NEWS | IN DEPTH laureate , then-president of The Rockefeller University in New York City, be- came CEO of the joint venture. To help build the structure, the government contributed £220 million; the also gave a significant amount. Construction went relatively smoothly for such a complicated project, observers say. Objections from worried neighbors did not prevent completion of a biosafety level 3 lab for studying diseases such as tuberculosis and influenza. Researchers have their own construction worries, however: The planned route of a new underground railway line was shifted closer to the Crick in 2013. Its vibrations and magnetic interference could affect sensitive devices such as those in MRC’s facility for biomolecular liquid-state nuclear magnetic resonance. The two founding institutes will shift about 900 scientists—including 180 Ph.D. students—and 250 support staff to the new DRUG DEVELOPMENT institute by early 2017. As tenured research- ers retire, they will be replaced with early career investigators who have a 6-year con- A painstaking overhaul for

tract that can be renewed only once. “There on September 1, 2016 will be continual renewal through new peo- ple coming in,” says Richard Treisman, a sci- cardiac safety testing entific director at the Crick. “It’s an effective way of keeping a research system lively and New methods may better predict arrhythmia risk in vitro vibrant,” says Iain Mattaj, director general of the European Labora- By Kelly Servick rest in more than 100 patients before the U.S. tory in Heidelberg, Germany. Food and Drug Administration (FDA) with- Another 200 scientists from University f any safety test should err on the side drew its approval in 1997. College London, , of caution, it’s one for whether a new It turned out that terfenadine blocked a and King’s College London—which each medicine might accidentally stop your key ion channel on the membrane of car- gave £40 million toward construction—will heart. But for years, researchers have diac muscle cells. Electrical signals from the bring in additional expertise during vis- worried that preclinical tests for cardiac flow of ions across these membranes rhyth- http://science.sciencemag.org/ its lasting from a few months to 6 years. risk are so simplistic and conservative mically contract the heart, and this particu- Pharmaceutical giant GlaxoSmithKline will Ithat they might lead drugmakers to aban- lar channel—encoded by a gene known as dispatch 20 researchers to work on joint don promising and safe treatments early hERG—helps restore equilibrium after a con- projects with Crick scientists. The proxim- in development. traction by letting potassium ions out. If it is ity to many London hospitals will make Now, an international team of regulators, blocked, heart cells are slower to “recharge” translational and clinical research easier. academic researchers, and drug companies after a beat, which can lead to an irregular

Some worried that the high cost of living is nearing completion of a project to validate rhythm and sometimes death. Downloaded from in London might discourage candidates. a new—and hopefully more accurate—set of ICH standards require that drug develop- Mattaj doubts that. “I think the Crick will cardiac safety tests, known as the Compre- ers test candidates for hERG-blocking activ- be a superattractive place.” The appeal for hensive in vitro Proarrhythmia Assay (CiPA). ity in cultured cells. Not all compounds with Ph.D. students is already clear: Thirteen In the coming weeks, a key component of hERG activity turn out to be dangerous; hundred applied for 44 positions this year. that assay, based on stem cell–derived heart more than a dozen ion channels interact to The United Kingdom’s future relation- cells, will undergo blind testing on drugs affect the rhythm of heart contractions. But ship with the European Union casts a big- with known risks at various academic and researchers spooked by an ominous hERG ger shadow (Science, 29 July, p. 437). “It is industry labs. assay sometimes “end up throwing drugs a blight that is hanging over the whole of Regulators in the United States, Europe, away that most likely would be good drugs,” British science, but especially the Crick, and Japan rely on standards for arrhythmia says Gary Gintant, a research fellow at drug- because it’s big and new,” says University risk testing set in 2005 by a body called the maker AbbVie Inc. in Chicago, Illinois, who of London neuroscientist and former MRC International Council for Harmonisation of leads one of the teams within CiPA. In phase chief executive Colin Blakemore. Restric- Technical Requirements for Pharmaceuticals II trials, most drugs get another safety test: tions on migration could make it more dif- for Human Use (ICH). Those standards were Patients’ electrocardiograms (ECGs) can ficult to recruit, Treisman says, and limited born from a series of grim surprises. In the reveal potentially dangerous lengthening access to EU research funds could squeeze 1990s, several drugs were yanked from the of the heart’s electrical cycle. That process the £130 million operating budget, about market for dramatic cardiac side effects. One costs millions of dollars, can throw up a red 5% of which comes from EU grants. “We prominent example, an antihistamine called flag after a company is heavily invested in j just don’t know what’s going to happen.” terfenadine, was reported to cause cardiac ar- a drug, and sometimes generates both false INSTITUTE CARDIOVASCULAR LAB/STANFORD WU JOSEPH PHOTO:

976 2 SEPTEMBER 2016 • VOL 353 ISSUE 6303 sciencemag.org SCIENCE

Published by AAAS Clusters of stem cell–derived cardiac muscle cells RESEARCH MISCONDUCT model how the heart may respond to a new drug. positives and false negatives, Gintant notes. Duke fraud case highlights Better ways to predict how heart muscle will behave would allow drug developers to advance more candidates into trials and financial risks for universities maybe avoid large-scale, late-phase ECG studies. Some drug companies and academic Whistleblower alleges doctored data were used to secure labs are already exploring whether channels $200 million in grants from NIH and other federal agencies other than hERG are good predictors of ar- rhythmia risk. They have also begun develop- ing more realistic models of the human heart By Alison McCook, Retraction Watch by a former colleague of Potts-Kant. It ac- in a dish, using induced pluripotent stem cuses the researcher, her former supervisor, (IPS) cells—reprogrammed adult cells capa- n a Friday in March 2013, a re- and the university of including fraudulent ble of differentiating into many types of cells. searcher working in the lab of a data in applications and reports involving The challenge now is to turn that new science prominent pulmonary scientist at more than 60 grants worth some $200 mil- into a well-validated, standard set of tests. Duke University in Durham, North lion. If successful, the suit—brought under The CiPA initiative, a partnership between Carolina, was arrested on charges the federal False Claims Act (FCA)—could FDA and several agencies and consortia, in- of embezzlement. The researcher, force Duke to return to the government up cluding Health Canada, the European Medi- Obiologist Erin Potts-Kant, later pled guilty to three times the amount of any ill-gotten cines Agency, and Japan’s National Institute of to siphoning more than $25,000 from the funds, and produce a multimillion-dollar Health Sciences, is an attempt to do just that. Duke University Health System, buying payout to the whistleblower. One group within CiPA is investigating seven merchandise from Amazon, Walmart, and The Duke case “should scare all [aca- channels (including hERG) known to regu- Target—even faking receipts to legitimize demic] institutions around the country,”

late heart rhythm to find which combination her purchases. A state judge ultimately lev- says attorney Joel Androphy of Berg & on September 1, 2016 of channel-blocking tests might best predict ied a fine, and sentenced her to probation Androphy in Houston, Texas, who special- safety. A second team is refining a computer and community service. izes in false claims litigation. It appears model of the human ventricle’s electrical be- Then Potts-Kant’s troubles got worse. to be one of the largest FCA suits ever to havior; it will turn ion channel data into es- Duke officials took a closer look at her work focus on research misconduct in academia, timates of arrhythmia risk. A third is testing and didn’t like what they saw. Fifteen of her he says, and, if successful, could “open the how well clusters of IPS-derived heart muscle papers, mostly dealing with pulmonary bio- floodgates” to other whistleblowing cases. cells mimic the behavior of the adult heart logy, have now been retracted, with many False claims lawsuits, also known as qui when exposed to various drugs. Some predict notices citing “unreliable” data. Several tam suits, are a growing part of the U.S. le- that this stem cell approach, after much re- others have been modified with either par- gal landscape. Under an 1863 law, citizen fining, could even eliminate the need to test tial retractions, expressions of concern, or whistleblowers can go to court on behalf individual ion channels. corrections. And last month, a U.S. district of the government to try to recoup federal In a first round of validation last year, aca- court unsealed a whistleblower lawsuit filed funds that were fraudulently obtained. http://science.sciencemag.org/ demic and industry labs blindly exposed IPS- derived cells to eight different compounds. Data across study sites “looked surprisingly Holding universities liable for research fraud similar,” says Joseph Wu, a cardiologist and Whistleblowers have a mixed record of success in False Claims Act (FCA) lawsuits against research universities stem cell biologist at Stanford University that involve allegations of scientific misconduct. Highlights from selected cases: in Palo Alto, California, who led one of the

testing efforts. Downloaded from A second round of FDA-funded tests for a YEAR WHISTLEBLOWER DEFENDANT ALLEGATIONS OUTCOME set of 28 compounds began this month. By 2009 Taryn Resnick, Weill Medical College of In grants totaling $14 mil- College settled for $2.6 mil- the end of next year, CiPA collaborators in- former employee Cornell University lion, researcher Lorraine J. lion, plus attorneys’ fees tend to propose the complete assay—a three- Gudas falsified data, failed and expenses. to disclose other funding, part process combining ion-channel assays, and misapplied funding. computer simulation, and IPS-derived cells— to a group within ICH that could choose to 2012 Daniel Feldman, Weill Medical College Misuse of research train- Defendants paid $887,714, fellowship program of Cornell University ing grant; deviated from plus $602,898.63 in attorneys’ revise the 2005 standard. participant and psychiatrist Wilfred submitted plan. fees and expenses. Even with this new set of tools, companies van Gorp may still decide that a compound that acts on multiple cardiac ion channels is too risky 2012 Kenneth Jones, Brigham and Women’s Including falsified data in Failed; whistleblower ulti- researcher Hospital, Massachu- application for Alzheimer’s mately lost at trial. to pursue, says Icilio Cavero, a retired cardio- setts General Hospital, disease research grant. vascular pharmacologist and safety con- and researchers sultant to drug companies who is based in Marilyn Albert and Paris. And it will take more than these initial Ronald Killiany 28 compounds to prove that IPS-derived 2014 Terri King, former University of Texas Falsifying research data. Failed. U.S. Supreme Court cells can be reliable safety predictors. “The associate professor Health Science Center upheld lower court ruling idea [of CiPA] is beautiful,” Cavero says, but that the public university was j DATA: RETRACTION WATCH RETRACTION DATA: “new things scare everybody.” exempt from FCA liability.

SCIENCE sciencemag.org 2 SEPTEMBER 2016 • VOL 353 ISSUE 6303 977

Published by AAAS A painstaking overhaul for cardiac safety testing Kelly Servick (September 1, 2016) Science 353 (6303), 976-977. [doi: 10.1126/science.353.6303.976]

Editor's Summary

This copy is for your personal, non-commercial use only.

Article Tools Visit the online version of this article to access the personalization and article tools: http://science.sciencemag.org/content/353/6303/976 on September 1, 2016 Permissions Obtain information about reproducing this article: http://www.sciencemag.org/about/permissions.dtl http://science.sciencemag.org/ Downloaded from

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.