ArticlePDF Available

A Randomised, Double Blind, Placebo-Controlled Pilot Study of Oral Artesunate Therapy for Colorectal Cancer

Authors:

Abstract and Figures

Artesunate is an antimalarial agent with broad anti-cancer activity in in vitro and animal experiments and case reports. Artesunate has not been studied in rigorous clinical trials for anticancer effects. To determine the anticancer effect and tolerability of oral artesunate in colorectal cancer (CRC). This was a single centre, randomised, double-blind, placebo-controlled trial. Patients planned for curative resection of biopsy confirmed single primary site CRC were randomised (n = 23) by computer-generated code supplied in opaque envelopes to receive preoperatively either 14 daily doses of oral artesunate (200 mg; n = 12) or placebo (n = 11). The primary outcome measure was the proportion of tumour cells undergoing apoptosis (significant if > 7% showed Tunel staining). Secondary immunohistochemical outcomes assessed these tumour markers: VEGF, EGFR, c-MYC, CD31, Ki67 and p53, and clinical responses. 20 patients (artesunate = 9, placebo = 11) completed the trial per protocol. Randomization groups were comparable clinically and for tumour characteristics. Apoptosis in > 7% of cells was seen in 67% and 55% of patients in artesunate and placebo groups, respectively. Using Bayesian analysis, the probabilities of an artesunate treatment effect reducing Ki67 and increasing CD31 expression were 0.89 and 0.79, respectively. During a median follow up of 42 months 1 patient in the artesunate and 6 patients in the placebo group developed recurrent CRC. Artesunate has anti-proliferative properties in CRC and is generally well tolerated.
Content may be subject to copyright.
(This is a sample cover image for this issue. The actual cover is not yet available at this time.)
This is an open access article which appeared in a journal published
by Elsevier. This article is free for everyone to access, download and
read.
Any restrictions on use, including any restrictions on further
reproduction and distribution, selling or licensing copies, or posting
to personal, institutional or third party websites are defined by the
user license specified on the article.
For more information regarding Elsevier’s open access licenses
please visit:
http://www.elsevier.com/openaccesslicenses
Author's personal copy
Original Article
A Randomised, Double Blind, Placebo-Controlled Pilot Study of Oral
Artesunate Therapy for Colorectal Cancer
,
☆☆
Sanjeev Krishna
a,
, Senthil Ganapathi
b
, Irina Chis Ster
a
, Mohamed E.M. Saeed
c
,MattCowan
d
,
Caroline Finlayson
a
, Hajnalka Kovacsevics
a
,HerwigJansen
e,1
,PeterG.Kremsner
f
,
Thomas Efferth
c
, Devinder Kumar
b
a
Institute of Infection and Immunity, Department of Pathology, United Kingdom
b
Department of Surgery, St. George's,University of London, Cranmer Terrace, SW17 0RE, United Kingdom
c
Department of Pharmaceutical Biology, Johannes Gutenberg-University, Staudinger Weg 5, 55128 Mainz, Germany
d
Department of Gastroenterology, Surrey and Sussex Healthcare NHS Trust, East Surrey Hospital, Canada Avenue, Redhill, Sussex RH1 5RH, United Kingdom
e
Dafra Pharma nv, 2300 Turnhout, Belgium
f
Institut für Tropenmedizin, Universitätsklinikum Tübingen, Wilhelmstraße 27, D-72074 Tübingen, Germany
abstractarticle info
Article history:
Received 6 October 2014
Received in revised form 13 November 2014
Accepted 13 November 2014
Available online 15 November 2014
Keywords:
Colorectal cancer
Artesunate
Dihydroartemisinin
Ki67
Neutropaenia
Background: Artesunate is an antimalarial agent with broad anti-cancer activity in in vitro and animal experi-
ments and case reports. Artesunate has not been studied in rigorous clinical trials for anticancer effects.
Aim: To determine the anticancer effect and tolerability of oral artesunate in colorectal cancer (CRC).
Methods: This was a single centre, randomised, double-blind, placebo-controlled trial. Patients planned for cura-
tive resection of biopsy conrmed single primary site CRC were randomised (n = 23) by computer-generated
code supplied in opaque envelopes to receive preoperatively either 14 daily doses of oral artesunate (200 mg;
n = 12) or placebo (n = 11). The primary outcome measure was the proportion of tumour cells undergoing
apoptosis (signicant if N7% showed Tunel staining). Secondary immunohistochemical outcomes assessed
these tumour markers: VEGF, EGFR, c-MYC, CD31, Ki67 and p53, and clinical responses.
Findings: 20 patients (artesunate = 9, placebo = 11) completed the trial per protocol. Randomization groups
were comparable clinically and for tumour characteristics. Apoptosis in N7% of cells was seen in 67% and 55%
of patients in artesunate and placebo groups, respectively. Using Bayesian analysis, the probabilities of an
artesunate treatment effect reducing Ki67 and increasing CD31 expression were 0.89 and 0.79, respectively.
During a median follow up of 42 months 1 patient in the artesunate and 6 patients in the placebo group devel-
oped recurrent CRC.
Interpretation: Artesunate has anti-proliferative properties in CRC and is generally well tolerated.
© 2014 Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/3.0/).
1. Introduction
Colorectal cancer (CRC) contributes 910% of the annual global can-
cer burden in men (746,000 cases) and women (614,000 cases) (Ferlay
et al., 2012). In the UK, 110 new cases are diagnosed daily, with older
patients particularly at risk of death (UK CR, 2014) and with N50% of
newly diagnosed cases having locally advanced disease (T3/T4). Resec-
tion is the only curative treatment for non-metastatic CRC but this has to
be combined with neo-adjuvant chemo- and/or radio-therapy, to
downstage more advanced presentations.
Prognosis with best available treatments does not increase disease
free or overall survival beyond ~ 60% at 5 years after diagnosis. For
most patients, access to advanced treatment modalities is lacking, too
expensive to be widely available, or associated with signicant morbid-
ity thereby further compromising their survival. There is therefore a
continuing and urgent need to develop new, cheap, orally effective
and safe CRC therapies. One approach is to study existing drugs that al-
ready have some anticancer properties in experimental settings, and to
assess their safety and efcacy in in vivo studies.
Artesunate is derived from artemisinin, which is extracted from
Artemisia annua L. and is a widely used antimalarial that can be ad-
ministered by oral, rectal and parenteral routes (Gomes et al., 2009;
Kremsner and Krishna, 2004; Kremsner et al., 2012; Nealon et al.,
2002; Hien et al., 1994, 1992; Jiang et al., 1982). Soon after the isolation
of artemisinin by a Chinese government's programme, the anticancer
properties of artemisinins were rst reported (Efferth et al., 2007;
EBioMedicine 2 (2015) 8290
Funding
☆☆ This study did not receive any specic funding. Drug and placebo was supplied by
Dafra pharmaceuticals.
Corresponding author.
1
Deceased.
http://dx.doi.org/10.1016/j.ebiom.2014.11.010
2352-3964/© 2014 Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/3.0/).
Contents lists available at ScienceDirect
EBioMedicine
journal homepage: www.ebiomedicine.com
Author's personal copy
Krishna et al., 2008). Subsequently, many studies of artemisinins using
in vitro and animal models have conrmed their remarkable capacity
to exert broad anti-cancer effects (Efferth et al., 2007). They reduce
cell proliferation and angiogenesis and trigger apoptosis (Anfosso
et al., 2006; Efferth et al., 2001, 1996).
There have only been isolated case reports in humans of anti-cancer
effects of artemisinins (reviewed Krishna et al., 2008). These include
cases of metastatic uveal melanoma (Berger et al., 2005) laryngeal squa-
mous cell carcinoma (Singh and Verma, 2002) and pituitary macro-
adenoma (Singh and Panwar, 2006). An open-label Chinese study
treated non-small cell lung cancer patients and showed prolonged
time to cancer progression compared with controls when artesunate
was added to conventional treatment (Zhang et al., 2008), but no
benet on mortality. An open-label pilot study of patients receiving
artesunate for advanced cervical cancer suggested that it was well toler-
ated and improved symptoms (Jansen et al., 2011). There has been a
phase II trial on the activity of artesunate in non-resectable tumours of
dogs (Rutteman et al., 2013)andefcacy of extracts of A. annua in 5 vet-
erinary sarcomas (Breuer and Efferth, 2014). This study examines anti-
CRC effects and tolerability of artesunate used as monotherapy in a
rigorous study design.
2. Methods
2.1. Ethics
The trial was approved by Wandsworth Ethics Committee (Wands-
worth UK, Ref: 08/H0803/3) and was registered (ISCRTN05203252).
2.2. Trial Design
This was a single-centre, double-blind, placebo-controlled trial with
balanced randomisation of patients (1:1) conducted at the St George's
University of London, UK and St. George's Healthcare NHS Trust.
2.3. Participants for Inclusion
Eligible participants were with biopsy conrmed single primary site
colorectal adenocarcinoma; aged 2190 years; with all stages amenable
to surgical treatment and not requiring neoadjuvant treatment; with
planned curative resection; and with written, informed consent.
