William Brown and Margareth Marques
The following questions
have been submitted by readers of Dissolution Technologies. Margareth
Marques, Ph. D. and Will Brown, United States Phamacopeia, authored
responses to each of the questions.
*Note: These are opinions and interpretations of the authors,
and are not necessarily the official viewpoints of the USP
Email for correspondence:
web@usp.org
Q
What is the impact on dissolution results when more than 5% organic solvent is used to prepare the standard solution?
A
It is not possible to provide a generalized response given the number of possible combinations involved. You will have to evaluate the impact using the solvent and substance in your hands. Based on the quantitative procedure that has been selected for the particular product, the various possible interferences need to be evaluated.The effect of the added solvent on the responses of the medium (blank) and the analyte are obvious avenues of investigation.
Q
We are analyzing an extended-release product that has sampling at three time points, 1 h, 4 h, and 8 h. The product did not meet the L1 acceptance criteria at 4 h as stated in the USP General Chapter <711> Dissolution, but it met the criteria at 1 h and 8 h. We then carried out the test with an additional six units and evaluated the results at L2. The results met the L2 criteria at 4 h. For the other two time points, 1 h and 8 h, should we have considered the release values obtained at the L1 stage only, or should we have evaluated the combined values at L1 and L2?
A
The instructions in the USP General Chapter <711> Dissolution direct to continue testing through the three levels unless the results conform at either L1 or L2. In the example provided, it is indicated that the results did not conform at the L1 level. In that case, the testing continues to the L2 level with an additional six units tested. The testing follows the procedure given with a complete sample taken. A complete sample includes all time points.
Q
How can the information obtained with the evaluation of intrinsic dissolution for a particular drug substance be used?
A
Intrinsic dissolution is a powerful tool during the development of a new product because it allows better understanding of the physicochemical properties of the drug substance. Comparison of the intrinsic dissolution for the API from different sources or prepared by different processes can help guide the decision whether to interchange them in the product. It is used as a develop- ment tool and not for QC routine analysis. There are several papers in the literature where more details on the applicability of this technique can be found.
Q
What linearity range should be evaluated for the validation of a dissolution method with the following characteristics: extended-release product with a 100-mg label claim; 900 mL of dissolution medium; and acceptance criteria of NMT 20% at 2 h, 30—50%at4h,55—65%at8h,andNLT70%at12h?
A
Please refer to the USP General Chapter <1092> The Dissolution Procedure: Development and Validation and to the paper Gray, V. A.; Brown, C. K.; Dressman, J. B.; Leeson, L. J. A New General Information Chapter on Dissolution. Pharm. Forum 2011, 27 (6), 3432—3439. The linearity range is typically established by selecting at least five concentrations from approximately �20% of the amount released at the first time point in the dissolution profile up to approximately �20% of the amount released in the last time point in the dissolution profile. Any additional strength that may be tested by the method should be taken into account when establishing this range.
Q
We have a product with a tolerance of NLT 85%. We tested six units and obtained an average result of 84.97%. Can we consider that the product meets the acceptance criteria, or should we continue the testing through S2 and S3?
A
General Notices section 7.20 says that results should be rounded to the number of decimal places in the limit expression. If the requirement was that the average value was NLT 85%, then the result, 84.97%, would be rounded accordingly to 85%. However, Acceptance Table 1 in the USP General Chapter <711> Dissolution uses only the individual values of the six units tested and not the average at the S1 stage of test. Therefore, you need to evaluate each individual result and verify if the testing should continue to S2 and S3 stages.
Q
We are developing an enteric-coated tablet. We have to perform comparative dissolution testing against the innovator product prior to the bioequivalence study. The comparative dissolu- tion evaluation should be performed in three dissolution media (i.e., pH 1.2 simulated gastric fluid without enzymes, pH 4.5 acetate buffer, and pH 6.8 simulated intestinal fluid without enzymes), and then the similarity factor, f2, is calculated for each case. The general procedure for dissolution of enteric-coated tablets requires dissolution in an acid stage followed by a buffer stage. We would
like to know how to perform the comparative dissolution evaluation in this case. Which one of the following options is the most appropriate: run the acid stage and then run the dissolution in the three different media or skip the acid stage and run the dissolution in the three different media?
A
To better answer this question, we need to break it into different parts. For a comparison of the dissolution profiles through the similarity factor, one of the first requirements is that the dissolution conditions must be exactly the same for both products being compared. That means the same medium volume and composition, same apparatus, and same rotation speed or flow rate. This also implies that the formulations being compared are almost identical, which is most often the case when they are immediate-release dosage forms. The similarity factor is usually not calculated for modified-release dosage forms because each product is likely to use a different release mechanism. This will be the case for different reasons, such as cost, technology, patent protection, and so forth. Each different release mechanism is going to require very specific dissolution conditions making the calculation
of f2 very difficult to apply and possibly meaningless. The similarity factor does have use for modified-release dosage forms in post-approval changes when the dissolu- tion conditions are going to be the same for the product before and after the changes.
Enteric dosage forms must have a dissolution test with at least two stages, the Acid stage, which demonstrates that the coating is acid resistant and is not going to release drug in the stomach, and the Buffer stage, where the drug must be released in a pH environment representing passage into the small intestine. In the case of product with colonic release, the test may have a Buffer stage with a dissolution medium pH of about 6.8 and a second Buffer stage with a dissolution medium pH around 7.2—7.5. The acid stage is going to be almost the same for all formulations, with the only difference being the time. For some products, the Acid stage lasts for 1 h, for others it lasts for 2 h.The conditions of the Buffer stage may vary according to the type of coating used.
Q
I am performing the performance verification test on my paddle apparatus. I have a six-position bath, and I am getting failing results with the new lot of prednisone tablets. I am working with the single-stage test and got the following values: 25, 27, 27, 30, 29, and 29 for the first run and 28, 30, 30, 33, 32, and 32 for the second run. The results pass for geometric mean but fail the requirement for %CV. I get a %CV of 8.1, and the limit is 6.8.
A
We ran your numbers through the online USP calcula- tor tool and found that the actual percent CV for the data set is 6.4. Please be careful to follow the instructions given in the certificate.The first step is to convert the results to the natural logarithm scale. Then take the variance of each set of numbers and average the two variances. Then convert the result to %CV as in the certificate procedure. You will get a different result, as your question shows, if you take the variance of all 12 log-transformed results and then convert that value to %CV.
USP recommends that the data be entered with at least three decimal points (e.g.,XX.YYY).The values in your question have been rounded to whole numbers. You may get different results using input values rounded to whole numbers and rounded to thousandths of percent dissolved.