7-Step Flow Synthesis of the HIV Integrase Inhibitor Dolutegravir

Article abstract of DOI:10.1002/anie.201802256

Dolutegravir (DTG), an important active pharmaceutical ingredient (API) used in combination therapy for the treatment of HIV, has been synthesized in continuous flow. By adapting the reported GlaxoSmithKline process chemistry batch route for Cabotegravir, DTG was produced in 4.5 h in sequential flow operations from commercially available materials. Key features of the synthesis include rapid manufacturing time for pyridone formation, one-step direct amidation of a functionalized pyridone, and telescoping of multiple steps to avoid isolation of intermediates and enable for greater throughput.

Full text of DOI:10.1002/anie.201802256

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201802256
German Edition: DOI: 10.1002/ange.201802256
Multistep Synthesis
7-Step Flow Synthesis of the HIV Integrase Inhibitor Dolutegravir
Robert E. Ziegler, Bimbisar K. Desai, Jo-Ann Jee, B. Frank Gupton, Thomas D. Roper,* and
Timothy F. Jamison*
Abstract: Dolutegravir (DTG), an important active pharma-
ceutical ingredient (API) used in combination therapy for the
treatment of HIV, has been synthesized in continuous flow. By
adapting the reported GlaxoSmithKline process chemistry
batch route for Cabotegravir, DTG was produced in 4.5 h in
sequential flow operations from commercially available mate-
rials. Key features of the synthesis include rapid manufacturing
time for pyridone formation, one-step direct amidation of
a functionalized pyridone, and telescoping of multiple steps to
avoid isolation of intermediates and enable for greater
throughput.
Figure 1. Integrase inhibitors for HIV treatment.
H
IV is a disease that currently affects around 37 million
people.[1] A number of innovative medicines has made HIV
a manageable disease; however, the cost of treatment is still
prohibitive for many patients in lower income countries.[2] As
part of the “Medicines for All” initiative funded by the Bill
and Melinda Gates Foundation,[3] our research groups are
focused on providing greater access to essential medicines for
serious diseases such as HIV, tuberculosis, and malaria. The
initiative has led to a number of reported studies.[4] We
became interested in Dolutegravir due to its importance in
HIV combination therapy.
resistance and minimal side effects have led DTG to be
placed on the World Health Organizationꢀs List of Essential
Medicines.[8] Recently, the first two drug combination therapy
for HIV (Dolutegravir and Rilpivirine) was approved by the
FDA.[9] A number of DTG analogues are currently in clinical
trials, including Cabotegravir 4 and Bictegravir 5; the latter
was recently approved by the FDA in a single tablet, three
drug regimen (Figure 1).[10] These analogues differ from
Dolutegravir 1 in the oxazine ring size; thus, a synthesis for
DTG should also be amenable to 4 or 5 if they emerge as the
integrase inhibitor of choice in the future.
GSK and Shinogi disclosed a number of approaches to the
synthesis of 1 and its analogues.[11] Initial medicinal chemistry
routes exploited commercially available heterocycles such as
nicotinic acid and maltol, and subsequently installed the
requisite functionalized N-H pyridone in ten or more
steps.[11b–d] Wang and co-workers from GSK later published
a highly efficient, chromatography-free approach to Cabote-
gravir 4 through rapid formation of the functionalized
pyridone core 8 and subsequent cyclization with (S)-alaninol
11 to synthesize the 5-membered oxazine ring (Sche-
me 1).[11e,f] We wished to optimize and adapt the synthesis to
a continuous flow system in order to streamline manufactur-
ing of the API.[12] Continuous flow reactions benefit from
Dolutegravir (DTG) 1 (Figure 1) is an HIV integrase
inhibitor co-developed by GlaxoSmithKline (GSK) and
Shinogi that was approved by the Food and Drug Admin-
istration (FDA) in 2013. Integrase inhibitors prevent the HIV
virus from inserting into cellular DNA by blocking transes-
terification, a process that is vital for replication and spread of
the disease.[5] Raltegravir 2 and Elvitegravir 3 were the first
integrase inhibitors to be approved and used in combination
therapy; however, these two drugs require large doses and
a pharmacokinetic booster, respectively, and have shown
vulnerability to HIV virus mutations.[6] DTG is an un-
boosted, once daily 50 mg tablet that is recommended as
a universal first-line treatment in combination therapy due to
its low dosage and limited side effects.[7] Only minimal
resistance has been observed thus far, which has not led to
significant spread of HIV virus after mutation.[6] The high
[*] Dr. R. E. Ziegler, Prof. Dr. T. F. Jamison
Department of Chemistry, Massachusetts Institute of Technology
77 Massachusetts Avenue, Cambridge, MA 02139 (USA)
E-mail: tfj@mit.edu
Dr. B. K. Desai, Dr. J. Jee, Prof. Dr. B. F. Gupton, Prof. Dr. T. D. Roper
Department of Chemical and Life Science Engineering
Virginia Commonwealth University, Biotech 8
737 N. 5th Street, Richmond, VA 23219 (USA)
E-mail: tdroper@vcu.edu
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201802256.
Scheme 1. GSK process synthesis of Cabotegravir 4.
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
increased mixing due to high surface area to volume ratio and
the ability to heat solvents well past their boiling point.[13] By
developing an efficient flow synthesis and telescoping steps to
avoid purifications, we felt we could achieve a short, scalable
synthesis of DTG 1.
both 6 and dimethyl oxalate 7 are solids. A solvent screen in
batch showed that CH3CN afforded the highest conversion
(93%) to product 16 compared to N-methylpyrrolidinone and
MeOH (both 85%); however the CH3CN condition suffered
from poor solubility that led to clogging in flow.[14] In addition,
when examining different bases for the deprotonation/cycli-
zation sequence, NaOMe in MeOH had much better sol-
ubility and conversion to 16 than LiOMe, which was utilized
in the GSK synthesis.[11e,f] These screening results in batch led
us to investigate Step 3 in flow using NaOMe as base and
MeOH as solvent in order to simplify the system through the
use of a single solvent. Following optimization of residence
time and temperature,[14] flow conditions of 858C and 30 min
tR led to a 91% isolated yield of 16 (Scheme 3).
We began our investigation with the condensation reac-
tion of methyl 4-methoxyacetoacetate 12 and dimethylforma-
mide dimethylacetal (DMF-DMA) 13. Our first attempt at
the analogous flow reaction proceeded with 60% conversion
to 14 when an equimolar ratio of the neat reactants 12 and 13
were streamed through a T-mixer at 308C with a residence
time (tR) of 10 min. Further optimization led to the discovery
that an elevated reaction temperature of 858C and 1.6 equiv
of DMF-DMA 13 resulted in full conversion of 12 to 14 via
HPLC with the same 10 min residence time.[14] With an
efficient approach to the dimethyl vinylogous amide 14, we
next sought to telescope Steps 1 and 2 in continuous flow
(Scheme 2). This was achieved by connecting Reactor I with
Next, we examined the synthesis of pyridone 16 from 12 in
a three-step telescope process to obviate time consuming
purification operations and work towards our goal of a fully
continuous synthesis (Scheme 4). After minor modifications
Scheme 4. Three-step telescoped synthesis of pyridone 16.
to the previously optimized flow conditions from Scheme 2
and Scheme 3, the telescoped synthesis of 16 was achieved. In
this setup, several 40 psi back pressure regulators (BPR) were
utilized as check valves and Reactor III was used to facilitate
premixing of solution 7 with the output from Reactor II prior
to the addition of NaOMe. A 55 min residence time in
Reactor IV was required for the reaction to reach completion.
The three-step telescoped synthesis of pyridone 16 led to
a 56% isolated yield in a total residence time of 74 min with
a throughput of 3.4 ghÀ1.
When considering ways to shorten the reported GSK
synthesis, the ester saponification to form 8 and subsequent
amide coupling stood out as an opportunity. Specifically, in
GSKꢀs synthesis of Cabotegravir 4 (Scheme 1),[11e,f] ester 16
was saponified to the 5-carboxylic acid 8 and then treated with
carbonyldiimidazole (CDI) and difluorobenzylamine 9 to
form amide product 17. The process required a filtration and
extraction, and took 17 h in total. We were encouraged by
a similar direct amidation that was reported in the midst of
our own studies.[15] Application of Kumar and co-workersꢀ
AcOH-catalyzed conditions led to high yield and chemo-
selectivity for amidation at the 5-position, producing amide 17
after 10 h (Scheme 5a). We reason that there is both an
electronic and steric effect which led to the selectivity. When
the corresponding N-H pyridone was reacted under the same
conditions, a 2:1 selectivity for 5-amidation over 2-amidation
was observed. The only byproducts observed on a gram scale
batch reaction of 16 were ring opened starting material 6 and
Scheme 2. Telescoped flow synthesis of vinylogous amide 6. PFA=per-
fluoroalkoxy, I.D.=inside diameter.
a T-mixer and adding neat aminoacetaldehyde dimethylacetal
15 directly to the output of Reactor I. Due to the high
crystallinity of the product 6, we conducted Step 2 at 858C to
increase solubility and avoid clogging. The optimized tele-
scoped flow process produced 6 in an isolated yield of 95%
and a throughput of 43 ghÀ1.