2.4. Exclusion Criteria
These were: contraindication to useof artesunate due to hypersensi-
tivity; pregnancy; history of hearing or balance problems; immunosup-
pression or concomitant medication known to interact with artesunate
(see below); weight b50 kg or N100 kg; severe anaemia (haemoglobin
b8 g/dL); other planned intervention, apart from standard of care;
inability to give informed consent; inability or unwillingness to take ef-
fective contraception in women of child-bearing age; chronic kidney
disease of NKF D/QOFI stage 3 or above (eGFR b60 mL/min); bilirubin
N2 of the upper limit of normal without haemolysis or known chronic
liver disease.
2.5. Recruitment
Recruitment was at St George's Healthcare NHS Trust in London
from 9 March 2009 to 15 October 2012.
2.6. Interventions
Patients received two weeks of experimental medication (artesunate
or placebo) just before surgery and standard care. Artesunate (Arinate®
100 mg) was manufactured by Famar Italia S.p.A and matching pla-
cebo tablets were manufactured by MPF in The Netherlands under a
manufacturing licence in accordance with EU cGMP certied by
Dafra Pharma (Belgium). Study medication was packaged, labelled
and certied by B&C CliniPack (Belgium) and was in pack sizes of
3 100mg andwasreceived,storedanddispensedbythePharmacy
at St George's Healthcare NHS Trust.
The dose of artesunate for the study was 200 mg orally, daily for
fourteen days, with medication stopped 4872 h prior to surgery.
Medication was provided in blister packs with one patient box pro-
vided 14 doses, sufcient for the duration of the study.
There was nodelay in surgery if patients entered into this study, nor
any otherchange in clinicalmanagement, and the 62 day rule (requiring
treatment within this time period after conrmation of diagnosis) was
strictly adhered to.
2.7. Outcomes
The primary endpoint of the trial was the presence or absence of
signicant apoptosis in the epithelial cells of the tumour specimen de-
ned as N7% of cells with apoptotic features.
Secondary outcomes included seven immunohistochemical stains
applied to the parafn-embedded tumour specimens and quantied
in both epithelial cells and broblasts: vascular endothelial growth fac-
tor (VEGF), c-MYC status and EGF-receptor status; microvessel density
determining the quantity of the cluster of differentiation 31 (CD31) pro-
tein; proliferative activity assessed with Ki67 staining and p53 tumour
suppressor protein expression. Each stain in each patient was generally
evaluated in 6 microscopic areas with a semiautomatic system (in a few
cases 7 or 8 areas were evaluated, and in some especially for bro-
blasts measurements less than 6 or no areas could be evaluated).
2.8. Blood Samples
Three blood samples were taken: (1) at baseline, (2) after one
week of medication (following protocol amendment for enhanced
safety monitoring) and (3) after ending the two week medication
(just before surgery). In each sample the safety measures included
assay of potassium, sodium, creatinine, urea, albumin, alkaline phospha-
tase, ALT, bilirubin, haemoglobin, platelet count and white cell count.
Carcinoembryonic antigen (CEA), was monitored where available in pa-
tients at baseline and after randomization.
2.9. Secondary Outcomes
These were measures of safety and tolerability (both clinical and
laboratory) according to conventional criteria assessed by comparing
baseline blood test results and those during or after treatment and anti-
cancer efcacy (with markers described above).
2.9.1. Changes to Outcomes
There were no changes to predened endpoints.
2.10. Sample Size
An indicative sample size calculation, given thepioneering nature of
this pilot study, was carried out on the primary outcome before starting
the trial based on the assumptionthat colorectal cancer is unlikely to ex-
hibit signicant apoptosis if untreated. Most patients in the placebo
group (more than 95%) were anticipated to have less than 7% of cells
with apoptotic features. The majority of patients (greater than 60%) in
the artesunate group were anticipated to have signicant apoptosis.
This large difference was derived from published baseline estimates of ap-
optotic indices (Yamamoto et al., 1998; Ikenaga et al., 1996; Bendardaf
et al., 2003). With equal group sizes a sample size of 2*11 was estimated
to have 80% power and accepting a Type I error of 5% for superiority, bear-
ing in mind that in most pilot studies the aim is to demonstrate proof-of-
concept (Arain et al., 2010) rather than exclusively test a hypothesis.
83S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
2.11. Randomisation
Subjects were randomised to receive either artesunate or placebo in
equal numbers. Randomisation was performed using a computer-
generated code, and results supplied in opaque and sealed envelopes
by Dafra Pharma. After enrollment and allocation of the next study
number in the series, participants were given their randomization
pack by a pharmacist. Copies of the key to the randomisation codes
were held by theClinical Trials Pharmacist only and allocation and con-
cealmentsteps were performedin Belgium. The code wasnot opened to
investigators, patients, data collectors until the data collection ended,
histological results had been analysed and datasets were locked.
2.12. Sequence Generation and Allocation Concealment Mechanism
Study medications were pre-packed in blister packs and consecu-
tively numbered for each participant according to the randomisation
schedule. Each participant was assigned an order number once they
had consented to the study and after eligibility checks, and they re-
ceived the medication pack with the corresponding randomization
number.
2.13. Blinding
It became necessary to unblind the allocation to 2 participants
during the course of the study at the request of the MHRA after receipt
of notication of Adverse Events. Blinding was maintained for all inves-
tigators and codes were supplied by the Sponsor's ofce (SGUL) to the
MHRA.
2.14. Data Collection and Structure
Immunohistopathological data generated multiple measurements
per individual as the number of slides differed according to the sizes
of the tumours. Each measurement represents an estimate of staining
of cancer cells found on a slide with 0 denoting no staining observed
on a section. Therefore, the dataset inherits a hierarchical structure
with patients at level one and within individual measurements as
level two. With the exception of the three individuals (Fig. 1)there
are no records considered missing for statistical analysis.
2.15. Exploratory Statistical Analysis
The nature of all data variables have been graphically assessed and
summarised accordingly with means/medians standard deviation for
continuous data or proportions for binary data. Correlations were ex-
plored with Spearman's coefcient.
2.16. Immunochemistry Results and Inferential Data Analysis
A random effects (variance components) model was employed for
immunochemistry data to capture their variabilities correctly, given
their inherent hierarchical structures.
Patients were followed-up with CEA measurements every 6 months
and annual CT scans for disease recurrence. Time since surgery to the
rst disease recurrence has been modelled with survival analysis.
Patient CRC06 has been included in survival analysis and although
there were no samples obtained for immunochemistry, the patient
was known to have survived. Patient CRC13 was initially randomised
Assessed for eligibility (n= 47)
I
A
Excluded (n= 30)
Not meeting inclusion criteria (n= 24)
Declined to participate (n= 6)
Analysed (n= 11)
Excluded from analysis (n= 0)
Lost to follow-up (n= 0)
Discontinued intervention (n= 0)
Allocated to intervention (n= 11)
Received allocated intervention (n= 11)
Did not receive allocated intervention (n= 0)
Lost to follow-up (n= 0)
Discontinued intervention (n= 0)
Allocated to intervention (n= 12)
Received allocated intervention (n= 10)
Did not receive allocated intervention
(travelled elsewhere for surgery) (n= 2)
Analysed (n= 9 )
Excluded from analysis (no cancer in
resected bowel) (n= 1)
Allocation
Follow-Up
Randomized (n= 23)
Enrollment
Analysis
Fig. 1. Patientow diagram. Assessment for eligibility was recorded after6 patients had been randomised, so that these added to the 17 patientsrandomised after screening give the total
number of randomised patients (n = 23).
84 S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
to artesunate but was included in the survival analysis as placebo as s/he
did not receive any drug (as per protocol analysis). Patient CRC21 was
deemed as missing and sensitivity analysis including them in either
artesunate or placebo groups is provided. The Cox proportional hazard's
(PH) model has been applied to investigate the hazard ratio of disease
recurrence for artesunate compared with placebo. and their pointwise
95% condence intervals are provided for each treatment group.
2.17. Statistical Inference
Model based Bayesian analysis and classical frequentist approaches
have been applied as appropriate. Frequentist statistical signicance is
conventionally associated with p-values lessthan 0.05 with uncertainty
of parameters assessed by the 95% condence intervals (CI). Parameter
estimates in Bayesian inference are summarised by their posterior
means and the corresponding 95% credible intervals (CrI). Initially, no
prior knowledge was assumed for the parameter that quanties the
treatment effect, i.e. the difference between the group means in terms
of immunochemistry measurements and the inference that has been
drawn. If prior information exists, then it is appropriate to assess how
the parameter values change based on this evidence. Ki67 and CD31
were the only stains with prior anti-CRC information available (Li
et al., 2007; Jansen et al., 2011). A sensitivity analysis was conducted
with Ki67. Statistical software used included OpenBUGS (Thomas
et al., 2006) STATA (StataCorp. 2013. Stata Statistical Software: Release
13. College Station, TX: StataCorp LP) and R (Team RDC, 2011).
3. Results
3.1. Participant Flow
Fig. 1 summarises patient ows. 12 patients were randomised to re-
ceive artesunate and 11 to placebo. 2 patients did not receive artesunate
despite randomization (one travelled for surgery outside the UK and
could not be followed up, and the drug expiry date for the relevant
batch had been reached when the other patient attended). One
artesunate recipient could not be evaluated for the primary endpoint
Table 1
Baseline demographic, clinical and laboratory characteristics.
