What are the minimum number of doses required to calculate ED50 of a sample in animal models? Is there any strict rules for minimum number of doses for ED50 calculation? For example, if there are only two doses one is showing less than 50% activity and another exhibiting more than 80% of activity, is it possible to calculate ED50 with these two doses? Sharing ideas and supporting documents are welcome.
Thanks in advance.
Hi , Md. Moniruzzaman
I think you must make a pilot experiment to determine the maximum effect and the dose which given no effect according to the studied substance. Dosing was set by the Dixon sequential up-down method; that is, a greater or less than certain % of reduction in effect decreased or increased the dose, respectively, by a fixed interval for the next subject. The median effective dose (ED50) was derived once 7 changes in dose direction occurred. A dose - response relation ship could be helpful in vitro.
Dear Dr. Nehal,
Thank you so much for your answer. However, I am wondering whether there is any strict rules for minimum number of doses for ED50 calculation. For example, if there are only two doses one is showing no activity and another exhibiting more than 80% of activity, is it possible to calculate ED50 with these two doses?
Have you tried entering this into something like Graphpad Prism to see of it will calculate it for you?
I could be wrong but I doubt that just these two values would be adequate to calculate it. You have one dose which is no activity, but you have no idea at what dose that activity begins. It could be at a dose fairly close to your first one, but then again it could be much closer to your 80% response dose. On that theoretical basis alone your ED50 cannot be calculated from your two values. I think you need a sigmoid curve in order to reflect the dynamics of the dose response. The minimum number of doses? I don't know, but certainly more than two.
Thank you Dr. David for sharing your opinion. I didn't put this data in Prism. Because only two doses even can't explain the dose-dependency. The fact is I found 33, 62, and 80% of activity with three doses and as I found more than 50% activity with the second dose, I calculated ED50. However, one of my friend suggested that 3 doses are not sufficient although I got more than 50% effect. Therefore, I was wondering for number of doses in my study to calculate ED50. What would be your comment?
I don't know if there is a rule about number of doses. I think it's more about the dynamic of the response over the doses you have administered and whether ED50 can be reliably calculated from it. If you imagine that you have just a few points (doses), how are you supposed to accurately plot the curve? The more points the better.
Yap. I also didn't find any strict rule. I agree with you. And I have one more point to focus that having more points is better but three points in above conditions should not be ignored.
Thanks a lot.
The standard method for LD50 is here:- https://ntp.niehs.nih.gov/iccvam/docs/acutetox_docs/udpproc/udpfin01/append/appi.pdf
However I question the need to do LD50 tests in this day and age. There are varous replacements i.e Fixed Dose Test, the Up and Down Procedure and the Acute Toxic Class methods.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology