiles of CAR T-cell therapies to date
an existing development approval for CAR T-cells from EPA
an existing licence to manufacture CAR T-cells from Medsafe
the IANZ accreditation of Wellington SCL
the experience of Wellington Hospital staff in the safe handling and disposal of medical
samples and waste
the lack of adverse effect on Māori relationships with their environment and taonga,
that it is highly improbable that CAR T-cells could form a self-sustaining population and have
significant adverse effects on the health of the public, any valued species, natural habitats, or the
environment.
Therefore, we recommend that the application is approved, subject to the proposed controls.
3
EPA staff assessment report for application APP203750
September 2019
Table of Contents
Executive summary and recommendation ......................................................................................... 2
Table of Contents .................................................................................................................................. 3
Purpose of this document .................................................................................................................... 4
Receptor T-cells (CAR T-cells) that will be developed under an existing containment approval
(APP203214). If approved for release, WZTL-002 cells are intended for use in the treatment of
patients with relapsed and refractory B-cell non-Hodgkin’s lymphomas in a Phase 1 dose
escalation clinical trial (the ENABLE trial) and for sterility testing in a clinical laboratory. The
application was lodged pursuant to section 34 of the HSNO Act.
B-cell non-Hodgkin’s lymphoma 3. Lymphomas are cancers that start in the bone marrow, in differentiating stem cells that are
destined to become lymphocytes, that is, immune cells, such as T-cells and B-cells (American
Cancer Society 2019a). Non-Hodgkin’s lymphoma (NHL) is a general descriptor for more than 50
different B-cell cancers (Swerdlow et al, 2016), including diffuse large B-cell lymphoma, the most
common type of B-cell NHL (Armitage et al, 2017).
4. There were 923 new registrations of NHL in New Zealand in 2017, which makes it among the 10
most prevalent cancers in the country (Ministry of Health 2018a). There were 317 NHL deaths in
2013, the most recent year from which data are available (Ministry of Health 2016). The disease is
approximately 35% more prevalent among men than women (Ministry of Health 2018a). Māori
have essentially equal incidences of NHL per 100,000 population as non-Māori (Ministry of Health
2019). The Ministry of Health does not distinguish between T-cell and B-cell NHL in its statistics,
so it is not clear what proportion of these registrations and deaths were the result of B-cell NHL.
5. Despite an increasing incidence of lymphomas in New Zealand (and around the world), the
prognosis for patients is generally improving, with decreasing death rates observed between 1995
(5.0 per 100,000) and 2009 (4.7 per 100,000). Death rates among Māori are somewhat higher
than among non-Māori at 5.2 deaths per 100,000 (National Lymphoma Tumour Standards
Working Group 2013).
Current Non-Hodgkin’s lymphoma treatments in New Zealand 6. Current standards of care outline the recommended treatments to be made available to all
lymphoma sufferers in New Zealand (National Lymphoma Tumour Standards Working Group
2013). These include various types of chemotherapy regimens, including both small molecules
and monoclonal antibodies, often in combination (Armitage et al, 2017), radiation therapy,
surgery, stem cell transplantation, and finally, palliative care when other treatment options have
failed (National Lymphoma Tumour Standards Working Group 2013). Treatment options vary
5
EPA staff assessment report for application APP203750
September 2019
widely, depending on the type of B-cell lymphoma being treated (American Cancer Society
2019b).
7. Although there are lymphoma and leukaemia patients currently residing in New Zealand who have
received some form of CAR T-cell therapy (see application), no CAR T-cell therapy of any kind is
currently available within New Zealand. Because CAR T-cells are human cells that carry a foreign
gene, they are considered to be genetically modified organisms under the HSNO Act. Therefore,
CAR T-cells are classified as new organisms under the HSNO Act, and any CAR T-cell therapy,
including the one described in the application, requires an approval for the release of a new
organism from EPA. This application, if approved, would provide new treatment options for eligible
patients in New Zealand in the Malaghan Institute’s ENABLE trial.
8. As discussed in the next section of this staff advice, most CAR T-cell therapies are still
experimental and are being assessed in a large number of clinical trials. As such, they are
generally only being used in individuals for whom other treatment options have failed. Even CAR
T-cell therapies approved by the Food and Drug Administration (FDA) in the United States are
only for use in patients for whom other treatment options have failed (generally referred to as
relapsed and refractory lymphomas and leukaemias). Consistent with worldwide best practice, the
ENABLE trial is only for patients that suffer from relapsed and refractory NHL, as noted in the
application.
