The n e w e ng l a n d j o u r na l of
Original Article
From the University of California, San
Francisco, San Francisco (M.D.C., S.G.,
J.L., G.D., P.J.R.); the Infectious Diseases
Research Collaboration (V.A., J.F.N., T.K.,
P.K.T., S.L.N., M.R.K.) and Makerere Uni-
versity (M.R.K.) — both in Kampala,
Uganda; the University of Tübingen,
Tübingen, Germany (V.A.); Brown Uni-
versity, Providence, RI (D.G., K.N., S.S.,
R.M.C., J.A.B.); and Dominican Universi-
ty of California, San Rafael (R.A.C.). Dr.
Conrad can be contacted at m ­ elissa​
.­conrad@​­ucsf​.­edu or at Box 3400, Uni-
versity of California, San Francisco, San
Francisco, CA 94110.
N Engl J Med 2023;389:722-32.
DOI: 10.1056/NEJMoa2211803
Copyright © 2023 Massachusetts Medical Society.
722
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
m e dic i n e
Evolution of Partial Resistance to
Artemisinins in Malaria Parasites in Uganda
Melissa D. Conrad, Ph.D., Victor Asua, M.Sc., Shreeya Garg, B.S.,
David Giesbrecht, Ph.D., Karamoko Niaré, Ph.D., Sawyer Smith, B.S.,
Jane F. Namuganga, M.H.S., Thomas Katairo, M.Sc., Jennifer Legac, B.S.,
Rebecca M. Crudale, B.S., Patrick K. Tumwebaze, M.Sc.,
Samuel L. Nsobya, Ph.D., Roland A. Cooper, Ph.D.,
Moses R. Kamya, M.B., Ch.B., Ph.D., Grant Dorsey, M.D., Ph.D.,
Jeffrey A. Bailey, M.D., Ph.D., and Philip J. Rosenthal, M.D.​​
A BS T R AC T
BACKGROUND
Partial resistance of Plasmodium falciparum to the artemisinin component of arte-
misinin-based combination therapies, the most important malaria drugs, emerged
in Southeast Asia and now threatens East Africa. Partial resistance, which mani-
fests as delayed clearance after therapy, is mediated principally by mutations in the
kelch protein K13 (PfK13). Limited longitudinal data are available on the emergence
and spread of artemisinin resistance in Africa.
METHODS
We performed annual surveillance among patients who presented with uncompli-
cated malaria at 10 to 16 sites across Uganda from 2016 through 2022. We sequenced
the gene encoding kelch 13 (pfk13) and analyzed relatedness using molecular meth-
ods. We assessed malaria metrics longitudinally in eight Ugandan districts from
2014 through 2021.
RESULTS
By 2021–2022, the prevalence of parasites with validated or candidate resistance
markers reached more than 20% in 11 of the 16 districts where surveillance was
conducted. The PfK13 469Y and 675V mutations were seen in far northern Uganda
in 2016–2017 and increased and spread thereafter, reaching a combined prevalence
of 10 to 54% across much of northern Uganda, with spread to other regions. The
469F mutation reached a prevalence of 38 to 40% in one district in southwestern
Uganda in 2021–2022. The 561H mutation, previously described in Rwanda, was first
seen in southwestern Uganda in 2021, reaching a prevalence of 23% by 2022. The 441L
mutation reached a prevalence of 12 to 23% in three districts in western Uganda
in 2022. Genetic analysis indicated local emergence of mutant parasites indepen-
dent of those in Southeast Asia. The emergence of resistance was observed predomi-
nantly in areas where effective malaria control had been discontinued or transmission
was unstable.
CONCLUSIONS
Data from Uganda showed the emergence of partial resistance to artemisinins in
multiple geographic locations, with increasing prevalence and regional spread over
time. (Funded by the National Institutes of Health.)
n engl j med 389;8 nejm.org August 24, 2023
The New England Journal of Medicine
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Partial Resistance to Artemisinins in Malaria Par asites
M
alaria, which is caused primari-
ly by Plasmodium falciparum, remains a
substantial health challenge, particularly
in Africa, where approximately 95% of malaria
cases and deaths occur.1 Malaria control focuses
on the use of long-lasting, insecticide-treated bed
nets and indoor residual spraying of insecticides
to limit mosquito vectors and on the use of ef-
ficacious drugs to treat and prevent malaria.
Resistance to older drugs limited the efficacy of
malaria treatment in Africa, where chloroquine
resistance probably contributed to millions of
excess malaria deaths.2 Early this century, treat-
ment shifted to artemisinin-based combination
therapy that was designed to limit drug resistance
with a rapidly effective, but short-acting, arte-
misinin component combined with a long-acting
partner drug.3 By 2005, artemisinin-based com-
binations, primarily artemether–lumefantrine and
artesunate–amodiaquine, were standard therapy
for uncomplicated malaria across Africa.4
Partial resistance to artemisinins, which man-
ifests clinically as delayed parasite clearance af-
ter treatment with artemisinins5 and in vitro as
enhanced parasite survival after exposure to an
artemisinin pulse,6 was previously reported in
Southeast Asia.5,6 The primary mediator was mu-
tations in the kelch protein (PfK13) propeller do-
main,7 with secondary determinants mediating
specific levels of resistance and fitness.8 Partial
resistance to artemisinins is of particular impor-
tance when it is accompanied by resistance to
partner drugs; this combined resistance led to
decreased efficacies of artemisinin-based combi-
nation therapies in Southeast Asia.9,10 Among the
many PfK13 mutations identified in P. falciparum,
approximately 20 have been associated with par-
tial resistance.11 These mutations have been re-
ported at low prevalence in some regions outside
Southeast Asia, including in Guyana, Papua New
Guinea, and India, but to date these appear to have
been sporadic outbreaks without stable PfK13
mutation prevalence or verified resistance.12
Of greatest concern has been the emergence
and spread of partial resistance to artemisinins
in Africa, where the effects are expected to be
profound. In older studies, PfK13 mutations were
seen in African parasites at low prevalence, but
with few exceptions these were not validated re-
sistance mediators.4,13 In East Africa, resistance-
mediating PfK13 mutations14-16 and in vitro en-
hanced survival of parasites after artemisinin
n engl j med 389;8
The New England Journal of Medicine
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
exposure17 were very uncommon,16 but this resis-
tance status changed recently. In Rwanda, the
PfK13 561H mutation, a validated resistance me-
diator, was first seen in 2014, had a prevalence of
approximately 20% among isolates obtained from
two sites in 2018, and was associated with clini-
cal delayed parasite clearance.18-20 In Uganda, a
single isolate with the 675V mutation and en-
hanced survival of the parasites after in vitro
artemisinin exposure was identified in 2016.21
Subsequently, the 469Y and 675V mutations were
seen at increasing prevalence in northern Ugan-
da,22,23 where they were associated with partial
resistance clinically and in vitro.24,25
Resistance to antimalarials has typically
emerged in regions in which the intensity of
malaria transmission has been relatively low,
facilitated by low population immunity (allow-
ing relatively unfit resistant parasites to spread)
and low complexity of infection (limiting within-
host competition between parasites).26 It is of in-
terest to determine why partial resistance to
artemisinins emerged in historically high-trans-
mission regions of Uganda. To better character-
ize the extent of resistance and explore reasons
for selection, we characterized parasite genotypes
and assessed malaria metrics over the period in
which resistance emerged and spread in Uganda.
Me thods
Genetic Surveillance of P. falciparum
in Uganda
At multiple sites (10 sites from 2016 through
2017 and 16 sites from 2018 through 2022), we
sequenced pf k13 in 50 isolates (2016–2019) or
100 isolates (2020–2022) annually from symp-
tomatic persons who presented with uncompli-
cated malaria at malaria reference centers around
Uganda (Fig. 1 and Table S1 in the Supplemen-
tary Appendix, available with the full text of this
article at NEJM.org). Symptomatic persons were
selected according to convenience at each refer-
ence center at approximately the same time of
year. Results from analyses of a subset of these
isolates obtained from 2017 through 2019 were
published previously.22,23
For our analyses, we used dideoxy sequencing
and molecular inversion probe technology, as
previously described.23 For all analyses, mixed
genotypes at any locus were categorized as mutant.
These studies and assessments of malaria met-
nejm.org August 24, 2023 723
 The n e w e ng l a n d j o u r na l of m e dic i n e

