International Journal of

Molecular Sciences

Review
Natural Killer Cell-Derived Extracellular Vesicles as a
Promising Immunotherapeutic Strategy for Cancer: A
Systematic Review
Alvin Man Lung Chan 1,2 , Jin Min Cheah 2 , Yogeswaran Lokanathan 1 , Min Hwei Ng 1 and Jia Xian Law 1, *

1 Centre for Tissue Engineering and Regenerative Medicine, Faculty of Medicine,

Universiti Kebangsaan Malaysia, Kuala Lumpur 56000, Malaysia
2 Ming Medical Sdn Bhd, D3-3 (2nd Floor), Block D3 Dana 1 Commercial Centre, Jalan PJU 1a/22,
Petaling Jaya 47101, Malaysia
* Correspondence: lawjx@ppukm.ukm.edu.my; Tel.: +60-391-457677

Abstract: Cancer is the second leading contributor to global deaths caused by non-communicable
diseases. The cancer cells are known to interact with the surrounding non-cancerous cells, including
the immune cells and stromal cells, within the tumor microenvironment (TME) to modulate the
tumor progression, metastasis and resistance. Currently, chemotherapy and radiotherapy are the
standard treatments for cancers. However, these treatments cause a significant number of side effects,
as they damage both the cancer cells and the actively dividing normal cells indiscriminately. Hence, a
new generation of immunotherapy using natural killer (NK) cells, cytotoxic CD8+ T-lymphocytes or
macrophages was developed to achieve tumor-specific targeting and circumvent the adverse effects.
However, the progression of cell-based immunotherapy is hindered by the combined action of TME
and TD-EVs, which render the cancer cells less immunogenic. Recently, there has been an increase in
interest in using immune cell derivatives to treat cancers. One of the highly potential immune cell
derivatives is the NK cell-derived EVs (NK-EVs). As an acellular product, NK-EVs are resistant to
the influence of TME and TD-EVs, and can be designed for “off-the-shelf” use. In this systematic
review, we examine the safety and efficacy of NK-EVs to treat various cancers in vitro and in vivo.

Citation: Chan, A.M.L.; Cheah, J.M.;

Keywords: natural killer cells; extracellular vesicles; exosomes; immunotherapy; cancer
Lokanathan, Y.; Ng, M.H.; Law, J.X.
Natural Killer Cell-Derived
Extracellular Vesicles as a Promising
Immunotherapeutic Strategy for
Cancer: A Systematic Review. Int. J.
1. Introduction
Mol. Sci. 2023, 24, 4026. https:// According to the World Health Organization (WHO), the COVID-19 pandemic has
doi.org/10.3390/ijms24044026 claimed over 6.6 million lives globally [1]. Nonetheless, the incidence of infections and
deaths has declined significantly with the availability of vaccines. Contradictorily, solu-
Academic Editor: Wioletta Olejarz
tions for cancer treatment still remain obscure today. In 2021 alone, the WHO recorded
Received: 26 December 2022 9.3 million global deaths from neoplasms [2]. Chemotherapy and radiotherapy are the stan-
Revised: 30 January 2023 dard treatment for cancerous tumors which cannot be removed entirely through surgery,
Accepted: 3 February 2023 despite the accompanying adverse effects [3,4]. These drugs and radiations damage the
Published: 16 February 2023 actively proliferating cancer cells and dividing normal cell indiscriminately [5–8], leading
to deterioration in patient health by causing severe weight loss, poor skin conditions and
hair loss (alopecia) [9,10]. The adverse effects of chemotherapy and radiotherapy in weak-
ening patients’ health and immune system have yet to be resolved despite many years of
Copyright: © 2023 by the authors.
refinement. In fact, criticisms from patients receiving these treatments discourages many
Licensee MDPI, Basel, Switzerland.
This article is an open access article
from enrolling as their physical and mental wellbeing are at risk. The percentage of patients
distributed under the terms and
refusing to have or continue their chemotherapy is reported to be as high as 3–19% [11].
conditions of the Creative Commons
All these limitations demonstrate the importance of developing novel treatment for cancer
Attribution (CC BY) license (https:// which is highly specific and effective in eradicating the cancer cells without affecting the
creativecommons.org/licenses/by/ healthy cells.
4.0/).

Int. J. Mol. Sci. 2023, 24, 4026. https://doi.org/10.3390/ijms24044026 https://www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2023, 24, 4026 2 of 25

Stem cells have been studied extensively in the field of regenerative medicine due
to their excellent regenerative and anti-inflammatory properties [12]. They have been
regarded as a viable solution for many age-related and chronic degenerative illnesses. This
development has piqued the interest of utilizing stem cells for other therapeutic applica-
tions, including cancer immunotherapy. The development of cell-based immunotherapy
can be achieved by establishing cell lines that express tumor selectivity and cytotoxic func-
tions. Several prospective candidates that have been rigorously studied are natural killer
(NK) cells, dendritic cells (DC), macrophages and T-lymphocytes [13–16]. These cells play
critical roles in alerting the immune system (e.g., chemotaxis), arming other immune cells
(e.g., antibody-dependent and cytokine-dependent activation) and exerting their cytotoxic
effect against the cancer cells (e.g., receptor-mediated cell apoptosis). DC and CD4+ T-cells
(T-helper cells) are important components of the immune system in that they serve as the
moderators of various immune responses [17,18]. They produce molecular signals, e.g.,
cytokines and chemokines, to guide and activate other immune cells. Conversely, NK cells,
macrophages and CD8+ T-cells (cytotoxic T-cells) are effector cells that exert a cytotoxic
effect against the cancer cells [19–22].
The CD8+ T-cell’s mechanism of action is through the antibody-dependent cellular
cytotoxicity (ADCC). ADCC is initiated by the binding of T-cell receptors (TCR) to the
antigen presented by major histocompatibility complex-1 (MHC-1) [23]. NK cells also
can eliminate the target cells via ADCC pathway through the cluster of differentiation-16
(CD16) or FcγRIII on its plasma membrane [24,25]. Upon activation, these cells will release
cytolytic enzymes (e.g., caspases, granzymes (Gzm) and perforin (PFN)) to damage the
target cells. Meanwhile, macrophages elicit antibody-dependent cellular phagocytosis
(ADCP), which involves engulfment and degradation of the internalized cells via phago-
some acidification [26,27]. Since mutational activities are frequent in cancer cells, antigen
or peptide reconfigurations or direct inhibition of cytotoxic receptor (e.g., CA-125 inhibits
Fcy-receptor) can disable ADCC/ADCP entirely [27–31]. In the event ADCC and ADCP
are inhibited, the immune cells still can eliminate the cancer cells via the TRAIL and Fas
signalling pathways. TRAIL and Fas receptors are “death receptors” which belong to the
receptor family of tumor-necrosis factors (TNF) that induce programmed cell death when
they are cross-linked by its ligands [32–34]. In addition, certain immune cells also secrete
lytic granules containing Gzm and PFN that induce lysis or apoptosis of target cells via
degranulation [35–37].
In spite of its potential, the progression of cell-based immunotherapy has been ham-
pered by the limitations that it may cause life-threatening adverse reactions, laborious
production procedure, limited efficacy against solid tumors due to insufficient trafficking
and homing, and variable efficacy owing to immune cell inactivation by the tumors [38,39].
Recently, purification of cell biologics was made possible by the new innovations in isola-
tion technique. By using these advanced isolation technologies, researchers have isolated
the extracellular vesicles (EVs) produced by the immune cells and use them as cancer
immunotherapy. In that regard, immune cells have been repurposed as biological manu-
facturers of these membrane-bound nanovesicles [40]. NK cells-derived EV (NK-EV) is
being explored in many studies as it is rich in cytotoxic proteins, cytokines and miRNAs
that selectively kill the tumor cells [41]. In the interest of this development, this systematic
review was performed to determine the potential of NK-EVs as cancer immunotherapy
based on data collected from the in vitro and in vivo experiments. EVs can be classified
based on their biogenesis and size into exosomes (EXO), microvesicles (MVs) and apoptotic
bodies [42,43]. In this review, NK-EVs are referring to the EXO and MVs produced by the
NK cells.
Int. J. Mol. Sci. 2023, 24, 4026 3 of 25

2. Methods
Search keywords were selected using medical subject headings (MESH) available from
PUBMED/MEDLINE. Common terms not available in MESH were also included in the
search. Access to SCOPUS, PUBMED and Web of Science (WOS) was provided by the Na-
tional University of Malaysia (UKM). All searches were filtered to either “research articles”
or “journal articles” published in the last 5 years (2017–2022). All search registers were
downloaded as bibliographies containing title, keywords and abstract. The bibliographies
were labelled appropriately as source, date of access and results (e.g., PUBMED_210622_455
results). The bibliographies were then exported into Mendeley (Elsevier, The Netherlands).
The merging of duplicates was performed automatically and manually. The first level of
screening was based on relevance of title, abstract and keywords with topic of interest.
After that, full-text research articles were downloaded for second level of screening. The
final screening was performed following the inclusion and exclusion criteria as briefly
described. Inclusion criteria: (i) NK cell, (ii) EV or EXO, (iii) cancer and (iv) controlled
experimental studies. Exclusion criteria: (i) other immune cells, (ii) no EV or EXO, (iii) com-
bined therapy and (iv) uncontrolled experimental study. Subsequently, the in vivo studies
were reviewed for their risk of bias using SYRCLE’s ROB tool for animal studies. The
protocol above has been registered (ID: CRD42022339710) in the international prospective
register of systematic reviews (PROSPERO) by the National Institute for Health Research
(United Kingdom).
We recorded a combined total of 1002 registers from three databases: PUBMED (330),
SCOPUS (438) and WOS (234). A total of 340 registers were removed as a result of duplicate
merging, leaving only 662 individual registers. The first process of screening removed
631 registers based on the title, abstract and keywords. This yielded 31 suitable articles
for full-text retrieval. One article was not successfully retrieved since no English-text or
translation was available. A total of 30 retrieved full-text articles were screened using the
inclusion and exclusion criteria stated above. Finally, 18 individual registers containing
either in vitro only, in vivo only or both in vitro and in vivo evidence were selected for data
extraction and analysis. The article selection was performed by two authors (A.M.L.C and
J.M.C.) and any disputes in article inclusion were resolved via discussion to reach mutual
consensus. The article selection process is summarized in Figure 1.
Int. J. Mol.
Int. Sci.Sci.
J. Mol. 2023,
2023,24,
24, x4026
FOR PEER REVIEW 4 of 25

Figure
Figure 1. PRISMA
1. PRISMA flowfor
flow chart chart for systematic
systematic review. review.

3. Results and Discussion

3. Results and Discussion
3.1. NK-EVs Express Cytotoxicity against Various Human Cancer Cell Lines
3.1.A NK-EVs
total of 17 Express Cytotoxicity
studies performed against
in vitro Various
experiments to Human Cancer
examine the Cell
efficacy Lines
of NK-EVs
against Aa panel of tumor cell lines, and all of them reported positive
total of 17 studies performed in vitro experiments to examine the outcomes (Table 1). efficac
The NK-EVs were found to be effective against a wide range of cancer cell lines, shown
EVs against a panel of tumor cell lines, and all of them reported positive outcom
in Figure 2 below, such as the brain [44–51], breast [44,45,47,49–55], blood [44,48,53,55–57],
1). The
cervix [57],NK-EVs were
colon [50,51], found
liver to be
[45,49,58], effective
lung against
[57], ovary [51,54],apancreas
wide range of cancer
[59], prostate [51],cell line
in [51,60],
skin Figurestomach/gastric
2 below, such[50,60], as theand brain [44–51],
thyroid [45,49].breast [44,45,47,49–55],
The dose of NK-EVs usedblood in [44
these studies ranged from low dose, between 0.3 − 5 µg, to median dose, between
57], cervix [57], colon [50,51], liver [45,49,58], lung [57], ovary [51,54], pancreas [ 10 − 25 µg,
and high dose, between 40−100 µg. The NK-EVs were found to kill the cancer cells in
tate [51], skin [51,60], stomach/gastric [50,60], and thyroid [45,49]. The dose of
a time- and dose-dependent manner [44–60]. Hence, higher cytotoxicity was recorded
used
when theincancer
thesecellstudies ranged
lines were fromtolow
exposed higherdose,
dosebetween
of NK-EVs 0.3−5
for aµg, to median
longer period. dose
10−25 µg,the
Importantly, and high showed
NK-EVs dose, between 40−100 µg.
selective cytotoxicity The NK-EVs
towards the cancerwere found
cell lines with-to kill t
cells
out in a time-
affecting and dose-dependent
the viability of the healthy or normal manner [44–60].
cell lines tested.Hence, higherclearly
This evidence cytotoxicity
showed that NK-EVs exhibited selectivity akin to their parental
orded when the cancer cell lines were exposed to higher dose of NK-EVs cells, as reported in the for
previous literature [61].
period. Importantly, the NK-EVs showed selective cytotoxicity towards the ca
lines without affecting the viability of the healthy or normal cell lines tested. This
clearly showed that NK-EVs exhibited selectivity akin to their parental cells, as
in the previous literature [61].

Table 1. In vitro evidence demonstrating cytotoxicity of NK-EVs against various cancer c

Dose of
Study Type of NK-EV Tested Cancer Cell Lines Key Findings
NK-EV
Human acute lymphoblastic
Int. J. Mol. Sci. 2023, 24, 4026 5 of 25

Table 1. In vitro evidence demonstrating cytotoxicity of NK-EVs against various cancer cell lines.

Dose of
Study Type of NK-EV Tested Cancer Cell Lines Key Findings
NK-EV
Human acute
lymphoblastic leukemia
Jong et al., NK-EVs from NK-EVs increased apoptosis of all
(NALM-6; SUPB15),
2017 PB-NK of healthy 20 or 40 µg tumor cell lines in a time- and
neuroblastoma (CHLA-136;
[44] donors dose-dependent manner.
CHLA-255) and breast
cancer (MCF-7)
All cell lines showed lower cell
viability in a time and
dose-dependent manner.
Mouse melanoma (B16F10);
Zhu et al., NK-EXO from NK-EXO did not illicit any response
Human gastric carcinoma
2017 human NK cell line 5 or 20 µg in healthy cells.
(SNU484) and colon cancer
[60] (NK92-MI) NK-EXO induced cancer cell
(HCT-5)
apoptosis through PFN and Gzm as
well as activation of Fas/FasL
pathway.
Human glioblastoma (D54), NK-EM showed greater anti-tumor
NK-EXO and breast carcinoma properties compared to NK-EXO.
Zhu et al.,
NK-EM from (MDA-MB-231), anaplastic Both treatment groups reduced BLI
2018 10, 20 or 30 µg/mL
human NK cell line thyroid cancer (CAL-62) signal intensity for all tested tumor
[45]
(NK92-MI) and hepatic carcinoma cell lines in a time and
(HepG2) dose-dependent manner.
IL-15 treated NK cells and its
exosomes successfully halted
growth of neuroblastoma cell lines.
Inactivation of NK cells via TGFβ1
Human MYCN-amplified
miR-186 enriched A series of 2-fold did not affect the functions of the
Neviani et al., (CHLA-136 and LAN-5)
NK-EVs from dilutions starting secreted exosomes despite
2019 and non-amplified
PB-NK of healthy from 4 × 1011 downregulation of cytotoxic
[46] (CHLA-255)
donors particles/mL proteins.
neuroblastoma
miR-186 delivery via NK-EXO
reduced expression of tumor escape
oncogenes, i.e., MYCN, AURKA,
TGFβ1R and TGFβ2R.
miR-3607-3p
Sun et al., NK-EVs inhibited growth, migration
enriched NK-EVs Human pancreatic cancer
2019 − and invasive properties of both
from PB-NK of (MIA PaCa-2 and PANC-1)
[59] cancer cell lines.
healthy donors
NK-EXOs reduced the viability of
NK-EXO and Human breast cancer both tumor cell lines in a
Wang et al.,
NN/NK-EXO from (MDA-MB-231) and dose-dependent manner.
2019 10, 20 or 40 µg
PB-NK of healthy neuroblastoma NN/NK-EXOs showed higher levels
[47]
donors (CHLA-255) of tumor cytotoxicity compared to
NK-EXOs.
NK-EVs were cytotoxic towards
both cancer cell lines.
PFN, GzmA, GzmB and GNLY were
NK-EVs from
Human acute found to work collectively to induce
PB-NK of healthy
Wu et al., 2019 lymphoblastic leukemia tumor cytotoxicity.
donors and human 40 µg/100 µL
[48] (SUPB15) and NK-EVs were able to enter
NK cell line
neuroblastoma (CHLA255) caspase-dependent and independent
(NK92-MI)
pathways, highlighting the
flexibility of NK-EVs to access
multiple cytotoxic pathways.
Int. J. Mol. Sci. 2023, 24, 4026 6 of 25

Table 1. Cont.

Dose of
Study Type of NK-EV Tested Cancer Cell Lines Key Findings
NK-EV
Both NK-EVIL-15 and NK-EVs
NK-EVs from Human breast cancer
displayed dose- and time-dependent
Zhu et al., human NK cell line (MDA-MB-231), anaplastic
cytotoxicity against all cancer cell
2019 (NK-92MI) with or 5, 10 or 15 µg thyroid cancer (CAL-62)
lines.
[49] without IL-15 and glioblastoma
NK-EVIL-15 had greater cytotoxic
treatment (U87/MG)
effect compared to NK-EVs.
Human hepatocarcinoma
(HEPG2), colon cancer
Choi et al., NK-EVs from
(SW-620), stomach cancer NK-EVs were cytotoxicity against all
2020 PB-NK of healthy 2, 5, 10 or 20 µg
(MKN-74), breast cancer cancer cell lines.
[50] donors
(MCF-7) and brain cancer
(T98G).
NK-EXO from IL-2 Human childhood B acute
Di Pace et al., NK-EXOs exerted cytotoxicity
or IL-15 stimulated 5, 20 or 50 µg/ lymphoblastic leukemia
2020 against both cancer cell lines in a
PB-NK of healthy 100 µL (NALM-18) and
[56] dose-dependent manner.
donors erythroleukemia (K562)
NK-EXOs were as cytotoxic as PTX
NK-EXO and 40 µg/mL of against the breast cancer cell line.
Han et al.,
PTX-NK-EXO from NK-EXO or Human breast cancer NK-EXOs showed great potential as
2020
human NK cell line 15 µg/mL of (MCF-7) cancer drug carriers with
[52]
(NK92-MI) PTX-NK-EXO PTX-NK-EXOs showed the highest
tumor cytotoxicity.
NK-EVs showed cytotoxicity against
NK-EVs from Human T cell leukemia both cancer cell lines in a time- and
Cochran et al.,
human NK cell 1, 10, 25, 50 (K562 and JURKAT) and dose-dependent manner.
2021
lines (NK3.3 and or 100 µg/mL breast cancer (HEK293, NK-EVs did not trigger apoptosis in
[53]
NK92-MI) MCF-7 and MDA-MB-231) normal cells, i.e., HEK293 and PB-
and CB-derived lymphocytes.
NK-EVs exerted time- and
Human T cell leukemia
Enomoto NK-EV from dose-dependent cytotoxic towards
(K562 and JURKAT), lung
et al., 2021 human NK cell line 0.3, 1 or 3 µg all the cancer cell lines albeit the
(A549) and cervical cancer
[57] (NK92-MI) dosage is much lower compared to
(HELA)
other studies.
NK92-EXO from
NK cells in hypoxic culture (24 and
human NK cell
48 h) doubled its EV production.
Jiang et al., lines (NK92-MI and Human breast cancer
Hypoxic and normoxic NK-EVs
2021 NK92-hIL-15) 25 or 50 µg/mL (MCF-7) and ovarian
demonstrated similar degree of
[54] cultured in either cancer (A2780)
cytotoxic towards the cancer cell
normal or hypoxic
lines.
condition
Transduced cells produced EVs
NK-EXO from Human T cell leukemia
enriched with BCL-2 siRNAs.
Kaban et al., human NK cell line (K562 and MEC1) and
Modified NK-EVs showed greater
2021 (NK92-MI) 200 µg/mL breast cancer (HEK293T,
cytotoxicity against the breast cancer
[55] overexpressing SKBR3, MCF-7, MCF-10A,
cell lines but have no effect on the
BCL-2 siRNAs T-47D and MDA-MB-231)
normal cells.
NK cell line (NK92-MI)
cervical cancer (HELA) dosage is much lower compared to other studies.
NK92-EXO from hu-
NK cells in hypoxic culture (24 and 48 h) dou-
man NK cell lines
bled its EV production.
(NK92-MI and NK92- 25 or 50 Human breast cancer (MCF-7)
Int. J. Mol.in
Sci.ei-
2023, 24, 4026 Hypoxic and normoxic NK-EVs demonstrated 7 of 25
hIL-15) cultured µg/mL and ovarian cancer (A2780)
similar degree of cytotoxic towards the cancer
her normal or hypoxic
cell lines.
condition
Table 1. TCont.
Human cell leukemia (K562 Transduced cells produced EVs enriched with
NK-EXO from human
and MEC1) and breast cancer BCL-2 siRNAs.
NK cell line (NK92-MI) Dose of
Study 200Type
µg/mLof NK-EV
(HEK293T, SKBR3, MCF-7, Tested CancerNK-EVs
Modified Cell Lines
showed greater Key Findings
cytotoxicity
overexpressing BCL-2 NK-EV
MCF-10A, T-47D and MDA- against the breast cancer cell lines but have no ef-
siRNAs Human colon cancer
NK-EVs from IL-15 MB-231) (HCT116 and fect on the normal
HCT-15), cells.isolated from IL-15 and
NK-EVs
Human colon cancer (HCT116 prostate cancer (DU145
or IL-12/15/18- IL-12/15/18-stimulated NK cells
NK-EVs from IL-15 or
Aarsund et al., stimulated PB-NK
and HCT-15), prostate cancer and PC3), breast
NK-EVs cancer
isolated and NK-92
from IL-15 cells were able to kill the
and IL-12/15/18-
L-12/15/18-stimulated
2022 of healthy donors 20 µg (SK-BR-3 and T-4D7), cancer cells in 2D andtospheroid
(DU145 and PC3), breast cancer stimulated NK cells and NK-92 cells were able
PB-NK of healthy [51]do- and human NK cell ovarian cancer (OVCAR-3), cultures.
20 µg (SK-BR-3 and T-4D7), ovarian kill the cancer cells in 2D and spheroid cultures.
nors and human NK line (NK92 and leukemia (KHYG-1), KHYG-1 EVs showed no tumor
KHYG-1) cancer (OVCAR-3), leukemia melanoma KHYG-1 (WM9)
EVs showed
and nocytotoxicity
tumor cytotoxicity
in both in in
vitro models.
cell line (NK92 and
(KHYG-1), melanoma (WM9) glioblastoma (U87). both in vitro models.
KHYG-1)
and glioblastoma (U87). NK-EXOs had a cytotoxicity
Kim et al., NK-EXO from Human hepatocellular
10, 20, 50, Human hepatocellular carci-
10, 20, 50, 100, 200 NK-EXOs had a cytotoxicitytowardstowards
Hep3B Hep3B compared to HepG2
NK-EXO from 2022 human human NK cell line
or 500
carcinoma (Hep3B, HepG2
µg and compared to HepG2 and Huh7 and Huh7 cells indose-de-
a dose-dependent
[58] 100, 200 or
(NK-92) noma (Hep3B, HepG2 and Huh7) cells in a
NK cell line (NK-92) but not time-dependent manner.
500 µg Huh7) pendent but not time-dependent manner.

Figure 2. Distribution of in-vitro

Figure studiesoffor
2. Distribution (A) type
in-vitro of cancer
studies for (A)cell
typeline studied
of cancer celland
line (B) tested
studied doses
and (B) tested doses
of NK-EVs for cancer cell cytotoxic
of NK-EVs assay.
for cancer cell cytotoxic assay.

The impressive
The impressive findings findings
from the fromstudies
in vitro the in vitro studies
merely merely demonstrated
demonstrated the fundamen-
the funda-
mental anti-cancer potential of NK-EVs without the interference of other biological sys- systems.
tal anti-cancer potential of NK-EVs without the interference of other biological
Thus, the use of mere in vitro data to show the anti-cancer potential of NK-EVs is totally
tems. Thus, the use of mere in vitro data to show the anti-cancer potential of NK-EVs is
insufficient, as the model does not replicate the physiological complexities that make up a
totally insufficient, as the model does not replicate the physiological complexities that
living organism [62]. For more effective translation, a highly defined and complex model of
make up a living pharmacodynamics
organism [62]. For(PD) moreandeffective translation,
pharmacokinetics (PK)a highly
is needed,defined
such asand
the com-
one constructed
plex model of pharmacodynamics (PD) and pharmacokinetics (PK) is needed,
by Bouhaddou et al. (2020) [63]. In term of cancer, the main debacle that hassuch as the
stumped the
one constructed by Bouhaddou
progression et al.medical
of many (2020) innovations
[63]. In termis of
thecancer,
existencetheof main debacle that
drug resistance, which is very
has stumped the difficult
progression of many
to replicate medical
in vitro innovations
[64]. Tumor cells areis the existence
known to interactof drug
with there-
surrounding
sistance, which is very difficult to replicate in vitro [64]. Tumor cells are known to interact survival
cells and tissues, also known as tumor microenvironment (TME), to ensure their
and the
with the surrounding development
cells and tissues,ofalso
drug resistance.
known The TME
as tumor provides important
microenvironment support
(TME), to needed
by the tumor to thrive, including the secretion of signalling molecules by the proximally
ensure their survival and the development of drug resistance. The TME provides im-
located cells, structural support and biochemical signals granted by the altered extracel-
portant support needed by the tumor to thrive, including the secretion of signalling
lular matrices (ECM), protection by the immunosuppressive local immune cells, nutrient
supply and waste removal by the sprouting blood vessels, and a dynamic microenviron-
ment (e.g., optimal temperature and pH level) that favor cancer cell growth, invasion and
metastasis [65–68].
 sels, and a dynamic microenvironment (e.g., optimal temperature and pH level) that favor
cancer cell growth, invasion and metastasis [65–68].

3.2. Tumor Microenvironment (TME): A Physicochemical Barrier against Immunotherapy

Int. J. Mol. Sci. 2023, 24, 4026 8 of 25
The TME is described as an extremely hostile ecosystem consisting of both physical
and chemical components, as illustrated in Figure 3 [69,70]. Similar to all other cell types,
cancerous cells are no
3.2.exceptions to the innate
Tumor Microenvironment ability
(TME): to secrete EVs.Barrier
A Physicochemical In fact, tumor-derived
against Immunotherapy
extracellular vesicles (TD-EVs) are expressed in significant quantities, making
The TME is described as an extremely hostile ecosystem consisting of both physical
them a
plausible biologicaland
marker andcomponents,
chemical progress indicator for cancer
as illustrated patients
in Figure [71–73].
3 [69,70]. SimilarCancer cells
to all other cell types,
are also more self-sufficient than
cancerous cells areregular cells to
no exceptions since they ability
the innate are capable ofEVs.
to secrete producing and
In fact, tumor-derived
responding to theirextracellular
own growth factors
vesicles [74]. As
(TD-EVs) are demonstrated by El-Fattah
expressed in significant Ibrahim
quantities, making etthem
al. a plau-
sible biological marker and progress indicator for cancer
(2019), these tumor-enabling secretions can operate in an autocrine (own cells), paracrine patients [71–73]. Cancer cells
are also more self-sufficient than regular cells since they are capable of producing and
(neighboring cells) and endocrine (distally located cells) manner [75]. TD-EVs were
responding to their own growth factors [74]. As demonstrated by El-Fattah Ibrahim et al.
known to alter the(2019),
surrounding microenvironment,
these tumor-enabling secretions canmaking
operate it in
hostile for immune
an autocrine cellsparacrine
(own cells),
and other healthy tissue, but their
(neighboring cells)utility and mechanism
and endocrine of action
(distally located cells) were
manner never fully com-
[75]. TD-EVs were known
prehended. However, they
to alter the do share a resemblance
surrounding microenvironment, to anti-inflammatory
making it hostile for secretions
immune cellsby and other
healthy tissue,
supporting cell proliferation andbut their utility and
angiogenesis, mechanism
while of action
inhibiting were nevermaturation
cell apoptosis, fully comprehended.
However,
or differentiation, and they do share
suppressing a resemblance
the recruitment oftoinflammatory-responsive
anti-inflammatory secretionscells by supporting
(e.g., cell
proliferation and angiogenesis, while inhibiting cell apoptosis, maturation or differenti-
NK cells, macrophages, B and T lymphocytes) [76]. When healthy cells are replaced or
ation, and suppressing the recruitment of inflammatory-responsive cells (e.g., NK cells,
destroyed, the surrounding
macrophages, environment is contaminated
B and T lymphocytes) [76]. Whenby healthy
the metabolic waste prod-
cells are replaced or destroyed,
ucts, inflammasomes the from cell lysis,
surrounding and cellisdebris.
environment Activebyglycolysis
contaminated the metabolic in waste
malignant cells
products, inflamma-
consumes the surrounding
somes from oxygen and
cell lysis, and releases acidic
cell debris. products
Active glycolysis (e.g., pyruvatecells
in malignant andconsumes
hy- the
surrounding
drogen ions) [77]. After depletingoxygentheand releases acidic
surrounding products
oxygen, (e.g., pyruvate
cancer cells willand hydrogen
undergo an-ions) [77].
aerobic glycolysis, After
which depleting the surrounding
contributes to furtheroxygen, cancer cells
acidification via will undergo
lactic anaerobic glycolysis,
acid production
which contributes to further acidification via lactic acid production [78]. These manifest
[78]. These manifest into the well-known acidic and hypoxic properties of TME, forming
into the well-known acidic and hypoxic properties of TME, forming a chemical barrier that
a chemical barrier that
limitslimits
immune immune cell penetration.
cell penetration.