Chimaeric antigen receptor T-cells – history and effectiveness 9. Cancer cells undergo many mutations relative to the normal cells in the body that would normally
lead to their destruction by the immune system (Muenst et al, 2016). Although such
immunosurveillance normally protects humans and other vertebrates from cancer, evasion of the
immune system that results in malignancies can occur by a wide variety of mechanisms (Muenst
et al, 2016). Thus, a great deal of clinical research involves exploring ways to overcome this
immune system evasion and re-engage immunosurveillance mechanisms in the destruction of
tumours (Maus et al, 2014; Muenst et al, 2016; ClinicalTrials.gov 2019c; Weinkove et al, 2019).
Such immune system approaches to cancer treatment are collectively known as immunotherapy.
10. One such promising line of immunotherapy research involves genetic modification of a specific
type of immune cell, called a T-cell, with a gene encoding a protein that will directly recognise and
kill tumour cells, called a chimaeric antigen receptor (CAR). The genetically modified T-cells are
known as CAR T-cells, and since they were first conceived and developed in the late 1980s
(Gross et al, 1989), a wide variety of CARs have been trialled. Their clinical effectiveness was
demonstrated for the first time in 2012 (Rosenbaum 2017).
11. Ordinarily, T-cell responses involve a complex series of steps involving recognition of a foreign
entity (such as an infecting bacterium or an immunosurveilled cancer cell), killing the cell, and
processing the killed cell in such a way that antigens (generally short processed parts of proteins
from the infecting entity) are displayed to T-cells in conjunction with the Major Histocompatibility
Complex (MHC; a group of proteins used by the immune system to distinguish ‘self’ from ‘non-
self’) by so-called antigen presenting cells (APCs; Fig. 1). The antigen is recognised by T-cell
receptors (TCRs) on a subset of T-cells in the body, in conjunction with co-stimulating signals
from the APC recognised by the CD28 receptor, which activates the intracellular signalling domain
or domains (also called co-stimulation domains) on the receptors. Activation of the co-stimulation
6
EPA staff assessment report for application APP203750
September 2019
domains signals the T-cell to activate its cell-killing functions, as well as to divide and proliferate
and recruit other immune cells to attack the invading organism (Fig. 1; Weinkove et al, 2019).
12. The idea behind CAR T-cell therapy is to bypass the normal mode of T-cell stimulation and equip
the T-cell with a receptor that allows the T-cell to directly kill cells carrying the CAR’s target
antigen (ie, tumour cells; Fig. 1), which bypasses the need for APCs, as well as many
mechanisms by which tumours evade immunosurveillance (Maus et al, 2014; Weinkove et al,
2019). Conceptually, CAR T-cells are relatively simple, as they are a specific type of immune cell
(the T-cell) that is genetically modified with a single gene construct, called a chimaeric antigen
receptor (CAR; Fig. 1, Fig. 2). Upon binding of the tumour cell antigen, the T-cell is directly
activated via one or more co-stimulation domains in the CAR (Fig. 1, Fig. 2) to kill the tumour cell,
as well as to divide and proliferate to attack other tumour cells.
13. CARs have a four segment structure consisting of 1) an extracellular antigen recognition domain,
2) a so-called hinge domain, 3) a transmembrane (ie, membrane-spanning) domain, and 4) one or
more intracellular signalling domains (Fig. 2). At a minimum, the co-stimulation domain activates
the cell killing response of the CAR T-cell, via the CD3ζ (zeta) stimulation domain as were used in
first generation CAR T-cells (Fig. 2; Maus et al, 2014).
14. First generation CAR T-cells often provided limited or temporary remissions of cancers, in part
because of limited proliferation (often referred to clinically as expansion) of the CAR T-cells in
patients. Therefore, additional co-stimulation domains were added to later generation CARs (Fig.
Figure 1. T-cell activation in wild-type and genetically modified CAR T-cells. (adapted from Weinkove et al, 2019). (a) a peptide antigen from a killed foreign or cancer cell is displayed in conjunction with a Major Histocompatibility Complex (MHC) cell surface protein by an antigen-presenting cell (APC), and is recognised by the T-cell receptor (TCR) in a subpopulation of
T-cells in the body. This recognition and antigen binding activates the T-cell’s cell-killing functions. Optimal activation and expansion of the T-cell population is dependent in part on co-stimulation via the CD28 receptor, which activates a separate stimulatory signal to the T-cell via recognition of CD80 or CD86 domains on the APC. (b) an antigen on a cell surface molecule of a tumour cell is recognised by an antigen recognition domain on a CAR, which directly activates the cell-killing functions of the T-cell as well as proliferation responses via multiple co-stimulation domains in the engineered CAR.
