Correlation of In-vitro and In-vivo Data

Introduction

Correlation of in-vitro and in-vivo is usually between the dissolution time/rate of the dosage form and the relevant in-vivo data.
The term “in-vitro/in-vivo correlation” in this context is related to the correlation between in-vitro dissolution and the release or uptake of the drug in-vivo. The establishment of such a correlation is one of the most important aspects of a dissolution test for a preparation under formation development.
This is because dissolution time is the rate-determining (or limiting) step in the absorption of drugs of low aqueous solubility.
A few studies have showed correlations between in-vitro disintegration time (of solid dosage forms) and in-vivo availability while,
very many others have shown no such correlation. It is highly probable that in those cases where disintegrations time correlated with in-vivo availability data, dissolution time determinations on the same dosage forms would also have demonstrated a correlation with in-vivo availability.
Variation in clinical response may occur between two or more orally administered dosage forms containing chemically equivalent amounts of a drug (usually of limited aqueous solubility). These may be as a result of differences in the dissolution rate of the drug from the dosage form which, in turn, affected the gastro-intestinal absorption of the drug.
Thus, tablets exhibiting inadequate clinical activity possess a low dissolution rate relative to those tablets giving the expected response or therapeutic activity.
Tablets having an enhanced availability or pharmacologic response invariably have a rapid rate of dissolution.

CLASSIFICATION OF IN-VITRO/IN-VIVO CORRELATIONS

In-vitro/in-vivo correlations may be classified into:

1. Pharmacological correlations, based on clinical observations
2. Semi-quantitative correlations, based on blood levels or urinary excretion data, and
3. Quantitative correlations arising from absorption kinetics.

Although most of the published correlations are of a semi-quantitative nature, the most valuable would be those based on absorption kinetics.
It is important to recognize that differences in the in-vitro dissolution rates of various formulations may not necessary be paralleled by differences in the in-vivo availability or response. Such a situation is most likely to occur with a drug that is well absorbed and has a long biological
half-life.
For instance, 5 different batches of Tolbutamide tablets (a sulphonylurea oral hypoglycaemic) which showed a 7-fold range of dissolution half￾lives but had no significant clinical differences (Lu et al, 1965; McKendry et al 1965).
Finally, it should be appreciated that it may not always be possible to obtain correlation between in-vivo data and data obtained from a model in-vitro system. This inability may result from one or more of several
reasons.

Correlation of In-vitro Dissolution Data with In-vivo Absorption Data and the Limitation of this

In-vitro dissolution tests are carried out on pharmaceutical products (mainly solid dosage forms) because it is rather too expensive to carry out in-vivo tests on all batches of pharmaceutical product.

Views on the roles of in-vitro Tests

View have been expressed by several researchers on the various roles of in-vitro dissolution tests. These include:

1. The in-vitro dissolution test should be regarded in the same manner as a disintegration test which merely places a minimum limit on the rate of dissolution of a product.
   This will be useful for quality control purposes and pre-formulation
   studies and also enable bad products to be eliminated.
2. It represents a more or less empirical approach to dissolution limited drug availability and thus must be correlated with a quantitative measurement of some in-vivo parameter.
3. In-vitro dissolution test should simulate the in-vivo dissolution of a drug in the GIT. It is postulated that such a test should enable one to obtain more information about those physico-chemical and physiocological conditions in the GIT which affect the dissolution of drugs.
4. Levy stated that the development of in-vitro dissolution tests capable of reflecting the absorption rate, in man, of drugs contained in various dosage forms is the most important task in biopharmaceutics.

He expressed the view that in-vitro tests should be able to correlate with in-vivo absorption of several different drugs in the
same dosage form or it should be able to correlate with in-vivo absorption for one drug in various dosage forms.
So far, the above views on the correlation of in-vitro dissolution tests with in-vivo absorption are beyond attainable objectives.

