Clinical trials: with the new US rules, cards are played face up

Fedaisf editorial staff

October 3, 2016

All results must be published, even negative ones. And then: detailed information on the protocol, the participants, the side effects of the drugs and the statistical tools used to analyze the data. All to respect the new watchword: transparency

Giovanna Dall'Ongaro 3 Ottobre 2016 – healthdesk

Transparency. This is the watchword of the new US regulations on clinical trials issued by the Department of Health and Human Services (HHS) and the National Institutes of Health (Nih) which will come into force starting next January 18th. Even the negative results will have to end in the light of day: everything will have to be published, even what will never end up on the market. Those who fail to do so will pay the sanctions provided for: first a letter of warning from the FDA arrived, now those who transgress risk up to 10,000 dollars a day or the withdrawal of public funding from the NIH.

Il giro di vite, voluto da entrambe le istituzioni, vuole correggere le storture del passato: da un’indagine condotta nel 2014 su 400 trial clinici scelti casualmente era emerso che nel 30 per cento dei casi dopo quattro anni dalla conclusione dello studio ancora non si trovava traccia dei risultati. Molti trial che fallivano nelle prime fasi non venivano pubblicati affatto. «Eppure – spiega su natures Christopher Gill della Boston University – dalla prospettiva dei consumatori e degli scienziati, i fallimenti sono altrettanto importanti dei successi».

The 2007 law required anyone who started clinical trials involving humans to register on the portal ClinicalTrials.gov, the world's leading clinical trial database, and to describe the method and results of the research program. But there were so many exceptions and loopholes that total transparency became a mirage difficult to achieve.

«Credo che a molte università – ha dichiarato a Nature Robert Califf, a capo della Food and Drug Administration (FDA) – sfugga il fatto che se si conduce un esperimento su una persona che ha dato il suo consenso si ha l’obbligo di rendere noti i risultati. È fondamentalmente una questione di etica».

The new rules provide that registration a ClinicalTrials.gov is carried out within 21 days of recruitment of the first patients. The researchers won't be able to wait for the trial results to publish the data, but they will have to do it along the way. The trial project will then have to be revealed in all the details: how it will be carried out, which statistical tools will be used to analyze the results, any changes to the protocol decided during the course of work. All of this has a very specific purpose: to prevent the phenomenon of "p-hacking", which, in a nutshell, means building customized results. One of the ways to do this is to analyze the data with different tools and report only the favorable ones.

Also new: Researchers are required to report the race and ethnicity of participants. They also need to provide as much information as possible about adverse and side effects that occur more than 5 percent of the time. Some privately funded studies are exempt from the new rules: small trials that only serve to analyze the safety of a drug or the feasibility of a medical tool and studies on behavior. However, the National Institute of Health requires all researchers, even those involved in studies of that type, to report data.

Trials prior to 2008 will continue to follow the old regulations.

Drugs: Scaccabarozzi, positive tsunami of new molecules is coming