|
Post by dreamboatcruise on Apr 27, 2018 19:53:28 GMT -5
You absolutely can have smaller studies. An example would be a medical student in a teaching hospital who wants to perform a study. You might only have a handful of people because that's what you can scare up without money (a friend of mine researching early detection of Alzheimer's used students who he paid in whiskey). If the study proves interesting then you go looking for the money to convert it to a full trial. I think this is the model STAT-1 is looking at; get good results from the pilot and use it as leverage to get money for STAT-2, a full blown trial. In my opinion, there is no sense in pouring money into a full blown trial for measures -- time in range, etc -- that insurers don't care about. Like it or not, insurers care about A1c and that's all they will care about for the foreseeable future. It would be so much simpler and a lot less expensive to conduct a large scale trial measuring just the change in A1c over 6 months using whatever dosing schedule Mannkind thinks would work best. Any trial should measure both. Adding a CGM to a trial protocol is comparatively small expense in running a trial. Whether the CGM should be blinded or not is separate issue. Having it blinded would be necessary to show insurance companies what could be achieved by patients that don't have access to CGM, but still useful to have the time in range data. I actually imagine FDA will be wanting to have ALL trials involve CGMs. With the new definitions of levels of hypoglycemia, a CGM would be required to properly record how many occurrences of the different levels happen.
|
|
|
Post by dreamboatcruise on Apr 27, 2018 20:05:24 GMT -5
Aged, I love you. I have read, your typing that says, there were not people in the STAT trial for ........ The awful human being in me, thought, there have been studies with less people, blah blah blah. Case in point. CLINICAL PHARMACOLOGY 12.1 Mechanism of Action Ezetimibe reduces blood cholesterol by inhibiting the absorption of cholesterol by the small intestine. In a 2-week clinical study in 18 hypercholesterolemic patients, ZETIA inhibited intestinal cholesterol absorption by 54%, compared with placebo. ZETIA had no clinically meaningful effect on the plasma concentrations of the fat-soluble vitamins A, D, and E (in a study of 113 patients), and did not impair adrenocortical steroid hormone production (in a study of 118 patients). www.merck.com/product/usa/pi_circulars/z/zetia/zetia_pi.pdf You absolutely can have smaller studies. An example would be a medical student in a teaching hospital who wants to perform a study. You might only have a handful of people because that's what you can scare up without money (a friend of mine researching early detection of Alzheimer's used students who he paid in whiskey). If the study proves interesting then you go looking for the money to convert it to a full trial. I think this is the model STAT-1 is looking at; get good results from the pilot and use it as leverage to get money for STAT-2, a full blown trial. You only need a large enough population to provide statistically significant outcomes, and that number varies by what you are trying to ascertain. Even the relatively large trials MNKD did in the past didn't have enough patients to have the data on "severe" hypoglycemia be statistically significant (95% probability of being causal), though logic (and human physiology) would almost dictate that if overall hypoglycemia was reduced (which was proven to the statistically significant standard) that severe hypos would also be reduced. Though I guess in medicine another issue is a population size large enough to represent the genetic diversity of the overall population. I think your researcher friend might have been invalidating his own study if he was testing for cognitive ability on a group selected for willingness to accept booze as pay
|
|
|
Post by digger on Apr 28, 2018 7:42:04 GMT -5
In my opinion, there is no sense in pouring money into a full blown trial for measures -- time in range, etc -- that insurers don't care about. Like it or not, insurers care about A1c and that's all they will care about for the foreseeable future. It would be so much simpler and a lot less expensive to conduct a large scale trial measuring just the change in A1c over 6 months using whatever dosing schedule Mannkind thinks would work best. Any trial should measure both. Adding a CGM to a trial protocol is comparatively small expense in running a trial. Whether the CGM should be blinded or not is separate issue. Having it blinded would be necessary to show insurance companies what could be achieved by patients that don't have access to CGM, but still useful to have the time in range data. I actually imagine FDA will be wanting to have ALL trials involve CGMs. With the new definitions of levels of hypoglycemia, a CGM would be required to properly record how many occurrences of the different levels happen. To me it is all about focusing all your energy and resources on the one thing that you need the most -- in this case, insurance reimbursement. While it might be worthy to add CGMs, they present only an added expense and "one more thing" that needs attention, but one which won't help your case with insurers.
|
|
|
Post by brotherm1 on Apr 28, 2018 8:06:53 GMT -5
No comment 🙄
|
|