Download Worries, confusion after cancer trial deaths

NEWS
CANCER IMMUNOTHERAPY
Worries, confusion after cancer trial deaths
edema is essentially an extreme version of
the brain issues doctors were already ren experimental cancer therapy is faccording or is something else. Either way,
ing its biggest setback yet, after an
the cause could lie in one or several specific
unexpected complication killed seven
features of the Juno trial. For one, most of
people over about a year, five of them
the deaths are in adults with acute leukemia;
in a single clinical trial. The company,
people with this disease tend to have more
Seattle, Washington–based Juno Therside effects from CAR-T therapy, in part beapeutics, has its most troubled trial on hold
cause the T cells can expand more quickly
and is racing to figure out why patients suffor reasons that are not well understood.
fered fatal brain swelling. Researchers elseThis increases effectiveness but also risks.
where are grappling with possible
Other variables include the age of
ramifications for the breakthrough
trial participants—children often
treatment, which goes up for drug
have fewer side effects—and the
approval next year.
chemotherapy given before the
“Why would we see this now?
T cells.
We don’t know, period,” says
Finally, there’s the design of the
Stephan Grupp, a pediatric oncoCAR-T cells themselves. All incorlogist at the Children’s Hospital
porate a “costimulatory molecule”
of Philadelphia in Pennsylvania.
that encourages proliferation. The
In several trials, he has treated
five deaths occurred in Juno’s trial
more than 100 children with
that uses a molecule called CD28.
the experimental approach, in
The company’s other trial, with
which a patient’s own immune
two edema deaths, relies on a difcells are genetically engineered
ferent molecule, called 4-1BB. Noto fight cancer. None experivartis, which is also running trials
enced the fluid buildup, known
in acute leukemia and has not reas cerebral edema, that killed
ported cerebral edema, relies on
adults in Juno’s trials. Grupp
4-1BB, too.
and others speculate that the exA third company, Kite Pharma,
planation may lie in the specific
based in Santa Monica, Califorproduct tested and the patient
nia, is using CD28 but is treating
population, rather than in the
adults with advanced lymphoma,
overall strategy itself. Still, they
not leukemia; it hasn’t seen any
are mostly in the dark. Cerebral
A new cancer treatment genetically engineers T cells (blue) to attack malignant
cases of cerebral edema, execuedema “wasn’t on anybody’s racells like this lymphocyte cell (orange).
tives say. Kite and Novartis plan to
dar,” Grupp admits. Company
apply for FDA approval next year.
officials declined to comment, saying only
ago, but it didn’t trigger alarm bells. Then
Better animal models could help solve
that they are investigating.
in July, the company revealed a cluster of
the mystery. The most popular CAR-T therSeveral hundred people with advanced
three more deaths in a different trial, for
apy model, in mice, failed to predict any
blood cancers, from toddlers to senior citiacute leukemia. Juno suggested these were
neurotoxicity. Earlier this week, a team at
zens, have received the treatment, called chidue to a chemotherapy drug called fludathe Seattle Children’s Research Institute in
meric antigen receptor (CAR)-T cell therapy,
rabine that the patients also received. Juno
Washington reported at the ASH meeting
with remarkable results (Science, 28 June
eliminated fludarabine from its protocol,
that they had tested a CAR-T therapy in
2013, p. 1514). Many have now been cancer
and the U.S. Food and Drug Administrarhesus monkeys, and that the animals defree for years, and the therapy is part of a new
tion (FDA) allowed the trial to resume.
veloped abnormal behaviors and tremors,
arsenal of immune-based cancer therapies
But late last month, edema killed two
suggesting neurologic effects.
(Science, 20 December 2013, p. 1432). Doctors
more patients in the same trial, which is
“It’s clearly sobering,” Mackall says of the
are beginning to test CAR-T therapy in solid
now back on hold. Neither the company nor
Juno deaths. “I’ve lived, eaten, and breathed”
tumors like lung cancer. Results there are
FDA has released much information since,
this therapy, and “you think you understand
mixed (Science, 2 September, p. 983).
including the total number of patients
it.” Still, she points out, all cancer drugs come
Like other cancer immunotherapies, CARtreated. Last weekend at the annual Ameriwith huge risks, and that’s something paT therapy can overstimulate the immune
can Society of Hematology (ASH) meeting,
tients, their doctors, and drug regulators gensystem, triggering an out-of-control prolifhowever, Juno reported that a seventh paerally accept given the severity of the disease.
eration of immune cells that can shut down
tient, with chronic leukemia, had also died
“Let’s get rid of the hype here: This is a new
vital organs. The first child treated with
from cerebral edema.
therapy that is very promising,” Mackall says,
CAR-T therapy, a 6-year-old girl with leukeMackall and others wonder whether the
“but we still have a lot to learn about it.” j
By Jennifer Couzin-Frankel
PHOTO: STEVE GSCHMEISSNER/SCIENCE PHOTO LIBRARY
A
mia, nearly died from this. “The community
recognized this as a huge threat to our patients and to the therapy” and found drugs to
manage it, says Crystal Mackall, who heads
the cancer immunology and immunotherapy
program at Stanford University in Palo Alto,
California. The therapy was also known to
carry neurological risks, including confusion,
delirium, seizures, and even a temporary inability to speak. But they have not been fatal.
Juno reported the first death about a year
SCIENCE sciencemag.org
9 DECEMBER 2016 • VOL 354 ISSUE 6317
Published by AAAS
1211
Downloaded from http://science.sciencemag.org/ on January 9, 2017
Experimental immune treatment linked to fatal brain swelling
Worries, confusion after cancer trial deaths
Jennifer Couzin-Frankel (December 8, 2016)
Science 354 (6317), 1211. [doi: 10.1126/science.354.6317.1211]
This copy is for your personal, non-commercial use only.
Article Tools
Permissions
Visit the online version of this article to access the personalization and
article tools:
http://science.sciencemag.org/content/354/6317/1211
Obtain information about reproducing this article:
http://www.sciencemag.org/about/permissions.dtl
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week
in December, by the American Association for the Advancement of Science, 1200 New York
Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for the
Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.
Downloaded from http://science.sciencemag.org/ on January 9, 2017
Editor's Summary