DMC post from Hayflick on dreaded YMB

[scottmayhew]

posted by hayflicklim 6/29/2016

dmc

indicated in trial NCT0259866 update on 6/22/162. DETERMINING NEED FOR A DMC
All clinical trials require safety monitoring, but not all trials require monitoring by a formal committee that may be external to the trial organizers, sponsors, and investigators. As noted earlier, DMCs have generally been established for large, randomized multisite studies that evaluate treatments intended to prolong life or reduce risk of a major adverse health outcome such as a cardiovascular event or recurrence of cancer. DMCs are generally recommended for any controlled trial of any size that will compare rates of mortality or major morbidity, but a DMC is not required or recommended for most clinical studies. DMCs are generally not needed, for example, for trials at early stages of product development. They are also generally not needed for trials addressing lesser outcomes, such as relief of symptoms, unless the trial population is at elevated risk of more severe outcomes (see Sections 4.4.1.5 and 4.4.2 for further discussion).
Although the value of a DMC is well accepted in settings such as those described above, it is important to recognize that DMCs add administrative complexity to a trial and require additional resources, so we recommend that sponsors limit the use of a DMC to the circumstances described in Section 2.1. There are several factors to consider when determining whether to establish a DMC for a particular trial. These factors, discussed below, relate primarily to safety, practicality, and scientific validity


(followed by)

posted by hayflicklim 6/29/2016

Does this mean they are getting ready for phase III OR HAS IT ALREADY STARTED

[scottmayhew]

so...please excuse my ignorance, but what is DMC and what is this trial? I couldnt find it by googling. I presume this is relevant -- was on GERN message board AND posted by Hayflick...

[Fishermangents]

IMerge is in Phase 3, because that is what is stated on the CT site (top right). Having an DMC installed means that things are getting serious. Here some clarification on DMC's:

(Clinical Trials Transformation Initiative)
Randomized clinical trials are often overseen by independent data monitoring committees (DMC or IDMCs). The concept was initially introduced in 1967 by the Greenberg Report (CCT, 1988) to provide oversight to randomized clinical trials sponsored by the National Heart Institute. Among its recommendations is that trials should not be modified or terminated without the input from an independent group of scientists otherwise not involved in the conduct of the trial. Trials might be terminated because the intervention’s benefit was clearly established, there was sufficient evidence of harm, the trial was no longer viable or of interest, or some other compelling reason. The application of these recommendations in early NIH sponsored cardiovascular trials led to the establishment of the independent data and safety monitoring board (DSMB) or more simply the Data Monitoring Committee (DMC). The charge to the DMC was to monitor trial conduct and safety and recommend trial changes or closure. Today DMCs are used across therapeutic areas and may oversee single trials, groups of related trials, or the entire portfolio of research related to an investigational product. Membership of the independent DMCs may be internal to the trial sponsor, totally external or a mix, depending on the nature of the trial. Responsibilities may be limited to analyses of efficacy and safety, or expanded to include review of data quality and other trial operations.

==============

DMC's can have different tasks and different compositions. It can be to monitor safety, to oversee changes to the protocol, monitor scientific quality etc. They can also being installed if the new drug is expected to have a major impact on clinical practice. The FDA recommends that DMC's have a trial specific character. NCBI:
"Establishing an appropriately qualified DMC can help ensure the safety of trial participants and uphold equipoise of the trial by continuing review of the safety data. It can also help assure the scientific validity and integrity of the trial."

A DMC is only useful for large trials. It makes no sense to put DMC at work if a trial doesn't has statistical significance. So the DMC on IMerge could be evidence that Phase 3 has started.

[biopearl]

Thanks to Hay and Fish. I believe that in this case having an independent DMC (presumably high level scientists, physicians and statisticians) will add validity to any decision to stop the study for efficacy. For example if a statistically significant drug effect is seen that results in a hard end point (like longevity?) it would be unethical to withhold drug from a control arm or from patients in dire need. To stop a study for this reason would be a big deal and having such an independent DMC recommendation would ensure validity and place that decision above politics and money. Lets hope that that's what they are thinking. Is the DMC just for the MDS study or was one also put in place for the MF study? (Hay shows true bravery in posting on YMB) Regards, bp

[Fishermangents]

Hi Bio, I don't see a DMC for IMbark. It seems IMerge is moving faster.

[biopearl]

Fish, isn't t that wild? I once speculated in an old message maybe on YMB that MDS trial would eclipse the MF trial re timing since it was phase III. (Perhaps I did that in a wine induce haze...) I would reaffirm that idea since it is a Phase III trial and has a larger potential patient population, a DMC, and a low bar for primary end point but the project trial end dates on CT site unfortunately does not support this WAG. But the DMB could feel otherwise as the study progresses...bp

[Fishermangents]

It certainly could be. If there is an unmet need, and if the drug is convincing in its efficacy related to it's risk profile, then earlier termination is a realistic option. In these cases a DMC comes in for assuring that the safety is adequately addressed and that all statistical data are scientifically sound and unbiased.