7
EPA staff assessment report for application APP203750
September 2019
1, Fig. 2) in attempts to aid in the T-cell’s ability to kill target cells, as well as the antigen
recognition-driven expansion of the CAR T-cell population, thus amplifying the anti-cancer
response (Maus et al, 2014; Weinkove et al, 2019). Stronger T-cell expansion responses may
enable the use of lower initial numbers of CAR T-cells, which is thought to lessen cytokine release
syndrome (a known adverse effect on the CAR T-cell recipient resulting from the killing of large
numbers of cancer cells; see paragraph 36) responses in recipients of the therapy (Lai et al, 2017;
Weinkove et al, 2019), see also paragraphs 36 and 48.
15. CAR T-cell therapies have proven sufficiently effective that the FDA approved two second-
generation CAR T-cell therapies in 2017: tisagenlecleucel (Kymriah™), for advanced leukaemia in
children and adults, and axicabtagene ciloleucel (Yescarta™) for the treatment of several
relapsed and refractory large B-cell lymphomas (see application section 2.3, p. 7; NCI Staff 2017).
16. Despite showing great strides in improved patient outcomes, current CAR T-cell therapies are not
completely effective, and many patients’ cancers return, despite initial remission even with third
generation CARs. CAR T-cells are still essentially bespoke and experimental therapies, and many
aspects of their function and action are still not well understood (Maus et al, 2014; Weinkove et al,
2019). Thus, research continues into improving and broadening the effectiveness of CAR T-cells
for a wider range of cancer sufferers, and there are currently 506 CAR T-cell clinical trials around
the world that are listed as active and/or recruiting on the US National Library of Medicine’s
Figure 2. First, second, and third-generation Chimaeric Antigen Receptors (CARs). (adapted from Maus et al, 2014). scFv: antibody single-chain variable fragment, which specifically recognises and binds a cell surface protein found on cancer cells; hinge: a “linker” region between the scFv and the transmembrane domain; transmembrane domain: a segment of hydrophobic (having poor water-solubility) amino acids that span the hydrophobic cellular membrane barrier; costimulation domain(s): one or more protein-binding domains derived from other immune cell receptor proteins that transduce a signal upon scFv binding of the cancer cell. The co-stimulation domains trigger the cell-killing response of the T-cell, and with later generation CAR T-cells, stimulate CAR T-cell proliferation, and or production of various cytokines (molecules that stimulate and attract other immune cells, enhancing the immune response).
Transmembrane domain
8
EPA staff assessment report for application APP203750
September 2019
The organism: autologous WZTL-002 human CAR T-cells 17. WZTL-002 CAR T-cells (referred to hereafter as WZTL-002 cells) are autologous (ie, patient-
derived) third generation CAR T-cells genetically modified with the CAR 1928T2z. The name of
the CAR specifies its antigen recognition domain, as well as its three co-stimulation domains
[CD19+CD28+Toll-like Receptor 2+CD3ζ (zeta)]. The CAR antigen recognition domain is derived
from the short-chain variable fragment (scFv) of an antibody against the human B-cell surface
marker CD19 (see application section 2.3). This antigen recognition domain recognises and binds
to any human B-cell, and this binding activates the intracellular co-stimulation domains to activate
the CAR T-cell to directly kill the B-cell (Maus et al, 2014; Weinkove et al, 2019) .
18. Like other anti-CD19 CAR T-cell therapies, WZTL-002 cells can kill healthy B-cells in addition to
cancerous B-cells. This can result in an immune deficiency syndrome known as
hypogammaglobulinaemia. Immunoglobulin infusions are generally given to patients exhibiting
this condition, which are readily available in New Zealand (see application section 3.0, p.13).
19. As a third generation CAR, 1928T2z has three intracellular T-cell signalling domains: CD28, Toll-
like-receptor 2 (TLR2), and CD3ζ (see application). A key feature of the CAR used in WZTL-002
cells that distinguishes it from most other CAR T-cell therapies to date is the TLR2 domain, which
is currently being tested for its efficacy in patients based on its stimulation of cytokine production
and T-cell expansion (Lai et al, 2017; Weng et al, 2018).
20. An earlier CAR, called 1928zT2 (differing from 1928T2z in the relative position of the CD3ζ and
TLR2 domains in the CAR) is currently being used in CAR T-cell clinical trials in patients with
relapsed B-cell acute lymphoblastic leukaemia (B-ALL; Lai et al, 2017; Weng et al, 2018). Strong
CAR T-cell expansion and complete B-ALL remission were reported in an early study (Lai et al,
2017), and complete remission against relapsed and refractory extramedullary B-ALL (ie
leukaemia in tissues outside the bone marrow) were reported in a subsequent, still ongoing,
clinical trial (Weng et al, 2018; ClinicalTrials.gov 2019a).
Manufacture, quality assurance and quality control of WZTL-002 CAR T-cells
Safety of lentiviral vectors
21. The WZTL-002 cells are manufactured at the applicant’s PC2 containment facility under an EPA
approval (application APP203214), and a licence to manufacture vaccines and sera from Medsafe
(Medsafe 2019b), using a replication-incompetent lentiviral vector that has been verified to lack
any revertant replication-competent lentiviral particles (RCLs). It is important to note that, due to
the safety features incorporated into lentiviral vector systems, there has never been a RCL
detected in the entire history of use of lentiviral vectors spanning more than 25 years of human
clinical trial work (Morgan & Boyerinas 2016; Cornetta et al, 2018a).
22. To elaborate on this point in the context of the manufacture and potential release of WZTL-002
cells, recombination could theoretically occur between the viral vector construct and the vector
packaging constructs. Such a recombined vector could theoretically infect another person via
transmission from a WZTL-002 cell recipient (Schambach et al, 2013). Such events have been
observed in the production of retroviral (ie, non-lentiviral) vectors, such as Moloney Murine
Leukaemia Virus (MLV) vectors (Cornetta et al, 2018b). To prevent such an occurrence using
lentiviral vectors, packaging is carried out in cells that carry the necessary replication and
packaging genes on separate plasmids. This decreases the probability of the re-creation of a RCL
because multiple independent recombination events would be required for this to occur. In the
9
EPA staff assessment report for application APP203750
September 2019
case of third generation lentiviral vectors, the packaging cells carry the requisite genes on three or
four separate plasmids (Schambach et al, 2013). In the case of the packaging system used in
APP203214, three separate plasmids are used (APP203750 application form, section 3.0, p 10).
23. An additional safety feature of the lentiviral vector production system used in APP203214 is that
packaging is carried out via transient transfection of the cells used in production (APP203214
application form, section 3.0, p 10). Therefore, the plasmids used in lentiviral vector production do
not replicate, and are only functional in the cells for protein production for a short time, limiting the
amount of time in which recombination may take place (Cornetta et al, 2018a).
24. Finally, third generation lentiviral vectors have precisely edited sequences that prevent the
translation of truncated viral vector proteins in regions of the vector that have some sequence
similarity, and therefore, the potential to recombine with viral sequences (Schambach et al, 2013).
The combination of multiple and redundant safety features into third generation lentiviral vectors
has rendered them so reliable that a recombination leading to a RCL has never been detected,
even in experimental systems designed to detect such low-frequency recombination events
(Schambach et al, 2013; Cornetta et al, 2018a).
25. Lentiviral vector safety was recently further confirmed through an extensive screening study of
nearly two dozen different T-cell products, after T-cell amplification in culture (Cornetta et al,
2018a). All the screened products were unsurprisingly negative for RCLs, since screening for
RCLs after production was also negative. These findings prompted the authors to suggest that the
requirement for screening for RCVs after T-cell amplification in approved trials in the United
States should be dropped if the vector product is found to be negative for RCL (Cornetta et al,
2018a).
26. Regardless of these findings and the unblemished safety record of lentiviral vectors, the applicant
states that they intend to test patient-derived CAR T-cells for replication-competent lentivirus in
containment at an external containment facility (Diatranz Otsuka) under its current development
approval APP203214 before their release to lymphoma patients (see application,section 3.0 p.
11).
Safety testing of WZTL-002 cells and culture supernatants
27. Clinical trials in New Zealand are expected by the Ministry of Health to comply with the European
Medicines Agency’s Guideline for good clinical practice E6 (European Medicines Agency 2016),
with certain modifications to ensure compliance with the Medicines Act 1981 (Ministry of Health
2018b). Under its Medsafe Licence to Manufacture approval (Medsafe 2019b), prior to their use
as a lymphoma treatment in patients in the ENABLE trial, the WZTL-002 cells must undergo
safety testing to ensure that the cultures are sterile (ie, they contain no biological contamination
such as bacteria or mycoplasmas), as well as to ensure that there is no residual replication-
defective LV-1928T2z viral vector.
28. To ensure compliance with these conditions, the applicant is applying to release the cells or cell
supernatants (ie, the liquid portion of the culture after the bulk of the WZTL-002 cells have been
separated by centrifugation) to a medical testing laboratory, specifically Wellington SCL, located
on-site at Wellington Hospital.
29. Under the HSNO Act, transfer of genetically modified organisms in containment requires transfer
to an MPI-approved containment or transitional facility that can comply with the controls imposed
on the relevant approval. However, because they generally conduct testing on human clinical
samples only, medical testing laboratories are not generally classified or certified as transitional or
10
EPA staff assessment report for application APP203750
September 2019
containment facilities under the Biosecurity Act 1993. Therefore, in order to carry out the
appropriate testing under its Licence to Manufacture WZTL-002 cells from Medsafe, MIMR
requires a release approval of the GMOs to Wellington SCL.
30. Medsafe has extensive requirements that it imposes on recipients of Licences to Manufacture
under the New Zealand Code of Good Manufacturing Practice (Medsafe 2019a), including:
a quality agreement detailing the responsibilities of the licence holder and the testing facility,
including disposal/destruction of the samples
assessment and approval of all contractors and suppliers
the traceability of all samples sent to the testing facility
secure containment of samples, including ensuring they are not contaminated or degraded in
transport
Thus, Wellington SCL’s testing activities are regulated by Medsafe under the Malaghan Institute’s
Medsafe-issued licence to manufacture CAR T-cells.
31. Moreover, Wellington SCL is accredited under the Testing Laboratory Registration Act 1972 by
the Crown entity International Accreditation New Zealand (IANZ; https://www.ianz.govt.nz/), to the
ISO 15189:2012 Standard (Wellington SCL 2015).
32. Based on its audits of the Malaghan Institute, Medsafe recommends that Malaghan’s Licence to
Manufacture should be renewed (Medsafe, personal communication), which would include
existing contractors, such as Wellington SCL.
Intended use and benefits of WZTL-002 CAR T-cells
33. The applicant states that they intend to administer WZTL-002 cells intravenously to patients with
relapsed and refractory B-cell non-Hodgkin lymphomas (a type of B-cell cancer) who are enrolled
in a Phase 1 dose escalation clinical trial. The purpose of a dose escalation trial is to assess both
the toxicity to the patient at a particular dose of the medicine being trialled, as well as effective
dosages of the medicine on the disease being treated (Office of New Drugs in the Center for Drug
Evaluation and Research 2013).
34. Data from previous clinical trials with second generation CAR T-cell therapies such as
tisagenlecleucel (CTL019; CART19; Kymriah™; ClinicalTrials.gov 2019b), or axicabtagene
ciloleucel (Yescarta™), reveal long-term remission in more than 50% of lymphoma patients
(Neelapu et al, 2017; Schuster et al, 2017). Although the WZTL-002 clinical trial is a Phase 1 trial
laboratory studies on the selectivity and effectiveness of WZTL-002 cells in vitro show that they
selectively kill cells that have the CD19 cell surface protein, because they only kill B-cells and not
cells that do not have the CD19 protein (see application, p. 13, and Figs. 2 and 3 therein). As
such, the potential immediate benefits of WZTL-002 cells to the lymphoma patients recruited to
the applicant’s Phase 1 dose escalation clinical trial include alleviation of lymphoma symptoms
and improved quality of life, cancer remission and prolonged survival.
35. WZTL-002 cells also have the potential to benefit the wider New Zealand community with non-
Hodgkin’s lymphoma, subject to positive results from the Phase 1 clinical trial described and
regulatory approval for commercial release.
Potential adverse effects from the use of WZTL-002 CAR T-cells
36. Treatment of various B-cell cancers with CAR T-cells is well known to cause a wide variety of
toxic effects in patients (Brudno & Kochenderfer 2016). Primary among these is the patient