Skip to main content
Loading Explore guides…
Dose-Response Relationships · Acedra
Home
Explore
Medicine
Dose-Response Relationships
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
63 segments
Copy
Hide video
0:06
Welcome to nonstop neuron dot com where learning medical concepts is as easy as watching cartoons. We are studying pharmacodynamics. It is basically a study of the effect of drugs on the body. In previous lectures on pharmacodynamics, we talked about the effects of drugs on the body. We also talked about how those effects are produced. Now in this lecture, we will talk about the quantitative aspect of the topic. That is, how much of an effect is produced at
0:40
different doses. In technical words, we call it a dose-response relationship. So let's get started. To be more specific, the response of the drug is not directly related to its dose. A particular dose first produces a certain plasma concentration. And depending on that concentration, a certain level of response is produced. Thus dose-response relationship actually has two parts. First is the dose plasma-concentration relationship. And second
1:17
is plasma-concentration, response relationship. The first part is part of pharmaco-kinetics. The plasma concentration of a drug depends on its absorption, distribution, metabolism, and excretion. And as you already know, all of these are part of pharmacokinetics. We have already studied these things in detail under pharmacokinetics. So this part is over. Now what is left, is plasma-concentration response relationship. And this is the
1:53
domain of pharmacodynamics. In fact, in routine language, the word DOSE-response relationship, refers to plasma-concentration response relationship only. So remember, everything that we will see in this lecture, refers to this part. With that thing clear, now let's get into the real quantitative aspect of the discussion. Here, we basically study dose-response curve in experimental animals or human subjects. In this graph, the x-axis shows the dose.
2:30
And Y axis shows the response. A response can be anything that we want to study. Like a rise in heart rate, or increase in urine output, etc. To obtain the graph, we start with a low dose. We give a small dose to the study subject and measure the response at that dose. Then we increase the dose And measure the response again. At a higher dose, the response would be higher. This way, we keep on increasing the dose and measuring the response. As we do so, eventually response reaches
3:10
a maximum level. After this, there is no further increase in the response, despite the increase in the dose. This is called maximal response, or ceiling response. And the dose where this response is achieved is called the maximal dose, or ceiling dose. When you connect these dots, you get this curve. Its called dose-response curve. As you can see, its shape is a hyperbola. Now comes a question: Why only this shape is produced? Why only hyperbola? Not any other shape?
3:51
So let's explore the answer. The reason why we get hyperbola is, the drug-receptor interaction, follows the law of mass action. In simple words, the receptors respond to the "fractional" change in the dose, rather than "absolute" change. Let's understand this with an example. Let's say, two men are working in a company. The salary of one person is 1 million per year. And that of the other one is 10 million. Now they both get an increment of 1 million per year. Who will get more
4:27
excited? Obviously the first one. Because for him, it's a 100% increase. But for the other person, it's just a 10% increase. So he won't get excited as much. Please note that the absolute change is the same for both. But the response does not depend on that. It's the fractional change, that decides how much the person gets excited. Receptors in our body, respond to an increase in dose, in a similar way. When fractional change is high, they respond more. But if the fractional
5:07
change is less, they don't respond much. If you understood this, come to the graph. Look here, the increase from 1 tablet to 2 tablets is an increase of 100%. But from 2 to 3, it's an increase of only 50%. So the receptors get more excited in the first case, as compared to the second one. And from 3 to 4 tablets, it's just 33% increase. So the excitement is even less. In this way, as we increase the dose, the fractional change gets lower and lower. So the increase
5:49
in response gets lesser and lesser. Because of such behavior, the increase in the response is very rapid in the beginning. But it progressively gets lesser and lesser. So we get a hyperbola. If you understood why we get this shape, I have a secret to share with you. Yes, I kept one thing secret from you in the discussion so far. But now it's time to reveal that secret. The fact is, the
6:19
experiments to obtain this graph, are done keeping this behavior of receptors in mind from the very beginning. So what is actually done, is during an experiment, we increase the dose in multiples, rather than addition. For example, if the first dose is 1 milligram per liter, we increase it to 2, then to 4, then to 8, and likewise. We don't go through all the doses like 1, 2, 3, 4, 5, 6, and so on. This way, we have enough fractional increase with each increment, to produce enough