Examples of Correlation of In-vitro and in-vivo data

1. GRISEOFULVIN

Dose in man = 125mg-500mg
H20 – solubility = 15ug/ml at 37ºC or mcg/ml
Kitchen & Symchowicz (1967) carried out in-vivo tests and in-vitro tests on 5 different preparations of griseofulvin tablets and capsules.
(a.) The In-vivo tests: were carried out on 10 human volunteers. The 5 preparations were administered at one week intervals to the volunteers.

• Each subject had 500mg tablet before breakfast (i.e. 8.00am)
• They ate a standard meal two hours later (i.e. 10.00AM).
• Blood samples were taken from the subjects at 0, 2, 4, 8 and 24hours.
• The index of drug absorption was taken as the mean blood level over the 24 hours, which is = (Area under 24hrs
  curve/24).

b.) The In-vitro Tests:
The Oscillatory tube apparatus was used with 20 Litres of distilled water or simulated intestinal fluid (pH 7.2) as dissolution medium and maintained at 37ºC.
Results
In water, the dissolution rate of griseofulvin was very low and 3 preparations could not be distinguished from one another.
But in the simulated intestinal fluid, it was possible to distinguish between the 5 preparations. The results obtained from both the in-vivo and in-vitro tests were then plotted on a graph as follows.

Note that dissolution were obtained on only 4 preparations. The 5th product clogged the screen of the dissolution apparatus.
From the figure above, it is seen that some form of correlation existed between the in-vitro and in-vivo tests results and this correlation depended on the type of dissolution medium used in the in-vitro test.
Note: The dissolution rate plotted in the figure above is not a rate constant but it is the rate of dissolution for 30mins only.

2. AMINOREX:

This is a sustained-release preparation. It is an Anorectic agent (i.e. causes a loss of appetite). Other examples are:
Mirapront Ponderax. Cressman et al (1969) carried out in-vivo and in-vitro tests on various formulations of this product.
(a.) The in-vivo tests

• They administered each formation to between 4 and 6 subjects before breakfast after an overnight fasting.
• Food was not consumed until 2 hours after administration and blood samples were taken from the subjects for 24 hours.

(b.) The In-vitro tests 
The Levy & Hayes Beaker method was used with 750mls of dissolution medium at 370C and stirred at 50rpm. Two different dissolution solutions were used for the test. Simulated gastric juice was used for the first one hour of the test and simulated intestinal
fluid for the remaining period of the experiment.
Results: These were expressed in terms of time for 25% and 50% dissolution of the tablets, plotted against the time for 25% and 50% absorption of the drug.
It was found that in some instances, the in-vitro dissolution test did not distinguish between products which had different rates of absorption.
The results obtained were plotted as follows.

3. ASPIRIN

Levy et al (1964) employed 3 different dosage forms of Aspirin preparations for their investigation. They used:

• Aspirin solution (200ml solution)
• Normal/Plain Aspirin tablets
• Microencapsulated Aspirin tablets

A dose of each preparation contained 650 mg of Aspirin.

(a.) The In-vivo tests:

• A test panel of 12 human healthy subjects were used.
• Administration was done in random and usually one week apart.
• Each dose was given with 200ml of water in the morning on an empty stomach and food was permitted after 2hours.
• Blood samples were taken from the subjects at 10, 20, 30, 45, 60 and 120 mins after administration.

(b.) The In-vitro tests:

Levy et al used the Levy & Hayes Beaker
method. 0.1N HCL was the dissolution medium used, maintained at 37ºC and a range of stirring speeds was used.

From the in-vivo results, Levy et al found that:

• i. The dissolution process is the rate determining step of aspirin absorption since there is a lag time before the absorption of aspirin begins.
• ii. The ratio of the rate of absorption of the plain tablets to the microencapsulated tablets was 3:1.

In-vitro Test Results
Levy et al then determined the rate of stirring in the in-vitro test that gave the ratio 3:1 for plain/microencapsulated tablets as in the in-vivo test results.
A stirring speed of 50 RPM was found to give the ratio of rate of dissolution of the plain tablets to the microencapsulated tablets of 3:1.
In order to demonstrate correlation, the % Dose of Aspirin absorbed at time t, following drug administration was plotted versus the % Dissolved in-vitro at time t-lag time/intensity factor.
The lag time was the observed delay in the in-vivo absorption of the drug.
The intensity factor = ratio of the rate constant for dissolution /
Ratio of the rate of absorption.
Under the conditions employed, the intensity factor had a value of 2.
They then obtained an excellent correlation as follows:

Note:
(a.) Attempt to achieve total correlation between all 3 dosage forms of Aspirin preparation using dissolution data obtained at speeds other than 50rpm were unsuccessful.
(b.) The above observation led the authors of the work to comment as follows:

• i.) The apparent failure of some dissolution test to reflect the in-vivo results may be due to improper agitation conditions within the model system.
• ii.) That by determining the in-vitro agitation conditions under which correlation is achieved, it may also be possible to extrapolate this to the agitation conditions that exist in-vivo.

Problems Involved in Obtaining In-vitro and In-vivo Correlation

1. The major problems are as follows: one is trying to correlate an in-vitro test whose conditions are vigorously controlled with a biological system.

• (a.) For example, there is subject to subject variation with regard to composition of Hormones, Electrolytes and Surface active agents in the GIT.
• (b.) With Aspirin, it has been shown that the gastric mucin can markedly
  affect the rate of dissolution of Aspirin products. The gastric mucin content of gastric juice varies from one individual to another. Its effects on the in-vivo dissolution/absorption of drugs are unpredictable.

1. Correlations are obtained with healthy subjects with carefully controlled dosage schedule. But in reality, the biological response to a dosage form may not be the same in healthy subjects as in unhealthy/sick subjects.

2. Stomach Emptying Rate 

The rate at which the dosage form passes through the GIT of an individual depends on the stomach emptying rate. This is largely
controlled by the type of food consumed and it has been shown that:
Starchy food (Carbohydrate) > Meat (Protein) > Vegetables > Fatty Food (Fatty Acids

• (a.) Fatty Acids inhibit gastric secretion and reduce the rate of
  stomach emptying.
• (b.) Although both Starch & Protein also have inhibitory effects on gastric secretion, their effects are less pronounced than those produced by Fatty Acids.
• (c.) Also, as the viscosity of the ingested meal or the gastric fluid is
  increased, there is a corresponding decrease in the rate of stomach emptying.
• (d.) pH of the environment. The stomach emptying rate has a great
  effect on the rate of dissolution of the drug especially for those drugs whose solubility is affected by the pH of the environment.

The pH of the stomach varies drastically. For example, a resting stomach has a neutral pH while someone who is very
apprehensive/scared has a very acidic stomach.
Thus, subjects who are used for the in-vivo tests are naturally or usually apprehensive since they are aware that samples of blood are going to be taken from them at periodic time intervals. This situation is quite different with unhealthy/sick persons.

1. In some instances there is adsorption of drugs on the GIT of subjects.

2. Dissolution Medium used for in-vitro tests

It has already been shown with griseofulvin tablets that the dissolution medium used for the test (water and simulated intestinal
fluid) greatly affects the rate of dissolution.
6. Degree of agitation
The degree of agitation influences the in-vitro dissolution rate. The degree of agitation in the G.I Tract is very variable and the nature of the mechanism of dissolution in the GIT is not at all clear. It is highly complex and some researchers have postulated a hybrid between the Diffusion Layer theory and the Danckwert`s theory.

1. All the in-vivo and in-vitro correlations have been made in Hindsight (i.e. the ability to see after the event is over, what should have been done). That is, the data obtained have been juggled from one dosage form and test to fit the in-vivo result.

Thus, it is unrealistic to expect to develop a generalized in-vitro test that simulates in-vivo conditions.
Therefore, the role of in-vitro tests are as follows:

• i. Carry out an in-vivo studies and also in-vitro dissolution test on a product initially and attempt to correlate the results.
• ii. Any modification of the formulation may be monitored with the in-vitro test.
• iii. When an acceptable dosage form has been obtained, then the final in-vivo studies is carried out.