Figure 3. The tumorigenic

Figure cells
3. Theare centeredcells
tumorigenic around faultyaround
are centered extracellular matrices,matrices,
faulty extracellular necroticnecrotic
bodies,bodies, and
and a hypoxic and acidified
a hypoxicmicroenvironment. This tumorThis
and acidified microenvironment. microenvironment
tumor microenvironmentdiscourages
discouragesimmu-immunoreac-
noreactive cells fromtive
responding or otherwise
cells from responding enteringentering
or otherwise the tumor to eradicate
the tumor thethe
to eradicate tumor
tumorcells.
cells. At
At the same
the same time, the hypoxic and
time, the acidicand
hypoxic environment will induce
acidic environment cell death
will induce and exacerbate
cell death and exacerbatethe the
inflam-
inflammation.
mation. Moreover, tumor-derived extracellular
Moreover, tumor-derived vesicles
extracellular will
vesicles willreprogram
reprogram the the host’s
host’s cellsendothelial
cells (e.g., (e.g., cells)
to support tumor development via increasing the secretion of growth factor, pro-angiogenic and
anti-inflammatory cytokines.
Int. J. Mol. Sci. 2023, 24, 4026 9 of 25

What is also seemingly peculiar about TD-EVs is the manipulation of host cells, akin
to a virus. TD-EVs are able to influence or reprogram the recipient cells to cooperate with
tumors [79]. For example, TD-EVs can falsely trigger the anti-inflammatory machinery
of the healthy cells to reduce the TME inflammation. On the other hand, death cells in
TME will continuously release inflammatory particles. Hence, TME paradoxically houses
both anti-inflammatory particles and pro-inflammatory particles. How these opposing
factors co-exists and modulate the TME inflammation is yet to be fully understood. In
addition, studies have shown that the circulating inflammatory bodies passively recruit
immunoreactive cells to the tumor region [80–83]. The immune cells recruited to the TME
will be modulated by the TD-EVs to deactivate its tumor cell killing function and/or to
maintain its status quo via overstimulation of the inhibition:activation signal ratio [84]. At
the same time, the aggregation of naïve/resting or inactivated immune cells interlocked
by faulty ECMs obstructs any movement and acts as a shield against activated immune
cells or medical interventions (e.g., chemotherapy drugs). Thus, the formation of this
pseudo-barrier reinforces the physical impenetrability of the tumor, contributing to drug
resistance.
TME and TD-EVs are known to work in harmony to disrupt NK cell function, thus,
reducing the effectiveness of NK cell therapy. The acidic pH, as well as the immunosup-
pressive myeloid derived suppressor cells’ (MDSCs) and M2 macrophages’ presence in the
TME, suppress the activation of NK cells, thereby compromising their cytotoxicity against
a range of tumor cells [85]. Similarly, the TD-EVs were found to impair the function of
NK cells through the transfer of multiple immunosuppressive factors (e.g., miRNAs and
TGF-β) [86,87]. The existence of physical and chemical barriers in TEM paired with the
inhibitory factors from TD-EVs render the NK cells and other immune cells almost entirely
incompetent. Cellular release and uptake of EVs are known to increase in acidic environ-
ments [88]. Additionally, the pH-related stress also increases the EVs’ protein content and
surface electrokinetic potential (zeta potential) [89]. The EVs with higher surface charges
can bind strongly to the cell membrane, therefore increasing its internalization via receptor
or non-receptor endocytosis. The higher EV absorption is not restricted to TD-EVs, but
applicable to other EVs as well. This implies the possibility of higher absorption of NK-EVs
and other immune cell-derived EVs by the cancer cells.

3.3. NK-EVs Show In Vivo Cytotoxicity in Tumor-Bearing Mice

The in vitro to in vivo extrapolation has seen numerous failures due to subpar experi-
mental design and the presence of many limitations that yet to be addressed [62]. Thus,
preclinical study needs to be performed to extrapolate animal data to humans despite that
it is also unreliable due to species differences. Table 2 describes studies that assess the
performance of NK-EV therapy using the in vivo model of tumor-bearing animals and the
range of doses per administration [45–47,50,53,60,90]. The four studies shown in Figure 4
that performed intra-tumoral (IT) infusion reported significant suppression (p < 0.05) of
tumor growth and lowered cancer cell viability in the neoplastic mass [45,53,60,90]. Similar
results were reported in the seven studies that performed intravenous (IV) delivery of
NK-EVs through the tail vein [45–47,49,50,58,90]. In all these studies, NK-EV infusion
significantly (p < 0.05) reduced the bioluminescent intensity (BLI) signals and/or physical
dimensions of the tumor mass compared to the untreated animals. The study by Zhu et al.
(2018) performed a direct comparison between IT and IV administration routes in their
animal models [45]. They found that IT administration is more effective (p < 0.05) compared
to IV administration in shrinking the tumor mass. Similar results have been reported in
previous studies examining the anti-tumor effect of immunotherapies in tumor-bearing
mice [91,92]. These findings indicate that the IT route is likely to be more efficacious for
NK-EV therapy but the IV route is still viable and may be used in specific situations when
the IT route is not accessible.
Int. J. Mol. Sci. 2023, 24, 4026 10 of 25

Table 2. In vivo evidence demonstrating the antitumor potential of NK-EVs.

Type of Cancer and Animal Model Dosage and Method of

Study Key Findings
(n = Sample Size per Group) Administration
B16F10/effluc cells (1 × 105 cells/ Tumor was effectively reduced
Zhu et al., 100 µL) were subcutaneously injected (3.5 folds) after 2 to 5 days.
20 µg/100 µL of human
2017 into the right thigh of pathogen-free In vivo and ex vivo BLI confirmed
NK-EXO via IT
[60] 6-week-old female C57BL/6 mice reduced signal intensity and reduced
(n = 6) tumor mass in the treatment group.
D54 cells (5 × 106 cells/100 µL) were
100 µg/150 µL (via IV) and
Zhu et al., subcutaneously injected into Both treatments reduced the tumor
30 µg/50 µL (via IT) of human
2018 pathogen-free 6-week-old female mass with IT route showed greater
NK-EM, 3 times at intervals of
[45] BALB/c mice reduction.
3 days
(n = 5)
BLI signal intensity of tumor cells and
CHLA-136-Fluc cells (1 × 106 cells)
weight of kidneys were decreased in
Neviani were intra-renally injected into left 1 mg/kg/d of miR-186
the animals received miR-186-enriched
et al., 2019 kidney of 4- to 8-week-old female and enriched human NK-EVs via
NK-EVs compared to the control group.
[46] male NSG mice IV 3 times per week
The treated animals showed improved
(n = 5 − 10)
survival rate.
CHLA-255-luc cells (1 × 107 cells/
100 µg of human NK-EXO NN/NK-EXOs showed better homing
Wang et al., 500 µL) were injected via IV into
with or without NN let-7a efficacy and greater suppression of
2019 specific pathogen-free, 6-week-old
loaded polyamidoamine tumor growth compared to the
[47] female NOD/SCID mice
dendrimer via IV route NK-EXOs.
(n = 3)
Both treatments significantly reduced
BLI signals and tumor weight
U87/MG/F cells were administered 50 µg of human NK-EVs or
Zhu et al., compared to the control group.
into specific pathogen-free, 6-week-old 50 µg of human NK-EVsIL-15
2019 NK-EVsIL-15 were significantly more
female BALB/c nude mice via IV for 5 times at intervals
[49] effective compared to NK-EVs. Both
(n = 15) of 2 days
treatments did not elicit toxic response
in tumor-bearing animal model.
MCF-7 cells (3 × 106 cells) were injected
Choi et al., The tumor dimension and mass
subcutaneously into the right flank of 50 µg/100 µL of human
2020 reduced significantly compared to the
5-week-old, female athymic nude mice NK-EVs via IV 3 times weekly
[50] control group after 2 weeks.
(n = 4)
GFP-expressing MDA-MB-231 cells
Cochran (2 × 106 cells) were injected into the 4th 50 µg of human NK3.3-EVs The NK3.3-EV treated animals showed
et al., 2021 mammary fat pads of female athymic via IT, 7 times at intervals of 3 a higher number of dead cells both
[53] nude mice to 4 days histologically and in TUNEL assay.
(n = 4–5)
Hep3B cells (1 × 107 cells/100 µL) was
NK-EXO exerted migratory and
subcutaneously (right back) or
Kim et al., 50, 100, 200 or 500 µg of targeting ability to inhibit tumor
(2 × 106 cells/50 µL) orthotopically
2022 human NK-EXO via IV 6 growth in dose-dependent manner in
(liver) xenografted into 6-week-old
[58] times at intervals of 2 days both subcutaneous and orthotopic
male BALB/c nude mice
animal models.
(n = 5)
Canine REM134 cells (1 × 104 cells)
Lee et al., 100 µg of canine NK-EXO via NK-EXO suppressed tumor growth and
xenografted into mammary fat pad of
2021 IT once and IV twice per week reduced the expression of
BALB/c nude mice via IT route
[90] for 6 weeks tumor-associated markers.
(n = n/a)
Abbreviations: BLI—Bioluminescent imaging; IT—Intra-tumoral; IV—Intravenous; NK-EM—Natural Killer
cell-derived exosome-mimetic vesicle; NSG—NOD-SCID gamma mice; PBS—Phosphate buffered saline.
023, 24, x Int.
FORJ. PEER REVIEW
Mol. Sci. 2023, 24, 4026 10 of 24 11 of 25

Figure 4. Distribution of in-vivo

Figure studiesof
4. Distribution for (A) type
in-vivo of cancer
studies for (A) studied, (B) administration
type of cancer route of
studied, (B) administration route of
NK-EVs, (C) tested NK-EVs,
dose of NK-EVs
(C) testedper
dosedose and the
of NK-EVs (D)
per frequency
dose of frequency
and the (D) dose administered through- throughout
of dose administered
out study phase. Numbers indicated in (A,B) refers to the frequency of reviewed studies that fall
study phase. Numbers indicated in (A,B) refers to the frequency of reviewed studies that fall into the
into the categories. categories.

The SYRCLE’s risk TheofSYRCLE’s

bias tool adopted
risk of bias from
toolHooijman
adopted from 2014Hooijman
was used2014 to assess the to assess
was used
risk of bias of the the risk ofstudies
selected bias of (Figure
the selected studies
5) [93]. (Figure most
Generally, 5) [93].of Generally,
the elementsmostevalu-
of the elements
evaluated risk
ated showed an unknown showed an unknown
of bias. While most riskofof the
bias.studies
Whiledemonstrated
most of the studies demonstrated
low risk
of bias for some of the elements investigated, a few studies were found to have high risk. to have
low risk of bias for some of the elements investigated, a few studies were found
high risk. Among the nine studies reviewed, five studies (55.6%) explicitly mentioned
Among the nine studies reviewed, five studies (55.6%) explicitly mentioned randomiza-
randomization [45–47,50,58] but three studies (33.3%) did not [49,53,60]. A single study
tion [45–47,50,58] (11.1%)
but threewasstudies
deemed(33.3%)
high-risk didduenotto[49,53,60].
undisclosedAsample singlesize
study
and(11.1%) was
allocation method [90].
deemed high-riskThe duebaseline
to undisclosed sampleprior
was presented sizetoand allocation
treatment in 6method [90]. The
studies (66.7%) baseline
but ambiguous in the
was presented prior to treatment
remaining in 6 studies
three studies (33.3%). (66.7%)
None of the butstudies
ambiguous in the
disclosed remaining
information for allocation
three studies (33.3%). None ofrandom
concealment, the studies
housing, disclosed
blindinginformation
of interventionsfor or
allocation conceal-
caregivers, random outcome
assessment,
ment, random housing, and blinding
blinding of outcome
of interventions or assessment.
caregivers, Nearly
randomalloutcome
studies (nassess-
= 8, 88.9%) have
unknown risk of bias for selective outcome reporting
ment, and blinding of outcome assessment. Nearly all studies (n = 8, 88.9%) have unknown bias, with one study has high-risk of
bias [50]. Five studies (55.5%) reported low risk of attrition
risk of bias for selective outcome reporting bias, with one study has high-risk of bias [50]. bias and three (33.3%) had an
unclear risk, with one (11.1%) having high-risk [50]. The unclear risk is mainly attributed
Five studies (55.5%) reported low risk of attrition bias and three (33.3%) had an unclear
to the inconsistency in the sample size of the study’s outcomes with the number of animals
risk, with one (11.1%) having high-risk [50]. The unclear risk is mainly attributed to the
allocated per group. Failure to disclosure the number of animals (alive or dead) excluded
inconsistency in the sample size ofor
from the analysis the study’s outcomes
parameter conductedwith werethealsonumber
took intoofaccount.
animalsAllallo-
studies were
cated per group. at Failure
low risktoofdisclosure the number
“other sources of bias” by ofdisclosure
animals of (alive
ethicsorapproval,
dead) excluded
potential conflict of
from the analysis or parameter
interest, or the conducted
collaborating were also
parties andtook into
their account. All studies were at
roles.
low risk of “other sources of bias” by disclosure of ethics approval, potential conflict of
interest, or the collaborating parties and their roles.
nt. J. Mol.Int.
Sci.J. 2023,
Mol. Sci.
24,2023,
x FOR24, 4026
PEER REVIEW 12 of 25 11 of 24

Figure 5. SYRCLE’s
Figure 5. SYRCLE’srisk
riskof
of bias tool:(A)
bias tool: (A)scoring
scoring of individual
of individual studies;
studies; (B)scores
(B) overall overall scores as
presented presented
as percentage of studies for in-vivo studies [45–47,49,50,53,60,90]. Symbols: yellow “?” indicates
percentage of studies for in-vivo studies [45–47,49,50,53,60,90]. Symbols: yellow “?” indicates unclear un-
clearrisk;
risk;green
green “+” indicates low risk; and red “-” indicates high
“+” indicates low risk; and red “-” indicates high risk of bias.risk of bias.

3.4. 3.4. NK-EVs

NK-EVs Overcomethe
Overcome theChemical
Chemical and
andPhysical
PhysicalBarriers of TME
Barriers of TME
NK-EVs inherit the components and functions of NK cells to exert cytotoxicity against
NK-EVs inherit the components and functions of NK cells to exert cytotoxicity
tumor cells in vitro and in vivo. NK-EVs not only express tumor-targeting ability but also
against tumor
possess cellsand
homing in vitro and in
migratory vivo. NK-EVs
properties not only
via chemotaxis express tumor-targeting
[45–47,50,58]. As depicted in ability
but Figure
also possess
6, NK-EVs’ framework is significantly more compact (30–200 nm) and they are As de-
homing and migratory properties via chemotaxis [45–47,50,58].
alsoin
picted more biostable
Figure than NK cells
6, NK-EVs’ [94,95]. These
framework properties enable
is significantly morethem to have(30–200
compact improved nm) and
infiltration and survival in the hostile TME. It has been demonstrated that NK-EVs
they are also more biostable than NK cells [94,95]. These properties enable them to have express
NKG2D, FasL and TRAIL just like the parent cells for receptor-mediated apoptosis. Lytic
improved infiltration and survival in the hostile TME. It has been demonstrated that NK-
proteins such as PFN and Gzm also exist ubiquitously in NK-EVs to initiate caspase-
EVs express NKG2D, FasL and TRAIL just like the parent cells for receptor-mediated
apoptosis. Lytic proteins such as PFN and Gzm also exist ubiquitously in NK-EVs to ini-
tiate caspase-dependent apoptosis pathway. All these functions show that the simple con-
figuration of NK-EVs does not compromise their capability to eliminate cancer cells
 Int. J. Mol. Sci. 2023, 24, 4026 13 of 25

, 24, x FOR PEER REVIEW dependent apoptosis pathway. All these functions show that the simple 12 configuration
of 24 of
NK-EVs does not compromise their capability to eliminate cancer cells through the multiple
cytotoxic pathways utilized by its parent cells. Instead, the NK-EVs might be more effective
without the interference of the TME and TD-EVs that halt the production of PFN and Gzm
NK-EVs in relative in
to NK
TD-EVs
cells. in the
It is TME will
believed thatlikely determine
the anti-cancer if theyofwill
potency be able
NK-EVs to exert
is mainly attributed
by these killer proteins. However, the concentration and ratio
anti-tumor effects. It is highly possible that the level of inhibitory signals induced by TD- of NK-EVs in relative to
EVs may overwhelm TD-EVs in the TMEsignals
the activating will likely
of determine
NK-EVs to if they will be
prevent anyable to exert anti-tumor
cytotoxic action [84].effects. It is
highly possible that the level of inhibitory signals induced by TD-EVs may overwhelm the
This means that the dose and delivery method of NK-EVs in vivo need to be critically
activating signals of NK-EVs to prevent any cytotoxic action [84]. This means that the dose
evaluated to ensureandtheir high method
delivery bioavailability
of NK-EVs ininthe tumor
vivo need in order
to be to exert
critically their to
evaluated cytotoxic
ensure their high
function. bioavailability in the tumor in order to exert their cytotoxic function.

Figure 6. TD-EVs suppress

Figure 6.the recruitment
TD-EVs suppressand migrationand
the recruitment as migration
well as reduce the
as well as proliferation,
reduce sur- survival
the proliferation,
vival and cytolytic function of NK cells. On the other hand, NK-EVs secreted by the NK cells
and cytolytic function of NK cells. On the other hand, NK-EVs secreted by the NK cells are are small
small and diligent enough to overcome the physical and chemical barriers of TME to reach and exert
and diligent enough to overcome the physical and chemical barriers of TME to reach and exert its
its cytolytic effect on cytolytic
the tumor cells.
effect on the tumor cells.

3.5. NK-EVs Are More Clinically Applicable than NK Cells

Results from the in vitro and in vivo experiments demonstrated that NK-EVs are an
effective and safe immunotherapy. However, there are several foreseeable issues that
need to be addressed in preparing this therapy for future clinical trials and registration.
To begin, the functions of NK cells are distinguishable by their cell activation status
[24,96,97]. Generally, both naïve and activated NK cells can secrete NK-EVs [98]. How-
ever, the activated NK cells are preferable for EV isolation due to their high cell number
Int. J. Mol. Sci. 2023, 24, 4026 14 of 25

3.5. NK-EVs Are More Clinically Applicable than NK Cells

Results from the in vitro and in vivo experiments demonstrated that NK-EVs are an
effective and safe immunotherapy. However, there are several foreseeable issues that need
to be addressed in preparing this therapy for future clinical trials and registration. To
begin, the functions of NK cells are distinguishable by their cell activation status [24,96,97].
Generally, both naïve and activated NK cells can secrete NK-EVs [98]. However, the
activated NK cells are preferable for EV isolation due to their high cell number in in vitro
culture. It has been reported that the NK-EVs derived from inactivated NK cells have lower
amounts of cytotoxic proteins [46]. In a separate study, the authors found that EVs secreted
by the NK92 cell line contain fewer cytotoxic proteins compared to the EVs secreted by
in vitro-expanded NK cells [48]. Additionally, Shoae-Hassani et al. (2017) reported that NK
cells exposed to neuroblastoma cells produced NK-EVs with higher cytotoxicity towards the
tumor cells in vitro and in vivo compared to the NK-EVs derived from naïve NK cells [99].
Priming of NK cells with IL-15 was also reported to increase the cytotoxicity of NK-EVs [49].
These observations show that NK-EVs are highly heterogeneous depending on the cellular
origin, culture environment and physiological status. Therefore, it is important to conduct
experiments to the optimize the NK cell culture protocol in order to harvest NK-EVs with
potent therapeutic potential. Importantly, the optimization should also consider increasing
the yield of NK-EVs in order to reduce the cost.
Unlike the stem cell, donor-derived NK cells have variable and limited expansion
potential in vitro [100]. Thus, efforts have been made to prepare immortalized NK cell
lines, such as the NK92-MI from the American Type Culture Collection (ATCC, USA). From
Table 1, eleven of the seventeen reviewed studies purchased this cell line to source their NK-
EVs. Wu et al. (2019) and Aarsund et al. (2022) discovered little-to-no difference between
the NK-EVs from NK92-MI cell line and those from freshly isolated peripheral blood-
derived NK (PB-NK) cells [48,51]. Cryopreservation is widely used for long-term storage of
NK cells. However, the cryopreserved NK cells have relatively poor viability albeit different
freezing medium have been tested. Many studies have reported cell viability less than
10% after revival [101]. More alarming is that the survived cells ceased proliferating and
failed to exert any cytotoxic functions [102]. Failure of cell cryopreservation will definitely
hamper the clinical translation of NK-EV therapy as it will limit the potential of producing
them on a large scale and consistently. At the moment, new donor-derived NK cells are
used to prepare the NK-EVs, which leads to high batch-to-batch variation. Furthermore, it
will also increase the production cost and lead time.
Manual cell culture protocol using tissue culture flask is prone to contamination,
technical errors and high batch-to-batch variation [103,104]. To overcome these limitations,
efforts should be made to shift the NK cell expansion to a bioreactor platform which
is highly automated, allows more control over the culture environment (e.g., glucose
concentration, oxygen concentration and pH) and requires less manpower [105]. Usage
of a highly efficient bioreactor system might also reduce the cost of production [44]. Not
least, it is important is to optimize the EV isolation protocol, which should allow quick and
reliable isolation of EV subpopulation.
Last but not least, long-term storage and off-the-shelf availability are highly desirable
for NK-EV therapy. To achieve this, researchers need to device a storage condition that
can preserve EV stability over a long period of time. As an acellular product, NK-EVs
can be kept chill (4 ◦ C) for immediate use or frozen (−20 ◦ C or −80 ◦ C) up to 12 months
with acceptable loss of proteins and RNAs [106–110]. More importantly, the NK-EVs retain
their anti-tumor effect. For easier storage and transportation, NK-EVs could be lyophilized.
Even though lyophilization has been widely explored for storage of stem cell-derived EVs,
it is still unknown whether lyophilization can be used to preserve the NK-EVs.
Int. J. Mol. Sci. 2023, 24, 4026 15 of 25

3.6. Defining Nomenclature, Isolation Technique, QA/QC Methods and Biomarkers for NK-EVs
In the very beginning, EVs were first thought as a mechanism of disposing cellular
waste products [111]. Later, studies found that these minute vesicles are rich in lipids,
proteins, nucleic materials (e.g., DNA, RNA, mRNA, miRNA) and other biologically
active molecules [112]. On their surface, they express different molecules and proteins
adopted from parent cell membrane during the budding or exocytosis process. These
physical hallmarks contribute to the diverse functions and heterogeneous properties of EV
molecules. Size is commonly used to classify EVs [113]. The largest EVs are the apoptotic
bodies (1000–5000 nm), followed by microvesicles (50–1000 nm), with exosomes (30–150 nm)
being the smallest [114]. The term “exosome” is often incorrectly used interchangeably with
“extracellular vesicle” to describe the small EV preparation [42]. This is because most of the
studies isolate the EVs based on their size instead of the specific phenotype that represented
their unique biogenesis pathway. Every EV subset has a specific protein phenotype and
molecular content. Thus, the Minimal Information for Studies of Extracellular Vesicles
(MISEV) 2018 guidelines have recommended the nomenclature “extracellular vesicles”
to describe the small membrane-bound particles secreted by the cells. It is crucial to
establish a uniform and specific standard for appropriate evaluation of different NK-EV
preparations as it would allow researchers to determine the strengths and weaknesses of
any modifications or novelties introduced.
Based on Table 3 and Figure 7, the most commonly used EV isolation method is
the ultracentrifugation (UC) [44,45,50,52–55,57,58,60,90]. UC is often used in combination
with other EV isolation techniques to improve the purity [45,50,53,58,60]. Alternative
isolation techniques used in the reviewed studies include density gradient centrifugation
(DGC) [45,49,50,53,60], polyethylene glycol-8000 (PEG8000) precipitation [44,48,53], differ-
ential centrifugation (DC) [47,56,58] and size exclusion chromatography (SEC) [46,49,51].
Generally, the existing techniques are unable to achieve both high recovery and high
specificity. With the rise in interest in EV-related therapeutics especially mesenchymal
stem cell-derived EVs (MSC-EVs) and EV-based immunotherapy, it is vital to develop a
novel isolation method which can give high recovery and high purity as well as can be
easily scaled up [42,113]. Currently, tangential flow filtration (TFF) is being revisited and
improved for more efficient recovery of EVs [58,115,116]. Watson et al. (2018) utilized
the combined platforms of TFF and SEC to isolate MSC-EVs and reported that TFF is a
reproducible and scalable technique to recover clinical-grade EVs compared to other known
techniques [117].

Table 3. Method of isolation, characterization and functional analysis of NK-EVs.

Size Range or Size Peak or

Isolation Methods of Identification
Study mean (±SEM) Mode Functional Markers
Method Assessment Markers
(nm) (nm)
Jong et al., CD63, FN,
Caspase assay, CD56, Cytochrome-C,
2017 PEG8000 155 ± 5.9 120 ± 6.4 GNLY, GzmA,
NTA, TEM, WB Rab5A
[44] GzmB, PFN
Caspase assay,
a ALIX, β-actin,
Zhu et al., Serial a a Cellular uptake
100–150 99.2 ± 21.5 CD63, AKT, ERK, PFN,
2018 extrusion + UC b 100–120 b 118 ± 33.1 assay,
Cytochrome-C, p-AKT, p-ERK
[45] or b UC + DGC Inhibition assay,
Fas, GM-130
NTA, TEM, WB
ALIX, β-actin,
Neviani AURKA, GzmA,
CANX, CD81,
et al., 2019 SEC 122.2 ± 1.3 92.5 ± 1.2 NTA, TEM, WB GzmB, MYCN, PFN,
FN, HSP70,
[46] TGFBR1, TGFBR2
TSG101
Int. J. Mol. Sci. 2023, 24, 4026 16 of 25

Table 3. Cont.

Size Range or Size Peak or

Isolation Methods of Identification
Study mean (±SEM) Mode Functional Markers
Method Assessment Markers
(nm) (nm)
Wang
BCA assay, DLS, ALIX, CD63, CD47, CXCR4,
et al., 2019 DC 100 −
NTA, TEM, WB TSG101 Cytochrome-C
[47]
β-actin,
Caspase assay,
Wu et al., Cytochrome-C
Cytochrome-C
2019 PEG8000 − − FasL, GzmA, −
assay, ELISA,
[48] GzmB, GNLY,
WB
PFN
NK-EVs: BLI, Caspase ALIX, β-actin,
Zhu et al., Membrane-FasL,
106.9 ± 21.6 assay, FC, MTT Calnexin, CD63,
2019 SEC + DGC − Cytoplasm-FasL,
NK-EVsIL-15 : assay, NTA, Cytochrome-C,
[49] PFN, GzmB
118.2 ± 20.3 TEM, WB GM-130
Apo A-IV, Apo E,
β-actin, DR4, DR5,
CD40L, CD49,
2-DE proteome DNAM-1, Fas, FasL,
CD51, CD63,
Choi et al., analysis, FGB, FGG, FN,
Integrin α1,
2020 DGC + UC − − Antibody HSP90 α/β, IFN-γ,
Integrin α3,
[50] blocking assay, IL-6, L-plastin,
Integrin β1,
BCA assay, WB NKG2D, NKP44,
L-selectin
NKP46, TRAIL,
TNF-α, VCP
Aarsund BCA assay, CD63, CD81,
DNAM-1, NKG2D,
et al., 2022 SEC 60–125 − LC-MS/MS, FasL, GzmB,
NKP46, NKP30
[51] NTA, TEM, WB PFN, TSG101
Han et al., DLS, HPLC,
ALIX, CD63, Bax, Bcl-2, β-actin,
2020 UC 80–110 − TEM, qRT-PCR,
TSG101 Cas-3,
[52] WB
BCA assay, ALIX, Annexin
DNAM1, GNLY,
Caspase assay, V, β-actin, CD9,
Cochran GzmA, GzmB,
PEG8000 + GO, CD63, HSP70,
et al., 2021 188.6 ± 2.7 133.4 ± 8.0 ICAM1, MHC-I,
DGC + UC LC-MS/MS, HSP90, LAMP1,
[53] MHC-II, PFN,
qPCR, NTA, NKLAM,
VCAM1
TEM, WB TSG101
BCA assay, FC,
Jiang CD63, FasL,
TEM, WB,
et al., 2021 UC 50–200 205.6 ± 29.65 GAPDH, GzmB, −
Wound healing
[54] PFN, TSG101
assay
Caspase assay,
7-AAD, Annexin V,
Kaban FC, NTA, TEM,
Bcl-2, Cas 3/7, Cas-9,
et al., 2021 UC 115.8–128.9 − Immunogold CD56, CD63
Cytochrome-C,
[55] staining,
TMRE
qRT-PCR
Int. J. Mol. Sci. 2023, 24, 4026 17 of 25

Table 3. Cont.

Size Range or Size Peak or

Isolation Methods of Identification
Study mean (±SEM) Mode Functional Markers
Method Assessment Markers
(nm) (nm)
CD3, CD16, CD19,
Bradford assay,
CD56, CD69,
Di Pace Caspase assay,
CANX, CD63, DNAM1, GzmA,
et al., 2020 DC 135.9 ± 0.5 88 ± 1.3 Cellular uptake
CD81, TSG101 GzmB, IFN-γ, LFA-1,
[56] assay, ELISA,
NKP44, NKG2D,
FC, NTA, WB
PFN, PD-1
BCA assay, FC,
Migration assay,
α-tubulin,
Enomoto miRNA CD226, FasL, GNLY,
β-actin, CD63,
et al., 2021 UC − 148.2 profiling (small GzmA, GzmB,
CD81,
[57] RNA-seq), MS, GzmH, PFN, TRAIL
Cytochrome-C
NTA, GO,
qRT-PCR, WB
Caspase assay,
ALIX, CD63, β-actin, Cas3, Cas7,
Kim et al., a Cellular uptake
106.1 ± 71.5 CD81, GzmB, Cas8, Cas9,
2022 UC + DC + TFF b 128.5 ± 33.3 − assay, a DLS,
PFN, TRAIL, Cytochrome-C, PARP,
[58] LDH assay, b
FasL p-AKT, p-ERK1/2
NTA, TEM, WB
FC, Migration
Sun et al.,
and invasion ACTIN, CD63,
2019 UC 50–200 − IL-26, mir-3607-3p
assay, SEM, TSG101
[59]
qRT-PCR
BCA assay,
Zhu et al., ALIX, CD63, Annexin V,
Caspase assay,
2017 DGC + UC 100–150 − β-actin, Cytochrome-C, FasL,
ELISA, FC,
[60] GM-130 PFN, p38, TNF-α
TEM, WB
ALIX, CD63, Bax, Bcl-xL, Bmi-1,
Lee et al.,
CD81, HSP70, CD133, IL-1β, IL-6,
2021 UC 136.6 ± 9.4 − NTA, WB
TSG101, GzmB, MDR, MMP-3, p53,
[90]
PFN PCNA, TNF-α, VEGF
Symbol: “−” indicates data not specified or available by authors. Abbreviations: 2-DE—2-dimensional gel
electrophoresis; 7-AAD—7-aminoactinomycin D; AKT—protein kinase B; ALIX—ALG-2-interacting protein
X; APO—apolipoprotein; Bax—Bcl-2-associated X protein; BCA—bicinchoninic acid assay; BCL2—B-cell lym-
phoma 2 gene; CANX—calnexin; Cas—caspase; CXCR4—C-X-C chemokine receptor type 4; DC—differential
centrifugation; DGC—density gradient centrifugation; DLS—dynamic light scattering; DNAM1—DNAX ac-
cessory molecule-1; DR4—death receptor 4; ELISA—enzyme-linked immunosorbent assay; FC—flow cytom-
etry; FGB—fibrinogen beta chain; FGG—fibrinogen gamma chain; FN—fibronectin; GFP—green fluorescent
protein; GM-130—golgi matrix protein; GNLY—granulysin; GO—gene ontology; GzmA—granzyme A; GzmB—
granzyme B; HPLC—high performance liquid chromatography; HSP70—70-kilodalton heat shock proteins;
ICAM1—intercellular adhesion molecule 1; IFN—interferon; LAMP1—lysosomal-associated membrane protein
1; LC-MS/MS—liquid chromatography–tandem mass spectrometry; LFA-1—lymphocyte function-associated
antigen 1; MHC—major histocompatibility complex; MS—mass spectrometry; NKG2D—natural killer group
2D; NKP44—natural killer cell P44-related protein; NN—nanoparticles; NTA—nanoparticle tracking analysis;
PARP—poly [ADP-ribose] polymerase; PD1—programmed cell death 1; PEG8000—polyethylene glycol 8000;
PFN—perforin; qPCR—quantitative polymerase chain reaction; qRT-PCR—quantitative real-time polymerase
chain reaction; SEC—size exclusion chromatography; SEM—scanning electron microscopy; TFF—tangential flow
filtration; TEM—transmission electron microscopy; TMRE—tetramethylrhodamine, ethyl ester; TNF—tumor
necrosis factor; TRAIL—TNF-related apoptosis inducing ligand; TSG101—tumor susceptibility gene 101; UC—
ultracentrifugation; VCAM-1—vascular adhesion molecule 1; VCP—valosin-containing protein; WB—Western
blot.
 Int.
ol. Sci. 2023, 24,J.xMol.
FORSci.
PEER2023, 24, 4026
REVIEW 17 of 24 18 of 25

Figure 7. Distribution
Figureof7. reviewed studies
Distribution for (A)studies
of reviewed source,for
(B)(A)
isolation techniques
source, (B) isolation(single or com-
techniques (single or combined),
bined), (C) common characterization assays and (D) common biological markers for NK-EVs. Num-
(C) common characterization assays and (D) common biological markers for NK-EVs. Numbers
bers indicated in (A) refer to the frequency of reviewed studies that fall into the categories.
indicated in (A) refer to the frequency of reviewed studies that fall into the categories.

3.7. Future Considerations of NK-EVs

According as a Tool
to MISEV for EVs
2018, Immunotherapy
should be categorized based on size and biochemi-
Table 4 summarizes the established
cal composition (e.g., CD9, benefits
CD81 and and TSG101)
challenges of NK cells
[118,119]. and
Most ofNK-EVs
the selected articles in
for immunotherapy. However,
this study used the
smalldiscussion
NK-EVsgoing in theforward shall emphasize
range 50–200 on improv-
nm [44–47,49,51–53,55–60,90]. Only
ing the EV-based Jiang et al. used a subset of EVs that are larger in diameter, i.e., 205.6 ± 29.65 nm [54].
immunotherapy as it has yet to be thoroughly explored but already ex-
hibiting significant advantages
To prevent compared
misreport to the former. One
or misrepresentation of of the author’s
each main benefits
work,ofonlycell- text-based evi-
based and EV-based
dence of therapies
the NK-EV is that
sizesthey arerecorded
were minimally invasive
since not allmedical
reviewed procedures
studies had specifically
[120]. They aredisclosed the diameter,
safer alternatives with mean or mode
potentially fewerof their NK-EVs.
procedural The frequently
complications to theused assessment
tools forsuch
high-risk populations EVsas in infants
these studies includes
and elderly. In nanoparticle
contrast, surgicaltracking analysis
treatment can(NTA)
lead to examine the
to many serioussize [44–47,49,51,53,55–58,90],
complications, including induced transmission
hemorrhage, electron microscopysepsis
post-operative (TEM) andto study the size
extended rehabilitation [121,122]. Selection of administration route or delivery technique and flow cy-
and morphology [44–47,49,51–55,58,60], Western blot (WB) [44–53,56–58,60,90]
tometry
is critical to ensure the (FC)
safety [49,54–57,59,60]
and efficacy oftoa determine the biochemical
therapy [123]. Although ITmarkers, and bicinchoninic acid
administration
would eliminate (BCA) protein assay
any concerns over [47,50,51,53,54,57,60]
the homing and migration to quantify
abilitytheofprotein
NK-EVs, concentration.
the IV The bio-
administrationchemical markers consistently
will be especially tested with
useful for regions to characterize
poor access theorEVs were CD63(e.g.,
penetrability [44,45,47,49–60,89],
brain) [124]. InCD81 [46,51,56,57,89],
actuality, a larger dose ALIX [45–47,49,52,53,58,60,90]
is typically needed for IV route and TSG101 [46,47,51–54,56,59,90].
to compensate for Fur-
thermore, specific markers, i.e., Fas/FasL or Fas/TRAIL
the accumulated loss of drugs due to unwanted distribution at the other tissues as well as [45,48,50,51,54,57,58], granulysin
metabolism and (GNLY) [44,48,52,57],
excretion primary atgranzymes
the liver and A kidney
(GzmA) and/or BNonetheless,
[125–127]. (GzmB) [44,46,48,49,51,53,54,56,57],
the con-
and PFN [44–46,48,49,51,53,54,56,57,59,60]
tinual IV administration of NK-EVs after the tumor elimination could were used to prove that the EVstooriginated from
be favourable
ensure complete NK cells. The
clearing characterization
of tumor of NK-EVs
cells to minimize using
the risk the above
of relapse techniques
[128,129]. To date,is crucial to verify
the NK-EVs, as well as to compare and standardize
it is still unclear on the optimal dosage regime of NK-EVs. More research is needed to the NK-EV research.
determine the ideal frequency, dosage and interval of NK-EV treatment.
3.7. Future Considerations of NK-EVs as a Tool for Immunotherapy
Table 4 summarizes the established benefits and challenges of NK cells and NK-EVs
for immunotherapy. However, the discussion going forward shall emphasize on improving
the EV-based immunotherapy as it has yet to be thoroughly explored but already exhibiting
significant advantages compared to the former. One of the main benefits of cell-based
and EV-based therapies is that they are minimally invasive medical procedures [120].
Int. J. Mol. Sci. 2023, 24, 4026 19 of 25

They are safer alternatives with potentially fewer procedural complications to the high-
risk populations such as infants and elderly. In contrast, surgical treatment can lead to
many serious complications, including induced hemorrhage, post-operative sepsis and
extended rehabilitation [121,122]. Selection of administration route or delivery technique
is critical to ensure the safety and efficacy of a therapy [123]. Although IT administration
would eliminate any concerns over the homing and migration ability of NK-EVs, the IV
administration will be especially useful for regions with poor access or penetrability (e.g.,
brain) [124]. In actuality, a larger dose is typically needed for IV route to compensate for
the accumulated loss of drugs due to unwanted distribution at the other tissues as well
as metabolism and excretion primary at the liver and kidney [125–127]. Nonetheless, the
continual IV administration of NK-EVs after the tumor elimination could be favourable to
ensure complete clearing of tumor cells to minimize the risk of relapse [128,129]. To date,
it is still unclear on the optimal dosage regime of NK-EVs. More research is needed to
determine the ideal frequency, dosage and interval of NK-EV treatment.

Table 4. Summary of pros and cons of NK cell vs. NK-EV for immunotherapy.

Pros Cons
1. Size and nature of cells limits
permeability through the
biological barriers.
2. Poor cell viability and stability
1. Origin of multiple and complex due to susceptibility from TME
immune functions that includes and TD-EVs.
NK-EVs. 3. Limited modification or drug
2. Able to incite stronger humoral delivery potential.
response (antibody-mediated) via 4. Possible graft rejection and
NK CELL cross-talks with DC. transmission of infectious agents
3. Have explicit evidence on safety with the use allogeneic cells.
and references to efficacy. 5. No known effective
4. Have GMP-compliant isolation cryopreservation methods despite
and expansion protocols. years of research and
development.
6. Cell therapy remain an ethical
issue.
7. Low yield/purity and high cost
of manufacturing.
1. Inherit the identity and properties
of the parent cells.
2. Nano-sized dimension improve
homing and migratory function.
3. Higher biostability, lower safety
risks and excellent treatment
efficacy. 1. Requires new isolation from
4. Unaltered by TME and/or tissue and cell cultures to
TD-EVs. replenish stock.
5. A viable drug delivery platform 2. Lack of GMP-compliant protocols
NK-EV for product isolation, testing and
through genetic engineering or
drug loading for an enhanced or preservation.
combined therapy. 3. Unknown clinical potential as a
6. Potential of storage with long relatively novel product.
shelf-life.
7. Allow standardization of dose for
targeted and precision medicine.
8. Its classification as “biologics” is
familiar to physicians and more
ethically acceptable.
Int. J. Mol. Sci. 2023, 24, 4026 20 of 25

Efforts are also needed to improve the efficacy of NK-EV production in order to reduce
the cost. Owing to its niche role, NK cells have high metabolic activity and need specific
cytokines to stimulate its proliferation. The commonly used cytokines to expand NK
cells are IL-2, IL-15, IL-18, and IL-21. The cytokines can be provided by the peripheral
blood mononuclear cells (PBMCs) that serve as the feeder cells [130]. Alternatively, the
culture medium can be supplemented with commercially available human recombinant
cytokines to maintain the NK cell culture. Although the use of recombinant cytokines
is simpler as no purification step is needed to remove the contaminating PBMCs and
safer as it is free from PBMC contamination, the cost of recombinant cytokines is very
high. The high concentration and cost of recombinant cytokines will significantly escalate
the production cost. On top of that, the requirement of using GMP-grade or clinical-
grade raw materials and consumables, skilled manpower, stringent quality assurance and
quality control (QA/QC), expansive equipment for scalable production, and the use of a
cGMP-certified production facility will escalate the treatment cost to unaffordable scales
for majority of the patients [131]. This, in turn, will affect the clinical translation and
commercial viability of NK-EV therapy.

4. Conclusions
NK-EVs show promising tumor cytotoxicity with no adverse or side effects in vitro
and in vivo. NK-EVs are not cytotoxic to the normal or healthy cells and can increase the
life span of tumor-bearing animals. Thus, this novel therapy has great clinical potential.
In addition, NK-EVs can circumvent some of the limitations of NK cell therapy, such as
impediment of NK cell functionality by the TME and TD-EVs. In fact, NK-EVs would see
potential benefits due to the increased cellular uptake in the TME. Although IT administra-
tion is superior, the IV administration will be meaningful for “difficult-to-reach” tumors
and high-risk populations. Additionally, NK-EVs can be readily available off-the-shelf as
they are more stable and easier to store compared to NK cells. Nonetheless, NK-EV therapy
and its production procedure need further investigation to improve its safety and efficacy
as well as to increase the yield and lower the production cost.

Author Contributions: Conceptualization, all authors; methodology, A.M.L.C. and J.M.C.; validation,
A.M.L.C. and J.M.C.; formal analysis, A.M.L.C. and J.M.C.; investigation, A.M.L.C. and J.M.C.; data
curation, A.M.L.C. and J.M.C.; writing—original draft preparation, A.M.L.C. and J.M.C.; writing—
review and editing, all authors; visualization, all authors. All authors have read and agreed to the
published version of the manuscript.
Funding: This work was supported by grant provided by Ming Medical Sdn Bhd (FF-2022-330).
The granting agencies played no role in activities pertaining to the execution of the study and the
submission of the manuscript.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The graphical abstract, Figures 3 and 6 were created using BioRender.
Conflicts of Interest: The authors declare that there is no conflict of interest.

References
1. World Health Organization (WHO). Coronavirus (COVID-19) Dashboard. 2022. Available online: https://covid19.who.int/
(accessed on 13 December 2022).
2. World Health Organization (WHO). Non Communicable Diseases. 2022. Available online: https://www.who.int/news-room/
fact-sheets/detail/noncommunicable-diseases (accessed on 14 July 2022).
3. Liao, G.S.; Apaya, M.K.; Shyur, L.F. Herbal medicine and acupuncture for breast cancer palliative care and adjuvant therapy. Evid.
Based Complement. Alternat. Med. 2013, 2013, 437948. [CrossRef] [PubMed]
4. Altun, İ.; Sonkaya, A. The Most Common Side Effects Experienced by Patients Were Receiving First Cycle of Chemotherapy. Iran.
J. Public Health 2018, 47, 1218–1219. [PubMed]
Int. J. Mol. Sci. 2023, 24, 4026 21 of 25

5. Alcindor, T.; Beauger, N. Oxaliplatin: A review in the era of molecularly targeted therapy. Curr. Oncol. 2011, 18, 18–25. [CrossRef]
[PubMed]
6. Corrie, P.G. Cytotoxic chemotherapy: Clinical aspects. Medicine 2011, 39, 717–722. [CrossRef]
7. Makovec, T. Cisplatin and beyond: Molecular mechanisms of action and drug resistance development in cancer chemotherapy.
Radiol. Oncol. 2019, 53, 148–158. [CrossRef] [PubMed]
8. Kalyanaraman, B. Teaching the basics of the mechanism of doxorubicin-induced cardiotoxicity: Have we been barking up the
wrong tree? Redox Biol. 2020, 29, 101394. [CrossRef]
9. Smoot, B.; Wampler, M.; Topp, K.S. Breast cancer treatments and complications: Implications for rehabilitation. Rehabilitation
Oncology 2009, 27, 16–26. [CrossRef]
10. Rossi, A.; Fortuna, M.C.; Caro, G.; Pranteda, G.; Garelli, V.; Pompili, U.; Carlesimo, M. Chemotherapy-induced alopecia
management: Clinical experience and practical advice. J. Cosmet. Dermatol. 2017, 16, 537–541. [CrossRef]
11. Frenkel, M. Refusing treatment. Oncologist 2013, 18, 634–636. [CrossRef]
12. Zakrzewski, W.; Dobrzyński, M.; Szymonowicz, M.; Rybak, Z. Stem cells: Past, present, and future. Stem Cell Res. Ther. 2019, 10,
68. [CrossRef]
13. Erhart, F.; Buchroithner, J.; Reitermaier, R.; Fischhuber, K.; Klingenbrunner, S.; Sloma, I.; Hibsh, D.; Kozol, R.; Efroni, S.; Ricken,
G.; et al. Immunological analysis of phase II glioblastoma dendritic cell vaccine (Audencel) trial: Immune system characteristics
influence outcome and Audencel up-regulates Th1-related immunovariables. Acta Neuropathol. Commun. 2018, 6, 135. [CrossRef]
14. Margolin, K.; Morishima, C.; Velcheti, V.; Miller, J.S.; Lee, S.M.; Silk, A.W.; Holtan, S.G.; Lacroix, A.M.; Fling, S.P.; Kaiser, J.C.; et al.
Phase I Trial of ALT-803, A Novel Recombinant IL15 Complex, in Patients with Advanced Solid Tumors. Clin. Cancer Res. 2018,
24, 5552–5561. [CrossRef] [PubMed]
15. Martins, F.; Oliveira, R.; Cavadas, B.; Pinto, F.; Cardoso, A.P.; Castro, F.; Sousa, B.; Pinto, M.L.; Silva, A.J.; Adão, D.; et al. Hypoxia
and Macrophages Act in Concert Towards a Beneficial Outcome in Colon Cancer. Cancers 2020, 12, 818. [CrossRef] [PubMed]
16. Gumrukcu, S.; Nguyen, T.X.; White, R.L.; Howell, G.T.; Musikanth, P. Allogeneic Natural Killer and Cytomegalovirus (CMV)-pp65
Pulsed Dendritic Cells Induced Complete Response Through 15 Months in a Patient with Recurrent Glioblastoma: A Case Study.
Am. J. Case Rep. 2021, 22, e931030. Retraction in Am. J. Case Rep. 2022, 23, e937680. [CrossRef] [PubMed]
17. Mellman, I. Dendritic cells: Master regulators of the immune response. Cancer Immunol. Res. 2013, 1, 145–149. [CrossRef]
[PubMed]
18. Luckheeram, R.V.; Zhou, R.; Verma, A.D.; Xia, B. CD4+ T cells: Differentiation and functions. Clin. Dev. Immunol. 2012, 2012,
925135. [CrossRef] [PubMed]
19. Mills, C.D.; Lenz, L.L.; Harris, R.A. A Breakthrough: Macrophage-Directed Cancer Immunotherapy. Cancer Res. 2016, 76, 513–516.
[CrossRef] [PubMed]
20. Farhood, B.; Najafi, M.; Mortezaee, K. CD8+ cytotoxic T lymphocytes in cancer immunotherapy: A review. J. Cell Physiol. 2019,
234, 8509–8521. [CrossRef]
21. Oh, S.; Lee, J.H.; Kwack, K.; Choi, S.W. Natural Killer Cell Therapy: A New Treatment Paradigm for Solid Tumors. Cancers 2019,
11, 1534. [CrossRef]
22. Komatsu, F.; Tamiya, H. Relationship between antibody-dependent cell-mediated cytotoxicity due to anti-HTLV-1 and negative
signal of major histocompatibility complex class I antigens on adult T-cell leukemia cell lines. Oncol. Res. 1998, 10, 59–67.
23. Høydahl, L.S.; Frick, R.; Sandlie, I.; Løset, G.Å. Targeting the MHC Ligandome by Use of TCR-Like Antibodies. Antibodies 2019, 8,
32. [CrossRef]
24. Paul, S.; Lal, G. The Molecular Mechanism of Natural Killer Cells Function and Its Importance in Cancer Immunotherapy. Front.
Immunol. 2017, 8, 1124. [CrossRef] [PubMed]
25. Lo Nigro, C.; Macagno, M.; Sangiolo, D.; Bertolaccini, L.; Aglietta, M.; Merlano, M.C. NK-mediated antibody-dependent cell-
mediated cytotoxicity in solid tumors: Biological evidence and clinical perspectives. Ann. Transl. Med. 2019, 7, 105. [CrossRef]
[PubMed]
26. Kurdi, A.T.; Glavey, S.V.; Bezman, N.A.; Jhatakia, A.; Guerriero, J.L.; Manier, S.; Moschetta, M.; Mishima, Y.; Roccaro, A.; Detappe,
A.; et al. Antibody-Dependent Cellular Phagocytosis by Macrophages is a Novel Mechanism of Action of Elotuzumab. Mol.
Cancer Ther. 2018, 17, 1454–1463. [CrossRef] [PubMed]
27. Li, H.; Somiya, M.; Kuroda, S. Enhancing antibody-dependent cellular phagocytosis by Re-education of tumor-associated
macrophages with resiquimod-encapsulated liposomes. Biomaterials 2021, 268, 120601. [CrossRef] [PubMed]
28. Li, M.; Zheng, H.; Duan, Z.; Liu, H.; Hu, D.; Bode, A.; Dong, Z.; Cao, Y. Promotion of cell proliferation and inhibition of ADCC by
cancerous immunoglobulin expressed in cancer cell lines. Cell Mol Immunol. 2012, 9, 54–61. [CrossRef]
29. Kline, J.B.; Kennedy, R.P.; Albone, E.; Chao, Q.; Fernando, S.; McDonough, J.M.; Rybinski, K.; Wang, W.; Somers, E.B.; Schweizer,
C.; et al. Tumor antigen CA125 suppresses antibody-dependent cellular cytotoxicity (ADCC) via direct antibody binding and
suppressed Fc-γ receptor engagement. Oncotarget 2017, 8, 52045–52060. [CrossRef]
30. Darwich, A.; Silvestri, A.; Benmebarek, M.R.; Mouriès, J.; Cadilha, B.; Melacarne, A.; Morelli, L.; Supino, D.; Taleb, A.; Obeck, H.;
et al. Paralysis of the cytotoxic granule machinery is a new cancer immune evasion mechanism mediated by chitinase 3-like-1. J.
Immunother. Cancer 2021, 9, e003224. [CrossRef]
31. Tsuchiya, H.; Shiota, G. Immune evasion by cancer stem cells. Regen. Ther. 2021, 17, 20–33. [CrossRef]
Int. J. Mol. Sci. 2023, 24, 4026 22 of 25

32. Cullen, S.P.; Martin, S.J. Fas and TRAIL “death receptors” as initiators of inflammation: Implications for cancer. Semin. Cell Dev.
Biol. 2015, 39, 26–34. [CrossRef]
33. Rossin, A.; Miloro, G.; Hueber, A.O. TRAIL and FasL Functions in Cancer and Autoimmune Diseases: Towards an Increasing
Complexity. Cancers 2019, 11, 639. [CrossRef] [PubMed]
34. Snajdauf, M.; Havlova, K.; Vachtenheim, J., Jr.; Ozaniak, A.; Lischke, R.; Bartunkova, J.; Smrz, D.; Strizova, Z. The TRAIL in the
Treatment of Human Cancer: An Update on Clinical Trials. Front. Mol. Biosci. 2021, 8, 628332. [CrossRef]
35. Clark, R.; Griffiths, G.M. Lytic granules, secretory lysosomes and disease. Curr. Opin. Immunol. 2003, 15, 516–521. [CrossRef]
[PubMed]
36. Krzewski, K.; Coligan, J.E. Human NK cell lytic granules and regulation of their exocytosis. Front. Immunol. 2012, 3, 335.
[CrossRef] [PubMed]
37. Kabanova, A.; Zurli, V.; Baldari, C.T. Signals Controlling Lytic Granule Polarization at the Cytotoxic Immune Synapse. Front.
Immunol. 2018, 9, 307. [CrossRef] [PubMed]
38. Ventola, C.L. Cancer Immunotherapy, Part 3: Challenges and Future Trends. Pharm. Ther. 2017, 42, 514–521.
39. Tan, S.; Li, D.; Zhu, X. Cancer immunotherapy: Pros, cons and beyond. Biomed. Pharmacother. 2020, 124, 109821. [CrossRef]
[PubMed]
40. Pugholm, L.H.; Bæk, R.; Søndergaard, E.K.; Revenfeld, A.L.; Jørgensen, M.M.; Varming, K. Phenotyping of Leukocytes and
Leukocyte-Derived Extracellular Vesicles. J. Immunol. Res. 2016, 2016, 6391264. [CrossRef]
41. Wu, F.; Xie, M.; Hun, M.; She, Z.; Li, C.; Luo, S.; Chen, X.; Wan, W.; Wen, C.; Tian, J. Natural Killer Cell-Derived Extracellular
Vesicles: Novel Players in Cancer Immunotherapy. Front. Immunol. 2021, 12, 658698. [CrossRef]
42. Ng, C.Y.; Kee, L.T.; Al-Masawa, M.E.; Lee, Q.H.; Subramaniam, T.; Kok, D.; Ng, M.H.; Law, J.X. Scalable Production of Extracellular
Vesicles and Its Therapeutic Values: A Review. Int. J. Mol. Sci. 2022, 23, 7986. [CrossRef]
43. Kee, L.T.; Ng, C.Y.; Al-Masawa, M.E.; Foo, J.B.; How, C.W.; Ng, M.H.; Law, J.X. Extracellular Vesicles in Facial Aesthetics: A
Review. Int. J. Mol. Sci. 2022, 23, 6742. [CrossRef] [PubMed]
44. Jong, A.Y.; Wu, C.H.; Li, J.; Sun, J.; Fabbri, M.; Wayne, A.S.; Seeger, R.C. Large-scale isolation and cytotoxicity of extracellular
vesicles derived from activated human natural killer cells. J. Extracell Vesicles 2017, 6, 1294368. [CrossRef] [PubMed]
45. Zhu, L.; Gangadaran, P.; Kalimuthu, S.; Oh, J.M.; Baek, S.H.; Jeong, S.Y.; Lee, S.W.; Lee, J.; Ahn, B.C. Novel alternatives to
extracellular vesicle-based immunotherapy—Exosome mimetics derived from natural killer cells. Artif. Cells Nanomed. Biotechnol.
2018, 46 (Suppl. 3), S166–S179. [CrossRef] [PubMed]
46. Neviani, P.; Wise, P.M.; Murtadha, M.; Liu, C.W.; Wu, C.H.; Jong, A.Y.; Seeger, R.C.; Fabbri, M. Natural Killer-Derived Exosomal
miR-186 Inhibits Neuroblastoma Growth and Immune Escape Mechanisms. Cancer Res. 2019, 79, 1151–1164. [CrossRef]
47. Wang, G.; Hu, W.; Chen, H.; Shou, X.; Ye, T.; Xu, Y. Cocktail Strategy Based on NK Cell-Derived Exosomes and Their Biomimetic
Nanoparticles for Dual Tumor Therapy. Cancers 2019, 11, 1560. [CrossRef]
48. Wu, C.H.; Li, J.; Li, L.; Sun, J.; Fabbri, M.; Wayne, A.S.; Seeger, R.C.; Jong, A.Y. Extracellular vesicles derived from natural killer
cells use multiple cytotoxic proteins and killing mechanisms to target cancer cells. J. Extracell Vesicles 2019, 8, 1588538. [CrossRef]
49. Zhu, L.; Kalimuthu, S.; Oh, J.M.; Gangadaran, P.; Baek, S.H.; Jeong, S.Y.; Lee, S.W.; Lee, J.; Ahn, B.C. Enhancement of antitumor
potency of extracellular vesicles derived from natural killer cells by IL-15 priming. Biomaterials 2019, 190–191, 38–50. [CrossRef]
50. Choi, J.W.; Lim, S.; Kang, J.H.; Hwang, S.H.; Hwang, K.C.; Kim, S.W.; Lee, S. Proteome Analysis of Human Natural Killer Cell
Derived Extracellular Vesicles for Identification of Anticancer Effectors. Molecules 2020, 25, 5216. [CrossRef]
51. Aarsund, M.; Segers, F.M.; Wu, Y.; Inngjerdingen, M. Comparison of characteristics and tumor targeting properties of extracellular
vesicles derived from primary NK cells or NK-cell lines stimulated with IL-15 or IL-12/15/18. Cancer Immunol. Immunother. 2022,
71, 2227–2238. [CrossRef]
52. Han, D.; Wang, K.; Zhang, T.; Gao, G.C.; Xu, H. Natural killer cell-derived exosome-entrapped paclitaxel can enhance its
anti-tumor effect. Eur. Rev. Med. Pharmacol Sci. 2020, 24, 5703–5713. [CrossRef]
53. Cochran, A.M.; Kornbluth, J. Extracellular Vesicles From the Human Natural Killer Cell Line NK3.3 Have Broad and Potent
Anti-Tumor Activity. Front. Cell Dev. Biol. 2021, 9, 698639. [CrossRef] [PubMed]
54. Jiang, Y.; Jiang, H.; Wang, K.; Liu, C.; Man, X.; Fu, Q. Hypoxia enhances the production and antitumor effect of exosomes derived
from natural killer cells. Ann. Transl. Med. 2021, 9, 473. [CrossRef] [PubMed]
55. Kaban, K.; Hinterleitner, C.; Zhou, Y.; Salva, E.; Kantarci, A.G.; Salih, H.R.; Märklin, M. Therapeutic Silencing of BCL-2 Using NK
Cell-Derived Exosomes as a Novel Therapeutic Approach in Breast Cancer. Cancers 2021, 13, 2397. [CrossRef]
56. Di Pace, A.L.; Tumino, N.; Besi, F.; Alicata, C.; Conti, L.A.; Munari, E.; Maggi, E.; Vacca, P.; Moretta, L. Characterization of Human
NK Cell-Derived Exosomes: Role of DNAM1 Receptor In Exosome-Mediated Cytotoxicity Against Tumor. Cancers 2020, 12, 661.
[CrossRef] [PubMed]
57. Enomoto, Y.; Li, P.; Jenkins, L.M.; Anastasakis, D.; Lyons, G.C.; Hafner, M.; Leonard, W.J. Cytokine-enhanced cytolytic activity of
exosomes from NK Cells. Cancer Gene Ther. 2022, 29, 734–749. [CrossRef]
58. Kim, H.Y.; Min, H.K.; Song, H.W.; Yoo, A.; Lee, S.; Kim, K.P.; Park, J.O.; Choi, Y.H.; Choi, E. Delivery of human natural killer
cell-derived exosomes for liver cancer therapy: An in vivo study in subcutaneous and orthotopic animal models. Drug Deliv.
2022, 29, 2897–2911. [CrossRef]
59. Sun, H.; Shi, K.; Qi, K.; Kong, H.; Zhang, J.; Dai, S.; Ye, W.; Deng, T.; He, Q.; Zhou, M. Natural Killer Cell-Derived Exosomal
miR-3607-3p Inhibits Pancreatic Cancer Progression by Targeting IL-26. Front. Immunol. 2019, 10, 2819. [CrossRef]
Int. J. Mol. Sci. 2023, 24, 4026 23 of 25

60. Zhu, L.; Kalimuthu, S.; Gangadaran, P.; Oh, J.M.; Lee, H.W.; Baek, S.H.; Jeong, S.Y.; Lee, S.W.; Lee, J.; Ahn, B.C. Exosomes Derived
From Natural Killer Cells Exert Therapeutic Effect in Melanoma. Theranostics 2017, 7, 2732–2745. [CrossRef]
61. Farcas, M.; Inngjerdingen, M. Natural killer cell-derived extracellular vesicles in cancer therapy. Scand. J. Immunol. 2020, 92,
e12938. [CrossRef]
62. Cirulli, E.T.; Goldstein, D.B. In vitro assays fail to predict in vivo effects of regulatory polymorphisms. Hum. Mol. Genet. 2007, 16,
1931–1939. [CrossRef]
63. Bouhaddou, M.; Yu, L.J.; Lunardi, S.; Stamatelos, S.K.; Mack, F.; Gallo, J.M.; Birtwistle, M.R.; Walz, A.C. Predicting In Vivo Efficacy
from In Vitro Data: Quantitative Systems Pharmacology Modeling for an Epigenetic Modifier Drug in Cancer. Clin. Transl. Sci.
2020, 13, 419–429. [CrossRef] [PubMed]
64. Mansoori, B.; Mohammadi, A.; Davudian, S.; Shirjang, S.; Baradaran, B. The Different Mechanisms of Cancer Drug Resistance: A
Brief Review. Adv. Pharm. Bull. 2017, 7, 339–348. [CrossRef] [PubMed]
65. Levine, M.S.; Holland, A.J. The impact of mitotic errors on cell proliferation and tumorigenesis. Genes Dev. 2018, 32, 620–638.
[CrossRef] [PubMed]
66. Rani, B.; Cao, Y.; Malfettone, A.; Tomuleasa, C.; Fabregat, I.; Giannelli, G. Role of the tissue microenvironment as a therapeutic
target in hepatocellular carcinoma. World J. Gastroenterol. 2014, 20, 4128–4140. [CrossRef] [PubMed]
67. Oliver, A.J.; Lau, P.K.H.; Unsworth, A.S.; Loi, S.; Darcy, P.K.; Kershaw, M.H.; Slaney, C.Y. Tissue-Dependent Tumor Microenviron-
ments and Their Impact on Immunotherapy Responses. Front. Immunol. 2018, 9, 70. [CrossRef]
68. Zhuang, X.; Zhang, H.; Hu, G. Cancer and Microenvironment Plasticity: Double-Edged Swords in Metastasis. Trend. Pharmacol.
Sci. 2019, 40, 419–429. [CrossRef]
69. Senthebane, D.A.; Rowe, A.; Thomford, N.E.; Shipanga, H.; Munro, D.; Mazeedi, M.A.M.A.; Almazyadi, H.A.M.; Kallmeyer,
K.; Dandara, C.; Pepper, M.S.; et al. The Role of Tumor Microenvironment in Chemoresistance: To Survive, Keep Your Enemies
Closer. Int. J. Mol. Sci. 2017, 18, 1586. [CrossRef]
70. Anderson, N.M.; Simon, M.C. The tumor microenvironment. Curr. Biol. 2020, 30, R921–R925. [CrossRef]
71. Xue, H.; Lu, B.; Lai, M. The cancer secretome: A reservoir of biomarkers. J. Transl. Med. 2008, 6, 52. [CrossRef]
72. Lin, J.; Ma, L.; Zhang, D.; Gao, J.; Jin, Y.; Han, Z.; Lin, D. Tumour biomarkers—Tracing the molecular function and clinical
implication. Cell Prolif. 2019, 52, e12589. [CrossRef]
73. Kartikasari, A.E.R.; Huertas, C.S.; Mitchell, A.; Plebanski, M. Tumor-Induced Inflammatory Cytokines and the Emerging
Diagnostic Devices for Cancer Detection and Prognosis. Front. Oncol. 2021, 11, 692142. [CrossRef] [PubMed]
74. Fouad, Y.A.; Aanei, C. Revisiting the hallmarks of cancer. Am. J. Cancer Res. 2017, 7, 1016–1036. [PubMed]
75. El-Fattah Ibrahim, S.A.; Abudu, A.; Johnson, E.; Aftab, N.; Conrad, S.; Fluck, M. Correction: The role of AP-1 in self-sufficient
proliferation and migration of cancer cells and its potential impact on an autocrine/paracrine loop. Oncotarget 2019, 10, 799.
Erratum for: Oncotarget 2018, 9, 34259–34278. [CrossRef] [PubMed]
76. Qiao, F.; Pan, P.; Yan, J.; Sun, J.; Zong, Y.; Wu, Z.; Lu, X.; Chen, N.; Mi, R.; Ma, Y.; et al. Role of tumor-derived extracellular vesicles
in cancer progression and their clinical applications (Review). Int. J. Oncol. 2019, 54, 1525–1533. [CrossRef] [PubMed]
77. Boedtkjer, E.; Pedersen, S.F. The Acidic Tumor Microenvironment as a Driver of Cancer. Annu. Rev. Physiol. 2020, 82, 103–126.
[CrossRef]
78. Kato, Y.; Ozawa, S.; Miyamoto, C.; Maehata, Y.; Suzuki, A.; Maeda, T.; Baba, Y. Acidic extracellular microenvironment and cancer.
Cancer Cell Int. 2013, 13, 89. [CrossRef]
79. Ribeiro Franco, P.I.; Rodrigues, A.P.; de Menezes, L.B.; Pacheco Miguel, M. Tumor microenvironment components: Allies of
cancer progression. Pathol. Res. Pract. 2020, 216, 152729. [CrossRef]
80. Li, P.; Lu, M.; Shi, J.; Hua, L.; Gong, Z.; Li, Q.; Shultz, L.D.; Ren, G. Dual roles of neutrophils in metastatic colonization are
governed by the host NK cell status. Nat. Commun. 2020, 11, 4387. [CrossRef]
81. Nowak, M.; Klink, M. The Role of Tumor-Associated Macrophages in the Progression and Chemoresistance of Ovarian Cancer.
Cells 2020, 9, 1299. [CrossRef]
82. Zhang, Y.; Zhang, Z. The history and advances in cancer immunotherapy: Understanding the characteristics of tumor-infiltrating
immune cells and their therapeutic implications. Cell Mol. Immunol. 2020, 17, 807–821. [CrossRef]
83. Cendrowicz, E.; Sas, Z.; Bremer, E.; Rygiel, T.P. The Role of Macrophages in Cancer Development and Therapy. Cancers 2021, 13,
1946. [CrossRef] [PubMed]
84. Hodge, G.; Barnawi, J.; Jurisevic, C.; Moffat, D.; Holmes, M.; Reynolds, P.N.; Jersmann, H.; Hodge, S. Lung cancer is associated
with decreased expression of perforin, granzyme B and interferon (IFN)-γ by infiltrating lung tissue T cells, natural killer
(NK) T-like and NK cells. Clin. Exp. Immunol. 2014, 178, 79–85. [CrossRef] [PubMed]
85. Cao, L.; Huang, T.; Chen, X.; Li, W.; Yang, X.; Zhang, W.; Li, M.; Gao, R. Uncovering the interplay between pH receptors and
immune cells: Potential drug targets (Review). Oncol. Rep. 2021, 46, 228. [CrossRef] [PubMed]
86. Zhao, J.; Schlößer, H.A.; Wang, Z.; Qin, J.; Li, J.; Popp, F.; Popp, M.C.; Alakus, H.; Chon, S.H.; Hansen, H.P.; et al. Tumor-Derived
Extracellular Vesicles Inhibit Natural Killer Cell Function in Pancreatic Cancer. Cancers 2019, 11, 874. [CrossRef] [PubMed]
87. Li, Q.; Cai, S.; Li, M.; Salma, K.I.; Zhou, X.; Han, F.; Chen, J.; Huyan, T. Tumor-Derived Extracellular Vesicles: Their Role in
Immune Cells and Immunotherapy. Int. J. Nanomed. 2021, 16, 5395–5409. [CrossRef]
Int. J. Mol. Sci. 2023, 24, 4026 24 of 25