6:57
change in response. Whereas, if we go with additions, the fractional changes at higher doses are very low. So the changes in response are also very small. And these changes are very difficult to measure accurately. So we choose doses by multiplication, rather than addition. Also this way, we can go through the entire dose-range more efficiently. So finally these are the doses that we give during the experiment. Now of course when we plot responses at these
7:33
doses, we get the hyperbolic curve. We have already seen the reason behind this shape earlier. But this time, I want to draw your attention to a different thing. See if you notice: using a linear scale for values that multiply, creates a problem. In this, most of the doses get crammed up in the initial part of the graph. You see, this tiny part contains four doses. But this huge part
8:03
contains 3 doses only. Because of this it becomes very difficult to do accurate evaluations between the doses in the initial portion. To get enough space here, we can try making a wider graph. But that would unnecessarily widen the right side also. And to make such a graph, we will need a huge graph paper. So stretching the graph is not a good solution. What else we can do? Well, a better solution is not using a linear scale at all. As the doses are multiplying,
8:42
a more suitable scale to plot them is the logarithmic scale. Here we are doubling the dose every time. So we can use log base 2 of doses. This converts 1 into 0, 2 into 1, 4 into 2, 8 into 3, and so on. Thus logarithmic scale converts multiplications into additions. So if we replace a simple arithmetic scale with a logarithmic scale, it solves many problems. First, look at the change in the shape. It has become sigmoid now. On the arithmetic scale,
9:24
it was hyperbola. But in logarithmic scale, it becomes sigmoid, or S-shaped. In this, the middle part is almost straight. And we also have enough space between all the doses. So we can do the evaluations more accurately. Second, on this scale, we can accommodate a wide range of doses in a small graph. For example, an increase in dose from 64 to 256 on the arithmetic scale, can be accommodated with only 2 more digits in the logarithmic scale. Thus we can display
10:05
a wide range of doses in a small graph. As we are displaying log values of the doses here, this curve is now called log dose-response-curve. So this completes our discussion on how the dose-response curve is obtained. Now let's see what information we can derive from the log-dose-response curve. First is the effective dose or ED50. It's the dose, that produces 50% of the maximum response. Here the maximal response is this much. 50%
10:45
of that response is produced at this dose. So this is the ED50. Now… so far in our discussion, we have seen DRC of only one drug. But from now on, we will start comparing the DRC of different drugs. So let's say we plot DRC for a particular response with two different drugs, we will now put together these DRCs on the same graph paper and do the comparison. The first concept that we will study under this, is potency.
11:23
So potency refers to the amount (or dose) of the drug, required to produce a certain response. And as we already know, the dose of the drug is shown on the X-axis. So the potency is indicated by the position of the curve on the X axis. For example, here the first drug requires less dose to produce the same response as compared to the second drug. So we can say the first drug is more potent. And
11:55
the second one is relatively less potent. As such, we can do the comparison at any level of response. But generally, it is done at the half-maximal response. Or in simple words, to know the relative potency, the ED50 of two drugs are compared. For example in producing analgesia, morphine is more potent than pethidine. Or in simple words, to produce the same level of analgesia, morphine is required in lesser quantity than pethidine. Now as you can infer from the ongoing discussion,
12:37
potency becomes more meaningful when we compare it with another drug. And it helps decide the dose of the drug. The more the potency, the less the required dose. OK... we get all these points. But why do different drugs have different potency for the same response? The answer is, that the potency depends on the affinity of receptors for that drug. And how strongly the drug-receptor interaction is coupled to response. The stronger these parameters, the more the potency. So these
13:16
were all the points regarding potency. A closely related parameter to potency is efficacy. So now let's talk about efficacy. Efficacy refers to the maximum response produced by the drug. As the response is shown on the Y axis, the efficacy is indicated by the height of the curve. For example, here first drug goes on to produce this much response. Whereas the second one stops at only this much response. Here we call the first drug more efficacious than the second one.