The next step in the reaction sequence was the pyridone
16 formation (Scheme 3). Given the complexity of adapting
to continuous flow, we first examined the independent flow
process by starting with purified vinylogous amide 6. The
choice of solvent for the reaction proved to be crucial given
Scheme 3. Pyridone 16 formation in flow.
2
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Next, we examined the acetal deprotection of 17 and
cyclization with (R)-3-aminobutan-1-ol 18 in flow. Initial
attempts were conducted on purified pyridone amide 17 with
an acid additive and amino alcohol 18. Ultimately, we found
that formic acid was required as a co-solvent to observe any
conversion of 17. Our best result afforded an 8% yield of
DTG-OMe 19 as a 5:1 mix of diastereomers favoring the
desired product (Scheme 7).[18] The major byproducts were
Scheme 5. Direct amidation comparison. SS=stainless steel.
Scheme 7. Initial cyclization attempts.
the analogous difluorobenzyl vinylogous amide (< 5% each).
An ethyl ester substrate gave comparable yield and AcOH
was found to be necessary for practical reaction time.[16]
A
brief screen of Brønsted and Lewis acids revealed that the
batch reaction could be accelerated compared to the AcOH
condition, but with a decrease in isolated yield due to other
byproducts and generally poorer solubility.
ring-opened 20, produced through elimination of the hemi-
aminal ether functionality, and the deprotected pyridone
aldehyde 10. We reasoned that a two-step flow procedure in
which an acid would deprotect acetal 17 in one reactor and
then meet amino alcohol 18 in a subsequent reactor may be
more fruitful for conversion and milder conditions. Gratify-
ingly, separating the steps gave full conversion and allowed
for stoichiometric amounts of p-TsOH·H2O instead of neat
formic acid with no observation of elimination byproduct 20.
We next sought to telescope the acid-mediated direct
amidation into the deprotection/cyclization steps to form
a three-step telescoped sequence. Initially, incomplete con-
version was observed due to issues with PhCH3 from Step 4
inhibiting the subsequent steps; however, it was eventually
found that more concentrated reaction solutions and longer
residence times in Steps 5 and 6 led to full conversion of
intermediate amide 17 and aldehyde 10. The optimized setup
proceeded in a total residence time of just over 3 h and a 48%
isolated yield of DTG-OMe 19 in 7:1 dr (Scheme 8). The
major diastereomer was separated by silica gel chromatog-
raphy to give analytically pure material.
The next step was to adapt the process to a flow setup.
Initially, we examined the reaction at 1208C with a residence
time of 1 h. Gratifyingly, product 17 was obtained with the use
of a 100 psi BPR albeit in low conversion of starting material
16. A systematic evaluation of temperature and residence
time led to good conversion at 1508C with a 1 h residence
time and at 1808C with a 30 min residence time.[14] We
screened other solvents and found that dioxane and CH3CN
were viable, but gave lower conversion than PhCH3 in the
flow setup. DCE led to clogging in the system, and DMF and
MeOH failed to produce any desired amide 17 after 1 h.
Under the optimal conditions of 2008C with a residence time
of 124 min, we obtained 96% isolated yield of amide 17 (2%
recovered 16) on a 3 mmol scale, which amounted to 3.5 ghÀ1
(Scheme 5b).
The analogous base-promoted amidation[17] was also
feasible using either LiOMe or NaOMe as base in a MeOH/
PhCH3 mixed solvent system. The optimized flow process led
to both shorter residence time and milder temperature
compared to the acid-mediated method (Scheme 6). How-
ever, attempts to telescope the basic amidation into the
subsequent downstream process led to extensive clogging
issues.
Finally, we examined the demethylation step in continu-
ous flow as a discrete operation using purified 19. Following
a batch screen of different demethylating reagents, it was
Scheme 6. Base-promoted direct amidation in flow.
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
Scheme 8. Telescoped synthesis of DTG-OMe 19.
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
found that GSKꢀs published reaction using LiBr gave the best
Keywords: amidation · continuous flow · HIV ·
multistep synthesis · pyridone
conversion and lowest amount of byproducts.[11e,f] Reaction
temperatures higher than 1208C led to formation of a similar
elimination byproduct to 20. The batch conditions translated
well to flow, with 89% yield of DTG observed at 1008C with
a residence time of 31 min (Scheme 9). The reaction concen-
[1] M. Vitoria, A. M. Hill, N. P. Ford, M. Doherty, S. H. Khoo, A. L.
Pozniak, J. Int. AIDS Soc. 2016, 19, 20504 – 20512.
[2] A. Gupta, S. Juneja, M. Vitoria, V. Habiyambere, B. D.
Nguimfack, M. Doherty, D. Low-Beer, PLoS One 2016, 11, 1 –
18.
[3] Medicines For All website: https://medicines4all.vcu.edu.
[4] For recent publications, see: a) C. E. Ocampo, D. Lee, T. F.
Jamison, Org. Lett. 2015, 17, 820 – 823; b) A. Sevenich, G.-Q. Liu,
A. J. Arduengo 3rd, B. F. Gupton, T. Opatz, J. Org. Chem. 2017,
82, 1218 – 1223; c) J. Verghese, C. J. Kong, D. Rivalti, E. C. Yu, R.
Krack, J. Alcazar, J. B. Manley, D. T. McQuade, S. Ahmad, K.
Belecki, B. F. Gupton, Green Chem. 2017, 19, 2986 – 2991; d) J.
Britton, T. F. Jamison, Angew. Chem. Int. Ed. 2017, 56, 8823 –
8827; Angew. Chem. 2017, 129, 8949 – 8953; e) J. Britton, T. F.
Jamison, Eur. J. Org. Chem. 2017, 6566 – 6574.
[5] For recent reviews, see: a) M. Su, J. Tan, C.-Y. Lin, Drug
Discovery Today 2015, 20, 1337 – 1348; b) S.-X. Gu, P. Xue, X.-L.
Ju, Y.-Y. Zhu, Bioorg. Med. Chem. 2016, 24, 5007 – 5016; c) N. A.
Meanwell, Bioorg. Med. Chem. Lett. 2017, 27, 5355 – 5372; d) M.
Psichogiou, G. Poulakou, D. Basoulis, D. Paraskevis, A. Marko-
giannakis, G. L. Daikos, Curr. Pharm. Des. 2017, 23, 2552 – 2567;
e) Y. Han, T. Mesplede, M. A. Wainberg, Expert Opin. Invest.
Drugs 2017, 26, 1207 – 1213.
Scheme 9. Demethylation to form DTG 1.
tration proved crucial to reproducible, extended running of
the continuous flow reactor. DTG 1 was insoluble upon
cooling to room temperature, which led to clogging at
concentrations higher than 0.5m THF. Attempts to telescope
this final demethylation with the previous three-step
sequence in Scheme 8 led to an 8% isolated yield over four
steps; however, the same clogging issue meant the system
could not be run for more than 10 h at a time.
[6] M. A. Wainberg, T. Mesplede, J. Int. AIDS Soc. 2015, 18, 20824 –
20826.
[7] P. Cahn, Curr. Opin. HIV AIDS 2017, 12, 318 – 323.
[8] WHO Model List of Essential Medicines. 20th list, March 2017.
[9] A. F. Capetti, M. V. Cossu, L. Paladini, G. Rizzardini, Expert
Opin. Pharmacother. 2018, 19, 65 – 77.
In conclusion, we have developed a continuous flow
synthesis of the HIV integrase inhibitor Dolutegravir 1. The
optimized process described involved seven total steps in
three separate flow operations in 24% overall yield (37%
overall when Step 3 was run as a separate flow operation).[19]
The key features of the flow route are rapid manufacturing
time, direct amidation of ester 16 to reduce the step count,
and separation of the acetal deprotection/oxazine formation
flow reactors to attain high reactivity and selectivity for
tricyclic product DTG-OMe 19. Importantly, our synthesis
should be adaptable to both Cabotegravir 4 and Bictegravir 5
by switching the benzylamine and amino alcohol used in the
synthesis. Further studies will focus on the telescoping of all
steps to achieve an end-to-end continuous flow synthesis as
well as formulation of the final API as its sodium salt and to
produce cGMP formulations in an engineered system without
the use of silica gel chromatography.
[10] a) J. Gallant, A. Lazzarin, A. Mills, C. Orkin, D. Podzamczer, P.
Tebas, P.-M. Girard, I. Brar, E. S. Daar, D. Wohl, J. Rockstroh, X.
Wei, J. Custodio, K. White, H. Martin, A. Cheng, E. Quirk,
Lancet 2017, 390, 2063 – 2072; b) P. E. Sax, A. Pozniak, M. L.
Montes, E. Koenig, E. DeJesus, H.-J. Stellbrink, A. Antinori, K.
Workowski, J. Slim, J. Reynes, W. Garner, J. Custodio, K. White,
D. SenGupta, A. Cheng, E. Quirk, Lancet 2017, 390, 2073 – 2082;
c) A. Markham, Drugs 2018, 78, 601 – 606.
[11] For a review, see: a) E. Schreiner, F. Richter, S. Nerdinger, Top.
Heterocycl. Chem. 2016, 44, 187 – 208; for selected articles, see:
b) T. Kawasuji, B. A. Johns, H. Yoshida, T. Taishi, Y. Taoda, H.
Murai, R. Kiyama, M. Fuji, T. Yoshinaga, T. Seki, M. Kobayashi,
A. Sato, T. Fujiwara, J. Med. Chem. 2012, 55, 8735 – 8744; c) T.
Kawasuji, B. A. Johns, H. Yoshida, J. G. Weatherhead, T.
Akiyama, T. Taishi, Y. Taoda, M. Mikamiyama-Iwata, H.
Murai, R. Kiyama, M. Fuji, N. Tanimoto, T. Yoshinaga, T.
Seki, M. Kobayashi, A. Sato, E. P. Garvey, T. Fujiwara, J. Med.
Chem. 2013, 56, 1124 – 1135; d) B. A. Johns, T. Kawasuji, J. G.
Weatherhead, T. Taishi, D. P. Temelkoff, H. Yoshida, T.
Akiyama, Y. Taoda, H. Murai, R. Kiyama, M. Fuji, N. Tanimoto,
J. Jeffrey, S. A. Foster, T. Yoshinaga, T. Seki, M. Kobayashi, A.
Sato, M. N. Johnson, E. P. Garvey, T. Fujiwara, J. Med. Chem.
2013, 56, 5901 – 5916; e) S. N. Goodman, M. D. Kowalski, D. M.
Mans, H. Wang, US 8889877B220141118, 2014; f) H. Wang,
M. D. Kowalski, A. S. Lakdawala, F. G. Vogt, L. Wu, Org. Lett.
2015, 17, 564 – 567.
Acknowledgements
We thank the Bill & Melinda Gates Foundation (“Medicines
for All” intitiative) for funding. We thank Dr. Rachel L.
Beingessner and Dr. Justin Lummiss (MIT) as well as Matt
Graham, Pooja Panchal, Megan Freeman and Dr. Daniel
Cook (VCU) for helpful discussions.
[12] For recent reviews, see: a) L. Malet-Sanz, F. Susanne, J. Med.
Chem. 2012, 55, 4062 – 4098; b) M. Baumann, I. R. Baxendale,
Beilstein J. Org. Chem. 2015, 11, 1194 – 1219; c) R. Porta, M.
Benaglia, A. Puglisi, Org. Process Res. Dev. 2016, 20, 2 – 25;
d) D. L. Hughes, Org. Process Res. Dev. 2018, 22, 13 – 20; for
recent examples, see: e) M. D. Hopkin, I. R. Baxendale, S. V.
Conflict of interest
The authors declare no conflict of interest.
4
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Ley, Org. Biomol. Chem. 2013, 11, 1822 – 1839; f) J. Hartwig, S.
Chem. 2015, 127, 3514 – 3530; c) B. Gutmann, D. Cantillo, C. O.
Ceylan, L. Kupracz, L. Coutable, A. Kirschning, Angew. Chem.
Int. Ed. 2013, 52, 9813 – 9817; Angew. Chem. 2013, 125, 9995 –
9999; g) S. Mascia, P. L. Heider, H. Zhang, R. Lakerveld, B.
Benyahia, P. I. Barton, R. D. Braatz, C. L. Cooney, J. M. B.
Evans, T. F. Jamison, K. F. Jensen, A. S. Myerson, B. L. Trout,
Angew. Chem. Int. Ed. 2013, 52, 12359 – 12363; Angew. Chem.
2013, 125, 12585 – 12589; h) T. Tsubogo, H. Oyamada, S.
Kobayashi, Nature 2015, 520, 329 – 332; i) A. D. Martin, A. R.
Siamaki, K. Belecki, B. F. Gupton, J. Org. Chem. 2015, 80, 1915 –
1919; j) C. A. Correia, K. Gilmore, D. T. McQuade, P. H.
Seeberger, Angew. Chem. Int. Ed. 2015, 54, 4945 – 4948;
Angew. Chem. 2015, 127, 5028 – 5032; k) A. Adamo, R. L.
Beingessner, M. Benham, J. Chen, T. F. Jamison, K. F. Jensen,
J.-C. M. Monbaliu, A. S. Myerson, E. M. Revalor, D. R. Snead,
T. Stelzer, N. Weeranoppanant, S. Y. Wong, P. Zhang, Science
2016, 352, 61 – 67; l) J. C. Yang, D. Niu, B. P. Karsten, F. Lima,
S. L. Buchwald, Angew. Chem. Int. Ed. 2016, 55, 2531 – 2535;
Angew. Chem. 2016, 128, 2577 – 2581; m) N. S. Suveges,
R. O. M. A. de Souza, B. Gutmann, C. O. Kappe, Eur. J. Org.
Chem. 2017, 6511 – 6517; n) P. Zhang, N. Weeranoppanant, D. A.
Thomas, K. Tahara, T. Stelzer, M. G. Russell, M. OꢀMahony,
A. S. Myerson, H. Lin, L. P. Kelly, K. F. Jensen, T. F. Jamison, C.
Dai, Y. Cui, N. Briggs, R. L. Beingessner, A. Adamo, Chem. Eur.
J. 2018, 24, 2776 – 2784.
Kappe, Angew. Chem. Int. Ed. 2015, 54, 6688 – 6728; Angew.
Chem. 2015, 127, 6788 – 6832; d) S. V. Ley, D. E. Fitzpatrick,
R. M. Myers, C. Battilocchio, R. J. Ingham, Angew. Chem. Int.
Ed. 2015, 54, 10122 – 10136; Angew. Chem. 2015, 127, 10260 –
10275; e) P. D. Morse, R. L. Beingessner, T. F. Jamison, Isr. J.
Chem. 2017, 57, 218 – 227; f) J. Britton, C. L. Raston, Chem. Soc.
Rev. 2017, 46, 1250 – 1271; g) M. B. Plutschack, B. Pieber, K.
Gilmore, P. H. Seeberger, Chem. Rev. 2017, 117, 11796 – 11893.
[14] See the Supporting Information for complete experimental
details.
[15] S. Sankareswaran, M. Mannam, V. Chakka, S. R. Mandapati, P.
Kumar, Org. Process Res. Dev. 2016, 20, 1461 – 1468.
[16] Less than 10% conversion was observed when 9 and 16 were
refluxed in PhCH3 overnight.
[17] For a continuous flow direct amidation mediated by base, see:
J. L. Vrijdag, F. Delgado, N. Alonso, W. M. De Borggraeve, N.
Perez-Macias, J. Alcazar, Chem. Commun. 2014, 50, 15094 –
15097.
[18] 5:1 is the thermodynamic ratio of diastereomers, see ref. [11f] for
calculations and experimental results.
[19] Note that reference 11f is for the synthesis of Cabotegravir.
GSKꢀs Dolutegravir synthesis has been noted in a patent
(reference [11e]), but not described in full detail.
[13] For selected recent reviews on the benefits of continuous flow,
see: a) D. T. McQuade, P. H. Seeberger, J. Org. Chem. 2013, 78,
6384 – 6389; b) S. V. Ley, D. E. Fitzpatrick, R. J. Ingham, R. M.
Myers, Angew. Chem. Int. Ed. 2015, 54, 3449 – 3464; Angew.
Manuscript received: February 20, 2018
Revised manuscript received: March 30, 2018
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Multistep Synthesis
R. E. Ziegler, B. K. Desai, J. Jee,
B. F. Gupton, T. D. Roper,*
T. F. Jamison*
&&&& — &&&&
7-Step Flow Synthesis of the HIV
Integrase Inhibitor Dolutegravir
It’s easy as D-T-G: We report a strategy
for the flow synthesis of the anti-HIV
pharmaceutical drug Dolutegravir in 7
steps with three discrete flow operations.
This resulted in a short synthesis of the
active pharmaceutical ingredient and two
examples of three-step telescoped flow
sequences.
6
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!