rics were approved by the Makerere University GEN Pf6K repository, collected in 2008–2013).28
Research and Ethics Committee, the Uganda Na- Parasites from Uganda were genotyped accord-
tional Council for Science and Technology, and ing to single-nucleotide polymorphisms that were
the University of California, San Francisco, Human
distributed across the genome.29 Equivalent vari-
Research Protection Program. ant sites were extracted from Pf6K whole genomes
after variant calling. After we filtered data for
Phylogenetic Analysis of Isolates missingness and coverage, a cladogram was cre-
with Partial Resistance Markers ated with the use of R software. Details of our
To assess the origins of mutant parasites, we analytic methods are provided in Section S1 in the
genotyped seven microsatellites flanking pf k13 Supplementary Appendix.
(Table S2) in parasites collected in Uganda from
2017 through 2021, removed samples missing Evaluation of Malaria Metrics at Sites
more than one genotype or with multiple geno- across Uganda
types at any locus, and generated a neighbor- Enhanced malaria surveillance was established
joining tree with the use of R software, version in 2006, and in 2014 it was extended to all sites
4.2.2.27 We also assessed phylogeny using geno- described in this article.30 All outpatients who
types of mutant parasites from Uganda (collected presented to malaria reference centers with sus-
in 2020) and Southeast Asia (from the Malaria- pected malaria underwent laboratory testing by

Koboko Lamwo

Kaabong
Arua
Agago

Kole
Katakwi
Amolatar
Hoima

Kapchorwa

Uganda
Tororo
Jinja
Mubende

Kasese Location of study clinic

Districts that received IRS every
6 mo from 2010 through 2014
and once in 2017
Districts that received four rounds
of IRS every 8 mo from 2016
through 2018
Kanungu
Districts that received regular
IRS beginning in 2015
Districts that did not receive IRS

Rukiga

Figure 1. Map of Uganda and Study Districts.

IRS denotes indoor residual spraying of insecticides.

724 n engl j med 389;8 nejm.org August 24, 2023

The New England Journal of Medicine

Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Partial Resistance to Artemisinins in Malaria Par asites
means of rapid diagnostic test or microscopy. To
evaluate the incidence of malaria, we considered
monthly laboratory-confirmed cases and test pos-
itivity (the percentage of patients tested who had
a positive malaria test) obtained from malaria
reference centers. Because immunity wanes and
the median age at presentation increases with
effective control and decreased force of infection,
we considered the median age of persons who
presented with malaria in order to evaluate anti-
malarial immunity in the population.31
Statistical Analysis
Analysis of 50 samples per site provided the study
with 92% power and 100 samples per site pro-
vided 99% power to detect a mutation at 5% preva-
lence and a 95% confidence level. Statistical
analyses were performed with the use of StataSE
software, version 14 (StataCorp). The prevalences
of mutations were compared with the use of the
two-sided Fisher’s exact test and assessed longi-
tudinally with the use of the chi-square test of
trend, implemented with the use of the ptrend
command. A P value of less than 0.05 was con-
sidered to indicate statistical significance.
R e sult s
Study Sites
We obtained isolates from patients presenting with
uncomplicated malaria beginning in 2016; ma-
laria metric data (monthly number of malaria
cases, test positivity, and median age at time of
presentation with malaria) were from a subset of
sites with enhanced surveillance data available
since 2014 (Fig. 1). Bed nets were distributed in
2013–2014, 2017–2018, and 2020–2021 at all study
sites. Indoor residual spraying, a highly effective
control measure that is limited in scope because
of cost, was implemented selectively, with districts
receiving one of the following: twice-yearly indoor
residual spraying (as part of a national campaign)
with the carbamate compound bendiocarb from
2010 through 2014 plus a single round of the
organophosphate pirimiphos-methyl in 2017, four
rounds of indoor residual spraying with pirimi-
phos-methyl in the context of a research study
every 8 months from 2016 through 2018, once-
or twice-yearly indoor residual spraying starting
in 2015 (bendiocarb, primarily in 2015, then
pirimiphos-methyl from 2016 through 2019 and
n engl j med 389;8
The New England Journal of Medicine
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
clothianidin or clothianidin–deltamethrin in 2020),
or no indoor residual spraying (Fig. 1).30,32
Prevalence of Resistance-Associated
Mutations
Analysis of the oldest available samples identified
modest prevalence of the PfK13 469Y and 675V
mutations in 2016, particularly in Lamwo District
in far north-central Uganda (Fig. 2 and Table S3).
Since 2019, these two mutations were maintained
at a combined prevalence greater than 10% in five
northern districts (Agago, Kaabong, Katakwi,
Koboko, and Lamwo), with the prevalence of each
of the individual mutations increasing over time
(test of trend, P<0.05) at four of the five sites. A
different mutation at the 469 codon, 469F, was
seen in Rukiga District in southwestern Uganda
in 2016 and reached high prevalence in 2021 (40%)
and 2022 (38%); a lower incidence of malaria at
this site in 2017–2018 yielded few samples for
study. The 561H mutation, which had reached a
high prevalence in Rwanda, was first detected in
2021 in Rukiga District at a prevalence of 16%,
increasing to 23% in 2022. The P441L mutation,
at the upstream boundary of the propeller domain,
reached a prevalence of 12 to 23% at three sites
in western Uganda in 2021–2022. The five noted
PfK13 mutations also occurred sporadically in
other regions — for example, in Mubende (469F
in 2018, 469Y in 2020, and 675V in 2020 and
2022) — but emergence was followed by stable
high prevalence only in northern Uganda (469Y
and 675V) and southwestern Uganda (469F, 561H,
and 441L).
Additional PfK13 propeller domain mutations
that are not candidate or validated resistance me-
diators were seen, mostly at a prevalence of less
than 10%, with some clusters suggesting local
emergence and spread (Table S4). Mutations out-
side the propeller domain were common (Table
S5). Overall, multiple PfK13 mutations, includ-
ing five candidate or validated resistance mark-
ers, have shown increased prevalence and spread
over time.
Relatedness of Isolates with
Partial-Resistance Markers
We explored the relatedness of mutant parasites
using two methods. First, we characterized micro-
satellites flanking pfk13 in parasites from Uganda.
Mutant parasites showed distinct haplotypes as-
nejm.org August 24, 2023 725
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Prevalence of Genotypes in Study Districts