13:54
In simple words, a less efficacious drug can never produce the level of response that a high-efficacy drug can. For example in producing analgesia, morphine is more efficacious than aspirin. This means the level of analgesia provided by morphine can never be reached with aspirin no matter the dose of aspirin. The reason we see a difference in efficacy is the different mechanisms or sites of action of different drugs. The efficacy helps choose a specific drug among many options. So
14:34
these are the basics regarding efficacy. Now, it seems that a more potent drug would be more efficacious too, doesn't it? But it is not necessarily so. Potency and efficacy are independent of each other. Let me explain it with examples. In all these comparisons, the first drug is more potent than the second one. But look at the efficacy. In the first example, it's more efficacious. In the second example, it's equally efficacious. And in the last one,
15:13
it's less efficacious. You see, we can have all kinds of combinations. Thus, potency and efficacy are independent of each other. Now if they are independent of each other a crucial question comes in. What is more important, Potency, or efficacy? The answer is: efficacy. See the potency is all about the dose of the drug. But the efficacy is about the response. Now in practice, our priority is that the desired therapeutic response is
15:52
achieved. It doesn't matter if it needs a large dose as long as the dose is safe. So efficacy is more important. However, low potency becomes disadvantageous if the size of the therapeutic dose is too inconvenient to administer. That completes our discussion of efficacy too. Ok… We have seen what we can learn from the position on the x-axis. We have also learned about the importance of height on the y-axis. Now let's talk about the slope of the curve. Yes,
16:32
the slope also says something. Here curve with one drug is very steep. And that with another is relatively flat. A steeper curve means a small change in dose produces a large variation in response. So dose of such drugs needs to be calculated more carefully for each patient. On the other hand, if the curve is flat, the same change in dose does not produce much change in response. With such a drug, a standard dose can be given to most of the patients. For example, a fall
17:12
in blood pressure is steeper with hydralazine as compared to that with hydrochlorothiazide.
17:20
So this is what we learn from the slope of the dose-response curve. That completes our list. Now, if you have understood everything so far, I have a big surprise for you. We are studying the relationship between dose-and- response, right? But so far in our discussion, we have ignored a very fundamental point about the response. And that is, the responses can actually be of two types. Graded and quantal. Graded responses are graded on a continuous
17:57
scale. Like a fall in blood pressure, an increase in urine output, etc. Here you have a range of values as responses. For example, you can have less fall or more fall in blood pressure. But quantal responses are all or none. For example death, anesthesia etc. Here you either have the response fully, or not at all. If an experimental animal is dead, it is fully dead. There is nothing like less dead or more dead! Such responses are called all or none (or quantal) responses. A
18:39
graded response also becomes quantal, if we look for a specific degree of response. For example, if we look for a 20% fall in blood pressure, the response becomes quantal. Because now it's either a 20% fall or not. There is no scope for a gray area. What we have seen so far in the lecture, was about the graded responses. In this, we had a continuous scale on the y-axis to note the degree of response. But in all-or-none responses, we don't have it. So these types of responses are
19:20
studied in a different way. But how? Let's see. In this, the study is done in a group of subjects, rather than on a single subject. First, they all are given a small dose of the drug. Then we see how many subjects show the response that we are looking for. Let's say at this dose, no animal is showing the response. Now we increase the dose in all the subjects and again count those showing
19:53
responses. Let's say this time one subject shows the response. Then we again increase the dose and count the subjects. This time more animals show the response. This way, we keep on increasing the dose, and count the subjects showing the response. Eventually, all the subjects show the response. With this data, we can now plot the percentage of subjects showing responses on the y-axis. This again gives a similar sigmoid curve. This is called the quantal dose-response-curve.
20:35
In this curve, we can study a few more parameters. So now let's see them one by one. First is the median effective dose. It is similar to the ED50 we saw earlier. But this time it means a dose where 50% of the study subjects show the response. Remember, in the graded dose-response-curve, we had the percentage of maximal response on the y-axis. And there, ED50 meant a dose where half of that maximal response was achieved. But in the quantal dose-response curve, on the Y axis,
21:16
we have a percentage of "subjects" showing response. So here ED50 means dose where half of the study population shows the response. So this is a small difference. But otherwise, they convey similar information about the drug. And that is, how potent is the drug, in producing the therapeutic response? Now, this is what we get when we observe therapeutic response. But if we observe toxic effects with the same drug we get a curve for the percentage of subjects
21:54
showing toxic effects. From this curve, we can find out the median toxic dose. It is a dose that produces toxic effects in 50% of the study population. A good drug shows toxic effects at higher than therapeutic doses. So the curve for toxic effect, mostly lies on the right side of the curve for therapeutic effect. Similarly, if we count deaths, we get a curve for that as well. And that gives the median "lethal" dose: the dose, at which half of the study animals die. Obviously,
22:36
this should happen at even higher doses. In fact, we study this gap explicitly. And this brings us to our next point: the therapeutic index (or margin of safety). It is equal to the median lethal dose, divided by the median effective dose. Or in a simple way, we divide "this" value by "this" value. This figure shows how safe the drug is. The larger the value, the safer the drug. For a drug to be clinically useful, its therapeutic index must be
23:15
at least more than 1. Or we can say LD50, must be more than ED50. In simple words, death should not happen at lower-than-effective doses. That will make the drug useless. The lethal dose must be higher than the effective dose. That is why the therapeutic index should be at least more than 1. And the more the value, the safer the drug. Now, although the therapeutic index gives very important information regarding a drug, it has a limitation too. See we are dealing with death
23:56
here. So obtaining this parameter is possible only with animal experiments. It's very difficult to find this value for human patients. And also when treating human patients in clinical practice, we don't have to take care of death only. We need to minimize toxic effects also. So in clinical practice, the therapeutic index becomes rather irrelevant. In practice, a different parameter called therapeutic range (or therapeutic window) is more
24:29
important. It is a range of doses between minimum effective dose and maximum tolerable dose. Here, this is the minimum dose required to produce a therapeutic effect. So we have to give doses higher than this. We cannot give a lower dose. Because that will not be helpful to the patient. And at this dose, the toxic effects are so much, that patients cannot tolerate any further toxic effects. So we have to keep the dose lower than this. Thus between these two doses,
25:07
we have a range that we can use. This is called the therapeutic range or therapeutic window. With this, we have completed all the core concepts related to the dose-response relationship. But there are a few topics related to this. So let's quickly see them as well. The first is specificity. It refers to the range of actions produced by a drug. If it produces just one action or a very limited type of action, its called specific for those
25:41
actions. But if it produces a wide range of actions, it's called non-specific. Specificity depends on two things. One, how many types of receptors do the drug act on? And second, how widely these receptors are distributed in the body. For example, omeprazole acts only on the hydrogen potassium ATPase pump. And these pumps are located only on the gastric parietal cells. So omeprazole specifically reduces the secretion of gastric acid. Opposite to this,
26:22
we have chlorpromazine. It acts on many types of receptors. So it produces a wide range of actions. So it's non-specific. Dexamethasone acts only on one type of receptor. However, these receptors are widely distributed in the body. So it also produces a wide range of actions. Or in other words, it does not produce specific action. So that was about specificity. A closely related term to specificity is selectivity. See most of the drugs produce
27:02
more than one type of response in our body. But when a drug produces a particular response preferentially over other responses, it is called selective for that action. In such cases, the DRC of selective action and other actions have a wide gap in between. For example, salbutamol can cause bronchodilation as well as cardiac stimulation. But bronchodilation occurs at a much lower dose than that which causes cardiac stimulation. In simple words, at a lower dose, the salbutamol
27:40
selectively causes bronchodilation. Now compare it with isoprenaline. With it, the curves lie close to each other. So the cardiac stimulation occurs almost parallel to the bronchodilation. Thus it does not cause bronchodilation selectively. So this is selectivity. Now, do you remember the discussion on efficacy? It refers to the maximum response achieved by the drug. A similar term to this is therapeutic efficacy. It refers to the degree of relief
28:22
or cure rate that can be achieved with the drug. If one drug provides more relief or cure rate than another drug, it's said to have higher therapeutic efficacy. For example in parkinsonism, levodopa carbidopa combination, provides more relief to the patient, than trihexyphenidyl. Thus they have higher therapeutic efficacy. See, the efficacy talks only about the changes in physiological parameters. However therapeutic efficacy talks about the therapeutic results as
29:03
a whole. So it is a broader term. It takes into account not only efficacy but relative potency, pharmacokinetics, and pathophysiological variables as well. So this was therapeutic efficacy. The final point in our discussion is the risk benefit ratio. It's an even broader term. It's a comparison of estimated harm (like adverse effects, cost, inconvenience, etc), to expected benefits (like relief in symptoms, cure, improvement in quality of life etc). In simple
29:44
words, it's putting all the knowledge together, to make clinical decisions for the patient. The drug should be prescribed only if the benefits are more than the harms. A problem with this parameter is that it depends on so many factors. On top of that, many of these parameters are subjective. For example, a quality of life means one thing for one person but a different thing for another. So it's not possible to produce such data for readymade reference. In general, a physician
30:20
has to rely on data from the use of drugs in a large population, and his own experience. This completes all the points we wanted to cover in the lecture. Now let's have a quick summary. In dose-response relationship, we study how much of an effect is produced by a drug. Strictly speaking, it's a study of the plasma-concentration, response relationship. As we increase the dose, the response also increases. The highest response that can be achieved with
30:56