88. Nakase, I.; Ueno, N.; Matsuzawa, M.; Noguchi, K.; Hirano, M.; Omura, M.; Takenaka, T.; Sugiyama, A.; Bailey Kobayashi, N.;
Hashimoto, T.; et al. Environmental pH stress influences cellular secretion and uptake of extracellular vesicles. FEBS Open Biol.
2021, 11, 753–767. [CrossRef]
89. Honary, S.; Zahir, F. Effect of Zeta Potential on the Properties of Nano-Drug Delivery Systems—A Review (Part 1). Trop. J. Pharm.
Res. 2013, 12, 255–264. [CrossRef]
90. Lee, J.; Lee, S.A.; Gu, N.Y.; Jeong, S.Y.; Byeon, J.S.; Jeong, D.U.; Ouh, I.O.; Lee, Y.H.; Hyun, B.H. Canine Natural Killer Cell-Derived
Exosomes Exhibit Antitumor Activity in a Mouse Model of Canine Mammary Tumor. Biomed. Res. Int. 2021, 2021, 6690704.
[CrossRef]
91. Kim, Y.I.; Ahn, B.C.; Ronald, J.A.; Katzenberg, R.; Singh, A.; Paulmurugan, R.; Ray, S.; Gambhir, S.S.; Hofmann, L.V. Intratumoral
versus intravenous gene therapy using a transcriptionally targeted viral vector in an orthotopic hepatocellular carcinoma rat
model. J. Vasc. Interv. Radiol. 2012, 23, 704–711. [CrossRef]
92. Baniel, C.C.; Sumiec, E.G.; Hank, J.A.; Bates, A.M.; Erbe, A.K.; Pieper, A.A.; Hoefges, A.G.; Patel, R.B.; Rakhmilevich, A.L.;
Morris, Z.S.; et al. Intratumoral injection reduces toxicity and antibody-mediated neutralization of immunocytokine in a mouse
melanoma model. J. Immunother. Cancer 2020, 8, e001262. [CrossRef]
93. Hooijmans, C.R.; Rovers, M.M.; de Vries, R.B.; Leenaars, M.; Ritskes-Hoitinga, M.; Langendam, M.W. SYRCLE’s risk of bias tool
for animal studies. BMC Med. Res. Methodol. 2014, 14, 43. [CrossRef] [PubMed]
94. van Dommelen, S.M.; Vader, P.; Lakhal, S.; Kooijmans, S.A.; van Solinge, W.W.; Wood, M.J.; Schiffelers, R.M. Microvesicles and
exosomes: Opportunities for cell-derived membrane vesicles in drug delivery. J. Control Release 2012, 161, 635–644. [CrossRef]
[PubMed]
95. Kao, C.Y.; Papoutsakis, E.T. Extracellular vesicles: Exosomes, microparticles, their parts, and their targets to enable their
biomanufacturing and clinical applications. Curr. Opin. Biotechnol. 2019, 60, 89–98. [CrossRef] [PubMed]
96. Bryceson, Y.T.; March, M.E.; Ljunggren, H.G.; Long, E.O. Activation, coactivation, and costimulation of resting human natural
killer cells. Immunol Rev. 2006, 214, 73–91. [CrossRef] [PubMed]
97. Hood, S.P.; Foulds, G.A.; Imrie, H.; Reeder, S.; McArdle, S.E.B.; Khan, M.; Pockley, A.G. Phenotype and Function of Activated
Natural Killer Cells From Patients With Prostate Cancer: Patient-Dependent Responses to Priming and IL-2 Activation. Front.
Immunol. 2019, 9, 3169. [CrossRef] [PubMed]
98. Lugini, L.; Cecchetti, S.; Huber, V.; Luciani, F.; Macchia, G.; Spadaro, F.; Paris, L.; Abalsamo, L.; Colone, M.; Molinari, A.; et al.
Immune surveillance properties of human NK cell-derived exosomes. J. Immunol. 2012, 189, 2833–2842. [CrossRef]
99. Shoae-Hassani, A.; Hamidieh, A.A.; Behfar, M.; Mohseni, R.; Mortazavi-Tabatabaei, S.A.; Asgharzadeh, S. NK Cell-derived
Exosomes From NK Cells Previously Exposed to Neuroblastoma Cells Augment the Antitumor Activity of Cytokine-activated
NK Cells. J. Immunother. 2017, 40, 265–276. [CrossRef]
100. Fehniger, T.A.; Bluman, E.M.; Porter, M.M.; Mrózek, E.; Cooper, M.A.; VanDeusen, J.B.; Frankel, S.R.; Stock, W.; Caligiuri, M.A.
Potential mechanisms of human natural killer cell expansion in vivo during low-dose IL-2 therapy. J. Clin. Invest. 2000, 106,
117–124. [CrossRef]
101. Ishikawa, T.; Okayama, T.; Sakamoto, N.; Ideno, M.; Oka, K.; Enoki, T.; Mineno, J.; Yoshida, N.; Katada, K.; Kamada, K.; et al.
Phase I clinical trial of adoptive transfer of expanded natural killer cells in combination with IgG1 antibody in patients with
gastric or colorectal cancer. Int. J. Cancer 2018, 142, 2599–2609. [CrossRef]
102. Sanchez-Martinez, D.; Allende-Vega, N.; Orecchioni, S.; Talarico, G.; Cornillon, A.; Vo, D.N.; Rene, C.; Lu, Z.Y.; Krzywinska, E.;
Anel, A.; et al. Expansion of allogeneic NK cells with efficient antibody-dependent cell cytotoxicity against multiple tumors.
Theranostics 2018, 8, 3856–3869. [CrossRef]
103. Manuel, J. Low cost tissue culture technology for the regeneration of some economically important plants for developing countries.
Int. J. Agric. Environ. Biotechnol. 2013, 6, 703–711.
104. Daniszewski, M.; Crombie, D.E.; Henderson, R.; Liang, H.H.; Wong, R.C.B.; Hewitt, A.W.; Pébay, A. Automated Cell Culture
Systems and Their Applications to Human Pluripotent Stem Cell Studies. SLAS Technol. 2018, 23, 315–325. [CrossRef] [PubMed]
105. Hassan, M.N.F.B.; Yazid, M.D.; Yunus, M.H.M.; Chowdhury, S.R.; Lokanathan, Y.; Idrus, R.B.H.; Ng, A.M.H.; Law, J.X. Large-Scale
Expansion of Human Mesenchymal Stem Cells. Stem Cells Int. 2020, 2020, 9529465. [CrossRef] [PubMed]
106. Wu, J.Y.; Li, Y.J.; Hu, X.B.; Huang, S.; Xiang, D.X. Preservation of small extracellular vesicles for functional analysis and therapeutic
applications: A comparative evaluation of storage conditions. Drug Deliv. 2021, 28, 162–170. [CrossRef] [PubMed]
107. Gelibter, S.; Marostica, G.; Mandelli, A.; Siciliani, S.; Podini, P.; Finardi, A.; Furlan, R. The impact of storage on extracellular
vesicles: A systematic study. J. Extracell Vesicles 2022, 11, e12162. [CrossRef]
108. Sivanantham, A.; Jin, Y. Impact of Storage Conditions on EV Integrity/Surface Markers and Cargos. Life 2022, 12, 697. [CrossRef]
109. Palay, S.L.; Palade, G.E. The fine structure of neurons. J. Biophys. Biochem. Cytol. 1955, 1, 69–88. [CrossRef]
110. Rashed, M.H.; Bayraktar, E.; Helal, G.K.; Abd-Ellah, M.F.; Amero, P.; Chavez-Reyes, A.; Rodriguez-Aguayo, C. Exosomes: From
Garbage Bins to Promising Therapeutic Targets. Int. J. Mol. Sci. 2017, 18, 538. [CrossRef]
111. Vidal, M. Exosomes: Revisiting their role as “garbage bags”. Traffic 2019, 20, 815–828. [CrossRef]
112. Doyle, L.M.; Wang, M.Z. Overview of Extracellular Vesicles, Their Origin, Composition, Purpose, and Methods for Exosome
Isolation and Analysis. Cells 2019, 8, 727. [CrossRef]
113. Koh, B.; Tan, K.L.; Chan, H.H.; Daniel Looi, Q.H.; Lim, M.N.; How, C.W.; Law, J.X.; Foo, J.B. A Simple Benchtop Filtration Method
to Isolate Small Extracellular Vesicles from Human Mesenchymal Stem Cells. J. Vis. Exp. 2022, 184, e64106. [CrossRef] [PubMed]
Int. J. Mol. Sci. 2023, 24, 4026 25 of 25

114. Al-Masawa, M.E.; Alshawsh, M.A.; Ng, C.Y.; Ng, A.M.H.; Foo, J.B.; Vijakumaran, U.; Subramaniam, R.; Ghani, N.A.A.; Witwer,
K.W.; Law, J.X. Efficacy and safety of small extracellular vesicle interventions in wound healing and skin regeneration: A
systematic review and meta-analysis of animal studies. Theranostics 2022, 12, 6455–6508. [CrossRef] [PubMed]
115. van Reis, R.; Leonard, L.C.; Hsu, C.C.; Builder, S.E. Industrial scale harvest of proteins from mammalian cell culture by tangential
flow filtration. Biotechnol. Bioeng. 1991, 38, 413–422. [CrossRef] [PubMed]
116. Busatto, S.; Vilanilam, G.; Ticer, T.; Lin, W.L.; Dickson, D.W.; Shapiro, S.; Bergese, P.; Wolfram, J. Tangential Flow Filtration for
Highly Efficient Concentration of Extracellular Vesicles from Large Volumes of Fluid. Cells 2018, 7, 273. [CrossRef] [PubMed]
117. Watson, D.C.; Yung, B.C.; Bergamaschi, C.; Chowdhury, B.; Bear, J.; Stellas, D.; Morales-Kastresana, A.; Jones, J.C.; Felber,
B.K.; Chen, X.; et al. Scalable, cGMP-compatible purification of extracellular vesicles carrying bioactive human heterodimeric
IL-15/lactadherin complexes. J. Extracell Vesicles 2018, 7, 1442088. [CrossRef]
118. Théry, C.; Witwer, K.W.; Aikawa, E.; Alcaraz, M.J.; Anderson, J.D.; Andriantsitohaina, R.; Antoniou, A.; Arab, T.; Archer, F.;
Atkin–Smith, G.K.; et al. Minimal information for studies of extracellular vesicles 2018 (MISEV2018): A position statement of
the International Society for Extracellular Vesicles and update of the MISEV2014 guidelines. J. Extracell Vesicles 2018, 7, 1535750.
[CrossRef]
119. Kazemi, N.Y.; Gendrot, B.; Berishvili, E.; Markovic, S.N.; Cohen, M. The Role and Clinical Interest of Extracellular Vesicles in
Pregnancy and Ovarian Cancer. Biomedicines 2021, 9, 1257. [CrossRef]
120. Witwer, K.W.; Théry, C. Extracellular vesicles or exosomes? On primacy, precision, and popularity influencing a choice of
nomenclature. J. Extracell Vesicles 2019, 8, 1648167. [CrossRef]
121. Ashammakhi, N.; Ahadian, S.; Darabi, M.A.; El Tahchi, M.; Lee, J.; Suthiwanich, K.; Sheikhi, A.; Dokmeci, M.R.; Oklu, R.;
Khademhosseini, A. Minimally Invasive and Regenerative Therapeutics. Adv. Mater. 2019, 31, e1804041. [CrossRef]
122. Pinto, A.; Faiz, O.; Davis, R.; Almoudaris, A.; Vincent, C. Surgical complications and their impact on patients’ psychosocial
well-being: A systematic review and meta-analysis. BMJ Open 2016, 6, e007224. [CrossRef]
123. Cheng, H.; Clymer, J.W.; Po-Han Chen, B.; Sadeghirad, B.; Ferko, N.C.; Cameron, C.G.; Hinoul, P. Prolonged operative duration is
associated with complications: A systematic review and meta-analysis. J. Surg. Res. 2018, 229, 134–144. [CrossRef]
124. Sultana, A.; Zare, M.; Thomas, V.; Kumar, T.S.; Ramakrishna, S. Nano-based drug delivery systems: Conventional drug delivery
routes, recent developments and future prospects. Med. Drug Discov. 2022, 15, 100134. [CrossRef]
125. Jin, J.F.; Zhu, L.L.; Chen, M.; Xu, H.M.; Wang, H.F.; Feng, X.Q.; Zhu, X.P.; Zhou, Q. The optimal choice of medication administration
route regarding intravenous, intramuscular, and subcutaneous injection. Patient Prefer. Adherence 2015, 9, 923–942. [CrossRef]
[PubMed]
126. Lam, W.J.; Bhowmick, T.; Gross, A.; Vanschooneveld, T.C.; Weinstein, M.P. Using higher doses to compensate for tubing residuals
in extended-infusion piperacillin-tazobactam. Ann. Pharmacother. 2013, 47, 886–891. [CrossRef] [PubMed]
127. Bolla, B.; Buxani, Y.; Wong, R.; Jones, L.; Dube, M. Understanding IV antimicrobial drug losses: The importance of flushing
infusion administration sets. JAC Antimicrob. Resist. 2020, 2, dlaa061. [CrossRef] [PubMed]
128. Gao, X.L.; Zhang, M.; Tang, Y.L.; Liang, X.H. Cancer cell dormancy: Mechanisms and implications of cancer recurrence and
metastasis. OncoTarget. Ther. 2017, 10, 5219–5228. [CrossRef]
129. Gomis, R.R.; Gawrzak, S. Tumor cell dormancy. Mol. Oncol. 2017, 11, 62–78. [CrossRef]
130. Becker, P.S.; Suck, G.; Nowakowska, P.; Ullrich, E.; Seifried, E.; Bader, P.; Tonn, T.; Seidl, C. Selection and expansion of natural
killer cells for NK cell-based immunotherapy. Cancer Immunol. Immunother. 2016, 65, 477–484. [CrossRef]
131. Bt Hj Idrus, R.; Abas, A.; Ab Rahim, F.; Saim, A.B. Clinical Translation of Cell Therapy, Tissue Engineering, and Regenerative
Medicine Product in Malaysia and Its Regulatory Policy. Tissue Eng. Part A 2015, 21, 2812–2816. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.