Artesunate
N=12
Mean (SD)
Placebo
n=11
Mean (SD)
Numbers
a
(artesunate/
placebo)
Demographics 3
Age (y) 69 (11) 66 (14) 12/10
Gender (%, F) 0.58 0.64
Ethnicity (% Caucasians) 0.83 0.82
Height (m) 1.70 (0.12) 1.62 (0.08) 10/9
Weight (kg) 74 (16.7) 75 (66, 85) 11/10
Biochemistry
Sodium (mmol/L) 140 (1)
a
139 (2.28) 11/11
Albumin (g/L) 38 (5.7) 36 (5.7) 11/11
ALT (U/L) 23 (8) 24 (10)
a
11/11
Bilirubin (μmol/L) 9 (3) 8 (3) 9/11
Creatinine (μmol/L) 73 (23) 65 (15) 9/11
Urea (mmol/L) 5 (2) 5 (1) 9/11
Haematology
Haemoglobin (g/dL) 11.5 (2) 12.2 (2)
a
9/11
White Cell count (/L) 5.3 (2.8) 6.7 (2) 9/11
Platelet Count (/L) 370000 (20000)
a
31300 (78000) 8/11
Dukes' stage
A22
B53
C1 2 6
C2 1 1
a
Numbers of patients contributing to each value are indicated in the last column.
Table 2
Predicted grand meansof immunohistochemistry results. Results are presented on a linear scale by treatment groups with their 95% credible intervals following a Bayesian analysis
which takes into account the variability within individual measurements as well as that of between different individuals in the cluster. The probability of an effect is the probability that
the difference between the grandmeans in the two groups (artesunateplacebo) is greaterthan 0 (and is not a p-value).*For Ki67 all corresponding estimates with informative priorsare
presented in after sensitivity analysis, with skeptical priors presented rst followed by informative priors. Results for some epithelial stains were correlated with each other in placebo
recipients (EGFR and c-MYC; r = 0.664; 0.026) and for artesunate recipients (EGFR and p53, r = 0.68; p = 0.04, and Tunel and p53 r = 0.87; p = 0.0025).
Marker Artesunate Placebo P of an effect Difference or change Sensitivity analysis (prior)
Mean 95% CrI Mean 95% CrI
Epithelial
cMYC 45 (27, 53) 43 (27, 64) 0.59 2.6 (22, 26)
CD31 8 (3,12) 5 (2,9) 0.79 2.6 (4, 9)
EGFR 32 (12, 58) 27 (9, 45) 0.66 5.4 (24, 32)
p53 25 (5, 45) 18 (3,35) 0.72 7 (20, 30)
Tunel 18 (4,32) 20 (7,32) 0.4 2(20, 17)
VEGF 38 (21, 54) 43 (40, 75) 0.45 5(28, 17)
Ki67* 33 (13, 52) 49 (31, 66) 0.11 16 (42, 10) N (0, 1000)
Non-informative
34 (14, 53) 48 (31, 65) 0.13 15 (40, 10) N (0, 100)
Vaguely informative
38 (22, 54) 44 (29, 59) 0.23 6(22, 10.1) N (0, 10)
Informative
32 (12, 51) 49 (32, 66) 0.08 18 (42, 8) N (15, 1000)
Non-informative
32 (13, 51) 49 (32, 66) 0.08 18 (42, 8) N (15, 100)
Vaguely informative
32 (17, 49) 48 (33, 63) 0.03 15 (31, 0) N (15, 10)
Informative
Fibroblast
c-MYC 7 (7, 21) 19 (7,32) 0.10 12 (30, 7)
CD31 13 (2, 26) 11 (0, 23) 0.58 1 (17, 19)
EGFR 0.1 (0.5, 0.7) 0.4 (0, 0.9) 0.21 0.3 (1, 1)
Ki67 5 (3, 12) 10 (3,16) 0.16 5(15, 5)
p53 0.7 (0.8, 2.1) 0.9 (0.3, 2) 0.40 0.2 (2.1, 1.2)
Tunel 3 (0, 5) 4 (1,6) 0.31 0.8 (4, 3)
VEGF 0.2 (0.02, 0.5) 0.2 (0.1, 0.3) 0.52 0.02 (0.30.3)
85S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
because no tumour was identied by histology after operation. Recruit-
ment ended after the planned numbers were randomised.
3.2. Baseline Data
Baseline demographic and clinical characteristics are summarised in
Table 1. There is comparability between groups, including for Dukes'
staging.
3.3. Primary Outcome
55% of placebo recipients and 67% of artesunate treated patients
achieved the primary outcome (patients in whom the proportion of ap-
optotic cells was N7%). In designing this trial, it was assumed that only
1/11 patients receiving placebo would have N7% of tumour cells
displaying apoptosis. The unpredicted higher baseline values in placebo
recipients precluded detection of an artesunate effect in the primary
outcome. Staining results for Tunel as a continuous variable are in
Table 2.
3.4. Secondary Outcomes
3.4.1. Immunohistochemistry Analyses
Table 2 presents analysis of immunohistochemistry results. A ran-
dom effects model scaled linearly provides the posterior distributions
of the group means, their 95% CrI and their estimated differences be-
tween groups. The estimated posterior distributions of differences be-
tween treatment groups lie on both sides of 0 for most measurements
suggesting that the two treatment groups do not differ markedly for
most analysed markers.
Interestingly, the probability of a reduction in Ki67 staining after
artesunate is 8992% (Table 2; a result that is 10.11, because Ki67 is
downregulated) using a non-informative prior on the Ki67 difference
between groups and resulting in a posterior difference of 16 (42,
10) for treatment. A similar result is conrmed if the condence around
the parameter value is increased as illustrated in Fig. 2a. With an opti-
mistic informative prior for artesunate of 15 (N (15,10)), the prob-
ability of an artesunate effect increases to 97% (10.03; Table 2), by
altering the parameter distributions as summarised in Fig. 2b.
To complete this analysis, we included a skeptical informative prior
for this parameter (N (0, 10)). Despite this, the probability of an
artesunate effect remains at a probability of 0.77 with a true mean
value estimated at 16 (22,10; Table 2). CD31 also provided a high
probability (0.79) for a treatment effect. Representative immuno-
stainings are shown in Fig. 3.
3.4.2. Survival Analysis
During a median follow up of 42 months, there were 6 recurrences in
the placebo group and 1 recurrence in an artesunate recipient. Fig. 4 il-
lustrates results from a Cox's proportional hazards model. The hazard
ratio of rst disease recurrence is 0.16 (95% CI (0.02, 1.3)) in the
artesunate group compared with placebo. The survival beyond 2 years
in the artesunate group is estimated at 91% (95% CI (54%, 98%)) whilst
surviving the rst recurrence in the placebo group is only 57%(95% CI
(28%, 78%)). A full sensitivity analysis for patient CRC21 is given in Sup-
plementary Table 1 and suggests that if this patient was in the
artesunate group without recurrence at 3 years, then a p-value for an
artesunate effect on survival would be p = 0.07 (95% CI 0.02, 1.21).
3.4.3. CEA Levels
Six artesunate and 4 placebo recipients had CEA levels measured be-
fore and after trial medication (and before resection, classied as re-
duced, stable or increased). No patients with artesunate had increased
CEA levels whereas 3 patients in the placebo group had increased values
(p = 0.03, Fisher's exact test).
3.5. Adverse Events
Six patients (26% for the ITT population) had adverse events (2 se-
vere, Table 3 and Supplementary Table 1). Two adverse events possibly
related to study drug are described in detail. In the remaining 4 cases, 2
complications (anastomotic leaks after surgery) were considered
unlikely to be related to artesunate, and one case of iron-deciency
anaemia (with no neutropaenia) was attributed to underlying disease.
There was one report of nausea. Detailed descriptions of 2 cases of
neutropaenia are given below and illustrated in Fig. 5.
CRC04: An 81 yo 51 kg female presented with anaemia and a change
in bowel habit, and was discovered to have a large, annular ascending
colon, polypoidal, carcinoma that was not producing obstruction on colo-
noscopy. She was randomised to receive artesunate. There was no evi-
dence of metastatic spread on staging scans (CEA = 3 μg/L). Her mid-
treatment review was unremarkable. She returned for surgery and was
found to be anaemic and neutropenic (Fig. 4a). She was transfused, mak-
ing this a Grade 3 adverse event according to CTCAE criteria (v4.0; http://
% positive
Placebo Artesunate
-40
-20
0
20
40
60
80
100
02
03
05
08
10
12
15
19
20
22
23
01
04
06
07
09
11
13
14
16
17
18
21
CRC Patients
Staining (%)
Probability density
0
.01
.02
.03
.04
.05
-80 -60 -40 -20 020 40
Treatment effect
a
b
Fig. 2. 2a Ki67 staining in treatment groups. Individual results for Ki 67 epithelial cell
staining (% positive), the grand means in each treatment group as predicted by the ran-
dom effects model and individual predications are displayed. Interquartile intervals are
presented for the raw measurements and the 95% credible intervals correspond to the in-
dividual and group predicted means. This analysis has been carried out under missing at
random assumption (MAR) with large uncertaintyaround the individualmeans of missing
individuals, as expectedand shown for 3 individuals randomisedto receive artesunate but
not able to be analysed fo r reasons given in results. These results correspond to non-
informative prior assumption with regards to the differencebetween the two groups. 2b
Sensitivity analysisto various prior information on thedifference betweenthe two groups
with respect to Ki67 the analysis has been justied by published experimental results.
The probability that the difference between artesunate and placebo is negative remains
high even under a skeptical prior of no effect (0.77).
86 S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_
5x7.pdf). Her neutropenia recovered the following day.