2016 2017 2018 2019 2020 2021 2022

C469Y

A675V

C469F

R561H

Prevalence (%)
>30 to 40
>20 to 30
P441L >10 to 20
>0 to 10
0

B Prevalence According to Study District over Time

100 100 100 100 100 100
Agago Amolatar Arua Hoima Jinja Kaabong
75 75 75 75 75 75

50 50 50 50 50 50

25 25 25 25 25 25

0 0 0 0 0 0

100 100 100 Kasese 100 Katakwi 100 Koboko 100 Kole
Kanungu Kapchorwa
75 75 75 75 75 75
Prevalence (%)

50 50 50 50 50 50

25 25 25 25 25 25

0 0 0 0 0 0
2016
2017
2018
2019
2020
2021
22

2016
2017
2018
2019
2020
2021
22
20

20

100 Lamwo 100 Mubende 100 Rukiga 100 Tororo

75 75 75 75 Genotype
675V
50 50 50 50 469Y
469F
25 25 25 25
561H
441L
0 0 0 0
2016
2017
2018
2019
2020
2021
22

2016
2017
2018
2019
2020
2021
22

2016
2017
2018
2019
2020
2021
22

2016
2017
2018
2019
2020
2021
22
20

20

20

20

726 n engl j med 389;8 nejm.org August 24, 2023

The New England Journal of Medicine

Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Partial Resistance to Artemisinins in Malaria Par asites
Figure 2 (facing page). Prevalence of Indicated PfK13
Mutations in Studied Ugandan Districts.
In Panel B, a horizontal dash along the x axis indicates
no prevalence, and no dash means no samples were
obtained in that year.
sociated with each mutation, a finding consistent
with singular origins (Fig. 3A and Table S6). The
less-defined clustering of 675V parasites proba-
bly represents a more distant origin, but multiple
origins cannot be ruled out.
Second, we characterized polymorphisms
across the P. falciparum genome in parasites from
Uganda and Southeast Asia (Fig. 3B). The mutant
parasites from Uganda were phylogenetically dis-
tant from parasites from Southeast Asia with the
675V mutation (parasites from Asia with other
relevant mutations were not available), consistent
with local emergence and spread within Uganda.
The studied mutations in parasites from Uganda
did not segregate in distinct genomic clusters, a
finding that suggests that none were limited to
a particular parasite background, and we found
extensive recombination after emergence in these
parasites as compared with mutations in para-
sites from Southeast Asia, where clonal transmis-
sion with fixed parasite backgrounds was gener-
ally observed.33
Malaria Metrics at Sites across Uganda
To gain insight into factors that might facilitate
resistance selection, we examined monthly malaria
burden, measured as case counts and test posi-
tivity, at the eight malaria reference centers from
which genomic data were also available from
2014 through 2021, capturing the interval during
which partial resistance to artemisinins emerged
in northern Uganda (Fig. 4 and Figs. S1, S2, and
S3). Patterns of malaria incidence were strongly
associated with the use of indoor residual spray-
ing, a highly effective control measure when carba-
mate or organophosphate insecticides are used.30
The incidence was cyclic at many sites, prob-
ably affected by seasonal rainfall, but patterns
that were associated with indoor residual spray-
ing interventions were clearly discernible. Districts
with regular indoor residual spraying from 2010
through 2014 and one additional round in 2017
had relatively low case counts and test positivity
after a sustained period of spraying,30,34 with
marked increases within approximately 4 to 8
n engl j med 389;8
The New England Journal of Medicine
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
months after indoor residual spraying ceased
(Fig. 4A). Increased incidence was necessarily
accompanied by more frequent treatment, pri-
marily with artemether–lumefantrine, and thus
increased selective pressure for drug resistance.35
Districts that had indoor residual spraying that
began in 2015 had marked decreases in case
counts and test positivity after implementation
until a change to different insecticides for spray-
ing and a marked increase in malaria incidence
in 2020 (Fig. 4B). Districts with no history of
indoor residual spraying, including sites with
diverse malaria epidemiologic profiles, had var-
ied case counts and test positivity over time,
including transient seasonal increases in both
(Fig. 4C).
We used the median age of patients at the
time of presentation with malaria as a surrogate
for antimalarial immunity, because as immunity
decreases with effective control, the disease bur-
den shifts to older persons.31 This analysis gen-
erally showed increased median age (indicating
decreased immunity) during periods of malaria
control, such that populations in northern Uganda
had relatively low immunity at the time of cessa-
tion of indoor residual spraying and subsequent
surges in malaria incidence (Fig. S1, S2, and S3).
It is noteworthy that selection of PfK13 mutations
occurred primarily in regions in which malaria-
control measures were discontinued (i.e., north-
ern Uganda) or where malaria transmission was
unstable (i.e., southwestern Uganda).
Discussion
The emergence of partial resistance to artemisi-
nins in P. falciparum in East Africa is a serious
challenge to the control of malaria. We exam-
ined the emergence and spread of five different
resistance-mediating PfK13 mutations in Ugan-
da since 2016. By 2021–2022, the prevalence of
parasites with candidate or validated resistance
markers exceeded 20% in 11 of the 16 districts
where surveillance was conducted, with foci of
more than 50% prevalence in both northern and
southwestern Uganda. The full clinical conse-
quences of these genetic changes are not yet
known, but experience from Southeast Asia in-
dicates that treatment of malaria may be com-
promised by the identified mutations. Improved
characterization of the emergence and spread of
artemisinin resistance and of factors that facili-
nejm.org August 24, 2023 727
 The n e w e ng l a n d j o u r na l of m e dic i n e

A Microsatellite Loci

Kas Kaabogago ’2 9
’1
Kas e ’2 ng ’21 0
ong
Ru a ’20
Ruki kiga ’20

Kaab
0
ga ’2
Rukig

Kase ese ’2 1
1
wo Kase se ’21
’21
es

se
Rukigaa ’20
o ’21
’20

0 ’21 e ’2 a ’2 ’20
0
Kapcho abong ’19

Rukig

’2
Katakwi ’19 Kole ’20

0
Agag

1 0 o ’2 a nd Rukigende
Agago ’20

Kaabong ’21

am

b
Lamwo ’21

Mu
rwa ’21

0
Ka

L
Katakwi

’19
Agago ’19
Lamw Agago ’17
Lam wo ’21 Kaab Agmwoi ’1 o ’1

ago
Kole ’20

Agago ’20
Katakwi ’21
Lamwo ’20
Kaab
Lambong

0
Ka

i ’2 ’20 ’2 g ’2 w orw be
a ’2
Ag
wo ’20

Lamwo ’20
a

kw a go on am ch Mu
o

Agago ’20
Kata wo ’21

’21
Agag
Kat ong ’2 a ’2

Katakwi ’21
ong Kole ’2

im
’20
’20

Lamwo ’21
Lam roro ’2
Kaa Hogago

Lamwo ’20
Lam i ’20 K

Ho
akw

kwi ’19
To wo
b

’19
o ’20

ta kig ga ab LKap

1
Ka 1 ro ’2
’20

8
A

i ’1
im ’19

’2 o
kw
ro or
0 0

ong ’20
Ka a ’2 a ’2 L i ’1 ’20 oim

0
on ago ’1 9

ro T
Ho pch Kat obo

ta
Agago ’20
Ka

Ka ong A
La takw’20 gag
ab

g ’2 ’2 7

Ka Ru A Ka
im orw akw ko H

ole

0
1 1

To
A 0

’20
1
Ka mo Kaa

Kaab
i ’2 a ’2

0 1 am 9
ab lata bo Kaa kw
o r n b ta ’20 ukig

K
Ka Ag ng ’2’21 g ’2 ong Ka ole R
Ag ab ag 1 1 L ’2 K ’20
T o o

8
wo
ag am 1 ro
o ’2 Aga oro ng ’2’21 wo ro
’20

’1
1 go ro ’2 1 To

7
’19 1 ’2 wi
0 tak

a
1 Ka

’2
a ’2

0
La kig 1
g ’2

0
R Ka mwo Ru

’2
R ukig ta ’20 ab
on ’21

ng
’2
Ruukkigaa ’21 Torokwi ’1 0 Ka wi i ’21

bo
le k
iga ’21 ro 9 Ko KataKatakw

a
Ka
’21 ’20
’21
Ru R wo
Rukikiga ’2ukiga ’21 Lam ’20
Ruk ga ’2 0 ’21 ese Arua
Ruk iga ’201 Kas le ’20
iga Ko a ’210
Kanu RuRu kiga’20 Rukig ’2
ng kiga
Rukig u ’21 Rukiga ’20 0
’2 Lamwo
Rukig a ’21 ’21 ’20
Lamwoago ’20
Rukig a ’21 Ag Lamwo
’19
Rukigaa ’21
’21 Lamwo ’21
Rukiga ’21
Rukiga ’21
Rukiga ’21 Kaabong ’21
Rukiga ’21
Rukiga ’21 Kole ’21
Rukiga ’21 Amolatar ’20
Rukiga ’21 Lamwo ’21
Rukiga ’20
Rukiga ’20 Rukiga ’21
Kata Agago ’18
Rukiga ’20 kwi
Rukiga ’21 ’20 Agago ’17
Lamwo ’19 Katakwi ’18
Lamwo ’17
Jinja ’20 Katakwi ’20 wo ’19 Lamwo ’20
Lam Agago ’20
Agago ’20
Lamwo ’20 go ’20 ’21 Ag Kaa Agago ’20
Aga Tororo ag Ka bon
o ab g ’2 Hoima ’20
’20 La 0
Agago ’21 ngu ’20 ’1
9 on
g ’1 mw
Hoim
Agago Kanu g ’21 Hoimaa ’20
’20 Kaabon 9 o ’2
Arua Ag 0 A Hoim ’20
’19 Ag gago Hoim a ’20
Kole ag
o
’2 Jin ag
o ’2 ’20 Kaab a ’20

Ka
1
ga ’2 0 ja La Lam ong ’20

ab
Ar
0
Ruki0 ’20 m Lam wo ’20

on
ua
wo
go ’2 Lam wo ’20

g
’2
’20

’2
Aga ’20

0
Aru wo ’20

0
wo Kaa a ’21
Lam
Kaa bong
bon ’21
0 To g ’21
de ’2 ’20 1 La roro
ben a ’20 ’21 iga se ’2 K mw ’21
Mu Aru kiga Ruk Kase ’20 7 Ko aabo o ’20
Ru wo o ’1 le ’2 ng
Lamo ’21 mw 1 Lam 0 ’21
w La u ’2 wo
Lam ng ’2
nu A 0
Ka Ka Ag Lam A gago
0
PfK13 Genotype
0 0 tak ag w ga
a ’2 0 ’2 ’20
0

’2 e 1 ’2 wi o ’2 o ’2 go
’2

im ’20 ago ases a ’2 1 a 0 ’20

Ho ’20 1
se

kig
roro Ag K Rukig a ’2 1
se

To Ru Ka
Ka

kig a ’2 0 K To Ka K tak
Ru kig e ’2 1 K obo ror abo aab wi
Ru end a ’2
ub k
ig
Ag aabo ko o ’2 ng ’2 ong ’21
ag ng ’21 1 0 ’20 469F
9

o
La Aga ’18 i ’1

M Ru ’1 ’21
on M gago’20 1

7
wo kw

La am am Ka ong Kas ese

A o ’2

0 0
e ’2
1 end 7
i ’1 0

Ag
o ’2 e ’2
m ata

L L
m wo wo ses ’2 ese ’21
m go

469Y
kw ’2

g ’2 ub ’1
8

ag

wo ’1 ’1 e 1
La K
ta a

d
Ka ukig

La
mw n
w

’2 8 8 ’21
’2
La ube

m
0
1
R

wo
M

ago 0

Ka
’21
u ’2 Ag nja ’2

’2
To

0
ab
675V

ro
ro
ab

’17
Ji

’20
Ka

wo

nde ’20

Ka
Ka
Lam0

Ka se ’2

s
’21 0

se
561H
Lam atakw ’20
wo i ’2

ab 1

’21
ung

o
K go

Ag

ng
Mube
0

Ka 8
’21 ’19
a

Agag mwo ’2 gago ’2

ago
g ’21
Kan
Ag

tak
Arua ’20

’1
Kaabong ’20
Lam mwo

wi
’19 ube Agag
441L
Kaabon

Rukiga ’21

Kaabong ’20
1

’21
Rukiga ’21
Agag Rukiga ’2
A
wo

Rukiga ’21
0

Kas a ’20
Ruk se ’20 1

Ag gago
La

Kase roro ’2 nungu horwa ’21

K
Kaabong g ’19
Katakwi ’21

ase 0
Rukiga ’21
Rukiga ’21
Rukiga ’21

Kaab

ago ’20
a ’21

Ruki Rukiga

A
ese
ig
o ’20

se
o ’21

To
Kaabon

nde
Jinja ’20

’2
Jinja ’20
Jinja ’20

’20

g
ong ’21 abong ’18
Jinja ’20

’20
Kaabon
Rukig

Wild type
La

’21 ’20
’21
’20
Mubende ’21

o
Ka
Lamwo

Ka
g ’19

’2

or other
Lamwo ’21

0
Kapc
’20

’21
Lamwo ’17

B Polymorphic Loci
PfK13 Genotype Lamwo
Agago
Katakwi
Kaabong
Kole
Kaabong
469F Agago
Agago
Agago
Agago
Katakwi

469Y Agago
Lamwo
Agago
Lamwo
Lamwo
Agago
675V Kole
Lamwo
Kole
Lamwo
Lamwo

561H Katakwi
Kaabong
Kanungu
Rukiga
Katakwi
Katakwi
Wild type Lamwo
Kole
Kanungu
Katakwi
or other Agago
Lamwo
Lamwo
Kaabong
Lamwo
Kaabong
Agago
Lamwo
Katakwi
Lamwo
Agago
Agago
Kaabong
Koboko
Agago
Kaabong
Lamwo
Kole
Kaabong
Lamwo
Agago
Koboko
Agago
Lamwo
Kole
Lamwo

Agago
Lamwo
Lamwo
Kole
Uganda
Lamwo
Lamwo
Agago
Kole
Agago
Lamwo
Kaabong
Lamwo
Kaabong
Lamwo
Katakwi
Kole
Koboko
Agago
Kaabong
Kaabong
Katakwi
Katakwi
Katakwi
Kole
Koboko
Katakwi
Kole
Lamwo
Agago
Katakwi
Rukiga
Rukiga
Kanungu
Rukiga
Rukiga
Rukiga
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Hoima
Kole
Koboko
Agago
Agago
Agago
Kole
Kaabong
Agago
Mubende
Mubende
Mubende
Mubende
Mubende
Koboko
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Kawthaung, MM
Mae Sot, TH
Mae Sot, TH
Southeast
Asia
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH
Mae Sot, TH

728 n engl j med 389;8 nejm.org August 24, 2023

The New England Journal of Medicine

Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Partial Resistance to Artemisinins in Malaria Par asites
Figure 3 (facing page). Phylogenetic Relatedness of
Parasites from Uganda.
Dendrograms show the relatedness of parasites from
Uganda with the indicated PfK13 mutations, wild-type
parasites, and parasites containing other PfK13 muta-
tions (other) according to the characterization of seven
microsatellite loci flanking PfK13 (Panel A) and the re-
latedness of mutant isolates from Uganda and South-
east Asia according to polymorphic loci distributed
across the genome (Panel B).
tate spread will assist policymakers in develop-
ing strategies to limit the spread and consequenc-
es of drug resistance.
Our findings with regard to the emergence of
partial resistance to artemisinins in northern
Uganda are notable in one important respect.
Resistance to antimalarial drugs has typically
first emerged in regions of relatively low malar-
ia-transmission intensity. This scenario was seen
with the emergence of resistance to chloroquine,
antifolates, mefloquine, and artemisinins, which
in each case was first observed in Southeast Asia
(and for chloroquine, independently in South
America). Emergence was followed by the spread
of chloroquine resistance from Asia to Africa36
and selective sweeps of antifolate resistance in
Africa.37 Our data, along with previous data from
Rwanda18,19 and Uganda,24 indicate multiple inde-
pendent emergences of PfK13 mutations in these
regions of relatively high malaria burden, rather
than the spread of resistant lineages from Asia.
Why has artemisinin resistance emerged and
spread in northern Uganda, a region that his-
torically has had a high incidence of malaria?
We hypothesize that emergence and spread have
been facilitated by events that have led to high
malaria-transmission intensity in populations with
relatively low malaria immunity. Key features of
malaria control in Uganda have included the
provision of effective antimalarial therapy with
artemisinin-based combinations, national distri-
butions of bed nets, and indoor residual spraying,
which is the most effective vector-control inter-
vention but one that has been limited in scope by
its high cost. Indoor residual spraying with highly
effective insecticides was used in 10 northern
districts from 2010 through 2014 and was accom-
panied by sustained decreases in malaria-test
positivity and parasite prevalence.34,38 The subse-
quent discontinuation of indoor residual spray-
ing in 2014 was associated with the incidence of
n engl j med 389;8 nejm.org
The New England Journal of Medicine
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
malaria increasing by a factor of five within 10
months, and a malaria epidemic was declared by
the Uganda Ministry of Health in June 2015.30,39
In the four studied districts where cessation
of indoor residual spraying was followed by re-
surgent malaria (Agago, Kole, and Lamwo, where
indoor residual spraying was stopped in 2014,
and Katakwi, a northern district that underwent
four rounds of indoor residual spraying from
2016 through 2018 in the context of a research
study),41 the prevalence of any of the five PfK13
mutations increased from 8.0% in 2016 to 32.0%
in 2022 (Table S7). The prevalence of these mu-
tations in these districts was significantly higher
than in districts where there was sustained in-
door residual spraying since 2015 (difference in
prevalence, 36.0 percentage points for 2020 and
24.1 percentage points for 2021) or at sites with-
out indoor residual spraying (difference in preva-
lence, 25.2, 14.8, and 11.3 percentage points for
2020, 2021, and 2022, respectively) for nearly all
the years studied. More recently, the prevalence
of the 469Y and 675V mutations has stabilized
in northern Uganda but increased in other regions,
which suggests a transition to stable prevalence
across much of the country. Rukiga, in south-
western Uganda, has had emergences of 469F
(an apparent independent emergence) and 561H
(most likely spread from Rwanda), probably fa-
cilitated by unstable malaria transmission in
that district. Finally, the 441L mutation, which is
a candidate resistance mediator, has recently
emerged in western Uganda.
Resistance selection in Uganda may also have
been furthered by the use of artemisinin mono-
therapy, with the artemisinin component unpro-
tected by a long-acting partner drug. The use of
artemisinin monotherapy to treat uncomplicated
malaria has been discouraged, but the practice
persists. In Nigeria, oral artemisinin monothera-
pies made up 2.5% of the market share among
studied drug outlets in 2015.40 The use of intra-
venous or rectal artesunate for severe malaria
should be followed by a full course of an arte-
misinin-based combination,42 but follow-up ther-
apy may be omitted. In addition, the use of paren-
teral drugs, including intravenous artesunate, to
treat uncomplicated malaria, although not recom-
mended, is a well-known practice that increases
the potential for selection of resistant parasites.43
Our study had important limitations. The num-
bers of samples collected per site per year were
August 24, 2023 729
 Copyright © 2023 Massachusetts Medical Society. All rights reserved.
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
The New England Journal of Medicine
August 24, 2023 nejm.org n engl j med 389;8 730
No. of Cases of Malaria No. of Cases of Malaria No. of Cases of Malaria
per Month per Month per Month
0
500
1000
1500
0
500
1000
1500
0
500
1000
1500

Jan. 2014 Jan. 2014 Jan. 2014

May 2014 May 2014 May 2014
Sept. 2014 Sept. 2014 Sept. 2014
Jan. 2015 Jan. 2015 Jan. 2015
May 2015 May 2015 May 2015
Sept. 2015 Sept. 2015 Sept. 2015
Jan. 2016 Jan. 2016 Jan. 2016
May 2016 May 2016 May 2016
Sept. 2016 Sept. 2016 Sept. 2016
Jan. 2017 Jan. 2017 Jan. 2017
May 2017 May 2017 May 2017
Sept. 2017 Sept. 2017 Sept. 2017
Jan. 2018 Jan. 2018 Jan. 2018
Arua
Agago

May 2018 May 2018 May 2018

Amolatar

Sept. 2018 Sept. 2018 Sept. 2018

Jan. 2019 Jan. 2019 Jan. 2019
May 2019 May 2019 May 2019
Sept. 2019 Sept. 2019 Sept. 2019
Jan. 2020 Jan. 2020 Jan. 2020
May 2020 May 2020 May 2020
B Indoor Residual Spraying of Insecticides, 2015–2021

Sept. 2020 Sept. 2020 Sept. 2020

C No History of Indoor Residual Spraying of Insecticides

Jan. 2021 Jan. 2021 Jan. 2021

May 2021 May 2021 May 2021
Sept. 2021 Sept. 2021 Sept. 2021
Jan. 2022 Jan. 2022 Jan. 2022
A Indoor Residual Spraying of Insecticides, 2010–2014 and 2017

0
0
0

20
40
80
20
40
80
20
40
80

100
100
100

0
500
60 1000
1500
0
500
60 1000
1500
0
500
60 1000
1500

Jan. 2014 Jan. 2014 Jan. 2014

May 2014 May 2014 May 2014
Sept. 2014 Sept. 2014 Sept. 2014
Jan. 2015 Jan. 2015 Jan. 2015
May 2015 May 2015 May 2015
Sept. 2015 Sept. 2015 Sept. 2015
Jan. 2016 Jan. 2016 Jan. 2016
May 2016 May 2016 May 2016
Sept. 2016 Sept. 2016 Sept. 2016
Jan. 2017 Jan. 2017 Jan. 2017
May 2017 May 2017 May 2017
Sept. 2017 Sept. 2017 Sept. 2017
Jan. 2018 Jan. 2018 Jan. 2018
Kole

May 2018 May 2018 May 2018

Tororo

Kanungu
Sept. 2018 Sept. 2018 Sept. 2018
Jan. 2019 Jan. 2019 Jan. 2019

Month and Year

Month and Year
Month and Year

May 2019 May 2019 May 2019

Sept. 2019 Sept. 2019 Sept. 2019
Jan. 2020 Jan. 2020 Jan. 2020
May 2020 May 2020 May 2020
Sept. 2020 Sept. 2020 Sept. 2020
Jan. 2021 Jan. 2021 Jan. 2021
May 2021 May 2021 May 2021
Sept. 2021 Sept. 2021 Sept. 2021
Jan. 2022 Jan. 2022 Jan. 2022

0
0
0

20
40
80
20
40
60
80
20
40
80

100
100
100

Prevalence of K13 Mutations (%)

0
500
60 1000
1500
0
500
60 1000
1500

Jan. 2014 Jan. 2014

May 2014 May 2014
Sept. 2014 Sept. 2014
Jan. 2015 Jan. 2015
May 2015 May 2015
Sept. 2015 Sept. 2015
Jan. 2016 Jan. 2016
May 2016 May 2016

per mo
Sept. 2016 Sept. 2016

of malaria
Jan. 2017 Jan. 2017

No. of cases
May 2017 May 2017
Sept. 2017 Sept. 2017
Jan. 2018 Jan. 2018

Bendiocarb
May 2018 May 2018
Lamwo

Mubende
Sept. 2018 Sept. 2018
Jan. 2019 Jan. 2019

Pirimiphos-methyl
May 2019 May 2019
Sept. 2019 Sept. 2019
Jan. 2020 Jan. 2020
May 2020 May 2020

Clothiandin–deltamethrin
Sept. 2020 Sept. 2020
Jan. 2021 Jan. 2021
May 2021 May 2021
Sept. 2021 Sept. 2021

No mutation detected
Jan. 2022 Jan. 2022

0
0

20
40
60
80
20
40
60
80

100
100

Prevalence of 441L mutation

Prevalence of 469F mutation
Prevalence of 469Y mutation
Prevalence of 675V mutation
Prevalence of K13 Mutations (%) Prevalence of K13 Mutations (%)
m e dic i n e of n e w e ng l a n d j o u r na l The
 Partial Resistance to Artemisinins in Malaria Par asites

