Review

published: 26 April 2017

doi: 10.3389/fimmu.2017.00458

Shaping of Natural Killer Cell

Antitumor Activity by Ex Vivo
Cultivation
Markus Granzin 1*, Juliane Wagner 2,3, Ulrike Köhl 4, Adelheid Cerwenka 5,6, Volker Huppert 7
and Evelyn Ullrich 2,3*
1
Clinical Research, Miltenyi Biotec Inc., Gaithersburg, MD, USA, 2 Division for Stem Cell Transplantation and Immunology,
Department for Children and Adolescents Medicine, Hospital of the Goethe University, Frankfurt, Germany, 3 LOEWE Center
for Cell and Gene Therapy, Cellular Immunology, Goethe University, Frankfurt, Germany, 4 Institute of Cellular Therapeutics,
Integrated Research and Treatment Center Transplantation, Hannover Medical School, Hannover, Germany, 5 Innate
Immunity Group, German Cancer Research Center, Heidelberg, Germany, 6 Division of Immunbiochemistry, Medical Faculty
Mannheim, Heidelberg University, Heidelberg, Germany, 7 R&D Reagents, Miltenyi Biotec GmbH,
Bergisch Gladbach, Germany

Natural killer (NK) cells are a promising tool for the use in adoptive immunotherapy, since
they efficiently recognize and kill tumor cells. In this context, ex vivo cultivation is an
attractive option to increase NK cells in numbers and to improve their antitumor potential
Edited by:
Hermann Einsele, prior to clinical applications. Consequently, various strategies to generate NK cells for
University of Würzburg, Germany adoptive immunotherapy have been developed. Here, we give an overview of different
Reviewed by: NK cell cultivation approaches and their impact on shaping the NK cell antitumor activity.
Reem Al-Daccak,
Institut national de la santé et
So far, the cytokines interleukin (IL)-2, IL-12, IL-15, IL-18, and IL-21 are used to culture
de la recherche médicale and expand NK cells. The selection of the respective cytokine combination is an import-
(INSERM), France
ant factor that directly affects NK cell maturation, proliferation, survival, distribution of
Stanislaw Stepkowski,
University of Toledo, USA NK cell subpopulations, activation, and function in terms of cytokine production and
*Correspondence: cytotoxic potential. Importantly, cytokines can upregulate the expression of certain
Markus Granzin activating receptors on NK cells, thereby increasing their responsiveness against tumor
[email protected];
Evelyn Ullrich
cells that express the corresponding ligands. Apart from using cytokines, cocultivation
[email protected] with autologous accessory non-NK cells or addition of growth-inactivated feeder cells
are approaches for NK cell cultivation with pronounced effects on NK cell activation
Specialty section:
and expansion. Furthermore, ex vivo cultivation was reported to prime NK cells for the
This article was submitted to
Alloimmunity and Transplantation, killing of tumor cells that were previously resistant to NK cell attack. In general, NK cells
a section of the journal become frequently dysfunctional in cancer patients, for instance, by downregulation of
Frontiers in Immunology
NK cell activating receptors, disabling them in their antitumor response. In such scenario,
Received: 28 January 2017
Accepted: 04 April 2017
ex vivo cultivation can be helpful to arm NK cells with enhanced antitumor properties to
Published: 26 April 2017 overcome immunosuppression. In this review, we summarize the current knowledge
Citation: on NK cell modulation by different ex vivo cultivation strategies focused on increasing
Granzin M, Wagner J, Köhl U, NK cytotoxicity for clinical application in malignant diseases. Moreover, we critically
Cerwenka A, Huppert V and Ullrich E
(2017) Shaping of Natural discuss the technical and regulatory aspects and challenges underlying NK cell based
Killer Cell Antitumor Activity by therapeutic approaches in the clinics.
Ex Vivo Cultivation.
Front. Immunol. 8:458. Keywords: natural killer cells, natural killer cell cultivation, natural killer cell expansion, natural killer cell therapy,
doi: 10.3389/fimmu.2017.00458 natural killer cell cytotoxicity, ex vivo stimulation

Frontiers in Immunology | www.frontiersin.org 1 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

INTRODUCTION cytokines such as transforming growth factor-β, interleukin

(IL)-10, or immunosuppressive enzymes, such as indoleamin
As an important part of the innate immune system, natural killer 2,3-dioxigenase, further impair antitumor NK cell responses of
(NK) cells are deployed as first line of defense against aberrant cancer patients (20–22).
cells caused by viral infections or malignancies. Human NK cells Ex vivo modulation of NK cell receptor expression is
can be identified via their morphology as large granular lym- therefore an important tool to overcome immune response
phocytes, and via their surface marker profile, as they express inhibition. A number of studies reported an upregulation of
by definition CD56, but not CD3. The NK cell compartment DNAM-1, NKG2D, and other NK cell-activating receptors
can be further divided into subpopulations. There are two main under certain culture conditions, mostly involving stimulation
NK cell subsets that can be distinguished, the CD56highCD16neg by IL-2 (23–26). In addition, other ILs such as IL-12, IL15, IL-18,
subpopulation, which has mostly immune modulatory function, or IL-21 and Type I IFNs shape the NK cell receptor expression
mainly accomplished by interferon (IFN)-γ secretion, and the profile (27–31).
CD56lowCD16pos fraction with direct cytotoxic capacity (1–3). Natural killer cells can play an important role for cellular
NK cell activation is based on a balanced system integrating immunotherapy and the adoptive transfer of NK cells represents
signals from activating and inhibitory receptors. Inhibitory sig- an attractive strategy to treat cancer patients (32, 33). In this
nals derive mainly from germ-line encoded inhibitory killer cell context, ex vivo expansion of NK cells prior to their clinical appli-
immunoglobulin-like receptors (KIRs). Ligands for inhibitory cation is not only required to increase the applicable cell doses but
KIRs, in humans major histocompatibility complex (MHC) class it is also reasonable to pre-activate and modify their antitumor
I molecules, are highly expressed by healthy cells and thereby features. For ex vivo cultivation, NK cells from different sources
prevent NK cell activation. Malignant cells often downregulate can be stimulated with different cytokines, and, to reach efficient
MHC class I molecules on their surface to evade T cell attack expansion rates, NK cells are cultured among autologous acces-
(4). However, these so-called “missing-self ” cells are recognized sory cells or together with different types of growth-inactivated
by NK cells through inhibitory receptors, and as signals from autologous or allogeneic feeder cells (Figure 1). Of note, it is
activating receptors prevail, NK cells become active and react possible to genetically engineer NK cells ex vivo to further aug-
against the encountered targets. Alternatively, NK cells can be ment their antitumor activity, for example, to integrate chimeric
activated by overexpression of stress-induced surface ligands on antigen receptors against distinct tumor antigens (34, 35). In this
infected or abnormal cells, which are recognized by activating review, we focus on the cultivation of NK cells without genetic
receptors, such as the natural cytotoxicity receptors (NCRs) modifications. Many different protocols exist for ex vivo expan-
NKp30, NKp44, and NKp46, and the so-called C-type lectin-like sion of NK cells, all with different features and capacities. Here, we
receptors, such as NKG2D (1, 5–9). In this case, activating signals give a comprehensive overview of strategies to obtain appropriate
outbalance inhibitory self-signals and lead to NK cell activation. amounts of functional NK cells. We will discuss starting material
Furthermore, NK cells become activated upon encounter of and culture systems as well as the use of cytokines, feeder cells,
antibody-coated targets by CD16, which binds to the Fc portion and other additives.
of the antibody and mediates a strong activating signal. By means
of activating and inhibitory receptors, NK cells, unlike T and
B-lymphocytes, can react immediately without prior priming or STARTING MATERIAL FOR NK CELL
antigen presentation. EXPANSION AND ROLE OF NK
Activated NK cells execute effector functions through dif- CELL PURITY
ferent mechanisms. NK cells mediate direct cytotoxicity via the
exocytosis pathway with release of cytotoxic granules, which Until recent, 92% of clinical studies used NK cells from peripheral
contain granzymes and perforin, resulting in lysis of the target blood, either donor (79% of recruiting trials) or patient derived
cell (10). In addition, NK cells induce apoptosis of target cells by (13% of recruiting trials) (36). Alternatives are the use of NK cell
expression of death receptor ligands, such as Fas ligand or tumor lines, or the differentiation of NK cells from umbilical cord blood
necrosis factor-related apoptosis-inducing ligand (TRAIL) (11). or pluripotent stem cells (37–39). NK cell lines, such as NK-92,
Production and release of IFN-γ by NK cells after activation also avoid the need for donor selection and enable the production
has multiple functional consequences, with particular relevance of large cell doses to treat patients on a flexible schedule (40).
in tumor surveillance, as IFN-γ inhibits tumor angiogenesis, has Nevertheless, NK cell lines require growth inactivation mainly
antimetastatic activity, and acts pro-apoptotic (12, 13). achieved by irradiation, possibly reducing their antitumor poten-
The ability of tumor cells to bypass the immune response is a tial due to short in vivo persistence. Differentiation of NK cells
basic prerequisite for cancer formation and progression. Within from cord blood CD34+ cells is attractive because of the “off-the-
immune editing, tumors undergo genetic, epigenetic, and pheno- shelf ” availability from a cord blood bank. Similarly, NK cells
typic changes, thereby becoming a heterogeneous cell population from pluripotent stem cells are a promising concept for the future
that is hardly visible to or assailable by immune cells due to down- but still in early development (39, 41). In this overview, we focus
regulation of tumor antigens and NCR ligands (14). Additionally, on peripheral blood-derived NK cells, currently the main source
malignant cells suppress NK cells by blocking the NKG2D recep- for NK cells for clinical use.
tor via shedding of NKG2D ligands (15–17) or upregulation of The NK cell purity, meaning the frequency of NK cells among
inhibitory MHC class I molecules (18, 19). Immunosuppressive other cells, is an important factor for the intended therapeutic

Frontiers in Immunology | www.frontiersin.org 2 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

Figure 1 | Scheme showing main components utilized for ex vivo natural killer (NK) cell activation and expansion procedures.

application. For ex vivo expansion, NK cells are often cultured THE ROLE OF IL-2
within a mixture of cells, such as PBMC, thereby avoiding further
purification. Whereas the cultivation of NK cells among other Interleukin-2 plays an important role in activation of NK cells
accessory cells is a practical strategy for autologous therapeutic via binding to the IL-2 receptor (IL-2R), a heterotrimeric protein
settings, it may be critical for allogeneic applications, since non- expressed on NK cells and other immune cells. This led to the
NK cells may induce unwanted side effects. Alloreactive T cells interest in both (i) using IL-2 for stimulation of autologous
are a major risk factor for the patient, as they mediate “graft- NK cells in cancer patients and (ii) ex vivo activation and expan-
versus-host disease” (GvHD), a severe complication following sion of allogeneic donor NK cells for adaptive immunotherapy.
allogeneic hematopoietic stem cell transplantation (HSCT) At the beginning of the 1980s, researchers around Rosenberg
(42). Furthermore, donor-derived B cells can lead to B cell lym- and colleagues showed that IL-2 exposed lymphokine-activated
phoproliferative disorder after reactivation of an Epstein–Barr killer (LAK) cells were able to attack autologous fresh tumor cells
virus (EBV) infection (43, 44), and they can cause the passenger and that this effect could mainly be ascribed to NK cells (71, 72).
lymphocyte syndrome (45), both critical side effects for the Nevertheless, in first clinical trials using adoptive transfer of LAK
patient. Therefore, purification of NK cells might be required cells and IL-2 therapy, the clinical response did not exceed the
and is realized so far in most clinical settings by magnetic cell efficacy of IL-2 monotherapy (73).
separation, for instance, by depletion of CD3-expressing cells and Importantly, during the last 20 years, it has been elaborated
subsequent enrichment for CD56-expressing cells (26, 46–49). that NK cells play a major role in the regulation of the balance
In addition, a first proof of concept is shown for good manu- between GvL and GvHD after allogeneic HSCT, especially haploi-
facturing practice (GMP)-compliant fluorescence-activated cell dentical HSCT (33, 74–78), demonstrating improved anticancer
sorting to purify for NK cell subsets, such as NK cells expressing activity while avoiding GvHD. In order to make haploidentical
a single KIR (50). NK cells available for clinical use, large-scale GMP-conform
manufacturing protocols were established. After starting with
a leukapheresis product that was depleted for CD3+ cells and
CYTOKINE-INDUCED NK CELL enriched for CD56+ cells, cultivation in medium containing IL-2
EXPANSION (1,000 U/mL) for up to 2 weeks yielded 0.1–3 × 109 CD56+CD3−
NK cells (Table 1), sometimes sufficient for multiple infusions to
Aims of adoptive transfer of ex vivo expanded NK cells are the patients with hematological malignancies (26). Median NK cell
enhancement of natural cytotoxicity and homing to tumor sites expansion was fivefold and median NK cell purity was >94
under maintenance of “self ” protection. Studies performed with with <0.1% T cell contamination (26). Ex vivo stimulation with
cytokine-stimulated NK cells or PBMC have shown the safety IL-2 induced elevated cytokine secretion by NK cells, enhanced
of this approach and indicated some clinical responses upon intracellular STAT3/AKT signaling, and upregulation of various
adoptive NK cell transfer following HSCT. In the next paragraph, NCRs and NKG2D receptors (52). Depletion of CD3 cells from
we summarize ex vivo NK cell expansion protocols starting with leukapheresis products without subsequent CD56 enrichment
purified NK cells (Table 1) or PBMC (Table 2). Concepts admin- and short-term activation with IL-2 overnight led to a final prod-
istering cytokines in the presence of growth-inactivated feeder uct containing 40% NK cells (Table 2). In all cases, IL-2-activated
cells will be discussed in later sections of this article. NK cells demonstrated a much higher cytotoxic activity against

Frontiers in Immunology | www.frontiersin.org 3 April 2017 | Volume 8 | Article 458

 Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

K562 target cells compared to unstimulated NK cells (26, 43, 52,

(26, 52, 53)

Reference

(56–59)
53, 60). In addition, after cryopreservation and thawing, NK cells
(51)

(54)

(55)
showed a moderate to high viability when activated with IL-2,
whereas the viability of unstimulated NK cells was low (26).
Setting

Transfer into the clinic in 2004 and 2005 with first patient stud-

Clinical
Clinical
In vitro

In vitro

In vitro

In vitro
ies using those IL-2-activated donor NK cells were performed in
parallel in Europe and the US, for both, haploidentical HSCT (53),

stimulation that is preserved during cell division

and in the non-transplant setting (43). In the latter one, Miller

Memory: increased IFN-γ production upon

and coworkers used IL-2 expanded haploidentical NK to treat

Responsive to picomolar concentrations

rhabdomyosarcoma cell lines via NCRs
43 patients with advanced cancer (43), with 19 of them suffering
Enhanced cytolysis of lymphoma and

Enhanced cytotoxicity against K562

from acute myeloid leukemia, followed by studies in patient with
Cytolysis of leukemia cell lines and

Improved cytotoxic activity against

ovarian and breast cancer and B-cell non-Hodgkin lymphoma

primary acute leukemic blasts

(60, 79). Importantly, the authors reported in vivo persistence

and even expansion of the alloreactive donor NK cells in patients
leukemia and tumors

pretreated with high dose preparative regimen, consisting of

NK cell function

5 days of 60 mg/kg intravenous cyclophosphamide and 25 mg/m2

intravenous fludarabine (43). Of note, successful NK cell engraft-
ment was dependent on the patients’ pretreatment regimen,
of IL-2 which was also responsible for the patients’ elevated IL-15 plasma
concentrations (43). In addition, it was demonstrated that in vivo
Upregulated: CD94, NKG2A, NKp30, NKp44,
increased CD56+CD16− frequency; increased

persistence of donor NK cells at day 7 after infusion and success-

p-STAT3 and p-AKT; increased lytic activity,
upregulation of CD69, NKG2D, and natural

ful in vivo expansion (more than 100 donor-derived NK cells per

Upregulated: CD69, NKp30, and NKp44

cytotoxicity receptors (NCRs); increasing

microliter of patient blood 14 days after transfer) correlated with

Upregulated: CD69, CD25; activation
Enhanced killing via NCRs, DNAM-1
amount of NK cells without killer cell

NKG2D, NKp46, CD69, and CD25

leukemia clearance (60). Expansion of host regulatory T cells was

associated with low numbers of NK cells (60). In parallel, Koehl
immunoglobulin-likereceptors

et al. reported on three pediatric patients with multiply relapsed

Downregulated: NKp80

leukemia (still in blast persistence at HSCT) treated with repeated

NK cell phenotype

transfusions of IL-2-activated donor NK cells post-haploidentical

HSCT (53), which led to complete remission remaining for sev-
and NKG2D

eral weeks up to some months. In the following clinical study,

of STAT3

they also demonstrated a small clinical benefit in patients with

various malignancies receiving IL-2-activated compared to
patients receiving resting NK cells only (80). Interestingly, IL-2-
NK cell purity

≤0.1% T cells

<0.1% T cells
~92–95% NK
75–100% NK

stimulated NK cells but not unstimulated NK cells promoted

94–99% NK

≥90% NK
Table 1 | Ex vivo cultivation of pure natural killer (NK) cells with cytokines only.

NK cell trafficking and changes in the distribution of leukocyte

subpopulations in the peripheral blood. In the meanwhile, safety
N/A

and feasibility using IL-2-activated and -expanded NK cells for

adaptive immunotherapy has been demonstrated in various
NK cell expansion

only; strongly with

4–5 (12–14 days)

clinical studies as summarized in a recent review by Koehl and

N/A (2–5 weeks)

None with IL-21

N/A (12–16 h)

others (33).
IL-2 + IL-21
>1 (5 days)
rate

IMPACT OF IL-15 ON NK CELL

CD3/CD19 depleted in flasks

EXPANSION
CD56 enriched in bags and
Starting material/culture

PBMC, CD3 depleted, and

PBMC, CD3 depleted and

PBMC, CD3 depleted and

Isolated CD3−CD56+ cells
CD56 enriched or PBMC,

Carson et al. postulated that NK cells might be dependent on

other cytokines than IL-2 such as IL-15 (81, 82). The trimeric
PBMC, sorted for

IL-12 + IL-15 + IL-18 PBMC, sorted for

CD3−CD56+ cells

CD56 enriched

IL-15 receptor on NK cells shares two subunits with the IL-2R,

CD3−CD56+

but not CD25 forming the high affinity IL-2R. Therefore, they
system

also share some functions, e.g., maintenance of NK cell survival

flasks

(82). Similarities and differences between IL-2 and IL-15 effects

on NK cells have been extensively reviewed elsewhere, and IL-15
might be the preferable cytokine for cancer therapy as it inhibits
activation-induced cell death and it is considered safe (83–85). In
IL-2 + IL-15

IL-2 + IL-21

addition, compared to IL-2, IL-15 leads to more sustained antitu-

Protocol
features

mor capacity of NK cells via signaling through mammalian target

IL-15
IL-2

of rapamycin and stress-activated gene expression (86). However,

Frontiers in Immunology | www.frontiersin.org 4 April 2017 | Volume 8 | Article 458

Frontiers in Immunology | www.frontiersin.org

Granzin et al.
Table 2 | Ex vivo cultivation of natural killer (NK) cells with accessory cells.

Protocol features Starting material/ NK cell expansion NK cell purity NK cell phenotype NK cell function Setting Reference
culture system rate

IL-2 PBMC, CD3 depleted N/A (overnight) 33% NK 0.1% T cells N/A N/A Clinical (60)
in bags and flasks N/A (14–16 h) 26.7% Enhanced cytotoxicity In vitro (61, 62)
N/A (overnight) 40% NK 0.9% T cells Enhanced cytotoxicity In vitro (43)
IL-15 PBMC, CD56 enriched 23 (20 days) 98% NK Expression of NKp30, NKp44, NKp46, Cytotoxic in vitro Clinical (63)
NKG2D, and 2B4
IL-15 PBMC, CD3 depleted 3.7 CD56+/CD122+ >90% CD56+/CD122+ 67% CD56+CD16+ Cytotoxic against K562 and patient bone Clinical (64)
+ IL-21 (2–3 weeks) <0.3% CD3+/CD56− marrow blasts
<3% CD3+/CD56+
OKT-3 PBMC in plates 193 (21 days) ~55% NK N/A Substantial cytotoxicity against K562 In vitro (65)
+ IL-2 ~22% T cells
PBMC in flasks 1,625 (20 days) ~65% NK Upregulated: 2B4, CD8, CD16, CD27, Increased cytotoxicity against tumor cell In vitro (25)
~22% T cells CD226, NKG2C, NKG2D, NKp30, NKp44, lines and primary MM cells In vitro
NKp46, LIR-1, KIR2DL3, and CXCR3
5

Downregulated: CCR7
PBMC 1,036 (total cells) ~30% NK Upregulated: NKG2A, LILR-B1, NKG2D, In vitro cytotoxicity increases during culture Clinical (66)
(19 days) ~40% T cells NKp30, NKp44, and NKp46
PBMC in a bioreactor, 77—bioreactor 38%—bioreactor Bioreactor compared to flasks: higher Bioreactor compared to flasks: higher In vitro (67)
flasks, and plates 530—bags 31%—bags expression of CD11b, NKG2D, and NKp44 cytotoxicity
770—flasks (20 days) 44%—flasks
OKT-3 PBMC in plates, flasks, 646 (14 days) 60% NK Upregulated: 2B4, NKG2D, NKp30, NKp44, Increased cytotoxicity Clinical (68)
+ IL-2 + Alemtuzumab and bags 1,537 (18 days) 37% T cells KIR2DL1, LIR-1, and CD16 In vitro and in vivo
<0.1% B cells Downregulated: CCR7

Antitumor Activity of Ex Vivo-Cultivated NK Cells

OKT-3 PBMC or PBMC: 112 With PBMC: 34% Upregulated: NKp30, NKp44, DNAM-1, Increased activity against neuroblastoma Preclinical (69)
+ IL-2 + IL-15 CD56+ + CD56− (1:1) 1:1 Mix: 89 (21 days) With “1:1 Mix”: 92% NKG2D, and CD11a cell lines in vitro and in vivo model
in flasks and bioreactor
(Cellbag)
April 2017 | Volume 8 | Article 458

aCD16 mAb PBMC in flasks and 637–5,712 (day 21) 79% NK 8.4% T cells Upregulated: NKG2D, NKp44, and CD69 Increased cytotoxicity against tumor cell In vitro (70)
+ OK432 + IL-2 bags (day 21) Downregulated: CD16 (transient) lines and primary cancer cells in vitro
ADCC activity
Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

recent data revealed that continuous IL-15 signaling causes production of IFN-γ by NK cells when added to PBMC (99, 100).
functional exhaustion of NK cells by decreased fatty acid oxidation, IL-12 is produced by DCs, macrophages, and B cells, and its
resulting in lower cytotoxicity in vitro and decreased tumor control receptor consists of two subunits (α and β), which mediate
in vivo (87). Thus, optimal dosing and timing of IL-15 is critical for signaling through members of the JAK–STAT family (101).
ex vivo NK cell activation. Purified NK cells expanded using IL-15 IL-2 enhances the response of NK cells to IL-12 by increasing
exhibit upregulation of NCRs and CD69 and cytolysis of leukemia the expression of the IL-12 receptor and STAT4, a relevant fac-
and primary ALL blasts (51). Enhanced cytotoxicity of IL-15- tor for IL-12 signaling (102). Furthermore, it was revealed that
stimulated NK cells against leukemia and rhabdomyosarcoma cell IL-12-mediated IFN-γ production of NK cells requires priming
lines could be attributed to NCRs, DNAM-1 and NKG2D (54). with IL-18, a cytokine also known to enhance IL-15-induced
Using IL-15 to expand NK cells from CD56-enriched PBMC for NK cell proliferation (103, 104). Due to the synergistic effects, it
20 days resulted in a 23-fold expansion of CD3−CD56+ NK cells with seems reasonable to combine the different cytokines for ex vivo
a final purity of about 98% (63). NK cells generated with the latter stimulation of NK cells. In this context, the combination of IL-12,
protocol were transferred to 15 non-small lung cancer patients in a IL-15, and IL-18 raised special interest, as it leads to the so-called
phase I clinical trial in two to four doses of 0.2–29 × 106 NK cells/ “cytokine-induced memory-like” (CIML) NK cells in mice and
kg, showing the safety of the approach (63). humans, which exhibit an increased capacity to produce IFN-γ
upon re-stimulation at later time points (56, 105). Importantly,
this memory response is a cell intrinsic effect that is passed on
IL-21 ENHANCES NK CELL EFFECTOR to offspring cells and is maintained up to several months (56).
FUNCTIONS In mice, the intrinsic ability for mediated IFN-γ production
coincided with demethylation of the conserved non-coding
The cytokine IL-21, in combination with IL-2 or IL-15, is utilized
sequence 1 in the IFN-γ locus (106). Furthermore, adoptive
in some protocols for NK cell stimulation (55, 64). IL-21 belongs
transfer of CIML NK cells had a clear antitumor activity against
to the IL-2 family and signals through a heterodimer consisting
established melanoma or lymphoma in vivo, which required
of the common γ-chain and the IL-21 receptor α-chain. Activated
IL-2 from CD4+ T cells (57, 106). For both, murine and human
CD4+ T cells are the main producers of IL-21 and IL-21 affects
NK cells, IL-12, IL-15, and IL-18 together induce an increased
many different cell types expressing the IL-21 receptor, including
expression of CD25, making CIML NK cells responsive to low
NK cells (88). IL-21 plays a role in the development of NK cells
concentrations of IL-2 in vitro and in vivo (57, 58). Thus, there is
from bone marrow progenitors (89), and, in mice, it dampens
a clear rationale to apply adoptive transfer of ex vivo-generated
the expansion of NK cells but is required for functional NK cell
CIML NK cells together with IL-2 injections as a combination
maturation (90, 91). Recently, expansion of “memory-like”
therapy. Recently, CIML NK cells together with low dose IL-2
NK cells has been shown to be IL-21 dependent in the context of
therapy were evaluated in a first-in-human phase I clinical trial
tuberculosis infection (92). Wendt et al. observed increased pro-
with promising results, as clinical response was observed in five
liferation of CD56bright human NK cells (55), but another group
of nine treated patients (107).
reported no effect of IL-21 on the proliferation of NK cells from
healthy human donors and from HIV patients (93). Moreover,
IL-21 is known to trigger apoptosis, resulting in a shorter lifespan AUTOLOGOUS ACCESSORY CELLS AND
of NK cells in vitro (90, 94). Thus, the time span NK cells are
exposed to IL-21 appears critical (95, 96). Besides its effect on
AUTOLOGOUS FEEDER CELLS FOR NK
NK cell proliferation, IL-21 enhances the effector functions of CELL EXPANSION
NK cells, including secretory and cytotoxic functions as well as
Although cytokines efficiently activate NK cells and result in cell
enhanced ADCC responses (93, 97, 98). Culturing CD3-depleted
products with advanced effector functions, cytokines alone do
PBMC for 13–20 days with IL-21 and IL-15 without additional
not allow pronounced ex vivo expansion (Table 1). Consequently,
feeder cells yields activated NK cells with a purity of >90%, which
in addition to the activation with cytokines, stimuli from autolo-
were applied in a clinical trial with 41 leukemia patients receiving
gous accessory cells can be used to further enhance the expansion
infusions of donor-derived NK cells 2–3 weeks after HSCT (64).
of NK cells to overcome the hurdle of limited NK cell doses for
Although the NK cells expanded weakly under this condition
adoptive NK cell therapy (Table 2). Outgrowth of NK cells from
(3.7-fold), they possessed potent cytotoxic activity against pri-
the whole PBMC fraction is more effective than cultivation of
mary bone marrow blasts prior to transplantation, and infusions
pure NK cells, because other cell types provide additional factors
with a median dose of 2 × 108 NK cells/kg were well tolerated and
for NK cell proliferation. CD14+ cells, for instance, enhance the
correlated with a reduction in leukemia progression compared to
ex vivo NK cell proliferation via direct cell contact and soluble
historical controls (64).
factors (108, 109). After activation, for instance by concanavalin
A, T cells also trigger NK cell proliferation (110).
IL-12/15/18 INDUCED MEMORY Stimulation of PBMC with IL-2 and the clinically approved
NK CELLS anti-CD3 antibody OKT-3 leads to a profound outgrowth of
NK cells (25, 65–67, 111), probably by activation of T cells and
Interleukin-12 was originally discovered as NK cell-stimulating this is utilized by several clinical protocols for NK cell cultivation.
factor, inducing proliferation, enhanced cytotoxicity, and Nevertheless, starting the culture from PBMC goes along with

Frontiers in Immunology | www.frontiersin.org 6 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

extensive coexpansion of unwanted CD3+/CD56− T cells and (115). Furthermore, whereas the availability of autologous feeder
CD3+/CD56+ NK-like T (NKT) cells, accounting for the majority cells is limited, as they have to be obtained directly from the
of cells in the final cellular product. Surprisingly, infusion of this patient, for allogeneic feeder cells it is possible to utilize estab-
heterogeneous cell product without removal of potentially allore- lished cell lines. Cell lines can be grown easily to sufficient num-
active T cells did not cause side effects, such as GvHD, in a safety bers and different cell lines in fact trigger NK cell proliferation,
trial with five cancer patients, evaluating the cultivated cellular such as HFWT, K562, RPMI 1866, Daudi, KL-1, MM-170, and
product in an allogeneic setting (66). This can be explained by different EBV-transformed lymphoblastoid cell lines (EBV-LCL)
the fact that T cells may lose their alloreactivity during extended (99, 119–122).
ex vivo expansion (112). Thus, low NK cell purities may be less Culturing PBMC together with the Wilms tumor cell line
critical for long-term cultivated cellular products compared to HFWT and IL-2 leads to significant NK cell expansion (124, 145),
NK cells directly obtained from a donor, but more clinical data and interestingly under this condition NK cells not only arise
are required to prove this hypothesis. Of note, the approach also from mature CD3−CD56+ NK cells but also from CD3−CD14−C
allows efficient expansion of functional patient-derived NK cells, D19−CD56− NK cell precursors expressing CD122 (146). In 2004,
as shown for B cell chronic lymphocytic leukemia and multiple early clinical data showed that adoptive transfer of autologous
myeloma patients, enabling therapy with autologous NK cells NK cells generated by coculture with irradiated HFWT is safe and
and further circumventing possible safety risks of therapy with patients with recurrent malignant glioma partially responded to
donor-derived cells (25, 111). the treatment (125).
Starting with PBMC enriched for CD56 cells together with the Another advantage of cell lines is that it is relatively easy to
corresponding non-CD56 PBMC in a 1:1 mixture favors a 89-fold genetically modify them and to integrate additional factors for
NK cell expansion with a final product consisting of 92% NK cells NK cell stimulation. In recent years, modified K562 cells have
after 21 days (69). Alternatively, adding irradiated autologous been utilized, such as K562 expressing membrane-bound IL-15
PBMC to the culture is a strategy to benefit from these “feeder and 41BBL (K562-mb15-41BBL) (126). While unmodified K562
cells” for NK cell activation and expansion but to avoid their only induce a weak NK cell proliferation (2.5-fold NK cell expan-
coexpansion (Table 3). Of note, to make a clear difference, we use sion in 1 week), with K562-mb15-41BBL the NK cell number can
the term “feeder cells” for all inactivated cells that are added to the be significantly increased by 20- or 1,000-fold in 1 or 3 weeks
culture, whereas cocultured non-NK cells that are not inactivated (126). In addition, stimulation of NK cells with K562-mb15-
are defined as “accessory cells.” Besides its growth inactivating 41BBL demonstrated that NK cells actually have a substantial
function, irradiation can induce upregulation of stress-regulated proliferative potential ex vivo, with up to 30 population doublings
surface molecules on PBMC, such as ULBP1–3, that further trig- and 5.9 × 104-fold NK cell expansion (147). NK cells expanded
ger NK cell activation, e.g., through NKG2D (113). Still, irradiated with K562-mb15-41BBL exhibit enhanced natural cytotoxicity
autologous PBMC induce only weak NK cell proliferation without against several allogeneic and autologous tumors in vitro, effi-
additional activation of the feeder cells (e.g., only 16-fold expan- ciently mediate ADCC and showed antitumor efficacy in mouse
sion within 2 weeks) (24). Whereas irradiated autologous PBMC xenograft models for the treatment of sarcoma and myeloma
previously activated with IL-2, OKT-3 and RetroNectin allow a (128, 148, 149). Of note, in a clinical trial assessing adoptive
median 4,720-fold NK cell expansion after 3 weeks with a NK cell transfer of K562-mb15-41BBL following HSCT, acute GvHD
purity of 91% starting from PBMC (114). To obtain a more pure occurred in five of nine patients, although the donors were
final product with 98% NK cells, it is possible to start the culture completely HLA matched and the doses of injected NK cells
with already CD3-depleted PBMC and add irradiated autologous and cotransferred T cells were low (1–10 × 105 and ≤2 × 104/kg)
PBMC as feeder cells together with IL-2 and OKT-3 (23). The (131). These observations suggested that the acute GvHD was
highest purity can be achieved by cell sorting, representing also T cell mediated, but NK cells apparently may promote this severe
the method of choice to expand defined NK cell subpopulations. side effect indirectly (150). Importantly, another group utilized
As demonstrated by Siegler et al., GMP-sorted and highly pure NK cells expanded with a similar K562 variant expressing 41BBL
single KIR+ NK cells can be expanded 160- to 390-fold in 19 days and IL-15 in another treatment setting and did not observe GvHD,
with IL-2, IL-15, OKT-3, and irradiated autologous PBMC (50). although up to 1 × 108 NK cells/kg were administered (129).
Furthermore, Denman and colleagues revealed that K562
NK CELL EXPANSION WITH ALLOGENEIC expressing 41BBL and membrane-bound IL-21 instead of IL-15
FEEDER CELLS are even more effective for ex vivo expansion of NK cells, and
weekly restimulation with this cell line supports a sustained
Using irradiated allogeneic cells as feeder cells is another option NK cell proliferation over several weeks (134). In coculture with
to stimulate NK cell expansion ex vivo (118) (Table 4). Compared K562 expressing membrane-bound IL-21 and 41BBL, NK cells
to autologous PBMC, allogeneic PBMC may be even more effi- show an increased telomere length and enhanced activation
cient as feeder cells for NK stimulation. Accordingly, in a study of the STAT-3 signaling pathway, explaining the positive effect
testing the expansion of NK cells from patients with advanced for sustained expansion of NK cells over long time (134, 151).
lymphomas or terminal solid tumors, 300-fold NK expansion Adoptive transfer of NK cells expanded with K562 expressing
was obtained with irradiated allogeneic PBMC feeder cells from membrane-bound IL-21 and 41BBL into tumor-bearing mice
healthy donors, whereas only 169-fold expansion was achieved improved the survival of the animals, indicating a therapeutic
with irradiated autologous PBMC feeder cells from the patients effect of these NK cells (135).

Frontiers in Immunology | www.frontiersin.org 7 April 2017 | Volume 8 | Article 458

Frontiers in Immunology | www.frontiersin.org

Granzin et al.
Table 3 | Ex vivo cultivation of natural killer (NK) cells with autologous feeder cells.

Protocol features Starting material/culture NK cell NK cell NK cell phenotype NK cell function Setting Reference
system expansion purity
rate

Irr. autologous PBMC, CD3 depleted, and CD56 16 (14 days) 97% NK Upregulated: NKG2D, DNAM-1, Efficient degranulation and lysis of K562 In vitro (24)
PBMC (depleted enriched in flasks 0.2% T cells NKp30, NKp44, CD158a, and
In vitro
for CD3−/CD56+ CD158e
cells) + IL-2 + IL-15
Irr. autologous PBMC PBMC in flasks and bags 4,720 91% NK Strong expression of NKG2D and Elevated cytotoxicity that is maintained for up to 4 weeks Clinical (114)
activated with OK432, (21–22 days) ~12% NK-like CD16 after infusion to patients
FN-CH296 and T and T
OKT-3 + IL-2
Irr. autologous PBMC, CD3 depleted, and CD56 169 (14 days) 84% NK Upregulated: CD16, CD56, NKG2D, Increased cytotoxicity against tumor cell lines in vitro In vitro (115)
PBMC + OKT-3 + IL-2 enriched in plates NKp30, and NKp44
PBMC, CD3 depleted in flasks 278–1,097 91–98% NK Most cells express NKG2D, CD16, Efficient lysis of tumor cell lines in vitro; persistence in Clinical (116)
and bags (21–26 days) CD94, NKp46, KIR2DL1, KIR3DL1, patients up to several months; cytotoxic potential is lost
and KIR2DL2/3 in vivo, while ability for ADCC is maintained
8

PBMC, CD3 depleted in bags 691 (14 days) 98% NK Upregulated: NKG2C, NKp30, Increased cytotoxicity against tumor cell lines in vitro; Preclinical (23)
0.06% T cells NK44, CXCR4, CD25, CD62L, and antitumor effect and ADCC activity in a leukemia xenograft model
CD69 mouse model; up to 4 days persistence in patients
758 (14 days) 98% NK Clinical (117)
0.4% T cells
Irr. autologous PBMC, CD3 depleted, and CD56 546 (14 days) 94.9% NK Upregulated: NKG2D, NKp30, Increased cytotoxicity against tumor cell lines in vitro In vitro (113)
PBMC (depleted enriched in plates and flasks 2.2% T cells NKp44, tumor necrosis factor-
for CD3−/CD56+ related apoptosis-inducing ligand,
cells) + OKT-3 + IL-2 and DNAM-1
Downregulated: NKp80
Irr. autologous PBMC, CD3 depleted, and CD56 117/63 in Bags: Upregulated: NKG2D, NKp44 High cytotoxicity against K562 and high productivity of In vitro (50)

Antitumor Activity of Ex Vivo-Cultivated NK Cells

PBMC + OKT-3 enriched in plates and bags bags (±IL-15) 45% NK IFN-γ
+ IL-2 ± IL-15 993 in plates 0.6% T cells
(19 days)
Good manufacturing practice 160–390 ~100% NK Single KIR + NK cells Anti-leukemic activity against primary acute myeloid Preclinical (50)
April 2017 | Volume 8 | Article 458

killer cell immunoglobulin-like >0.01% leukemia cells in vitro and in vivo model
receptor (KIR) sorted NK cells T cells
in bags
 Table 4 | Ex vivo cultivation of natural killer (NK) cells with allogeneic feeder cells.
Frontiers in Immunology | www.frontiersin.org

Granzin et al.
Protocol features Starting material/culture NK cell NK cell NK cell phenotype NK cell function Setting Reference
system expansion purity
rate

Irr. allogeneic PBMC activated In vivo IL-2 primed PBMC 1–148 (14 days) 64–98% NK N/A Cytotoxic activity against leukemic Clinical (123)
with ConA + IL-2 depleted for non-NK cells cell lines
in flasks
Irr. allogeneic PBMC activated PBMC, depleted for CD3, 80–200 91% CD56 Upregulated: CD16, CD25 Increased cytotoxicity against tumor In vitro (118)
with ConA, PHA and CD4, CD19, and CD33 (15 days) 0.3% CD3 cell lines in vitro; decreased frequency
ionomycin + IL-2 + IL-15 in bags (day 12) of INF-g producing cells

Irr. allogeneic PBMC, CD3 depleted, and 300 (14 days) 94% NK Upregulated: CD16, CD56, NKG2D, Increased cytotoxicity against tumor cell In vitro (115)
PBMC + OKT-3 + IL-2 CD56 enriched in plates NKp30, and NKp44 lines in vitro
Irr. HFWT + IL-2 PBMC in flasks 113 (2 weeks) 86% CD56+/ N/A Cytotoxic against tumor cell lines in vitro Clinical (124, 125)
CD16+
Irr. Jurkat/KL-1 + IL-2 PBMC in flasks ~130 (2 weeks) 40–90% NK Upregulated: CD54, CD11a, CD48, CD2, Increased cytotoxicity against tumor cell Preclinical (121)
CD49d, CD58, NKp30, NKp44, 2B4, lines in vitro and antitumor activity in vivo model
DNAM-1, NKG2D, CD25, and CD69
Downregulated: CD16
Irr. K562 expressing PBMC in plates 1,089 (3 weeks) “Virtually N/A N/A In vitro (126)
membrane-bound IL-15 and pure”
41BBL + IL-2
PBMC in bags 23, 152, and 96.8% NK Marked differences of gene expression Increased cytotoxicity against tumor cell Preclinical (127)
277 after 7, 14, 3.1% T cells profile compared to unstimulated or IL-2- lines in vitro and antitumor activity in vivo model
9

and 21 days (day 21) stimulated NK cells

PBMC 447 (days 88% NK Upregulated genes for cytolytic activity, Increased cytotoxicity against primary MM cells Preclinical (128)
10–14) 2.2% T cells cytokines, chemokines, activating in vitro and in vivo; high productivity of IFN-γ model
(day 14) receptors, adhesion molecules, cell cycle
regulators, and multiple pathways
PBMC in G-Rex, bags 442—G-Rex 70% NK Upregulated: NKp30, NKp44, NKG2D, Increased cytotoxicity and ADCC against primary Clinical (129, 130)
227—bags 5–35% CD26, CD70, and CXCR3 tumor cells in vitro; robust in vivo proliferation
(10 days) T cells Downregulated: CD16, CD62L post-infusion

Irr. K562 expressing PBMC, CD3 depleted, and 1,000 (21 days) N/A Upregulated: CD56, NKG2D, tumor Increased cytotoxicity in vitro independent Clinical (131, 132)
membrane-bound IL-15 and CD56 enriched necrosis factor-related apoptosis-inducing of killer cell immunoglobulin-like receptor mismatch;

Antitumor Activity of Ex Vivo-Cultivated NK Cells

41BBL + IL-15 ligand (TRAIL), CD158a, CD158b, and NK infusion contributed to acute graft-versus-
CD158e1 host disease in first clinical trial
Plasma membrane particles PBMC in plates and flasks 1,265 (17 days) 86% Upregulated: NKp30, NKp44, NKp46, Increased cytotoxicity against leukemic cell lines In vitro (133)
of K562 expressing IL-15 and NK cells NKG2D, 2B4, NKG2A, TRAIL, and Fas and primary acute myeloid leukemia (AML)
April 2017 | Volume 8 | Article 458

41BBL + IL-2 9% T cells ligand (FasL) cells in vitro

2% NK- Downregulated: CD16
like T
Irr. K562 expressing PBMC in flasks 4.8 × 104 21.7% High expression of natural cytotoxicity Cytotoxic against tumor cell lines in vitro; In vitro (134)
membrane-bound IL-21, (21 days) T cells receptors, CD16, and NKG2D capable of ADCC; increased telomere length
41BBL, CD64, CD86, and
2,363 83% NK Upregulated: DNAM-1, NKG2D, CD16, Cytotoxic and capable of ADCC against Preclinical (135)
CD19 + IL-2
(14 days) 9.1% T cells and CD56 neuroblastoma cell lines in vitro and in vivo model

(Continued)
Frontiers in Immunology | www.frontiersin.org

Granzin et al.
TABLE 4 | Continued

Protocol features Starting material/culture NK cell NK cell NK cell phenotype NK cell function Setting Reference
system expansion purity
rate

Plasma membrane particles PBMC 825 (14 days) >90% NK N/A Increased cytotoxicity against leukemic Preclinical (136)
of K562 expressing >105 (28 days) (day 14) cell lines and primary AML cells in vitro; model
membrane-bound IL-21 and enhanced proliferation in vivo
41BBL + IL-2
Irr. allogeneic PBMC; irr. EBV PBMC depleted for CD3 ~43 90% NK N/A Increased cytotoxicity against tumor Clinical (137, 138)
transformed lymphoblastoid and monocytes in bags (31–21 days) <5% T cells cell lines in vitro
cell lines (EBV-LCL) (LAZ 388 and plates
cells) + PHA + IL-2
Irr. EBV-LCL (TM-LCL) + IL-2
10

PBMC, CD3 depleted, and 800–1,000 98% NK Upregulated: TRAIL, FasL, NKG2D, Increased cytotoxicity against tumor In vitro (139, 140)
CD56 enriched in bags (2 weeks) NKp30, NKp44, NKp46, CD48, CD25, cell lines in vitro
LTB, MX1, and BAX
Irr. EBV-LCL (SMI-LCL) + IL-2 PBMC, CD3 depleted, and 3,637 99.7% NK Clinical (141)
CD56 enriched in bags (24–27 days)
PBMC, CD3 depleted, 850 (14 days) >99% NK Upregulated: TRAIL, FasL, NKG2D, Increased cytotoxicity and ADCC In vitro (142)
and CD56 enriched in NKp30, NKp44, and DNAM-1 against tumor cell lines in vitro
CliniMACS Prodigy
Irr. EBV-LCL PBMC depleted for non- 2,900 (14 days) >99% NK Upregulated: TRAIL, NKG2D, and Cytotoxic against tumor cell lines in vitro Preclinical (96)
(SMI-LCL) + IL-2 + IL-21 NK cells (research kit) in 2.7 × 1011 DNAM-1 and in vivo; enhanced and model
plates and flasks (46 days) sustained production of IFN-γ and TNF-α
Lysate of CTV-1 PBMC, CD3 depleted, and N/A (overnight) 97–98% NK Upregulated: CD69 Cytotoxic against NK-resistant leukemia Clinical (143, 144)

Antitumor Activity of Ex Vivo-Cultivated NK Cells

CD56 enriched Downregulated: CD16 cell lines and primary tumors in vitro
April 2017 | Volume 8 | Article 458
Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

The stimulatory effect of EBV-LCL on NK cell proliferation NK cell culture, ranging from simple cell culture plates for small
was discovered more than 30 years ago (152). In 1994, an early scale experiments to highly standardized and automated systems
clinical trial already evaluated the adoptive transfer of autologous for clinical scale. The selection of the adequate culture system is
NK cells expanded with the LAZ 388 cell line to treat 10 patients based on the intended application of the cells. Most preclinical
with metastatic renal cell adenocarcinoma (137). More recently, experimental studies grow NK cells in cell culture plates or tissue
the cell lines TM-LCL and SMI-LCL were reported for NK cell culture (T) flasks. These are commonly used and very convenient
expansion, allowing around 800-fold expansion of highly pure to test and compare different culture additives in parallel, e.g.,
NK cells within 2 weeks (139–142). NK cells generated with these different cytokine concentrations. However, for clinical applica-
EBV-LCL feeder cells are currently applied in a study testing them tions in large scale, cultivation in plates and flasks is rather inap-
for adoptive transfer in an autologous setting with intended doses propriate for different reasons. First, due to the small volume of
up to 1 × 109 NK cells/kg (141). Recently, it was reported that T flasks, numerous T flasks have to be handled at the same time,
repeated stimulation with SMI-LCL in IL-2-containing medium with for instance 51 T flasks for the treatment of a single patient
and adding IL-21 only at start of cultivation enables 1011-fold (116). In addition, T flasks have to be opened from time to time
NK cell expansion after 6 weeks, to our knowledge representing for medium exchange or harvesting of cells, bearing the risk of
the most efficient protocol to expand NK cells at the moment contaminating the cellular product. Although the likelihood of
(96). NK cells generated with the latter method are highly cyto- contamination for each T flask is reduced to a minimum by sterile
toxic in vitro, show a sustained high productivity of IFN-γ and workflows in safety cabinets, the remaining risk potentates by the
TNF-α, similar to CIML NK cells, and they efficiently controlled number of flasks.
melanoma in a xenograft mouse model (96). To overcome the drawbacks of small cell culture vessels, clini-
Although feeder cells, and allogeneic feeder cell lines in cal NK cell cultivation is often done in cell culture bags, which
particular, make it possible to generate substantial numbers of make it possible to culture high volumes in a closed system,
NK cells for adoptive therapy, from a regulatory point of view this as all required steps can be done by sterile welding of tubing
strategy has drawbacks as feeder cell lines must be qualified as connections for the transfer of media, harvesting of cells, etc.
safe for human use. The cell line qualification of modified K562 Unfortunately, different reports describe that the NK cell expan-
cells, for instance, includes costly viral testing and assays to prove sion performance is reduced after transition of a protocol from
absence of bacterial and Mycoplasma contamination (153). In this T flasks to larger scale in cell culture bags (50, 67). In addition,
context, lysates from cell lines containing the NK cell-stimulating bag systems still require several labor-intensive interventions
factors could be an alternative to the intact feeder cells to minimize during the culture, especially when different cultures are set up
regulatory concerns. It was demonstrated that short cultivation of in parallel.
NK cells with lysate of the leukemia cell line CTV-1 primes NK cells The G-Rex vessel is another system avoiding frequent process-
to specifically lyse cell lines that are resistant to resting NK cells ing steps for exchange of medium during the culture. In contrast
(143). Interestingly, the priming effect of CTV-1 on NK cells is KIR to normal cell culture flasks, the bottom of the G-Rex is highly gas
independent and does not require supplementation of cytokines, permeable, ensuring optimal CO2 exchange and O2 supply for the
such as IL-2 or IL-15, making this an unique approach for NK cell cells. Thus, by its design, G-Rex flasks can be filled directly with a
activation (154). NK cells primed with CTV-1 were evaluated in high level of cell culture medium and exchange of medium is not
the first UK clinical trial of a cell therapy regulated as a medicine, necessary for long time. For NK cell culture, G-Rex were used for
with an anti-leukemia effect in four of seven treated patients and no example for 10 days of culture without any cell manipulation or
evidence of NK cell infusion-related toxicities (144). Another step feeding, and resulted in higher fold expansion of NK cells com-
forward from a regulatory standpoint could be to add only specific pared to cell culture bags (130). Unfortunately, although G-Rex
fragments of feeder cells to the culture that are responsible for the are scalable in general, multiple G-Rex flasks are still required to
desired NK cell activation, instead of using intact feeder cells or achieve high cell numbers for clinical trials, which can be cum-
their lysates. Of note, NK cells can be expanded ex vivo with IL-2 bersome and costly, and G-Rex flasks are still an open system and
and plasma membrane particles prepared from K562-expressing may require adaption to a closed system (156).
membrane-bound IL-15 and 41BBL with a rate of expansion that Automated systems combine the need for reduced interven-
is comparable to stimulation with intact feeder cells and far bet- tions during the culture with a closed system. Automation of the
ter than stimulation with soluble IL-15, 41BBL, and IL-2 (133). cell manufacturing ensures constant product quality without the
Plasma membrane particles from K562 expressing membrane- need for highly skilled experts, is finally cost saving, and may
bound IL-21 and 41BBL work for ex vivo NK cell expansion as well be required for cellular therapy to become available beyond
and may be an option for in vivo NK cell expansion, as demon- specialized academic centers (157). Although early integration of
strated in a first proof of concept using a mouse model (136). automation is associated with higher capital costs in the develop-
ment phase, it allows a smooth transition at later stages of clinical
TECHNICAL ASPECTS OF NK CELL development (158). A first feasibility study of automated NK cell
EXPANSION cultivation with a stirred bioreactor was already published in
1996, showing advantages of the bioreactor culture over manu-
In general, one encounters technical challenges and opportunities ally handled controls (159). More recently, different investigators
when manufacturing NK cells as medicinal products, as reviewed report automated NK cell expansion procedures with a rocking
recently (155). In this section, we focus on technical options for motion bioreactor (67, 69, 156, 160), yielding 2–10 × 109 NK cells

Frontiers in Immunology | www.frontiersin.org 11 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

under GMP-compliant conditions. However, the latter system No 1394/2007; Directive 2001/83/EC and Regulation (EC)
still needs preceding manual cultivation, because relatively high No 726/2004]. Quality aspects related to somatic cell therapy
cell numbers are required as inoculum for the automated culture medicinal product as defined in guidelines (CPMP/BWP/3088/
(67, 69, 156, 160). Alternatively, fully automated NK cell expan- 99; EMEA/CHMP/410869/2006; Ph. Eur. 0784: Ph. Eur. 5.14) will
sion with an automated cell processing device can be performed apply to the identity, potency, and activity. The establishment of
for clinical use, with as little as 106 NK cells being sufficient correspondingly adequate in process and quality controls as well
to initiate the automated culture that can yield up to 2.7 × 109 as of process target values and product specifications will have
NK cells after stimulation with clinical grade feeder cells (142). Of to take into account the variability of the primary effector cell as
note, in addition to the culture process, the cell processing device the starting material (162).
is designed for GMP-compliant cell separation, concentration,
and washing applications, so that combined NK cell purification,
cultivation, and final formulation of the cellular product is pos-
CONCLUSION AND OUTLOOK
sible fully automated (161). Thus, the whole processing, from the Comparing different protocols for NK cell cultivation in detail is
starting material, such as a leukapheresis product, to the finally challenging as these are extremely heterogeneous. The duration
expanded NK cells, readily prepared for infusion, can be covered of ex vivo NK cell cultivation ranges from a few hours for short
by a single instrument. NK cell activation up to several weeks for long-term expansion,
Centralized processing of NK cell products probably will be different starting materials are in use with varying NK cell purities,
carried out mainly in specialized centers for manufacturing of different cytokines are combined at different doses, and NK cells
cellular products. Consequently, after ex vivo cultivation, storage often are cocultured with different feeder cells at different NK-to-
of the NK cell product and shipment to the location of use will feeder ratios. Nevertheless, overall differently ex vivo expanded
be needed. Compared to naive NK cells, IL-2-activated NK cells NK cells exhibit some common characteristics.
are less sensitive to freezing, as they show higher recovery and In general, ex vivo cultivated NK cells show an increased cyto-
viability after thawing (26). Still, different groups state that cryo- toxicity and may become even responsive against tumor targets
preservation of cultivated NK cells goes along with a drop in cell previously appearing resistant to NK cell lysis. This explains the
viability and cytotoxicity, whereas the latter can be restored by a use of IL-2 or IL-15 in virtually every protocol, as it is known
short re-stimulation, e.g., by a short resting in IL-2-containing since a long time that both cytokines amplify NK cell activity
medium (139, 156). Poor survival of the NK cells can be an issue (81, 163). However, upon NK cell activation with different stimuli,
during further in vitro culture post thawing, so that shipping of including IL-2 and IL-15, downregulation of CD16 surface
freshly formulated cells for direct infusion may be advantageous levels occurs by metalloproteases-mediated shedding of CD16
(129). Interestingly, some groups recently claim that freezing (164–166). The Fc receptor CD16 is crucial for NK cells to per-
and thawing does not influence the cytotoxicity or the prolifera- form ADCC and would be of particular importance for potential
tive ability of cultivated NK cells in their hands (24, 68). These combination therapies using NK cells together with therapeutic
divergent observations possibly result from different cultivation antibodies. Of note, although reduced levels of CD16 on NK cells
methods and different protocols for freezing and thawing, which are observed for several NK cell cultivation protocols the NK cells
should be investigated further. Without freezing, transport of still mediate ADCC (70, 129, 142). Nevertheless, inhibition of the
the readily prepared cells in an appropriate time frame is chal- relevant metalloproteases to maintain CD16 on NK cells could
lenging, and any delay during the shipment affects the quality of be an option to further increase the ADCC function of ex vivo
the cellular product with critical consequences for the patient. activated NK cells (164, 167).
Alternatively, automated and closed systems for cell processing Another clinically highly relevant aspect is the tumor-induced
open the way for scale out strategies and de-centralized NK cell immunosuppression as important challenge for all cell therapeu-
manufacturing directly at the location of intended use, avoiding tic strategies. Remarkably, it ruled out from most preclinical and
the freezing and shipment process (142). But, although de-cen- clinical NK cell studies that NK cells may gain the capability to
tralized manufacturing in the clinics seems promising, cellular overcome tumor immunosuppression. Different research groups
therapeutics are very complex and still in early development, so have reported signs of NK cell suppression in cancer patients such
that manufacturing by well-trained specialists in specific facilities as a lower expression of NK cell receptors, e.g., NCRs, NKG2D,
is reasonable at that state. DNAM-1, and 2B4 (22, 25, 168–170), the shedding of tumor cell
ligands, such as NKp30 and NKG2D (171–174), or the release
REGULATORY ASPECTS OF NK CELL of blocking NKG2D ligands, such as MICA and ULBP3, via
CULTIVATION FOR CLINICAL USE tumor-derived exosomes (175, 176). Notably, ex vivo cultivation
of patient-derived NK cells is often possible with same efficacy
Apart from technical difficulties, one has to consider regulatory as for donor-derived NK cells (25, 111) and can normalize the
aspects for the use of ex vivo-generated NK cells with regulations NK cell phenotype and activation (25). Additionally, elevated
varying in time and geographical policies (153). In Europe, levels of NKG2D on ex vivo-activated NK cells can scavenge
for instance, cytokine-activated and -expanded NK cells are shed NKG2D ligands and counter their inhibitory effect (177).
currently classified as advanced therapy medicinal products Furthermore, the high cytotoxicity of ex vivo expanded NK cells
and will be regulated accordingly either centralized or under has been shown to be independent of KIR inhibition for some
the hospital exemption by the member states [Regulation (EC) protocols (107, 132).

Frontiers in Immunology | www.frontiersin.org 12 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

In comparison to other cell therapeutic approaches using, and efficacy of the NK cell product for clinical application. Finally,
e.g., T cells, donor-derived allogeneic NK cells mediate GVL with- with regard to the possible tumor-mediated immunosuppression,
out an elevated risk for GVHD or even with a GVHD-reducing therapeutic concepts have to be developed that either directly
effect, as reported in mice and men (74, 75, 77, 78). However, con- strengthen NK cells to deal with the hostile tumor environment
tradictory results regarding GVHD induction have been reported and/or specifically counteract tumor-induced immunosuppres-
in clinical trials assessing adoptive transfer of NK cells expanded sive mechanisms.
with K562 feeder cell variants expressing 41BBL and IL-15 (129,
131). These reports show that there are still open questions that AUTHOR CONTRIBUTIONS
have to be unraveled to better understand the complex role of
NK cells and their specific subsets in the bidirectional regulation MG, JW, and EU extensively reviewed the current literature on ex
of GVL and GVHD. vivo NK cell cultivation and expansion and prepared a compre-
In conclusion, many different protocols are in use to expand hensive overview that is listed in the tables. MG, JW, UK, AC, VH,
NK cells in vitro, each with its specific advantages and disadvan- and EU wrote and critically reviewed the manuscript.
tages in regard of cell numbers, function, and handling efforts.
The data summarized in this review underline the complexity ACKNOWLEDGMENTS
related to the design of an optimal NK cell therapeutic protocol
that should be not only reliable and safe in use but also highly The authors apologize to all investigators whose works were
efficient in targeting different forms of malignancies. With this not cited in this article due to space limitations. EU and JW
in mind, additional studies need to be envisioned that not only have been supported by the LOEWE Center for Cell and
further address ex vivo NK cell purification, expansion, and Gene Therapy, Frankfurt, funded by the Hessian Ministry
activation strategies but also the final clinical setting including of Higher Education, Research and the Arts, Germany (III L
pre-conditioning, dosing, and timing of the NK cell application. 4-518/17.004). AC’s work on cytokine-induced NK cells is
Efforts for harmonization of protocols at the European and supported by the Grant of German Cancer Aid 111455 and the
worldwide level should be undertaken to ensure highest quality GRK 2099.

REFERENCES to interferon gamma-dependent natural killer cell protection from tumor

metastasis. J Exp Med (2001) 193:661–70. doi:10.1084/jem.193.6.661
1. Moretta A, Bottino C, Vitale M, Pende D, Cantoni C, Mingari MC, et al. 12. Smyth MJ, Crowe NY, Pellicci DG, Kyparissoudis K, Kelly JM, Takeda K,
Activating receptors and coreceptors involved in human natural killer et al. Sequential production of interferon-gamma by NK1.1(+) T cells
cell-mediated cytolysis. Annu Rev Immunol (2001) 19:197–223. doi:10.1146/ and natural killer cells is essential for the antimetastatic effect of alpha-
annurev.immunol.19.1.197 galactosylceramide. Blood (2002) 99:1259–66. doi:10.1182/blood.V99.4.1259
2. Voskoboinik I, Smyth MJ, Trapani JA. Perforin-mediated target-cell death 13. Chawla-Sarkar M, Lindner DJ, Liu Y-F, Williams BR, Sen GC, Silverman RH,
and immune homeostasis. Nat Rev Immunol (2006) 6:940–52. doi:10.1038/ et al. Apoptosis and interferons: role of interferon-stimulated genes
nri1983 as mediators of apoptosis. Apoptosis (2003) 8:237–49. doi:10.1023/A:
3. Vivier E, Tomasello E, Baratin M, Walzer T, Ugolini S. Functions of natural 1023668705040
killer cells. Nat Immunol (2008) 9:503–10. doi:10.1038/ni1582 14. Costello RT, Sivori S, Marcenaro E, Lafage-Pochitaloff M, Mozziconacci M-J,
4. Garrido F, Ruiz-Cabello F, Cabrera T, Pérez-Villar JJ, López-Botet M, Reviron D, et al. Defective expression and function of natural killer cell-
Duggan-Keen M, et al. Implications for immunosurveillance of altered HLA triggering receptors in patients with acute myeloid leukemia. Blood (2002)
class I phenotypes in human tumours. Immunol Today (1997) 18:89–95. 99:3661–7. doi:10.1182/blood.V99.10.3661
doi:10.1016/S0167-5699(96)10075-X 15. Salih HR, Rammensee H-G, Steinle A. Cutting edge: down-regulation
5. Bottino C, Castriconi R, Moretta L, Moretta A. Cellular ligands of activating of MICA on human tumors by proteolytic shedding. J Immunol (2002)
NK receptors. Trends Immunol (2005) 26:221–6. doi:10.1016/j.it.2005.02.007 169:4098–102. doi:10.4049/jimmunol.169.8.4098
6. Pende D, Cantoni C, Rivera P, Vitale M, Castriconi R, Marcenaro S, 16. Salih HR, Goehlsdorf D, Steinle A. Release of MICB molecules by tumor
et al. Role of NKG2D in tumor cell lysis mediated by human NK cells: cells: mechanism and soluble MICB in sera of cancer patients. Hum Immunol
cooperation with natural cytotoxicity receptors and capability of recog- (2006) 67:188–95. doi:10.1016/j.humimm.2006.02.008
nizing tumors of nonepithelial origin. Eur J Immunol (2001) 31:1076–86. 17. Kaiser BK, Yim D, Chow I-T, Gonzalez S, Dai Z, Mann HH, et al. Disulphide-
doi:10.1002/1521-4141(200104)31:4<1076:AID-IMMU1076>3.0.CO;2-Y isomerase-enabled shedding of tumour-associated NKG2D ligands. Nature
7. Pende D, Spaggiari GM, Marcenaro S, Martini S, Rivera P, Capobianco A, (2007) 447:482–6. doi:10.1038/nature05768
et al. Analysis of the receptor-ligand interactions in the natural killer- 18. Rouas-Freiss N, Moreau P, Ferrone S, Carosella ED. HLA-G proteins in
mediated lysis of freshly isolated myeloid or lymphoblastic leukemias: evi- cancer: do they provide tumor cells with an escape mechanism? Cancer Res
dence for the involvement of the Poliovirus receptor (CD155) and Nectin-2 (2005) 65:10139–44. doi:10.1158/0008-5472.CAN-05-0097
(CD112). Blood (2005) 105:2066–73. doi:10.1182/blood-2004-09-3548 19. Urosevic M, Dummer R. Human leukocyte antigen-G and cancer immunoed-
8. Bauer S, Groh V, Wu J, Steinle A, Phillips JH, Lanier LL, et al. Activation of iting. Cancer Res (2008) 68:627–30. doi:10.1158/0008-5472.CAN-07-2704
NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science 20. Castriconi R, Dondero A, Bellora F, Moretta L, Castellano A, Locatelli F, et al.
(1999) 285:727–9. doi:10.1126/science.285.5428.727 Neuroblastoma-derived TGF-β1 modulates the chemokine receptor reper-
9. Ullrich E, Koch J, Cerwenka A, Steinle A. New prospects on the NKG2D/ toire of human resting NK cells. J Immunol (2013) 190:5321–8. doi:10.4049/
NKG2DL system for oncology. Oncoimmunology (2013) 2:e26097. jimmunol.1202693
doi:10.4161/onci.26097 21. Stringaris K, Sekine T, Khoder A, Alsuliman A, Razzaghi B, Sargeant R, et al.
10. Smyth MJ, Cretney E, Kelly JM, Westwood JA, Street SEA, Yagita H, Leukemia-induced phenotypic and functional defects in natural killer cells
et al. Activation of NK cell cytotoxicity. Mol Immunol (2005) 42:501–10. predict failure to achieve remission in acute myeloid leukemia. Haematologica
doi:10.1016/j.molimm.2004.07.034 (2014) 99:836–47. doi:10.3324/haematol.2013.087536
11. Smyth MJ, Cretney E, Takeda K, Wiltrout RH, Sedger LM, Kayagaki N, et al. 22. Pietra G, Manzini C, Rivara S, Vitale M, Cantoni C, Petretto A, et al.
Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) contributes Melanoma cells inhibit natural killer cell function by modulating the

Frontiers in Immunology | www.frontiersin.org 13 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

expression of activating receptors and cytolytic activity. Cancer Res (2012) haploidentical NK cells in patients with cancer. Blood (2005) 105:3051–7.
72:1407–15. doi:10.1158/0008-5472.CAN-11-2544 doi:10.1182/blood-2004-07-2974
23. Lim O, Lee Y, Chung H, Her JH, Kang SM, Jung M-Y, et al. GMP-compliant, 44. Ho VT, Soiffer RJ. The history and future of T-cell depletion as graft-versus-
large-scale expanded allogeneic natural killer cells have potent cytolytic host disease prophylaxis for allogeneic hematopoietic stem cell transplanta-
activity against cancer cells in vitro and in vivo. PLoS One (2013) 8:e53611. tion. Blood (2001) 98:3192–204. doi:10.1182/blood.V98.12.3192
doi:10.1371/journal.pone.0053611 45. Skeate R, Singh C, Cooley S, Geller M, Northouse J, Welbig J, et al. Hemolytic
24. Torelli GF, Rozera C, Santodonato L, Peragine N, D’Agostino G, Montefiore anemia due to passenger lymphocyte syndrome in solid malignancy patients
E, et al. A good manufacturing practice method to ex vivo expand natural treated with allogeneic natural killer cell products. Transfusion (2013)
killer cells for clinical use. Blood Transfus (2015) 13:464–71. doi:10.2450/ 53:419–23. doi:10.1111/j.1537-2995.2012.03942.x
2015.0231-14 46. McKenna DH, Sumstad D, Bostrom N, Kadidlo DM, Fautsch S,
25. Alici E, Sutlu T, Björkstrand B, Gilljam M, Stellan B, Nahi H, et al. Autologous McNearney S, et al. Good manufacturing practices production of natural
antitumor activity by NK cells expanded from myeloma patients using killer cells for immunotherapy: a six-year single-institution experience.
GMP-compliant components. Blood (2008) 111:3155–62. doi:10.1182/ Transfusion (2007) 47:520–8. doi:10.1111/j.1537-2995.2006.01145.x
blood-2007-09-110312 47. Iyengar R, Handgretinger R, Babarin-Dorner A, Leimig T, Otto M,
26. Koehl U, Brehm C, Huenecke S, Zimmermann S-Y, Kloess S, Bremm M, Geiger TL, et al. Purification of human natural killer cells using a
et al. Clinical grade purification and expansion of NK cell products for an clinical-scale immunomagnetic method. Cytotherapy (2003) 5:479–84.
optimized manufacturing protocol. Front Oncol (2013) 3:118. doi:10.3389/ doi:10.1080/14653240310003558
fonc.2013.00118 48. Passweg JR, Tichelli A, Meyer-Monard S, Heim D, Stern M, Kuhne T,
27. Colombo MP, Trinchieri G. Interleukin-12 in anti-tumor immunity and et al. Purified donor NK-lymphocyte infusion to consolidate engraftment
immunotherapy. Cytokine Growth Factor Rev (2002) 13:155–68. doi:10.1016/ after haploidentical stem cell transplantation. Leukemia (2004) 18:1835–8.
S1359-6101(01)00032-6 doi:10.1038/sj.leu.2403524
28. Becknell B, Caligiuri MA. Interleukin-2, interleukin-15, and their roles in 49. Meyer-Monard S, Passweg J, Siegler U, Kalberer C, Koehl U, Rovó A,
human natural killer cells. Adv Immunol (2005) 86:209–39. doi:10.1016/ et al. Clinical-grade purification of natural killer cells in haploidentical
S0065-2776(04)86006-1 hematopoietic stem cell transplantation. Transfusion (2009) 49:362–71.
29. Theofilopoulos AN, Baccala R, Beutler B, Kono DH. Type I interferons doi:10.1111/j.1537-2995.2008.01969.x
(alpha/beta) in immunity and autoimmunity. Annu Rev Immunol (2005) 50. Siegler U, Meyer-Monard S, Jorger S, Stern M, Tichelli A, Gratwohl A,
23:307–36. doi:10.1146/annurev.immunol.23.021704.115843 et al. Good manufacturing practice-compliant cell sorting and large-scale
30. Burgess SJ, Marusina AI, Pathmanathan I, Borrego F, Coligan JE. IL-21 expansion of single KIR-positive alloreactive human natural killer cells for
down-regulates NKG2D/DAP10 expression on human NK and CD8+ multiple infusions to leukemia patients. Cytotherapy (2010) 12:750–63.
T cells. J Immunol (2006) 176:1490–7. doi:10.4049/jimmunol.176.3.1490 doi:10.3109/14653241003786155
31. Ortaldo JR, Winkler-Pickett R, Wigginton J, Horner M, Bere EW, Mason AT, 51. van Ostaijen-ten Dam MM, Prins H-J, Boerman GH, Vervat C, Pende D,
et al. Regulation of ITAM-positive receptors: role of IL-12 and IL-18. Blood Putter H, et al. Preparation of cytokine-activated NK cells for use in adoptive
(2006) 107:1468–75. doi:10.1182/blood-2005-04-1579 cell therapy in cancer patients: protocol optimization and therapeutic poten-
32. Terme M, Ullrich E, Delahaye NF, Chaput N, Zitvogel L. Natural killer tial. J Immunother (2016) 39:90–100. doi:10.1097/CJI.0000000000000110
cell-directed therapies: moving from unexpected results to successful strate- 52. Huenecke S, Zimmermann SY, Kloess S, Esser R, Brinkmann A, Tramsen L,
gies. Nat Immunol (2008) 9:486–94. doi:10.1038/ni1580 et al. IL-2-driven regulation of NK cell receptors with regard to the distribu-
33. Koehl U, Kalberer C, Spanholtz J, Lee D, Miller J, Cooley S, et al. Advances tion of CD16+ and CD16- subpopulations and in vivo influence after hap-
in clinical NK cell studies: donor selection, manufacturing and quality con- loidentical NK cell infusion. J Immunother (2010) 33:200–10. doi:10.1097/
trol. Oncoimmunology (2015) 5(4):e1115178. doi:10.1080/2162402X.2015. CJI.0b013e3181bb46f7
1115178 53. Koehl U, Sörensen J, Esser R, Zimmermann S, Grüttner HP, Tonn T, et al.
34. Glienke W, Esser R, Priesner C, Suerth JD, Schambach A, Wels WS, et al. IL-2 activated NK cell immunotherapy of three children after haploidentical
Advantages and applications of CAR-expressing natural killer cells. Front stem cell transplantation. Blood Cells Mol Dis (2004) 33:261–6. doi:10.1016/j.
Pharmacol (2015) 6:21. doi:10.3389/fphar.2015.00021 bcmd.2004.08.013
35. Klingemann H. Challenges of cancer therapy with natural killer cells. 54. Boerman GH, van Ostaijen-ten Dam MM, Kraal KCJM, Santos SJ, Ball LM,
Cytotherapy (2015) 17:245–9. doi:10.1016/j.jcyt.2014.09.007 Lankester AC, et al. Role of NKG2D, DNAM-1 and natural cytotoxicity
36. Cany J, Dolstra H, Shah N. Umbilical cord blood-derived cellular products receptors in cytotoxicity toward rhabdomyosarcoma cell lines mediated
for cancer immunotherapy. Cytotherapy (2015) 17:739–48. doi:10.1016/j. by resting and IL-15-activated human natural killer cells. Cancer Immunol
jcyt.2015.03.005 Immunother (2015) 64:573–83. doi:10.1007/s00262-015-1657-9
37. Spanholtz J, Tordoir M, Eissens D, Preijers F, van der Meer A, Joosten I, 55. Wendt K, Wilk E, Buyny S, Schmidt RE, Jacobs R. Interleukin-21 differentially
et al. High log-scale expansion of functional human natural killer cells from affects human natural killer cell subsets. Immunology (2007) 122:486–95.
umbilical cord blood CD34-positive cells for adoptive cancer immunother- doi:10.1111/j.1365-2567.2007.02675.x
apy. PLoS One (2010) 5:e9221. doi:10.1371/journal.pone.0009221 56. Romee R, Schneider SE, Leong JW, Chase JM, Keppel CR, Sullivan RP, et al.
38. Suck G, Odendahl M, Nowakowska P, Seidl C, Wels WS, Klingemann HG, Cytokine activation induces human memory-like NK cells. Blood (2012)
et al. NK-92: an “off-the-shelf therapeutic” for adoptive natural killer 120:4751–60. doi:10.1182/blood-2012-04-419283
cell-based cancer immunotherapy. Cancer Immunol Immunother (2015) 57. Ni J, Miller M, Stojanovic A, Garbi N, Cerwenka A. Sustained effector func-
65:485–92. doi:10.1007/s00262-015-1761-x tion of IL-12/15/18-preactivated NK cells against established tumors. J Exp
39. Knorr DA, Kaufman DS. Pluripotent stem cell-derived natural killer cells for Med (2012) 209:2351–65. doi:10.1084/jem.20120944
cancer therapy. Transl Res (2010) 156:147–54. doi:10.1016/j.trsl.2010.07.008 58. Leong JW, Chase JM, Romee R, Schneider SE, Sullivan RP, Cooper MA, et al.
40. Klingemann H, Boissel L, Toneguzzo F. Natural killer cells for immunotherapy – Preactivation with IL-12, IL-15, and IL-18 induces CD25 and a functional
advantages of the NK-92 cell line over blood NK cells. Front Immunol (2016) high-affinity IL-2 receptor on human cytokine-induced memory-like natural
7:91. doi:10.3389/fimmu.2016.00091 killer cells. Biol Blood Marrow Transplant (2014) 20:463–73. doi:10.1016/j.
41. Eguizabal C, Zenarruzabeitia O, Monge J, Santos S, Vesga MA, Maruri N, bbmt.2014.01.006
et al. Natural killer cells for cancer immunotherapy: pluripotent stem 59. Berrien-Elliott MM, Wagner JA, Fehniger TA. Human cytokine-induced
cells-derived NK cells as an immunotherapeutic perspective. Front Immunol memory-like natural killer cells. J Innate Immun (2015) 7:563–71.
(2014) 5:439. doi:10.3389/fimmu.2014.00439 doi:10.1159/000382019
42. Ferrara JLM, Levine JE, Reddy P, Holler E. Graft-versus-host disease. Lancet 60. Geller MA, Cooley S, Judson PL, Ghebre R, Carson LF, Argenta PA, et al. A
(2009) 373:1550–61. doi:10.1016/S0140-6736(09)60237-3 phase II study of allogeneic natural killer cell therapy to treat patients with
43. Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, recurrent ovarian and breast cancer. Cytotherapy (2011) 13:98–107. doi:10.3
Fautsch SK, et al. Successful adoptive transfer and in vivo expansion of human 109/14653249.2010.515582.A

Frontiers in Immunology | www.frontiersin.org 14 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

61. Klingemann H, Grodman C, Cutler E, Duque M, Kadidlo D, Klein AK, supporting graft-versus-leukemia- and reducing graft-versus-host-effects.
et al. Autologous stem cell transplant recipients tolerate haploidentical Oncoimmunology (2015) 4:e981483. doi:10.4161/2162402X.2014.981483
related-donor natural killer cell-enriched infusions. Transfusion (2013) 79. Bachanova V, Burns LJ, McKenna DH, Curtsinger J, Panoskaltsis-Mortari A,
53:412–8. doi:10.1111/j.1537-2995.2012.03764.x Lindgren BR, et al. Allogeneic natural killer cells for refractory lym-
62. Koepsell SA, Kadidlo DM, Fautsch S, McCullough J, Klingemann H, phoma. Cancer Immunol Immunother (2010) 59:1739–44. doi:10.1007/
Wagner JE, et al. Successful “in-flight” activation of natural killer s00262-010-0896-z
cells during long-distance shipping. Transfusion (2013) 53:398–403. 80. Brehm C, Huenecke S, Quaiser A, Esser R, Bremm M, Kloess S, et al. IL-2
doi:10.1111/j.1537-2995.2012.03695.x stimulated but not unstimulated NK cells induce selective disappearance of
63. Iliopoulou EG, Kountourakis P, Karamouzis MV, Doufexis D, Ardavanis A, peripheral blood cells: concomitant results to a phase I/II study. PLoS One
Baxevanis CN, et al. A phase I trial of adoptive transfer of allogeneic natural (2011) 6:e27351. doi:10.1371/journal.pone.0027351
killer cells in patients with advanced non-small cell lung cancer. Cancer 81. Carson WE, Giri JG, Lindemann MJ, Linett ML, Ahdieh M, Paxton R, et al.
Immunol Immunother (2010) 59:1781–9. doi:10.1007/s00262-010-0904-3 Interleukin (IL) 15 is a novel cytokine that activates human natural killer
64. Choi I, Yoon SR, Park S-Y, Kim H, Jung S-J, Jang YJ, et al. Donor-derived cells via components of the IL-2 receptor. J Exp Med (1994) 180:1395–403.
natural killer cells infused after human leukocyte antigen-haploidentical doi:10.1084/jem.180.4.1395
hematopoietic cell transplantation: a dose-escalation study. Biol Blood 82. Carson WE, Fehniger TA, Haldar S, Eckhert K, Lindemann MJ, Lai CF, et al.
Marrow Transplant (2014) 20:696–704. doi:10.1016/j.bbmt.2014.01.031 A potential role for interleukin-15 in the regulation of human natural killer
65. Carlens S, Gilljam M, Chambers BJ, Aschan J, Guven H, Ljunggren HG, cell survival. J Clin Invest (1997) 99:937–43. doi:10.1172/JCI119258
et al. A new method for in vitro expansion of cytotoxic human CD3-CD56+ 83. Waldmann TA. The biology of interleukin-2 and interleukin-15: implications
natural killer cells. Hum Immunol (2001) 62:1092–8. doi:10.1016/ for cancer therapy and vaccine design. Nat Rev Immunol (2006) 6:595–601.
S0198-8859(01)00313-5 doi:10.1038/nri1901
66. Barkholt L, Alici E, Conrad R, Sutlu T, Gilljam M, Stellan B, et al. Safety 84. Fehniger TA, Cooper MA, Caligiuri MA. Interleukin-2 and interleukin-15:
analysis of ex vivo-expanded NK and NK-like T cells administered to immunotherapy for cancer. Cytokine Growth Factor Rev (2002) 13:169–83.
cancer patients: a phase I clinical study. Immunotherapy (2009) 1:753–64. doi:10.1016/S1359-6101(01)00021-1
doi:10.2217/imt.09.47 85. Waldmann TA, Lugli E, Roederer M, Perera LP, Smedley JV, Macallister RP,
67. Sutlu T, Stellan B, Gilljam M, Quezada HC, Nahi H, Gahrton G, et al. et al. Safety (toxicity), pharmacokinetics, immunogenicity, and impact on ele-
Clinical-grade, large-scale, feeder-free expansion of highly active human nat- ments of the normal immune system of recombinant human IL-15 in rhesus
ural killer cells for adoptive immunotherapy using an automated bioreactor. macaques. Blood (2011) 117:4787–95. doi:10.1182/blood-2010-10-311456
Cytotherapy (2010) 12:1044–55. doi:10.3109/14653249.2010.504770 86. Mao Y, van Hoef V, Zhang X, Wennerberg E, Lorent J, Witt K, et al. IL-15
68. Masuyama J, Murakami T, Iwamoto S, Fujita S. Ex vivo expansion of natural activates mTOR and primes stress-activated gene expression leading to
killer cells from human peripheral blood mononuclear cells co-stimulated prolonged antitumor capacity of NK cells. Blood (2016) 128:1475–89.
with anti-CD3 and anti-CD52 monoclonal antibodies. Cytotherapy (2015) doi:10.1182/blood-2016-02-698027
18:1–6. doi:10.1016/j.jcyt.2015.09.011 87. Felices M, Lenvik A, Chu S, McElmurry R, Cooley S, Tolar J, et al. Continuous
69. Rujkijyanont P, Chan WK, Eldridge PW, Lockey T, Holladay M, Rooney B, IL-15 signaling leads to functional exhaustion of human natural killer cells
et al. Ex vivo activation of CD56+ immune cells that eradicate neuroblas- through metabolic changes that alters their in vivo anti-tumor activity. Blood
toma. Cancer Res (2013) 73:2608–18. doi:10.1158/0008-5472.CAN-12-3322 (2016) 128:551.
70. Deng X, Terunuma H, Nieda M, Xiao W, Nicol A. Synergistic cytotoxicity of 88. Davis MR, Zhu Z, Hansen DM, Bai Q, Fang Y. The role of IL-21 in immunity
ex vivo expanded natural killer cells in combination with monoclonal anti- and cancer. Cancer Lett (2015) 358:107–14. doi:10.1016/j.canlet.2014.12.047
body drugs against cancer cells. Int Immunopharmacol (2012) 14:593–605. 89. Parrish-novak J, Dillon SR, Nelson A, Hammond A, Sprecher C,
doi:10.1016/j.intimp.2012.09.014 Gross JA, et al. Interleukin 21 and its receptor are involved in NK cell
71. Lotze MT, Grimm EA, Mazumder A, Strausser JL, Rosenberg SA. Lysis of expansion and regulation of lymphocyte function. Nature (2000) 408:57–63.
fresh and cultured autologous tumor by human lymphocytes cultured in doi:10.1038/35040504
T-cell growth factor. Cancer Res (1981) 41:4420–5. 90. Kasaian MT, Whitters MJ, Carter LL, Lowe LD, Jussif JM, Deng B, et al. IL-
72. Grimm EA, Mazumder A, Zhang HZ, Rosenberg SA. Lymphokine-activated 21 limits NK cell responses and promotes antigen-specific T cell activation:
killer cell phenomenon. Lysis of natural killer-resistant fresh solid tumor cells a mediator of the transition from innate to adaptive immunity. Immunity
by interleukin 2-activated autologous human peripheral blood lymphocytes. (2002) 16:559–69. doi:10.1016/S1074-7613(02)00295-9
J Exp Med (1982) 155:1823–41. doi:10.1084/jem.155.6.1823 91. Brady J, Hayakawa Y, Smyth MJ, Nutt SL. IL-21 induces the functional
73. Burns LJ, Weisdorf DJ, DeFor TE, Vesole DH, Repka TL, Blazar BR, et al. IL-2- maturation of murine NK cells. J Immunol (2004) 172:2048–58. doi:10.4049/
based immunotherapy after autologous transplantation for lymphoma and jimmunol.172.4.2048
breast cancer induces immune activation and cytokine release: a phase I/II 92. Venkatasubramanian S, Cheekatla S, Paidipally P, Tripathi D, Welch E,
trial. Bone Marrow Transplant (2003) 32:177–86. doi:10.1038/sj.bmt.1704086 Tvinnereim AR, et al. IL-21-dependent expansion of memory-like NK cells
74. Asai O, Longo DL, Tian ZG, Hornung RL, Taub DD, Ruscetti FW, et al. enhances protective immune responses against Mycobacterium tuberculosis.
Suppression of graft-versus-host disease and amplification of graft-versus- Mucosal Immunol (2016):1–12. doi:10.1038/mi.2016.105
tumor effects by activated natural killer cells after allogeneic bone marrow 93. Strbo N, de Armas L, Liu H, Kolber MA, Lichtenheld M, Pahwa S. IL-21
transplantation. J Clin Invest (1998) 101:1835–42. doi:10.1172/JCI1268 augments natural killer effector functions in chronically HIV-infected
75. Olson JA, Leveson-Gower DB, Gill S, Baker J, Beilhack A, Negrin RS. individuals. AIDS (2008) 22:1551–60. doi:10.1097/QAD.0b013e3283089367
NK cells mediate reduction of GVHD by inhibiting activated, alloreactive 94. Li Q, Ye L-J, Ren H-L, Huyan T, Li J, Shi J-L, et al. Multiple effects of IL-21
T cells while retaining GVT effects. Blood (2010) 115:4293–301. doi:10.1182/ on human NK cells in ex vivo expansion. Immunobiology (2015) 220:876–88.
blood-2009-05-222190 doi:10.1016/j.imbio.2015.01.009
76. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, Tosti A, 95. Lim D-P, Jang Y-Y, Kim S, Koh SS, Lee J-J, Kim J-S, et al. Effect of exposure
et al. Effectiveness of donor natural killer cell alloreactivity in mismatched to interleukin-21 at various time points on human natural killer cell culture.
hematopoietic transplants. Science (2002) 295:2097–100. doi:10.1126/ Cytotherapy (2014) 16:1419–30. doi:10.1016/j.jcyt.2014.04.008
science.1068440 96. Granzin M, Stojanovic A, Miller M, Childs R, Huppert V, Cerwenka A.
77. Velardi A, Ruggeri L, Mancusi A, Aversa F, Christiansen FT. Natural killer cell Highly efficient IL-21 and feeder cell-driven ex vivo expansion of human
allorecognition of missing self in allogeneic hematopoietic transplantation: a NK cells with therapeutic activity in a xenograft mouse model of mela-
tool for immunotherapy of leukemia. Curr Opin Immunol (2009) 21:525–30. noma. Oncoimmunology (2016) 5:e1219007. doi:10.1080/2162402X.2016.
doi:10.1016/j.coi.2009.07.015 1219007
78. Meinhardt K, Kroeger I, Bauer R, Ganss F, Ovsiy I, Rothamer J, et al. 97. McMichael EL, Jaime-Ramirez AC, Guenterberg KD, Luedke E, Atwal LS,
Identification and characterization of the specific murine NK cell subset Campbell AR, et al. IL-21 enhances natural killer cell response to

Frontiers in Immunology | www.frontiersin.org 15 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

cetuximab-coated pancreatic tumor cells. Clin Cancer Res (2017) 23:489–502. in patients with malignant lymphoma or advanced solid tumors. Cancer
doi:10.1158/1078-0432.CCR-16-0004 Immunol Res (2016) 4:215–24. doi:10.1158/2326-6066.CIR-15-0118
98. Iannello A, Boulassel M-R, Samarani S, Tremblay C, Toma E, Routy J-P, et al. 118. Luhm J, Brand JM, Koritke P, Hoppner M, Kirchner H, Frohn C. Large-scale
IL-21 enhances NK cell functions and survival in healthy and HIV-infected generation of natural killer lymphocytes for clinical application. J Hematother
patients with minimal stimulation of viral replication. J Leukoc Biol (2010) Stem Cell Res (2002) 11:651–7. doi:10.1089/15258160260194794
87:857–67. doi:10.1189/jlb.1009701 119. Bae DS, Lee JK. Development of NK cell expansion methods using feeder
99. Kobayashi M, Fitz L, Ryan M, Hewick RM, Clark SC, Chan S, et al. cells from human myelogenous leukemia cell line. Blood Res (2014) 49:154.
Identification and purification of natural killer cell stimulatory factor doi:10.5045/br.2014.49.3.154
(NKSF), a cytokine with multiple biologic effects on human lymphocytes. 120. Fuchshuber PR, Lotzova E. Feeder cells enhance oncolytic and proliferative
J Exp Med (1989) 170:827–45. doi:10.1084/jem.170.3.827 activity of long-term human bone marrow interleukin-2 cultures and induce
100. D’Andrea A, Rengaraju M, Valiante NM, Chehimi J, Kubin M, Aste M, et al. different lymphocyte subsets. Cancer Immunol Immunother (1991) 33:15–20.
Production of natural killer cell stimulatory factor (interleukin 12) by peri­ doi:10.1007/BF01742522
pheral blood mononuclear cells. J Exp Med (1992) 176:1387–98. doi:10.1084/ 121. Lim SA, Kim T-J, Lee JE, Sonn CH, Kim K, Kim J, et al. Ex vivo expansion
jem.176.5.1387 of highly cytotoxic human nk cells by cocultivation with irradiated tumor
101. Vignali DAA, Kuchroo VK. IL-12 family cytokines: immunological playmak- cells for adoptive immunotherapy. Cancer Res (2013) 73(8):2598–607.
ers. Nat Immunol (2012) 13:722–8. doi:10.1038/ni.2366 doi:10.1158/0008-5472.CAN-12-2893
102. Wang KS, Frank DA, Ritz J. Interleukin-2 enhances the response of natural 122. Warren HS, Kinnear BF, Skipsey LJ. Human natural killer (NK) cells: require-
killer cells to interleukin-12 through up-regulation of the interleukin-12 ments for cell proliferation and expansion of phenotypically novel subpop-
receptor and STAT4. Blood (2000) 95:3183–90. ulations. Immunol Cell Biol (1993) 71(Pt 2):87–97. doi:10.1038/icb.1993.9
103. Chaix J, Tessmer MS, Hoebe K, Fuseri N, Ryffel B, Dalod M, et al. Cutting edge: 123. Lister J, Rybka WB, Donnenberg AD, DeMagalhaes-Silverman M,
priming of NK cells by IL-18. J Immunol (2008) 181:1627–31. doi:10.4049/ Pincus SM, Bloom EJ, et al. Autologous peripheral blood stem cell transplan-
jimmunol.181.3.1627 tation and adoptive immunotherapy with activated natural killer cells in the
104. French AR, Holroyd EB, Yang L, Kim S, Yokoyama WM. IL-18 acts syner- immediate posttransplant period. Clin Cancer Res (1995) 1:607–14.
gistically with IL-15 in stimulating natural killer cell proliferation. Cytokine 124. Harada H, Saijo K, Watanabe S, Tsuboi K, Nose T, Ishiwata I, et al. Selective
(2006) 35:229–34. doi:10.1016/j.cyto.2006.08.006 expansion of human natural killer cells from peripheral blood mononuclear
105. Cooper MA, Elliott JM, Keyel PA, Yang L, Carrero JA, Yokoyama WM. cells by the cell line, HFWT. Jpn J Cancer Res (2002) 93:313–9. doi:10.1111/j.
Cytokine-induced memory-like natural killer cells. Proc Natl Acad Sci U S A 1349-7006.2002.tb02174.x
(2009) 106:1915–9. doi:10.1073/pnas.0813192106 125. Ishikawa E, Tsuboi K, Saijo K, Harada H, Takano S, Nose T, et al. Autologous
106. Ni J, Hölsken O, Miller M, Hammer Q, Luetke-Eversloh M, Romagnani C, natural killer cell therapy for human recurrent malignant glioma. Anticancer
et al. Adoptively transferred natural killer cells maintain long-term antitumor Res (2004) 24:1861–71.
activity by epigenetic imprinting and CD4(+) T cell help. Oncoimmunology 126. Imai C, Iwamoto S, Campana D. Genetic modification of primary natural
(2016) 5:e1219009. doi:10.1080/2162402X.2016.1219009 killer cells overcomes inhibitory signals and induces specific killing of
107. Romee R, Rosario M, Berrien-Elliott MM, Wagner JA, Jewell BA, leukemic cells. Blood (2005) 106:376–83. doi:10.1182/blood-2004-12-4797
Schappe T, et al. Cytokine-induced memory-like natural killer cells exhibit 127. Fujisaki H, Kakuda H, Shimasaki N, Imai C, Ma J, Lockey T, et al. Expansion
enhanced responses against myeloid leukemia. Sci Transl Med (2016) of highly cytotoxic human natural killer cells for cancer cell therapy. Cancer
8:357ra123. doi:10.1126/scitranslmed.aaf2341 Res (2009) 69:4010–7. doi:10.1158/0008-5472.CAN-08-3712
108. Miller JS, Oelkers S, Verfaillie C, McGlave P. Role of monocytes in the 128. Garg TK, Szmania SM, Khan JA, Hoering A, Malbrough PA,
expansion of human activated natural killer cells. Blood (1992) 80:2221–9. Moreno-Bost A, et al. Highly activated and expanded natural killer cells
109. Rölle A, Pollmann J, Ewen E, Le VTK, Halenius A, Hengel H, et al. IL-12- for multiple myeloma immunotherapy. Haematologica (2012) 97:1348–56.
producing monocytes and HLA-E control HCMV-driven NKG2C+ NK cell doi:10.3324/haematol.2011.056747
expansion. J Clin Invest (2014) 124:5305–16. doi:10.1172/JCI77440 129. Szmania S, Lapteva N, Garg T, Greenway A, Lingo J, Nair B, et al. Ex vivo-
110. Rabinowich H, Sedlmayr P, Herberman RB, Whiteside TL. Increased pro- expanded natural killer cells demonstrate robust proliferation in vivo in high-
liferation, lytic activity, and purity of human natural killer cells cocultured risk relapsed multiple myeloma patients. J Immunother (2015) 38:24–36.
with mitogen-activated feeder cells. Cell Immunol (1991) 135:454–70. doi:10.1097/CJI.0000000000000059
doi:10.1016/0008-8749(91)90290-R 130. Lapteva N, Durett AG, Sun J, Rollins LA, Huye LL, Fang J, et al. Large-scale
111. Guven H, Gilljam M, Chambers BJ, Ljunggren HG, Christensson B, ex vivo expansion and characterization of natural killer cells for clinical
Kimby E, et al. Expansion of natural killer (NK) and natural killer-like applications. Cytotherapy (2012) 14:1131–43. doi:10.3109/14653249.2012.7
T (NKT)-cell populations derived from patients with B-chronic lymphocytic 00767
leukemia (B-CLL): a potential source for cellular immunotherapy. Leukemia 131. Shah NN, Baird K, Delbrook CP, Fleisher TA, Kohler ME, Rampertaap S,
(2003) 17:1973–80. doi:10.1038/sj.leu.2403083 et al. Acute GVHD in patients receiving IL-15/4-1BBL activated NK cells
112. Sutlu T, Alici E. Ex vivo expansion of natural killer cells: a question of following T cell depleted stem cell transplantation. Blood (2014) 125:784–92.
function. Cytotherapy (2011) 13:767–8. doi:10.3109/14653249.2011.563295 doi:10.1182/blood-2014-07-592881
113. Ahn Y-O, Kim S, Kim TM, Song EY, Park MH, Heo DS. Irradiated and activated 132. Zhang H, Cui Y, Voong N, Sabatino M, Stroncek DF, Morisot S, et al. Activating
autologous PBMCs induce expansion of highly cytotoxic human NK cells signals dominate inhibitory signals in CD137L/IL-15 activated natural killer
in vitro. J Immunother (2013) 36:373–81. doi:10.1097/CJI.0b013e3182a3430f cells. J Immunother (2011) 34:187–95. doi:10.1097/CJI.0b013e31820d2a21
114. Sakamoto N, Ishikawa T, Kokura S, Okayama T, Oka K, Ideno M, et al. Phase 133. Oyer JL, Igarashi RY, Kulikowski AR, Colosimo DA, Solh MM, Zakari A,
I clinical trial of autologous NK cell therapy using novel expansion method et al. Generation of highly cytotoxic natural killer cells for treatment of AML
in patients with advanced digestive cancer. J Transl Med (2015) 13:277. using feeder-free, particle based approach. Biol Blood Marrow Transplant
doi:10.1186/s12967-015-0632-8 (2015) 21:632–39. doi:10.1016/j.bbmt.2014.12.037
115. Kim E-K, Ahn Y-O, Kim S, Kim TM, Keam B, Heo DS. Ex vivo activation 134. Denman CJ, Senyukov VV, Somanchi SS, Phatarpekar PV, Kopp LM,
and expansion of natural killer cells from patients with advanced cancer Johnson JL, et al. Membrane-bound IL-21 promotes sustained ex vivo prolif-
with feeder cells from healthy volunteers. Cytotherapy (2013) 15:231–41.e1. eration of human natural killer cells. PLoS One (2012) 7:e30264. doi:10.1371/
doi:10.1016/j.jcyt.2012.10.019 journal.pone.0030264
116. Parkhurst MR, Riley JP, Dudley ME, Rosenberg SA. Adoptive transfer 135. Liu Y, Wu H-W, Sheard MA, Sposto R, Somanchi SS, Cooper LJN, et al.
of autologous natural killer cells leads to high levels of circulating natural Growth and activation of natural killer cells ex vivo from children with
killer cells but does not mediate tumor regression. Clin Cancer Res (2011) neuroblastoma for adoptive cell therapy. Clin Cancer Res (2013) 19:2132–43.
17:6287–97. doi:10.1158/1078-0432.CCR-11-1347 doi:10.1158/1078-0432.CCR-12-1243
117. Yang Y, Lim O, Kim TM, Ahn Y-O, Choi H, Chung H, et al. Phase I study 136. Oyer JL, Pandey V, Igarashi RY, Somanchi SS, Zakari A, Solh M, et al. Natural
of random healthy donor-derived allogeneic natural killer cell therapy killer cells stimulated with PM21 particles expand and biodistribute in vivo:

Frontiers in Immunology | www.frontiersin.org 16 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

clinical implications for cancer treatment. Cytotherapy (2016) 18:653–63. natural cytotoxicity receptors. J Immunol (2011) 187:6227–34. doi:10.4049/
doi:10.1016/j.jcyt.2016.02.006 jimmunol.1101640
137. Escudier B, Farace F, Angevin E, Charpentier F, Nitenberg G, Triebel F, et al. 155. Chabannon C, Mfarrej B, Guia S, Ugolini S, Devillier R, Blaise D, et al.
Immunotherapy with interleukin-2 (IL2) and lymphokine-activated natural Manufacturing natural killer cells as medicinal products. Front Immunol
killer cells: improvement of clinical responses in metastatic renal cell carci- (2016) 7:504. doi:10.3389/fimmu.2016.00504
noma patients previously treated with IL2. Eur J Cancer (1994) 30A:1078–83. 156. Lapteva N, Szmania SM, van Rhee F, Rooney CM. Clinical grade purification
doi:10.1016/0959-8049(94)90460-X and expansion of natural killer cells. Crit Rev Oncog (2014) 19:121–32.
138. Hercend T, Farace F, Baume D, Charpentier F, Droz JP, Triebel F, et al. doi:10.1615/CritRevOncog.2014010931
Immunotherapy with lymphokine-activated natural killer cells and recombi- 157. Kalos M, June CH. Adoptive T cell transfer for cancer immunotherapy
nant interleukin-2: a feasibility trial in metastatic renal cell carcinoma. J Biol in the era of synthetic biology. Immunity (2013) 39:49–60. doi:10.1016/j.
Response Mod (1990) 9:546–55. immuni.2013.07.002
139. Berg M, Lundqvist A, McCoy P Jr, Samsel L, Fan Y, Tawab A, et al. Clinical-grade 158. Rafiq QA, Thomas RJ. The evolving role of automation in process devel-
ex vivo-expanded human natural killer cells up-regulate activating receptors opment and manufacture of cell and gene-based therapies. Cell Gene Ther
and death receptor ligands and have enhanced cytolytic activity against Insights (2016) 2:473–9. doi:10.18609/cgti.2016.058
tumor cells. Cytotherapy (2009) 11:341–55. doi:10.1080/14653240902807034 159. Pierson BA, Europa AF, Hu WS, Miller JS. Production of human natural killer
140. Park KU, Jin P, Sabatino M, Feng J, Civini S, Khuu H, et al. Gene expression cells for adoptive immunotherapy using a computer-controlled stirred-tank
analysis of ex vivo expanded and freshly isolated NK cells from cancer patients. bioreactor. J Hematother (1996) 5:475–83. doi:10.1089/scd.1.1996.5.475
J Immunother (2010) 33:945–55. doi:10.1097/CJI.0b013e3181f71b81 160. Spanholtz J, Preijers F, Tordoir M, Trilsbeek C, Paardekooper J, de Witte
141. Childs RW, Berg M. Bringing natural killer cells to the clinic: ex vivo manip- T, et al. Clinical-grade generation of active NK cells from cord blood
ulation. Hematology Am Soc Hematol Educ Program (2013) 2013:234–46. hematopoietic progenitor cells for immunotherapy using a closed-system
doi:10.1182/asheducation-2013.1.234 culture process. PLoS One (2011) 6:e20740. doi:10.1371/journal.pone.
142. Granzin M, Soltenborn S, Müller S, Kollet J, Berg M, Cerwenka A, et al. Fully 0020740
automated expansion and activation of clinical-grade natural killer cells 161. Apel M, Brüning M, Granzin M, Essl M, Stuth J, Blaschke J, et al. Integrated
for adoptive immunotherapy. Cytotherapy (2015) 17:621–32. doi:10.1016/j. clinical scale manufacturing system for cellular products derived by magnetic
jcyt.2015.03.611 cell separation, centrifugation and cell culture. Chemie-Ingenieur-Technik
143. North J, Bakhsh I, Marden C, Pittman H, Addison E, Navarrete C, et al. (2013) 85:103–10. doi:10.1002/cite.201200175
Tumor-primed human natural killer cells lyse NK-resistant tumor targets: 162. Schüle S, Renner M, Longhurst S, Narayanan G. Regulatory requirements
evidence of a two-stage process in resting NK cell activation. J Immunol for clinical trial and marketing authorisation application for gene ther-
(2007) 178:85–94. doi:10.4049/jimmunol.178.1.85 apy medicinal products. Bundesgesundheitsblatt Gesundheitsforschung
144. Kottaridis PD, North J, Tsirogianni M, Marden C, Samuel ER, Gesundheitsschutz (2010) 53:30–7. doi:10.1007/s00103-009-0988-0
Jide-Banwo S, et al. Two-stage priming of allogeneic natural killer cells for the 163. Caligiuri MA, Zmuidzinas A, Manley TJ, Levine H, Smith KA, Ritz J. Functional
treatment of patients with acute myeloid leukemia: a phase I trial. PLoS One consequences of interleukin 2 receptor expression on resting human lym-
(2015) 10:e0123416. doi:10.1371/journal.pone.0123416 phocytes. Identification of a novel natural killer cell subset with high affinity
145. Peng B-G, Liang L-J, He Q, Huang J-F, Lu M-D. Expansion and activation receptors. J Exp Med (1990) 171:1509–26. doi:10.1084/jem.171.5.1509
of natural killer cells from PBMC for immunotherapy of hepatocellular 164. Romee R, Foley B, Lenvik T, Wang Y, Zhang B, Ankarlo D, et al. NK cell
carcinoma. World J Gastroenterol (2004) 10:2119–23. doi:10.3748/wjg.v10. CD16 surface expression and function is regulated by a disintegrin and
i14.2119 metalloprotease-17 (ADAM17). Blood (2013) 121:3599–608. doi:10.1182/
146. Harada H, Watanabe S, Saijo K, Ishiwata I, Ohno T. A Wilms tumor cell line, blood-2012-04-425397
HFWT, can greatly stimulate proliferation of CD56+ human natural killer 165. Peruzzi G, Femnou L, Gil-Krzewska A, Borrego F, Weck J, Krzewski K, et al.
cells and their novel precursors in blood mononuclear cells. Exp Hematol Membrane-type 6 matrix metalloproteinase regulates the activation-induced
(2004) 32:614–21. doi:10.1016/j.exphem.2004.03.011 downmodulation of CD16 in human primary NK cells. J Immunol (2013)
147. Fujisaki H, Kakuda H, Imai C, Mullighan CG, Campana D. Replicative 191:1883–94. doi:10.4049/jimmunol.1300313
potential of human natural killer cells. Br J Haematol (2009) 145:606–13. 166. Lajoie L, Congy-Jolivet N, Bolzec A, Gouilleux-Gruart V, Sicard E,
doi:10.1111/j.1365-2141.2009.07667.x Sung HC, et al. ADAM17-mediated shedding of FcγRIIIA on human
148. Cho D, Shook DR, Shimasaki N, Chang Y-H, Fujisaki H, Campana NK cells: identification of the cleavage site and relationship with activation.
D. Cytotoxicity of activated natural killer cells against pediatric solid J Immunol (2014) 192:741–51. doi:10.4049/jimmunol.1301024
tumors. Clin Cancer Res (2010) 16:3901–9. doi:10.1158/1078-0432.CCR- 167. Wiernik A, Foley B, Zhang B, Verneris MR, Warlick E, Gleason MK, et al.
10-0735 Targeting natural killer cells to acute myeloid leukemia in vitro with a CD16
149. Voskens CJ, Watanabe R, Rollins S, Campana D, Hasumi K, Mann DL. x 33 bispecific killer cell engager and ADAM17 inhibition. Clin Cancer Res
Ex-vivo expanded human NK cells express activating receptors that mediate (2013) 19:3844–55. doi:10.1158/1078-0432.CCR-13-0505
cytotoxicity of allogeneic and autologous cancer cell lines by direct recogni- 168. Fauriat C, Just-Landi S, Mallet F, Arnoulet C, Sainty D, Olive D, et al.
tion and antibody directed cellular cytotoxicity. J Exp Clin Cancer Res (2010) Deficient expression of NCR in NK cells from acute myeloid leukemia: evo-
29:134. doi:10.1186/1756-9966-29-134 lution during leukemia treatment and impact of leukemia cells in NCRdull
150. Lee DA. The off-target effects of nonspecific NK cells. Blood (2015) 125:744–5. phenotype induction. Blood (2007) 109:323–30. doi:10.1182/blood-2005-08-
doi:10.1182/blood-2014-12-616359 027979
151. Wang X, Lee DA, Wang Y, Wang L, Yao Y, Lin Z, et al. Membrane-bound 169. Le Maux Chansac B, Moretta A, Vergnon I, Opolon P, Lécluse Y,
interleukin-21 and CD137 ligand induce functional human natural killer Grunenwald D, et al. NK cells infiltrating a MHC class I-deficient lung ade-
cells from peripheral blood mononuclear cells through STAT-3 activation. nocarcinoma display impaired cytotoxic activity toward autologous tumor
Clin Exp Immunol (2013) 172:104–12. doi:10.1111/cei.12034 cells associated with altered NK cell-triggering receptors. J Immunol (2005)
152. Hercend T, Meuer S, Reinherz EL, Schlossman SF, Ritz J. Generation of a 175:5790–8. doi:10.1182/blood-2005-08-027979
cloned NK cell line derived from the “null cell” fraction of human peripheral 170. Carlsten M, Norell H, Bryceson YT, Poschke I, Schedvins K, Ljunggren H-G,
blood. J Immunol (1982) 129:1299–305. et al. Primary human tumor cells expressing CD155 impair tumor targeting
153. Lee DA. Regulatory considerations for NK cells used in human immuno- by down-regulating DNAM-1 on NK cells. J Immunol (2009) 183:4921–30.
therapy applications. Methods Mol Biol (2016) 1441:347–61. doi:10.1007/ doi:10.4049/jimmunol.0901226
978-1-4939-3684-7_29 171. Doubrovina ES, Doubrovin MM, Vider E, Sisson RB, O’Reilly RJ, Dupont B,
154. Sabry M, Tsirogianni M, Bakhsh IA, North J, Sivakumaran J, et al. Evasion from NK cell immunity by MHC class I chain-related mol-
Giannopoulos K, et al. Leukemic priming of resting NK cells is killer Ig-like ecules expressing colon adenocarcinoma. J Immunol (2003) 171:6891–9.
receptor independent but requires CD15-mediated CD2 ligation and doi:10.4049/jimmunol.171.12.6891

Frontiers in Immunology | www.frontiersin.org 17 April 2017 | Volume 8 | Article 458

Granzin et al. Antitumor Activity of Ex Vivo-Cultivated NK Cells

172. Hilpert J, Grosse-Hovest L, Grünebach F, Buechele C, Nuebling T, Raum T, glycosylphosphatidylinositol (GPI)-anchored NKG2D ligands. J Biol Chem
et al. Comprehensive analysis of NKG2D ligand expression and release in (2010) 285:8543–51. doi:10.1074/jbc.M109.045906
leukemia: implications for NKG2D-mediated NK cell responses. J Immunol 177. Kloess S, Huenecke S, Piechulek D, Esser R, Koch J, Brehm C, et al. IL-2-
(2012) 189:1360–71. doi:10.4049/jimmunol.1200796 activated haploidentical NK cells restore NKG2D-mediated NK-cell
173. Reiners KS, Topolar D, Henke A, Simhadri VR, Kessler J, Sauer M, et al. cytotoxicity in neuroblastoma patients by scavenging of plasma MICA. Eur
Soluble ligands for NK cell receptors promote evasion of chronic lymphocytic J Immunol (2010) 40:3255–67. doi:10.1002/eji.201040568
leukemia cells from NK cell anti-tumor activity. Blood (2013) 121:3658–65.
doi:10.1182/blood-2013-01-476606 Disclaimer: MG and VH were employed by company Miltenyi Biotec.
174. Schlecker E, Fiegler N, Arnold A, Altevogt P, Rose-John S, Moldenhauer G,
et al. Metalloprotease-mediated tumor cell shedding of B7-H6, the ligand Conflict of Interest Statement: All authors, including MG and VH, declare that
of the natural killer cell-activating receptor NKp30. Cancer Res (2014) they have no commercial, proprietary, or financial conflict of interest.
74:3429–40. doi:10.1158/0008-5472.CAN-13-3017
175. Ashiru O, Boutet P, Fernández-Messina L, Agüera-González S, Skepper JN, Copyright © 2017 Granzin, Wagner, Köhl, Cerwenka, Huppert and Ullrich. This
Valés-Gómez M, et al. Natural killer cell cytotoxicity is suppressed by is an open-access article distributed under the terms of the Creative Commons
exposure to the human NKG2D ligand MICA*008 that is shed by tumor Attribution License (CC BY). The use, distribution or reproduction in other forums
cells in exosomes. Cancer Res (2010) 70:481–9. doi:10.1158/0008-5472. is permitted, provided the original author(s) or licensor are credited and that the
CAN-09-1688 original publication in this journal is cited, in accordance with accepted academic
176. Fernández-Messina L, Ashiru O, Boutet P, Agüera-González S, Skepper practice. No use, distribution or reproduction is permitted which does not comply
JN, Reyburn HT, et al. Differential mechanisms of shedding of the with these terms.

Frontiers in Immunology | www.frontiersin.org 18 April 2017 | Volume 8 | Article 458