She underwent a laparoscopic right-sided hemicolectomy with an
uneventful recovery. She was offered post-operative chemotherapy in
light of her moderately differentiated pT3, Dukes' stage B adenocarcino-
ma, which she declined. This was dened as a non-severe adverse event
as there was no prolongation in hospitalization.
CRC07: A 79 yo 50 kg lady presented with anaemia, rectal bleeding
and a change in bowel habit in the preceding few months. She had a
history of endometrial carcinoma and underwent a total abdominal
hysterectomy and bilateral salpingo-oophorectomy followed by radio-
therapy 11 years before.
Colonoscopy conrmed an adenocarcinoma with impassable stric-
ture at the splenic exure. Staging CT scan of the chest, abdomen and
pelvis excluded metastasis. She was randomised to receive artesunate.
Her CEA was 212 μg/L but fell steeply following artesunate (to a nadir
of 56 μg/L, Fig. 4b) and no other intervention. She had a persistent
thrombocytosis, but she developed anaemia and leucopenia, which
A
B
C D
E F
G H
Fig. 3. Immunohistochemical staining of biomarkers in colorectal cancer. (a) Detection of apoptotic cells by the Tunel assay, (b) Ki67, (c) p53, (d) EGFR, (e) c-MYC, (f) CD31, (g) VEGF,
(h) negative control (without primary antibody). Magnication: ×250. For determination of protein expression the UltraVision polymer detection method (kit from Thermo Fisher Sci-
entic GmbH, Dreieich, Germany) was used as detailed in Supplementary methods. The immunostained slides were scannedby Panoramic Desk (3D Histotech Pannoramic digital slide
scanner,Budapest, Hungary)and interpreted(Quantication of immunostained slides) bypanoramic viewersoftware (NuclearQuant and membraneQuant, 3DHISTECH) in which positive
stainednucleus or membrane werecounted in each denedannotated area. Evaluation parameters includednumber of overall detected objects(nucleus or membrane)in each annotated
area, average of positivity and intensity.Nuclear stainings(Ki67, p53, c-MYC,TUNEL) were quantied using the Nuclear Qant software and Membrane-bound and cytosolic stainings were
quantied by the MembraneQuant software (3D histoQuant). Results are in Table 2.
87S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
was noted on the day of her planned surgery (Fig. 4c). Her pre-
surgical screening carried out 5 days before surgery also showed
anaemia and leucopenia. Her surgery was delayed and after ap-
propriate expert consultation and bone marrow examination her
anaemia was treated with blood transfusions and her persistent
neutropenia with G-CSF, making this also a Grade 3 adverse event
according to CTCAE criteria.
Bone marrow aspiration showed normal erythropoiesis, with dys-
plastic granulopoiesis and maturation arrest. Some myeloid precursors
were vacuolated with nucleocytoplasmic asynchrony, but without ex-
cess myeloblasts. Megakaryocytes were plentiful. There was a slight in-
crease in plasma cells. Findings were consistent with drug induced
myelosuppression.
The neutrophil count rose after 2 days of G-CSF, which was
stopped. The patient underwent a left hemi-colectomy with en
bloc resection of a small segment of involved small bowel and pri-
mary colonic and small bowel anastamosis 11 days after artesunate,
without any post-operative complications, and a fall in platelet
count (Fig. 4c). The adenocarcinoma was staged as a Dukes' C1 and
T4abN2M0. The tumour was found to be poorly differentiated signet
ring type adenocarcinoma with extramural vascular invasion. She
was advised to have adjuvant chemotherapy as the risk of recur-
rence was ~50%. She declined and opted for surveillance alone.
After 3 years of follow up she is conrmed to be symptom and dis-
ease free and continues to lead an independent life.
4. Discussion
This is the rst randomised, double blind study to test the anti-CRC
properties of oral artesunate. Escape from apoptosis is a hallmark of tu-
mour cells (Fiandalo and Kyprianou, 2012), with higher apoptotic indi-
ces being associated with more aggressive CRCs (Alcaide et al., 2013).
The pre-dened primary endpoint (proportion of patients with N7%
Tunel positive staining of tumour cells) after artesunate treatment
was not informative, perhaps because an unexpectedly high proportion
(55%) of placebo recipients exceeded the pre-dened threshold. Never-
theless, several secondary endpoints have given encouraging results,
despite limitations of a small study size and an inherent variability in
quantitating immunohistochemical markers.
Artesunate has a very high probability (0.97, calculated with an in-
formative prior in Bayesian analysis, Fig. 2b) of effect on Ki67 staining
of tumour cells. This is consistent with a high probability of artesunate
effect on Ki67 staining of broblasts (0.84; Table 2). Ki67 is a marker
of tumour cell proliferation whose upregulation is associated with a
poorer prognosis in colorectal cancer. Other markers of tumour biology
were also affected by artesunate, although with lower probabilities (for
example, 0.79 probability for increased CD31 expression). In one case
(Fig. 4b) there was a ~ 75% fall in circulating CEA levels after 2 weeks
of artesunate treatment alone.
The recurrence-free survival probability was also higher after
artesunate compared with placebo (at 3 years 0.89 compared with
0.5; Fig. 3) although condence intervals for these estimates overlap
(HR 0.16,p = 0.091, Supplementary Table 1) because of the small num-
bers of patients and therefore events included in this study. Till this
analysis, there have been no deaths in artesunate recipients (despite
some patients having relatively poor prognosis), and 3 deaths in place-
bo recipients.
Two patients who were at the lower weight limit for inclusion in this
study (50 kg, giving an effective dose of 4 mg/kg of artesunate/day)
developed leucopenia (Fig. 4). In one case this reversed shortly after
stopping artesunate, whereas in the other G-CSF may have hastened
recovery. Bone-marrow examination suggested a toxic effect of arte-
sunate. These ndings are consistent with the recent observations in
malaria of a dose-dependent neutropenia with artesunate (N4 mg/kg)
(Bethell et al., 2010), although bone marrow examinations have not
been carried out before. We instituted mid-treatment monitoring for
neutropenia after observations on malaria but did not note this compli-
cation in other patients. Artesunate associated leucopenia may be dose-
dependent in cancer patients as it is in malaria, and although delayed
haemolysis has been observed after artemisinin use (Rolling et al.,
2014, 2012) it was not a complication in our patients. In future studies,
it may be safer to restrict daily dose of artesunate to b4 mg/kg and to
monitor for haematological complications. A recent publication on
an artesunate dose-nding study in metastatic breast cancer disease
suggests that 200 mg once a day can be tolerated for up to 3 weeks
(Ericsson et al., 2014).
Liver recurrence was commonest in our patients, followed by perito-
neal and ovarian sites, suggesting that seeding is mainly haematogenous
and trans-peritoneal. As patients had clear circumferential and longitudi-
nal margins at surgery, and detectable metastases were not identied at
Placebo Artesunate
Surv(1yrs)=0.78(95%CI(0.46, 0.92))
Surv(2yrs)=0.63(95%CI(0.32, 0.83))
Surv(3yrs)=0.45(95%CI(0.18, 0.70))
0
.2
.4
.6
.8
1
0
200
400
600
800
1000
1200
1400
Days (12 patients)
Probability of surviving reoccurence
Surv(1yrs)=0.96(95%CI(0.74, 0.99))
Surv(2yrs)=0.93(95%CI(0.61, 0.99))
Surv(3yrs)=0.89(95%CI(0.47, 0.98))
0
.2
.4
.6
.8
1
0
200
400
600
800
1000
1200
1400
Days (10 patients)
Fig. 4. Survival recurrence curves predicted by Cox proportional hazards model. Patient
CRC21 was assumed to be missing completely at random (please see Supplementary
Table 1 for a full sensitivity analysis). In the placebo group 2 patients died within a year
(108, 170) days leaving 10 (83%) in the study, another 2 within the next year (383, 663
days) leaving 8 (66%) in the study and the other two died within the third year of the follow
up (749 and 990, respectively) leaving 50% patients beyond the third year. The only death in
the artesunate group happened after 552 days leaving 9 patients (90%) surviving beyond the
third year. These crude estimates support the estimates from the data above.
Table 3
Adverse events. For treatment allocations, please see Fig. 2a.
Study number Event Due to pre-existing illness Related to study drug Serious? Study treatment Outcome
CRC 004 Neutropaenia and anaemia No Possibly N Resolved
CRC 007 Neutropaenia and anaemia No Possibly Y Stopped Resolved
CRC 017 Nausea, but no vomiting No Possibly N Continued unchanged Resolved
CRC 018 Anastomotic leak No Not related Y Resolved
CRC 019 Anastomotic leak No Unlikely Y Stopped Resolved
CRC 022 Anaemia Yes Unlikely N Continued unchanged Resolved
88 S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
randomisation, it is likely that micrometastases spread through vascular
invasion (VI) caused recurrence. PreviousexperiencesuggeststhatVIpre-
dicts decreased survival in CRC (Ganapathi et al., 2011; Liang et al., 2007;
Talbot et al., 1980; Betge et al., 2012). Patients with cryptic dissemination
of CRC may benet from systemic neo-adjuvant therapy and artesunate
may be particularly suitable because it does not usually delay surgery. It
also reduces liver metastases in an animal model (Li et al., 2007).
These observations provide critical information for the design of fur-
ther studies. In assessing its neo-adjuvant properties, we have also ex-
amined artesunate's mechanisms of action in human CRC. Artesunate
does not restore apoptosis in tumour cells in our study, but rather de-
creases the expression of a Ki67. Ki67 is also an important marker of
prognosis in CRC unlike CD31, which is increased in expression. These
ndings are consistent with uncontrolled observations made in cervical
cancer (Jansen et al., 2011), where decreased Ki67 staining was also ob-
served, although decreased CD31 staining of blood vessels in that study
contrasts with our observations. Detailed laboratory observations on
anti-cancer mechanisms of artesunate such as on proliferation (Efferth
et al., 2007, 2003; Yeung et al., 2013) and expression of tumour cell
markers (Konkimalla et al., 2009; Efferth et al., 2004; Konkimalla and
Efferth, 2010) (including for angiogenesis) can now be interpreted in
light of in vivo observations. Larger clinical studies with artesunate
that aim to provide well tolerated and convenient anticancer regimens
should be implemented, and may provide an intervention where
none is currently available, as well as synergistic benets with current
regimens.
Funding and Acknowledgements
Artesunate and placebo were donated by Dafra Pharma (Turnhout,
Belgium). We thank Dr. Jan Poliniecki for sample size estimations and
statistical support to the Data Safety Monitoring Committee, Mr.
Nikolaos Katsoulas for clinical support and Mrs.Doris Rohr for technical
assistance with immunohistochemistry. This study did not receive any
direct funding. MEMS was funded by a PhD stipend from the National
Research Council, Sudan. This study is dedicated to the memory of
Yasmin Krishna.
Author contributions
SK and HJ conceived the study and designedit together with PGK,DK
and TE; ICS carried out statistical analysis, SG, MC, DK and SK recruited
and managed patients; HK managed the study; CF provided clinical histo-
pathological analysis; TE and MEMS analysed immunohistochemistry; SK
wrote the rst draft to which all authors contributed to produce the nal
version. No author declares a conict of interest.
Appendix A. Supplementary Data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.ebiom.2014.11.010.
References
Alcaide, J., Funez, R., Rueda, A., et al., 2013. The role and prognostic value of apoptosis in
colorectal carcinoma. BMC Clin. Pathol. 13 (1), 24.
Anfosso, L., Efferth, T., Albini, A., Pfeffer, U., 2006. Microarray expression proles
of angiogenesis-related genes predict tumor cell response to artemisinins.
Pharmacog enomics J. 6 (4), 269278.
Arain, M., Campbell, M.J., Cooper, C.L., Lancaster, G.A., 2010. What is a pilot or feasibility
study? A review of current practice and editorial policy. BMC Med. Res. Methodol.
10, 67.
Bendardaf, R., Ristamaki,R., Kujari, H., et al., 2003.Apoptotic index and bcl-2expression as
prognostic factors in colorectal carcinoma. Oncology 64 (4), 435442.
Berger, T.G., Dieckmann, D., Efferth, T., et al., 2005. Artesunate in the treatment of metastatic
uveal melanomarstexperiences.Oncol.Rep.14(6),15991603.
Betge, J., Pollheimer, M.J., Lindtner, R.A., et al., 2012. Intramural and extramural vascular
invasion in colorectal cancer: pro gnostic signicance and quality of pat hology
reporting. Cancer 118 (3), 628638.
Bethell, D., Se, Y., Lon, C., et al., 2010. Dose-dependent risk of neutropenia after 7-day
courses of artesunate monotherapy in Cambodian patients with acute Plasmodium
falciparum malaria. Clin. Infect. Dis. 15, e105e114.
Breuer, E., Efferth, T., 2014. Treatment of iron-loaded veterinary sarcoma by Artemisia
annua. Nat. Prod. Bioprospect. 4, 113118.
Efferth, T., Rucker, G., Falkenberg, M., et al., 1996. Detection of apoptosis in KG-1a leuke-
mic cells treated with investigational drugs. Arzneimittelforschung 46 (2), 196200.
Fig. 5. Adverse events. a. PatientCRC 04's haemoglobin(lled red circles, g/dl),total white
cell count (lled blue squares,× 10
9
/L), neutrophil count (greentriangles, ×10
9
/L) and
platelet count (grey diamonds, x) are shown from the start of the study (Day 0). b. CRC
07's serum carcinoembryonic antigen levels are shown from the start of the study (Day
0). c. CRC 07's haematological results are shown from the start of the study (Day 0) with
symbols as in a. Tx is transfusion of red cells.
89S. Krishna et al. / EBioMedicine 2 (2015) 8290
Author's personal copy
Efferth, T., Dunstan, H., Sauerbrey, A., Miyachi, H., Chitambar, C.R., 2001. The anti-malarial
artesunate is also active against cancer. Int. J. Oncol. 18 (4), 767773.
Efferth, T., Sauerbrey, A., Olbrich, A., et al., 2003. Molecular modes of action of artesunate
in tumor cell lines. Mol. Pharmacol. 64 (2), 382394.
Efferth, T., Ramirez, T., Gebhart, E., Halatsch, M.E., 2004. Combination treatment of glio-
blastoma multiforme cell lines with the anti-malarial artesunate and the epidermal
growth factor receptor tyrosine kinase inhibitor OSI-774. Biochem. Pharmacol. 67
(9), 16891700.
Efferth, T., Giaisi, M., Merling, A., Krammer, P.H., Li-Weber, M., 2007. Artesunate induces
ROS-mediated apoptosis in doxorubicin-resistant T leukemia cells. PLoS One 2 (8),
e693.
Ericsson,T., Blank, A., von Hagens, C., Ashton, M., Abelo, A., 2014. Population pharmacoki-
netics of artesunate and dihydroartemisinin during long-term oral administration of
artesunate to pat ients with metasta tic breast cancer. Eu r. J. Clin. Pharmacol. 70,
14531463.
Ferlay, J., Soerjomataram, I., Ervik,M., et al., 2012. GLOBOCAN 2012 v1.0, CancerIncidence
and Mortality Worldwide: IARC CancerBase No. 11 [Internet], (accessed 11 July
2014).
Fiandalo, M.V., Kyprianou, N., 2012. Caspase control: protagonists of cancercell apoptosis.
Exp. Oncol. 34 (3), 165175.
Ganapathi, S., Kumar, D., Katsoulas, N., et al., 2011. Colorectal cancer in the young: trends,
characteristics and outcome. Int. J. Color. Dis. 26 (7), 927934.
Gomes, M.F., Faiz, M.A., Gyapong, J.O., et al.,2009. Pre-referral rectal artesunate to prevent
death and disability in severe malaria: a placebo-controlled trial. Lancet 373 (9663),
557566.
Hien, T.T., Arnold, K., Vinh, H., et al., 1992. Comparison of artemisinin suppositories with
intravenous artesunate and intravenous quinine in the treatment of cerebral malaria.
Trans. R. Soc. Trop. Med. Hyg. 86, 582583.
Hien, T.T., Arnold, K., Hung, N.G., et al., 1994. Single dose artemisininmeoquine treat-
ment for acute uncomplicated malaria. Trans. R. Soc. Trop. Med. Hyg. 88, 688691.
Ikenaga, M., Takano, Y., Saegusa, M., et al., 1996. Apoptosis of colon cancers assessed by in
situ DNA nick end-labeling method. Pathol. Int. 46 (1), 3337.
Jansen, F.H., Adoubi, I., Comoe, J.C., et al., 2011. First study of oral artenimol-R in advanced
cervical cancer: clinical benet, tolerability and tumor markers. Anticancer Res. 31
(12), 44174422.
Jiang, J.B., Li, G.Q., Guo, X.B., Kong, Y.C., Arnold, K., 1982. Antimalarial activity of meo-
quine and qinghaosu. Lancet ii, 285288.
Konkimalla, V.B., Efferth, T., 2010. Inhibition of epidermal growth factor receptor over-
expressing cancer cells by the aphorphine-type isoquinoline alk aloid, dicentrine.
Biochem. Pharmacol. 79 (8), 10921099.
Konkimalla, V.B., McCubrey, J.A., Efferth, T., 2009. The role of downstream signaling path-
ways of the epidermal growth factor receptor for artesunate's activity in cancer cells.
Curr. Cancer Drug Targets 9 (1), 7280.
Kremsner, P.G., Krishna, S., 2004. Antimalarial combinations. Lancet 364, 285294.
Kremsner, P.G., Taylor, T., Issifou, S., et al., 2012. A simplied intravenous artesunate reg-
imen for severe malaria. J. Infect. Dis. 205 (2), 312319.
Krishna, S., Bustamante, L., Haynes, R.K., Staines, H.M., 2008. Artemisinins: their growing
importance in medicine. Trends Pharmacol. Sci. 29 (10), 520527.
Li, L.N., Zhang, H.D., Yuan, S.J., Tian, Z.Y., Wang, L., Sun, Z.X., 2007. Artesunate attenuates
the growth of human colorectal carcinom a and inhibits hyper active Wnt/beta-
catenin pathway. Int. J. Cancer 121 (6), 13601365.
Liang, P., Nakada, I., Hong, J.W., et al ., 2007. Prognosti c signicance of immuno-
histochemically detected blood and lymphatic vessel invasion in colorectal carcinoma:
itsimpactonprognosis.Ann.Surg.Oncol.14(2),470477.
Nealon, C., Dzeing, A., Muller-Romer, U., et al., 2002. Intramuscular bioavailability and
clinical efcacy of artesunate in gabonese children with severe malaria. Antimicrob.
Agents Chemother. 46 (12), 39333939.
Rolling, T., Schmiedel, S., Wichmann, D., Wittkopf, D., Burchard, G.D., Cramer, J.P., 2012.
Post-treatment haemolysis in severe imported malaria after intravenous artesunate:
case report of three patients with hyperparasitaemia. Malar. J. 11, 169.
Rolling, T., Agbenyega, T., Issi fou, S., et al., 201 4. Delayed hemolysis after treatment with
parenteral artesunate in African children with severe malariaa double-center pro-
spective study. J. Infect. Dis. 209 (12), 19211928.
Rutteman, G.R., Erich, S.A., Mol, J.A., et al., 2013. Safety and efcacy eld study of
artesunate for dogs with non-resectabletumours. Anticancer Res. 33 (5), 18191827.
Singh, N.P., Panwar, V.K., 2006. Case report of a pituitary macroadenoma treated with
artemether. Integr. Cancer Ther. 5 (4), 391394.
Singh, N., Verma, K., 2002. Case report of a laryngeal squamous cell carcinoma treated
with artesunate. Arch. Oncol. 2002, 279280.
Talbot, I.C., Ritchie, S., Leighton, M.H., Hughes, A.O., Bussey, H.J., Morson, B.C., 1980. The
clinical signicance of invasion of veins by rectal cancer. Br. J. Surg. 67 (6), 439442.
Team RDC, 2011. R: A Language and Environment for Statistical Computing. http://www.
R-project.org/ (accessed 12 Jul y 2014).
Thomas, A., O' Hara, R., Ligges, U., Sturtz, S., 2006. Making BUGS open. R News 6 (1),
1217.
UK CR, 2014. Bowel Can cer Statistics. http://www.cancerresearchuk.org/cancer-info/
cancerstats/types/bowel/ (accessed 11 July 2014).
Yamamoto, T., Igarashi, N., Kato, Y., Kobayashi, M., Kawakami, M., 1998. Apoptosis in
adenoma and early adenocarcinoma of the colon. Histol. Histopathol. 13 (3), 743749.
Yeung, T.M., Buskens, C., Wang, L.M., Mortensen, N.J., Bodmer, W.F., 2013. Myobroblast
activation in col orectal cancer lymph node metastases . Br. J. Cancer 108 (10 ),
21062115.
Zhang, Z.Y., Yu, S.Q., Miao, L.Y., et al., 2008. Artesunate co mbined with vinorelbine
plus cisplatin in treatment of advanced non-small cell lung cancer: a randomized
controlled trial. Zhong Xi Yi Jie He Xue Bao 6 (2), 134138.
90 S. Krishna et al. / EBioMedicine 2 (2015) 8290
... Artemisinins have also shown the ability to enhance the cytotoxic effects of platinum agents in ovarian cancer cells [46][47][48]. While most evidence of anticancer efficacy comes from preclinical studies, artesunate has exhibited clinical activity in early-phase clinical trials [18,49,50]. Moreover, artemisinins, including artesunate, are well tolerated, having been widely used throughout the world as frontline treatment for malaria without reports of any serious adverse ...
... Artemisinins have also shown the ability to enhance the cytotoxic effects of platinum agents in ovarian cancer cells [46][47][48]. While most evidence of anticancer efficacy comes from preclinical studies, artesunate has exhibited clinical activity in early-phase clinical trials [18,49,50]. Moreover, artemisinins, including artesunate, are well tolerated, having been widely used throughout the world as frontline treatment for malaria without reports of any serious adverse events [51]. ...
Article
Full-text available
Simple Summary Modern ovarian cancer treatment has not substantially improved outcomes, and superior therapeutic strategies are needed. The aim of this study was to evaluate the efficacy of artesunate and navitoclax drug combination in ovarian cancer. We determined the combination of these two drugs was extraordinarily effective in multiple models of ovarian cancer, in vitro, inhibiting cancer cell proliferation more than expected based on single-agent activities. Unfortunately, we were unable to validate these findings using a mouse model of metastatic ovarian cancer. These data provide valuable information regarding the potential utility and challenges associated with the artesunate/navitoclax drug combination for ovarian cancer therapy. Abstract Artesunate belongs to a class of medications derived from the sweet wormwood plant (Artemisia annua) known as artemisinins. Artesunate has traditionally been used as a frontline treatment for severe malaria but has also demonstrated antineoplastic activity against various malignancies, including ovarian cancer. Data suggest that artesunate exacerbates cellular oxidative stress, triggering apoptosis. In the current study, we investigated the ability of navitoclax, an inhibitor of the antiapoptotic Bcl-2 protein family, to enhance artesunate efficacy in ovarian cancer cells. Artesunate and navitoclax both demonstrated antiproliferative effects on 2D and 3D ovarian cancer cell models as single agents. Upon combination of navitoclax with artesunate, antineoplastic drug synergy was also observed in each of the 2D cell lines and ovarian tumor organoid models tested. Further investigation of this drug combination using intraperitoneal CAOV3 xenograft models in BALB/scid mice showed that the artesunate/navitoclax doublet was superior to single-agent artesunate and vehicle control treatment. However, it did not outperform single-agent navitoclax. With optimization, this drug combination could provide a new therapeutic option for ovarian cancer and warrants further preclinical investigation.
... Interestingly, a clinical study revealed that apoptosis was present in 67% of the tumor cells of patients with colorectal cancer in the artesunate group. Additionally, Ki67 expression decreased and CD31 expression increased in patients in the artesunate group [79]. ...
Article
Full-text available
Background In recent decades, natural compounds have been considered a significant source of new antitumor medicines due to their unique advantages. Several in vitro and in vivo studies have focused on the effect of terpenoids on apoptosis mediated by mitochondria in malignant cells. Recent findings In this review article, we focused on six extensively studied terpenoids, including sesquiterpenes (dihydroartemisinin and parthenolide), diterpenes (oridonin and triptolide), and triterpenes (betulinic acid and oleanolic acid), and their efficacy in targeting mitochondria to induce cell death. Terpenoid‐induced mitochondria‐related cell death includes apoptosis, pyroptosis, necroptosis, ferroptosis, autophagy, and necrosis caused by mitochondrial permeability transition. Apoptosis and autophagy interact in meaningful ways. In addition, in view of several disadvantages of terpenoids, such as low stability and bioavailability, advances in research on combination chemotherapy and chemical modification were surveyed. Conclusion This article deepens our understanding of the association between terpenoids and mitochondrial cell death, presenting a hypothetical basis for the use of terpenoids in anticancer management.
... ART and its derivatives have been effective against cervical (Jansen et al. 2011) and colorectal carcinoma (Krishna et al. 2014) without serious side effects. In a phase I clinical trial, 23 patients with metastatic breast cancer received daily oral administration of ART (100,150,or 200 mg), and only four patients experienced treatment-related, non-serious adverse effects on the auditory system (König et al. 2016). ...
Article
Full-text available
Objectives Chemoprevention can be a treatment for potentially malignant lesions (PMLs). We aimed to evaluate whether artemisinin (ART) and cisplatin (CSP) are associated with apoptosis and immunogenic cell death (ICD) in vitro, using oral leukoplakia (OL) and oral squamous cell carcinoma (OSCC) cell lines, and whether these compounds prevent OL progression in vivo. Methods Normal keratinocytes (HaCat), Dysplastic oral cells (DOK), and oral squamous cell carcinoma (SCC-180) cell lines were treated with ART, CSP, and ART + CSP to analyze cytotoxicity, genotoxicity, cell migration, and increased expression of proteins related to apoptosis and ICD. Additionally, 41 mice were induced with OL using 4NQO, treated with ART and CSP, and their tongues were histologically analyzed. Results In vitro, CSP and CSP + ART showed dose-dependent cytotoxicity and reduced SCC-180 migration. No treatment was genotoxic, and none induced expression of proteins related to apoptosis and ICD; CSP considerably reduced High-mobility group box-1 (HMGB-1) protein expression in SCC-180. In vivo, there was a delay in OL progression with ART and CSP treatment; however, by the 16th week, only CSP prevented progression to OSCC. Conclusion Expression of proteins related to ICD and apoptosis did not increase with treatments, and CSP was shown to reduce immunogenic pathways in SCC-180, while reducing cell migration. ART did not prevent the malignant progression of OL in vivo; CSP did despite significant adverse effects.
... Clinical trials have shown that the anti-tumor effects of artemisinin and its derivatives on cervical cancer, advanced NSCLC, metastatic breast cancer, and colorectal cancer are being translated into clinical practice [63][64][65]. A phase II clinical trial explored the tumor inhibition effect of the combination treatment of DHA with icotinib in EGFR-positive NSCLC patients (NCT03402464) [30]. ...
Article
Full-text available
Background Immunosurveillance is pivotal in the effectiveness of anticancer therapies and tumor control. The ineffectiveness of cisplatin in activating the immunosurveillance is attributed to its lack of adjuvanticity resulting from its inability to stimulate endoplasmic reticulum stress. Dihydroartemisinin demonstrates the anti-tumor effects through various mechanisms, including the activation of the endoplasmic reticulum stress. This study aimed to develop a novel strategy to enhance the immunogenicity of dying tumor cells by combining cisplatin with dihydroartemisinin, thereby triggering effective anti-tumor immunosurveillance and improving the efficacy of cisplatin in clinical practice. Methods Lewis lung carcinoma (LLC) and CT26 colon cancer cell lines and subcutaneous tumor models were used in this study. The importance of immunosurveillance was validated in both immunocompetent and immunodeficient mouse models. The ability of dihydroartemisinin and cisplatin therapy to induce immunogenic cell death and tumor growth control in vivo was validated by prophylactic tumor vaccination and therapeutic tumor models. The underlying mechanism was elucidated through the pharmaceutical or genetic intervention of the PERK/eIF2α pathway in vitro and in vivo. Results Dihydroartemisinin enhanced the generation of reactive oxygen species in cisplatin-treated LLC and CT26 cancer cells. The combination treatment of dihydroartemisinin with cisplatin promoted cell death and ensured an optimal release of damage-associated molecular patterns from dying cancer cells, promoting the phagocytosis of dendritic cells. In the tumor vaccination model, we confirmed that dihydroartemisinin plus cisplatin treatment induced immunogenic cell death. Utilizing immunocompetent and immunodeficient mouse models, we further demonstrated that the combination treatment suppressed the tumor growth of CT26 colon cancer and LLC lung cancer, leading to an improved prognosis through the restoration of cytotoxic T lymphocyte responses and reinstatement of anti-cancer immunosurveillance in vivo. Mechanistically, dihydroartemisinin restored the immunogenicity of cisplatin by activating the adjuvanticity of damage-associated molecular patterns, such as calreticulin exposure, through the PERK/eIF2α pathway. Additionally, the inhibition of eIF2α phosphorylation attenuated the anti-tumor efficiency of C + D in vivo. Conclusions We highlighted that dihydroartemisinin acts as an immunogenic cell death rescuer for cisplatin, activating anticancer immunosurveillance in a PERK/eIF2α-dependent manner and offering a strategy to enhance the anti-tumor efficacy of cisplatin in clinical practice.
... Clinically, similar chemicals as Artenimol-R could downregulate p53, EGFR, Ki-67, and CD31 in biopsy samples among patients with advanced cervical cancer after oral treatment (Jansen et al. 2011). Likewise, Sanjeev Krishna et al. observed inspiring results including higher recurrence-free survival probability, lower mortality, and strong effect on Ki-67 of tumor cells during pilot study of oral Artesunate therapy for colorectal cancer (Krishna et al. 2015). In conclusion, DHA inhibit the malignant biological features of hepatocarcinoma, including proliferation, apoptosis, and abnormal transcription through multiple mechanisms and multiple targets. ...
Article
Full-text available
With high incidence of hepatocarcinoma and limited effective treatments, most patients suffer in pain. Antitumor drugs are single-targeted, toxicity, causing adverse side effects and resistance. Dihydroartemisinin (DHA) inhibits tumor through multiple mechanisms effectively. This study explores and evaluates safety and potential mechanism of DHA towards human hepatocarcinoma based on network pharmacology in a comprehensive way. Adsorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of DHA were evaluated with pkCSM, SwissADME, and ADMETlab. Potential targets of DHA were obtained from SwissTargetPrediction, Drugbank, TargetNET, and PharmMapper. Target gene of hepatocarcinoma was obtained from OMIM, GeneCards, and DisGeNET. Overlapping targets and hub genes were identified and analyzed for Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome pathway. Molecular docking was utilized to investigate the interactions sites and hydrogen bonds. Cell counting kit-8 (CCK8), wound healing, invasion, and migration assays on HepG2 and SNU387 cell proved DHA inhibits malignant biological features of hepatocarcinoma cell. DHA is safe and desirable for clinical application. A total of 131 overlapping targets were identified. Biofunction analysis showed targets were involved in kinase activity, protein phosphorylation, intracellular reception, signal transduction, transcriptome dysregulation, PPAR pathway, and JAK-STAT signaling axis. Top 9 hub genes were obtained using MCC (Maximal Clique Centrality) algorithm, namely CDK1, CCNA2, CCNB1, CCNB2, KIF11, CHEK1, TYMS, AURKA, and TOP2A. Molecular docking suggests that all hub genes form a stable interaction with DHA for optimal binding energy were all less than − 5 kcal/mol. Dihydroartemisinin might be a potent and safe anticarcinogen based on its biological safety and effective therapeutic effect.
Article
Full-text available
Artemisinin is a natural sesquiterpene lactone obtained from the traditional Chinese medicinal herb Artemisia annua L. (qinghao). Artemisinin and its derivatives share an unusual endoper-oxide bridge and are extensively used for malaria treatment worldwide. In addition to antimalarial activities, artemisinin and its derivatives have been reported to exhibit promising anticancer effects in recent decades. In this review, we focused on the research progress of artemisinin and its derivatives with potential anticancer activities. The pharmacological effects, potential mechanisms, and clinical trials in cancer therapy of artemisinin and its derivatives were discussed. This review may facilitate the future exploration of artemisinin and its derivatives as effective anticancer agents.
Article
Cancer cells have significantly higher intracellular free-metal ions levels than normal cells, and it is well known that artemisinin (ART) molecules or its derivatives sensitize cancer cells when its endoperoxide moiety combines with metal ions, resulting in the production of reactive oxygen species, lysosomal degradation of ferritin, or regulation of system Gpx4 leading to apoptosis, ferroptosis or cuproptosis. Artemisinin derivatives (ADs) are reported to interfere more efficiently with metal-regulatory-proteins (MRPs) controlling iron/copper homeostasis by interacting with cytoplasmic unbound metal ions and thereby promoting the association of MRP to mRNA molecules carrying the respective sequences. However, the simple artemisinin analogues are required to be administered in higher doses with repeated administration due to low solubility and smaller plasma half-lives. To overcome these problems, amino ARTs were introduced which are found to be more stable, and later on, a series of ARTs derivatives containing sugar moiety was developed in search of analogues having good water solubility and high pharmacological activity. This review focuses on the preparation of N-glycosylated amino-ART analogues with their application against cancer. The intrinsic capability of glycosylated ART compounds is to give sugar-- containing substrates, which can bind with lectin galectin-8 receptors on the cancer cells making these compounds more specific in targeting cancer. Various AD mechanism of action against cancer is also explored with clinical trials to facilitate the synthesis of newer derivatives. In the future, the latest nano-techniques can be used to create formulations of such compounds to make them more target-specific in cancer.
Chapter
Despite more than 200 approved anticancer agents, cancer remains a leading cause of death worldwide due to disease complexity, tumour heterogeneity, drug toxicity, and the emergence of drug resistance. Accordingly, the development of chemotherapeutic agents with higher efficacy, a better safety profile, and the capability of bypassing drug resistance would be a cornerstone in cancer therapy. Natural products have played a pivotal role in the field of drug discovery, especially for the pharmacotherapy of cancer, infectious, and chronic diseases. Owing to their distinctive structures and multiple mechanistic activities, natural products and their derivatives have been utilized for decades in cancer treatment protocols. In this review, we delve into the potential of natural products as anticancer agents by targeting cancer’s hallmarks, including sustained proliferative signalling, evading growth suppression, resisting apoptosis and cell death, enabling replicative immortality, inducing angiogenesis, and activating invasion and metastasis. We highlight the molecular mechanisms of some natural products, in vivo studies, and promising clinical trials. This review emphasizes the significance of natural products in fighting cancer and the need for further studies to uncover their fully therapeutic potential.
Article
Full-text available
The increase in cases of various cancers, the number of cancer patients, and the serious side effects of current chemical treatments have encouraged researchers to discover novel and more effective drugs from plant sources. In this review, PubMed and Scopus were searched for English-language peer-reviewed articles published since 1994, using the keywords: Medicinal plants, Egypt, and Cancer. The inclusion criteria for this systematic review were English peer-reviewed original research articles. A total of 286 studies were included. Studies have shown that the active compounds in certain plants can have anticancer activities by various mechanisms, such as cell cycle arrest, apoptosis induction, and antioxidant effects. Additionally, it is evident that medicinal plants can act as inhibitors of cancer cells. Plants can also act as immune checkpoint inhibitors, which inhibit proteins such as PD-L1 on tumor cells, PD-1 and CTLA-4 on T cells, and help to regulate immune responses. This review also discusses the efficacy of nanoparticles of these plants against cancer cells. The findings indicate the high potential of investigating the bioactive anticancer components of Egyptian medicinal plants for advancing novel treatments that are more effective in combating cancer. The extracts and active compounds of the medicinal plants detailed in this review could provide the foundation for further clinical trials to be conducted, to develop new anticancer drugs.
Article
ZUSAMMENFASSUNG Artesunat ist ein klinisch etabliertes Medikament zur Malariabehandlung. Präklinische Daten weisen darauf hin, dass Artesunat und Artemisinin, die Leitsubstanz aus der chinesischen Heilpflanze Artemisia annua L., auch gegen Krebs eingesetzt werden könnten. Im Allgemeinen wurden Artemisinin und Artesunat nach den bisher vorliegenden klinischen Befunden gut vertragen, wobei vereinzelt immer wieder auch signifikante Nebenwirkungen wie Hämatotoxizität, Hepatotoxizität oder Neurotoxizität auftraten. Auf das Vorkommen schwerer Nebenwirkungen muss auch weiterhin besonderes Augenmerk gelegt werden, insbesondere wenn Artesunat zusammen mit anderen Medikamenten verabreicht wird. Die veterinärmedizinischen Studien zeigen, dass spontane Tiertumoren nicht nur ein interessantes Modell für die Translation präklinischer Daten in die Klinik sind, sondern dass Artesunat- und Artemisia-annua-Präparate eine neue Behandlungsoption für Haustiere mit Tumorerkrankungen darstellen könnten. Die Wirksamkeit von Artesunat gegen verschiedene solide humane Tumoren ist bisher als eher mäßig einzustufen, wobei das Zervixkarzinom möglicherweise besser anzusprechen scheint. Insgesamt liegen noch zu wenige Daten vor, um die klinischen Ansprechraten bei verschiedenen Tumortypen zuverlässig beantworten zu können. Weitere größere klinische Phase-II- und -III-Studien sind nötig.
Article
Full-text available
Purpose: The purpose of this study were firstly to characterize the population pharmacokinetics of artesunate (ARS) and its active metabolite dihydroartemisinin (DHA) in patients with metastatic breast cancer during long-term (>3 weeks) daily oral ARS administration and secondly to study the relationship between salivary and plasma concentrations of DHA. Methods: Drug concentration-time data from 23 patients, receiving oral ARS (100, 150, or 200 mg OD), was analyzed using nonlinear mixed effects modeling. A combined drug-metabolite population pharmacokinetic model was developed to describe the plasma pharmacokinetics of ARS and DHA in plasma. Saliva drug concentrations were incorporated as being directly proportional to plasma concentrations. Results: A first-order absorption model for ARS linked to a combined two-compartment disposition model for ARS and one-compartment disposition model for DHA provided the best fit to the data. No covariates were identified that could explain between-subject variability. A time-dependent increase in apparent elimination clearance of DHA was observed. Salivary DHA concentrations were proportionally correlated with total DHA plasma concentrations, with an estimated slope factor of 0.116. Conclusions: Population pharmacokinetics of ARS and DHA in patients with breast cancer was well described by a combined drug-metabolite model without any covariates and with an increase in apparent elimination clearance of DHA over time. The estimated DHA saliva/plasma ratio was in good agreement with the reported DHA unbound fraction in human plasma. Saliva ARS concentrations correlated poorly with plasma concentrations. This suggests the use of saliva sampling for therapeutic drug monitoring of DHA. However, further studies are warranted to investigate the robustness of this approach.
Article
Full-text available
Artemisinin, a constituent of Artemisia annua L., is a well-known antimalarial drug. Artemisinin-type drugs also inhibit cancer growth in vitro and in vivo. Herbal extracts of A. annua inhibit the growth of cancer cell lines. Here, we report on the use of capsules containing powder of Herba Artemisiae annuae to treat pet sarcoma. The surgical tumor removal as standard treatment was supplemented by adjuvant therapy with A. annua. One cat and one dog with fibrosarcoma survived 40 and 37 months, respectively, without tumor relapse. Two other dogs suffering from fibrosarcoma and hemangioendothelial sarcoma also showed complete remission and are still alive after 39 and 26 months, respectively. A. annua was well tolerated without noticeable side effects. These four cases indicate that A. annua may be a promising herbal drug for cancer therapy.
Article
Full-text available
Background. Parenteral artesunate is recommended as first-line therapy for severe malaria. While its efficacy is firmly established, data on safety are still incomplete. Delayed hemolysis has been described in hyperparasitemic nonimmune travelers, but it is unknown if African children are equally at risk. Methods. Children aged 6 to 120 months with severe malaria were followed up after treatment with parenteral artesunate in Lambaréné, Gabon, and Kumasi, Ghana. The primary outcome was incidence of delayed hemolysis on day 14. Results. In total, 72 children contributed complete data sets necessary for primary outcome assessment. Delayed hemolysis was detected in 5 children (7%), with 1 child reaching a nadir in hemoglobin of 2.8 g/dL. Patients with delayed hemolysis had higher parasite counts on admission (geometric mean parasite densities (GMPD) 306 968/µL vs 92 642/µL, P = .028) and were younger (median age: 24 months vs 43 months, P = .046) than the rest of the cohort. No correlation with sickle cell trait or glucose-6-phosphate-dehydrogenase deficiency was observed. Conclusions. Delayed hemolysis is a frequent and relevant complication in hyperparasitemic African children treated with parenteral artesunate for severe malaria. Physicians should be aware of this complication and consider prolonged follow-up. Clinical Trials Registration. Pan-African Clinical Trials Registry: PACTR201102000277177 (www.pactr.org).
Article
Full-text available
Alterations to apoptosis are a common occurrence in human tumours. The aim of our study was to determine the influence of apoptotic variations on the carcinogenesis and prognosis of colorectal carcinomas (CRCs). A TUNEL assay was performed on archival material from 103 colorectal carcinomas, 26 adenomas and 20 samples of normal epithelia. The number of apoptotic cells was higher in CRCs (1.09 +/- 0.13) than in adenomas (0.38 +/- 0.23, p = 0.059) and normal epithelium (0.06 +/- 0.04, p = 0.001). In addition, the apoptotic index (AI) was greater in metastatic disease (stage IV) than in other stages (p = 0.017). No relationship was found between apoptotic rates and age, gender or tumour grade. However, patients with tumours that showed higher AI values had a significantly lower disease-free survival (DFS) and overall survival (OS) than those with tumours that had lower AIs (p = 0.020 and p = 0.027). In a multivariate Cox proportional hazards model, AI remained a significant independent predictor of survival. We conclude that disregulated apoptosis is an important event during CRC development and progression. Higher AIs are associated with more aggressive tumours and a poorer prognosis for patients with CRC.
Article
Full-text available
Background: Myofibroblasts have an important role in regulating the normal colorectal stem cell niche. While the activation of myofibroblasts in primary colorectal cancers has been previously described, myofibroblast activation in lymph node metastases has not been described before. Methods: Paraffin-embedded lymph node sections from patients with macrometastases, micrometastases and isolated tumour cells were stained to identify myofibroblasts and to characterise the distribution of different cell types in tumour-containing lymph nodes. The extent of myofibroblast presence was quantified and compared with the size of the metastasis and degree of proliferation and differentiation of the cancer cells. Results: We show substantial activation of myofibroblasts in the presence of colorectal metastases in lymph nodes, which is intimately associated with glandular structures, both in micro- and macrometastases. The degree of activation is positively associated with the size of the metastases and the proportion of Ki67+ve cancer cells, and negatively associated with the degree of enterocyte differentiation as measured by CK20 expression. Conclusion: The substantial activation of myofibroblasts in tumour-containing lymph nodes strongly suggests that these metastatic cancer cells are still significantly dependent on their microenvironment. Further understanding of these epithelial–mesenchymal interactions could lead to the development of new therapies in metastatic disease.
Article
Full-text available
The anti-malarial drug artesunate has shown anticancer activity in vitro and in preliminary animal experiments, but experience in patients with cancer is very limited. Pre-clinical studies in dogs indicated morbidity at high dosage levels. This study evaluated the effects of artesunate in canine cancer cell lines and in canine cancer patients. Four canine cell lines were tested in vitro for sensitivity towards artesunate and dihydroartemisinin (DHA; active metabolite of artesunate). The half-maximal inhibitory concentration (IC50) values for artesunate or DHA were 2-60 μM in three cell lines, while one cell line was much less sensitive to artesunate (IC50 337 μM) than to DHA (IC50 50 μM). A safety/efficacy field study with artesunate was conducted in 23 dogs with non-resectable tumours. Artesunate was administered for 7-385 days at a dosage of 651-1178 (median 922) mg/m(2). No neurological or cardiac toxicity was observed and seven dogs exhibited no adverse effects at all. Fever and haematological/gastrointestinal toxicity, mostly transient, occurred in 16 dogs. One dog died from pneumonia. Plasma artesunate and DHA levels fell below the limit of detection within 8-12 h after artesunate administration, while levels after two hours were close to 1 μM. Artesunate produced a long-lasting complete remission in one case of cancer and short-term stabilization of another seven cases.
Article
Emergence of castration-resistant metastatic prostate cancer is due to activation of survival pathways, including apoptosis suppression and anoikis resistance, and increased neovascularization. Thus targeting of apoptotic players is of critical significance in prostate cancer therapy since loss of apoptosis and resistance to anoikis are critical in aberrant malignant growth, metastasis and conferring therapeutic failure. The majority of therapeutic agents act through intrinsic mitochondrial, extrinsic death receptor pathways or endoplasmic reticulum stress pathways to induce apoptosis. Current therapeutic strategies target restoring regulatory molecules that govern the pro-survival pathways such as PTEN which regulates AKT activity. Other strategies focus on reactivating the apoptotic pathways either by down-regulating anti-apoptotic players such as BCL-2 or by up-regulating pro-apoptotic protein families, most notably, the caspases. Caspases are a family of cystine proteases which serve critical roles in apoptotic and inflammatory signaling pathways. During tumorigenesis, significant loss or inactivation of lead members in the caspase family leads to impairing apoptosis induction, causing a dramatic imbalance in the growth dynamics, ultimately resulting in aberrant growth of human cancers. Recent exploitation of apoptosis pathways towards re-instating apoptosis induction via caspase re-activation has provided new molecular platforms for the development of therapeutic strategies effective against advanced prostate cancer as well as other solid tumors. This review will discuss the current cellular landscape featuring the caspase family in tumor cells and their activation via pharmacologic intervention towards optimized anti-cancer therapeutic modalities. This article is part of a Special Issue entitled "Apoptosis: Four Decades Later".