Wednesday, April 22, 2020

For multi-center, open label clinical trials, can individual investigation sites disclose results from their sites while study is ongoing?

The golden standard for clinical trials is RCT (randomized, control trials) including three key components: randomization, including a control group (preferably concurrent control), and blinding (or concealment of the treatment assignment). The blinding can minimize the biases (conscious or unconscious) introduced into the study by investigators, patients, and the study team.

However, in some situations, blinding is not possible or is very difficult to implement. A randomized, open-label study may be more appropriate. In the randomized, open-label study, while patients are randomized assigned to receive the experimental new drug or control, the investigator, patients, (in some cases the sponsor study team) are aware of the treatment assignment. 

In an early posting "Some Blinding Techniques in Clinical Trial", some blinding techniques can be applicable even for the randomized, open-label study. On the conservative side, the sponsor can follow the same procedures in handling the randomization, open-label study as the double-blind study - i.e., establishing a firewall to keep the sponsor study team or anyone with the capability to do the aggregate analyses and decision making from accessing the data during the study. 

With a randomized open-label study, both site investigators and patients participating in the study know which treatment group the patient is receiving - knowing the treatment assignment itself may cause biases in efficacy and safety assessments, in concomitant medication selection, in subsequent therapy selection,...

In the last several days, we see an extreme case (presumably due to the special situation in fighting the Covid-19 pandemic). In multi-center, randomized, open-label studies for remdesivir, while the studies are ongoing, the individual sites start to disclose the results / experiences from patients in their perspective sites. 

"The University of Chicago Medicine recruited 125 people with Covid-19 into Gilead’s two Phase 3 clinical trials. Of those people, 113 had severe disease. All the patients have been treated with daily infusions of remdesivir."
"The best news is that most of our patients have already been discharged, which is great. We’ve only had two patients perish,” said Kathleen Mullane, the University of Chicago infectious disease specialist overseeing the remdesivir studies for the hospital.
 Komo News (04/21/2020) Experimental anti-viral drug Remdesivir used to treat COVID-19 patients
Dr. Vinay Malhotra said at his hospital, 40 patients have been treated with the anti-viral drug.
According to Dr. Malhotra, 32 patients were in severe condition. 8 were in moderate condition. Of the 40 patents treated, Malhotra said 17 of them have already been discharged.
“Everyone is talking about the discharge of these patients. Everyone is talking about how good the effect is in three days,” Malhotra said. "We were fortunate to be chosen for this trial. And that has helped our population tremendously."
Both the University of Chicago Hospital in Illinois and the Multicare Institute for Research and Innovation in Tacoma, Washington are investigation sites participating in two pivotal clinical trials sponsored by Gilead:
Both studies (one for severe Covid-19 and one for moderate Covid-19 patients) were designed as randomized, open-label studies with one group receiving Remdesivir (on top of standard of care) and one group receiving standard of care only. According to clinicaltrials.gov, there are 179 and 178 investigation sites for these two studies.

If all sites do the same as the University of Chicago hospital and Multicare Institute for Research and Innovation to disclose their experiences, we could see many reports: some may be good, some may not be very good. The sites with good results may be eager to get their information out; the sites with results not very good may keep silent.

In both of these reports, there was no mention of how many patients in the control group (receiving standard of care) - maybe these patients in the control group are doing pretty well too (unless there are selection biases in determining which patients are allocated to receiving remdesivir or not receiving remdesivir (standard of care)). 

It is not clear if there is an agreement between the sponsor (Gilead) and the investigation sites about disclosing the data while the study is ongoing. Disclosing the data for individual sites while the study is still ongoing should certainly be discouraged or prohibited even in the public health emergency situation. The better way will be that the sponsor designs the study as group sequential design and perform the interim analyses using the data from all sites - if the results are indeed great, the pre-specified boundaries may be crossed and the sponsor may be able to stop the study for efficacy. In this way, the study results may be disclosed earlier while the study integrity is maintained. 

There are already reports that the motive and wrongdoing should be investigated: 

There are also reports that criticize the study design (no adequate control group and no blinding) for Gilead's designed studies: 

1 comment:

Redman said...

Totally agree. It was rather bizarre seeing that happen. This unexpected disclosure from the investigator also had major impact to the stock price. Surely, there would have been a clause in the agreement between the PI and Sponsor that would have prevented that. It is unethical and worrisome.