Figure 4 (facing page). Relationship between Malaria
Metrics and Prevalence of PfK13 Mutations.
Malaria incidence and mutation prevalence are shown
for sites that received indoor residual spraying of insec-
ticides. The metrics shown were assessed monthly and
are displayed with locally weighted scatterplot smooth-
ing. Arrows indicate times of insecticide application;
the darkest arrows indicate an apparently ineffective in-
secticide (clothiandin–deltamethrin). Histogram bars
indicate the prevalence of PfK13 mutations. A horizon-
tal dash along the x axis indicates sample collection oc-
curred but no mutations were detected. Other metrics
of malaria (test positivity and median age of presenta-
tion with malaria) are shown in Figures S1, S2, and S3.
small, which limited the precision of prevalence
estimates. Assessments of associations between
ecologic factors and drug-resistance markers re-
lied on data collected for different purposes and
did not cover all sites that were studied for drug
resistance. Metrics for malaria incidence (month-
ly case counts and test positivity) and immunity
(median age at presentation with malaria) were
necessarily imprecise, but exact measures were
challenging to obtain and were beyond the ca-
pacity of our surveillance network. Overall, our
proposal that the establishment of resistance
genotypes was facilitated by a high incidence of
malaria after periods of low incidence in north-
ern Uganda should be considered a hypothesis in
need of additional testing.
Our results show worrisome, sustained preva-
lence in Uganda of P. falciparum with artemisinin-
resistance–mediating PfK13 mutations. The 469Y
and 675V mutations are extending their range,
and three other mutations, 469F, 561H, and 441L
have emerged in western Uganda. Resistance
selection is possible under many scenarios, but
our results suggest that in a population with a
low level of immunity to malaria (owing to a
sustained low burden of disease), a malaria
epidemic increased the likelihood of resistance
selection. In northern Uganda, this scenario
occurred after the withdrawal of effective ma-
laria control.
The epidemiology of malaria in Africa is highly
varied, with many regions having major fluctua-
tions in disease incidence, as seen in Uganda.
Hence, there is concern about multiple additional
emergences of partial resistance to artemisinins.
Further study is needed to determine whether
efforts to blunt malaria epidemics, including the
maintenance of effective malaria control inter-
ventions and prompt attention to regions with
increasing incidence, may decrease the emergence
and spread of antimalarial drug resistance in
Africa.
Supported by grants (AI075045, AI089674, and AI139520)
from the National Institutes of Health.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
We thank Adrienne Epstein, Bryan Greenhouse, and Isabel
Rodriguez-Barraquer at the University of California, San Fran-
cisco, for helpful discussions; Deborah Chin at Brown University
for assistance with sequencing; Venkatachalam Udhayakumar,
Naomi Lucchi, and Samaly Svigel at the Centers for Disease Con-
trol and Prevention for providing advice and reagents for micro-
satellite analyses; the staff at Ugandan health centers for sample
and data collection; and all the study participants for generously
providing blood samples for genomic surveillance.

References
1. World Health Organization. World
malaria report 2021. December 6, 2021
(https://www​.­who​.­int/​­teams/​­global​
-­malaria​-­programme/​­reports/​­world​
-­malaria​-­report​-­2021).
2. Trape JF. The public health impact of
chloroquine resistance in Africa. Am J Trop
Med Hyg 2001;​64:​Suppl:​12-7.
3. Nosten F, White NJ. Artemisinin-based
combination treatment of falciparum ma-
laria. Am J Trop Med Hyg 2007;​77:​Suppl:​
181-92.
4. Conrad MD, Rosenthal PJ. Antima-
larial drug resistance in Africa: the calm
before the storm? Lancet Infect Dis 2019;​
19(10):​e338-e351.
5. Dondorp AM, Nosten F, Yi P, et al. Ar-
temisinin resistance in Plasmodium falci-
parum malaria. N Engl J Med 2009;​361:​
455-67.
6. Witkowski B, Amaratunga C, Khim
N, et al. Novel phenotypic assays for the
detection of artemisinin-resistant Plasmo-
dium falciparum malaria in Cambodia: in-
vitro and ex-vivo drug-response studies.
Lancet Infect Dis 2013;​13:​1043-9.
7. Ariey F, Witkowski B, Amaratunga C,
et al. A molecular marker of artemisinin-
resistant Plasmodium falciparum malaria. Na-
ture 2014;​505:​50-5.
8. Stokes BH, Dhingra SK, Rubiano K, et
al. Plasmodium falciparum K13 mutations in
Africa and Asia impact artemisinin resis-
tance and parasite fitness. Elife 2021;​10:​
e66277.
9. Amaratunga C, Lim P, Suon S, et al.
Dihydroartemisinin-piperaquine resistance
in Plasmodium falciparum malaria in Cam-
bodia: a multisite prospective cohort study.
Lancet Infect Dis 2016;​16:​357-65.
10. Phyo AP, Ashley EA, Anderson TJC, et
al. Declining efficacy of artemisinin com-
bination therapy against P. falciparum ma-
laria on the Thai-Myanmar border (2003-
2013): the role of parasite genetic factors.
Clin Infect Dis 2016;​63:​784-91.
11. WWARN K13 Genotype-Phenotype
Study Group. Association of mutations in
the Plasmodium falciparum Kelch13 gene
(Pf3D7_1343700) with parasite clearance
rates after artemisinin-based treatments
— a WWARN individual patient data meta-
analysis. BMC Med 2019;​17:​1.
12. Dhorda M, Amaratunga C, Dondorp
AM. Artemisinin and multidrug-resistant
Plasmodium falciparum — a threat for ma-
laria control and elimination. Curr Opin
Infect Dis 2021;​34:​432-9.
13. Ndwiga L, Kimenyi KM, Wamae K, et
al. A review of the frequencies of Plasmo-
dium falciparum Kelch 13 artemisinin resis-
tance mutations in Africa. Int J Parasitol
Drugs Drug Resist 2021;​16:​155-61.
14. Conrad MD, Bigira V, Kapisi J, et al.
Polymorphisms in K13 and falcipain-2 as-
sociated with artemisinin resistance are

n engl j med 389;8 nejm.org August 24, 2023 731

The New England Journal of Medicine
Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Partial Resistance to Artemisinins in Malaria Par asites

not prevalent in Plasmodium falciparum iso- artemisinin resistance across Uganda. 34. Raouf S, Mpimbaza A, Kigozi R, et al.
lated from Ugandan children. PLoS One J Infect Dis 2021;​223:​985-94. Resurgence of malaria following discon-
2014;​9(8):​e105690. 24. Balikagala B, Fukuda N, Ikeda M, et tinuation of indoor residual spraying of
15. Conrad MD, Nsobya SL, Rosenthal PJ. al. Evidence of artemisinin-resistant ma- insecticide in an area of Uganda with pre-
The diversity of the Plasmodium falciparum laria in Africa. N Engl J Med 2021;​385:​ viously high-transmission intensity. Clin
K13 propeller domain did not increase 1163-71. Infect Dis 2017;​65:​453-60.
after implementation of artemisinin- 25. Tumwebaze PK, Conrad MD, Okitwi 35. Mpimbaza A, Babikako H, Rutazanna
based combination therapy in Uganda. M, et al. Decreased susceptibility of Plas- D, et al. Adherence to malaria manage-
Antimicrob Agents Chemother 2019;​ modium falciparum to both dihydroartemis- ment guidelines by health care workers in
63(10):​e01234-e19. inin and lumefantrine in northern Uganda. the Busoga sub-region, eastern Uganda.
16. Tacoli C, Gai PP, Bayingana C, et al. Nat Commun 2022;​13:​6353. Malar J 2022;​21:​25.
Artemisinin resistance-associated K13 26. Masserey T, Lee T, Golumbeanu M, et 36. Wellems TE, Plowe CV. Chloroquine-
polymorphisms of Plasmodium falciparum al. The influence of biological, epidemio- resistant malaria. J Infect Dis 2001;​184:​
in southern Rwanda, 2010-2015. Am J logical, and treatment factors on the es- 770-6.
Trop Med Hyg 2016;​95:​1090-3. tablishment and spread of drug-resistant 37. Pearce RJ, Pota H, Evehe MS, et al.
17. Cooper RA, Conrad MD, Watson QD, Plasmodium falciparum. Elife 2022;​11:​11. Multiple origins and regional dispersal of
et al. Lack of artemisinin resistance in 27. Talundzic E, Okoth SA, Congpuong resistant dhps in African Plasmodium fal-
Plasmodium falciparum in Uganda based K, et al. Selection and spread of artemis- ciparum malaria. PLoS Med 2009;​ 6(4):​
on parasitological and molecular assays. inin-resistant alleles in Thailand prior to e1000055.
Antimicrob Agents Chemother 2015;​59:​ the global artemisinin resistance contain- 38. Kigozi R, Baxi SM, Gasasira A, et al.
5061-4. ment campaign. PLoS Pathog 2015;​11(4):​ Indoor residual spraying of insecticide
18. Uwimana A, Legrand E, Stokes BH, et e1004789. and malaria morbidity in a high transmis-
al. Emergence and clonal expansion of in 28. Ahouidi A, Ali M, Almagro-Garcia J, sion intensity area of Uganda. PLoS One
vitro artemisinin-resistant Plasmodium fal- et al. An open dataset of Plasmodium falci- 2012;​7(8):​e42857.
ciparum kelch13 R561H mutant parasites parum genome variation in 7,000 worldwide 39. Okullo AE, Matovu JKB, Ario AR, et
in Rwanda. Nat Med 2020;​26:​1602-8. samples. Wellcome Open Res 2021;​6:​42. al. Malaria incidence among children less
19. Uwimana A, Umulisa N, Venkatesan 29. Verity R, Aydemir O, Brazeau NF, et than 5 years during and after cessation of
M, et al. Association of Plasmodium falci- al. The impact of antimalarial resistance indoor residual spraying in Northern Ugan-
parum kelch13 R561H genotypes with de- on the genetic structure of Plasmodium fal- da. Malar J 2017;​16:​319.
layed parasite clearance in Rwanda: an ciparum in the DRC. Nat Commun 2020;​ 40. ACTwatch Group, Ujuju C, Anyanti J,
open-label, single-arm, multicentre, ther- 11:​2107. Newton PN, Ntadom G;​ACTwatch Group.
apeutic efficacy study. Lancet Infect Dis 30. Namuganga JF, Epstein A, Nank- When it just won’t go away: oral artemis-
2021;​21:​1120-8. abirwa JI, et al. The impact of stopping inin monotherapy in Nigeria, threatening
20. Straimer J, Gandhi P, Renner KC, and starting indoor residual spraying on lives, threatening progress. Malar J 2017;​
Schmitt EK. High prevalence of Plasmodi- malaria burden in Uganda. Nat Commun 16:​489.
um falciparum K13 mutations in Rwanda is 2021;​12:​2635. 41. Mulebeke R, Wanzira H, Bukenya F, et
associated with slow parasite clearance 31. Kigozi SP, Kigozi RN, Epstein A, et al. al. Implementing population-based mass
after treatment with artemether-lumefan- Rapid shifts in the age-specific burden of drug administration for malaria: experi-
trine. J Infect Dis 2022;​225:​1411-4. malaria following successful control in- ence from a high transmission setting in
21. Ikeda M, Kaneko M, Tachibana S-I, et terventions in four regions of Uganda. North Eastern Uganda. Malar J 2019;​18:​
al. Artemisinin-resistant Plasmodium falci- Malar J 2020;​19:​128. 271.
parum with high survival rates, Uganda, 32. Epstein A, Maiteki-Sebuguzi C, 42. World Health Organization. WHO
2014-2016. Emerg Infect Dis 2018;​24:​718- Namuganga JF, et al. Resurgence of ma- guidelines for malaria. March 14, 2023
26. laria in Uganda despite sustained indoor (https://www​.­who​.­int/​­publications/​­i/​­item/​
22. Asua V, Vinden J, Conrad MD, et al. residual spraying and repeated long last- ­g uidelines​-­for​-­malaria).
Changing molecular markers of antima- ing insecticidal net distributions. PLOS 43. Wang LT, Bwambale R, Keeler C, et al.
larial drug sensitivity across Uganda. An- Glob Public Health 2022;​2(9):​e0000676. Private sector drug shops frequently dis-
timicrob Agents Chemother 2019;​ 63:​ 33. Parobek CM, Parr JB, Brazeau NF, et pense parenteral anti-malarials in a rural
e01818-e18. al. Partner-drug resistance and popula- region of Western Uganda. Malar J 2018;​
23. Asua V, Conrad MD, Aydemir O, et al. tion substructuring of artemisinin-resis- 17:​305.
Changing prevalence of potential media- tant Plasmodium falciparum in Cambodia. Copyright © 2023 Massachusetts Medical Society.
tors of aminoquinoline, antifolate, and Genome Biol Evol 2017;​9:​1673-86.

732 n engl j med 389;8 nejm.org August 24, 2023

The New England Journal of Medicine

Downloaded from nejm.org on February 21, 2024. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.