a drug is called maximal response or ceiling response. The shape of a simple dose-response curve is a hyperbola. This is because the receptors respond to fractional changes in the dose, rather than absolute change. Because of this, during the experiment, the dose is increased by multiplication rather than addition. But if we plot the multiplying numbers on a linear scale, most of the doses get crammed up in the small initial part of the graph. This makes evaluation
31:35
very difficult in this part. To solve this problem, we use a logarithmic scale on the x-axis. On this scale, the shape we get is sigmoid. In this log dose-response curve, the middle part is almost straight. This makes the evaluation of response at different doses, as well as, the comparison between drugs easy. Also, we can accommodate large doses in a small graph. ED50 is the dose that produces half of the maximal response. Potency refers to the amount of drug
32:18
required to produce a certain response. It is indicated by the position of the curve on the x-axis. It helps choose the dose of the drug. Efficacy refers to the highest response that can be achieved with a drug. It is indicated by the height of the curve on the Y-axis. It helps make the choice of drug. Potency and efficacy are independent of each other. And in clinical practice, efficacy is more important. Because when treating a patient,
32:55
it is more important that the desired therapeutic effect is achieved. It does not matter if we need to increase the dose for that, as long as the dose is safe. A steeper curve means a small change in dose produces a large variation in response. With such a drug, the dose needs to be calculated more carefully. But if the curve is flatter, a standard dose can be given to most of the patients. This was all about the graded responses. But if the responses are all-or-none type,
33:33
we cannot plot them in the same way on the Y axis. So to study them, increasing doses of a drug are given to a group of subjects, and the percentage of subjects showing a specified response is plotted. This produces a similar sigmoid curve. Here the ED50 means a dose that produces a specified response in 50% of the study population. The median toxic dose is the dose that produces toxic effects in 50% of the study population. And median "lethal" dose is the dose that kills 50% of
34:15
the recipients. Therapeutic index is equal to LD50 divided by ED50. It shows how safe is the drug. To be clinically useful, it needs to be at least more than one. However, in clinical practice, a more relevant parameter (than the therapeutic index) is the therapeutic range. It is a range between minimum effective dose to maximum tolerated dose. Specificity refers to how many types of effects a drug can produce in the body. Selectivity refers
34:57
to whether a drug can produce one effect more preferentially over other effects. If DRCs for different effects are wide apart, the drug is called selective. Therapeutic efficacy refers to the degree of relief or cure rate achieved with a drug. Thus it is a broader term than simply efficacy. The risk-benefit ratio weighs all the risks against all the benefits to make decisions in clinical practice. So this completes our
35:32
discussion of the dose-response relationship. But as you know, we have bonus points at the end of each lecture. If you are learning this topic for the first time, I recommend not watching the bonus points. Because that might create unnecessary confusion to you. But if you are doing revision, you can watch them. So the first point is: yes, it's good if a drug has high efficacy. But sometimes, low efficacy can also be advantageous, especially in case
36:08
of overdose. If the high-efficacy drug is taken in high doses by accident, there will be a huge rise in its effects. And this may be dangerous depending on the situation. On the other hand, in case of an overdose of low efficacy drug, there won't be that much increase in the response. This makes such drugs safe in case of overdose. For example, an accidental overdose of diazepam is relatively safer than overdose of phenobarbitone, because diazepam is less efficacious. But again,
36:47
if the drug is used carefully, a high-efficacy drug is usually better. Next point, if two drugs act by the same mechanism, their dose-response curves tend to be parallel to each other. On the contrary to this, if the curves are not parallel, it's indicative of different mechanisms of action. Next. When studying quantal responses, most of the subjects convert from nonresponsive to responsive in the middle range of doses. However,
37:24
a few subjects start to respond at very low doses. These are very sensitive people. In practice too, there may be some patients that may respond to an unusually low dose of the drug. Opposite to this, we have less sensitive subjects too. They respond to only very high doses of the drug. And again in practice too, there may be some patients, that respond only to unusually high doses of the drug. But as I said in the beginning, most of the subjects fall in the mid portion.
38:04
The last bonus point is the certain safety factor. Remember the definition of the therapeutic index? Its ratio of LD50 to ED50. It shows the safety margin of the drug. Although useful, it's still a gross number. Certain-safety-factor narrows down this margin even more aggressively. It's a ratio of dose where 99% of the recipients show the desired therapeutic response, to the dose that kills only 1% of the subjects. It takes into account doses where 99% of the
38:41
recipients show the desired therapeutic response and only 1% of subjects die. In simple words, if we get its value of more than one, and we use the drug in this range, almost all the patients will show the therapeutic response, and almost no one will die. In short, this factor is even more certain regarding safety. Maybe this is the reason why its called a certain safety factor! At least, that's how you can remember it. So finally we have completed the
39:09
entire lecture. If you liked it, please share it with your friends and colleagues too. And don't forget to subscribe because lots more to come. Thanks for watching. See you in the next video.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat