Stability Studies: Preclinical, Clinical and Postapproval

Stability Studies: Preclinical, Clinical and Postapproval

By Rex Roberts, Ph.D.

Introduction

This article presents an overview of stability studies from initial, preclinical development of the drug substance and drug
product to routine manufacture of the FDA approved drug substance and drug product. Proper design, implementation,
monitoring and evaluation are crucial for obtaining useful and accurate stability data. Stability studies are linked to the
establishment and assurance of safety, quality and efficacy of the drug product from early phase development through the
lifecycle of the drug product. Stability data for the drug substance are used to determine the physico-chemical stability
profile for the compound. From this information, optimal storage and packaging conditions can be assessed; the appropriate
retest intervals and criteria can be established for bulk lots of the material. The stability studies for the drug product are
designed to determine the expiration date (or shelf life). In order to assess stability, the appropriate physical, chemical,
biological and microbiological testing must be performed. Usually this testing is a subset of the release testing. The
specifications for stability testing may be different from release specifications if properly justified.

Preclinical Studies1,2,3,4

The drug substance characterization and stability is usually determined as part of pre-formulation studies. Studies are setup
to degrade the solid drug substance and appropriate solutions and determine the degradation profile. The drug substance is
usually challenged under a variety of accelerated environmental conditions to evaluate its intrinsic stability and degradation
profile.

HPLC is the predominant tool used to analyze the drug substance and the impurities, particularly for small molecules.
Frequently, the same HPLC method may be used for drug substance and drug product, although different sample
preparation methods would normally be required. Often the assay and impurity testing can be performed using a single
HPLC method. However, the assay and purity determinations may also be separate methods. At least in the US, full
validation of the analytical method is not required until the end of Phase 2 clinical trials. However, establishment of
specificity, linearity and limit of quantification (for impurities) are important at the earliest stages, since verification of
stability hinges on a suitable method for separating impurities from the active ingredient and at least quantifying the
impurities relative to the drug substance.

Stress studies at elevated temperature (e.g. 50°C, 60°C and 70°C) for several weeks may be performed to assess thermal
stability. Provided the degradation mechanism is the same at the different temperatures used, kinetic or statistical models
can be used to determine the rate of degradation at other temperatures (e.g., 25°C). The solid stability should also be
performed in the presence and absence of water vapor to assess the dependence of stability on humidity.

Degradation studies should also be performed in solution. The solvent used for the solution testing will depend on the
solubility of the drug substance and should include water, if the drug substance is water-soluble. Other solutions or solvent
systems may be evaluated depending on the anticipated formulation or the synthetic process. A series of buffered solutions
in the pH range 2-9 are useful in assessing the impact of solution pH on the degradation. Photostability should also be
evaluated. A xenon light source can be used as a stress condition. Alternatively, one can use an accelerated version of either
Options 1 or 2 as described in the ICH guideline for determination of photostability.5 Oxidation of the drug substance
under accelerated conditions (e.g. hydrogen peroxide), may also be performed to establish oxidation products that could be
formed and sensitivity to oxidative attack.

Early drug product stability studies are designed to help establish a suitable formulation for delivery of the drug substance.
Compatibility studies of the drug substance with excipients should be performed to eliminate excipients that are not
compatible with the drug substance. Factorial design may be useful to reduce the number of experiments. Studies similar to
the solid drug substance stress studies may be performed. In addition, thermogravimetric analysis (TGA) and differential

http://www.southerntesting.com/StabilityStudies.htm (1 of 5)5/5/2011 3:14:20 PM

 Stability Studies: Preclinical, Clinical and Postapproval

scanning calorimeter (DSC) may be used to assess the stability of formulations.

Clinical Studies

Stability testing must be continued throughout clinical trials to support the safety, quality and efficacy of materials released
for clinical trials. Stability data must be submitted as part of the IND filing prior to initiating the Phase 1 clinical trial. Prior
to the first Phase 1 stability study, the pre-clinical studies should provide information on the appropriate long-term
condition and the appropriate container/closure system. ICH Q1A6 provides the guidance for design of clinical stability
studies. Selection of batches, the container closure system, specifications, testing frequency and storage conditions are the
most important issues to consider when designing a stability study.

Futscher and Schumacher12 proposed that the world could be divided into four zones based on temperature and humidity:
Zone I (temperate), Zone II (Mediterranean, sub-tropical), Zone III (hot dry), Zone IV (hot humid). The United States,
European Union and Japan are primarily Zone II.13 The provisions for Zones I and II as stated in ICH Q1A are summarized
in Tables 1 and 2.

Table 1: ICH Q1A Summary of Stability Parameters

Material Selection of Batches Container Closure Specifications Testing Frequency
at least 3 batches, the same as or
Drug Substance acceptance criteria for physical, chemical, Longterm (months):
minimum pilot plant simulates
biological and microbiological attributes 0,3,6,9,12,18,24,36...
scale, simulates container for
that measure changes in quality, safety intermediate: 0,6,9,12
production scale storage and
process distribution and/or efficacy. See ICH Q6A and Q3A 7,8 accelerated: 0, 3, 6
See above. Reduced
Drug Product at least 3 batches, the same acceptance criteria for physical, chemical, designs by applying
minimum of two pilot container closure biological and microbiological attributes bracketing and
plant scale, one can system proposed that measure changes in quality, safety matrixing which can
be smaller if justified for marketing and/or efficacy. See ICH Q6A and Q3B9 be used if justified

The container closure system must be evaluated for compatibility with the drug substance and drug product to ensure that
the container does not contribute to degradation or contamination. 10

The testing frequency represents the minimum data required for filing. It may be advisable to pull and test a one-month
sample for each storage condition to ensure that the study is proceeding as expected.

During Phase 1 it may be necessary to evaluate multiple formulations, dosage strengths and container closure systems.
Using bracketing and/or matrixing can frequently reduce the resource allocation for these studies. These two design
approaches are discussed in ICH Q1D.14 Bracketing uses the extremes to provide data for the entire study. For example, if
dosage strengths of 10, 25, 50 and 100 mg are to be evaluated the study may include testing of all strengths at the initial
and final time points with only the 10 and 100 mg strengths being tested at the intermediate time points. Matrixing might
be used to evaluate the same strength in multiple container/closure system by selecting only certain container closure
systems for testing at each time-point. This selection is usually done in a random fashion.

At the end of Phase 1, the process for manufacture of the drug substance, and the drug product should be established
(although refinements will typically continue for much longer).

The time period in Table 2 represents the minimum data required for the NDA. The studies must continue until the
longterm stability study is completed for the shelf life and retest period proposed in the NDA submission. Temperature
cycling studies and in-use stability studies may be needed for certain types of formulations (particularly liquid
and semisolid formulations). In early Phase 3 studies one should expect to be placing the batches on stability (at
least three drug substance and drug product lots) that will be used for filing the NDA. These may be the validation batches
if process validation is performed early enough. Process validation may be performed near the end of Phase 3 and adequate

http://www.southerntesting.com/StabilityStudies.htm (2 of 5)5/5/2011 3:14:20 PM

 Stability Studies: Preclinical, Clinical and Postapproval

stability data for these batches may not be available at the time of filing. shelflife and retest periods may be determined
statistically with adequate quantitative data.

Table 2: ICH Q1A Summary of Stability Parameters

Study Storage Condition Minimum Comments
Time Period
Must cover retest or shelflife period at a
General Case: Long-term 25 °C ± 2°C/60% RH ±5% RH or 12 months minimum and includes storage, shipment and
30°C ± 2°C/65% RH ± 5% RH subsequent use.
General Case: Intermediate 30°C ± 2°C/65% RH ± 5% RH 6 months Must cover retest or shelflife period at a
minimum and includes storage, shipment and
subsequent use.
Must cover retest or shelflife period at a
General Case: Accelerated 40°C ± 2°C/75% RH ± 5% RH 6 months minimum and includes storage, shipment and
subsequent use.
Refrigeration: Long-term 5°C± 3°C 12 months Must cover retest or shelflife period at a
minimum and includes storage, shipment and
subsequent use.
Must cover retest or shelflife period at a
Refrigeration: Accelerated 25°C ± 2°C/60% RH ± 5% RH 6 months
minimum and includes storage, shipment and
subsequent use.
Must cover retest or shelflife period at a
Freezer: Long-term - -20°C± 5°C 12 months minimum and includes storage, shipment and
subsequent use.

Aqueous products stored in semi-permeable containers must undergo studies designed to determine if water might be lost
during storage. Storage conditions for these studies are summarized in Table 3.

Table 3. ICH Q1A Stability Study for Semi-permeable Containers

Study Storage Condition Minimum Time Period
Long-term 25°C ± 2°C/40% RH ± 5% RH 12 months
Intermediate 30°C ± 2°C/60% RH ± 5% RH 6 months
Accelerated 40°C ± 2°C/ NMT 25% RH 6 months

The second revision of Q1A (Q1A (R2)), which addresses stability studies required for Zones I and II, resulted in several
changes intended to easily allow incorporation of storage conditions in Zones I, II, III and IV into the same study design.
The intermediate storage condition changed from 30°C ± 2°C/60% RH ± 5% RH to 30°C ± 2°C/65% RH ± 5% RH for
drug substance storage, drug product and drug products packaged in semi-permeable containers. The 30°C ± 2°C/65% RH
± 5% RH condition may be used as a longterm condition instead of 25°C ± 2°C/60% RH ± 5% RH. However, keep in mind
that this would result in no backup condition for the accelerated storage condition of 40°C ± 2°C/75% RH ± 5% RH.
11
Stability studies for Zones III and IV are covered in a separate ICH document.

ICH Q1E15 addresses the evaluation of stability data. The scenarios for room temperature (RT) and refrigerated (RF)
conditions are summarized in Table 4.

The time period in Table 2 represents the minimum data required for the NDA. The studies must continue until the
longterm stability study is completed for the shelf life and retest period proposed in the NDA submission.

Table 4: Summary of Evaluation Scenarios for Room Temperature and Refrigerated Conditions
(RF) Retest
Scenario Storage Condition Statistical Analysis (RT) Retest Period Period or
or Shelf Life (Y) Shelf Life (Y)

http://www.southerntesting.com/StabilityStudies.htm (3 of 5)5/5/2011 3:14:20 PM

 Stability Studies: Preclinical, Clinical and Postapproval

No significant change in accelerated condition over 6 Room temperature Usually not required Y= NMT 2X and Y= NMT 1.5X
mo; little or no change over time and little or no or Refrigerated NMT X+12 mo and NMT X+6
variability for accelerated and longterm mo
No significant change in accelerated condition over 6 Long-term data
mo, but either change or variability or both; little or Room temperature amenable to statistical Y= NMT 2X and Y= NMT 1.5X
no change over time and little or no variability for or Refrigerated analysis and statistical NMT X+12 mo and NMT X+6
longterm analysis performed mo
Long-term data not
No significant change in accelerated condition over 6 amenable or statistical
mo, but either change or variability or both; little or Room temperature analysis not performed, Y= NMT 1.5X and Y= NMT 1.5X
no change over time and little or no variability for or Refrigerated but relevant supporting NMT X+6 mo and NMT X+6
longterm data mo
Significant change in accelerated condition over 6 mo Room Temperature No extrapolation NMT longterm data; N/A
and significant change for intermediate possibly less
Long-term data
Significant change in accelerated condition Room Temperature amenable to statistical Y=NMT 1.5X and N/A
and no significant change for intermediate over 6 mo analysis and statistical NMT X+6 mo
analysis performed
Long-term data not
Significant change in accelerated condition Room Temperature amenable to statistical Y= NMT X+3 mo N/A
and no significant change for intermediate over 6 mo analysis, but relevant
supporting data
NMT
Significant change in accelerated condition over 6 mo Refrigerated No extrapolation N/A
longterm data;
and no significant change for intermediate over 3 mo
possibly less
NMT
longterm data;
possibly less;
data may be
Significant change in accelerated condition over 6 mo Refrigerated No extrapolation; N/A needed to
and significant change for intermediate over 3 mo support
excursions

In Table 4, ‘X’ represents the number of months of data being evaluated. No extrapolation is allowed if the longterm
condition is –20ºC (freezer). In this case, the shelflife and retest period will not be approved beyond the time period for the
available supporting data. Concepts of change over time, significant change and variability must be defined in order to
determine the retest and shelflife using Table 4. The concepts of change over time and significant change are defined in
Q1E. A p-value > 0.25 for the slope of the regression line indicates insignificant change; otherwise, the p-value represents
change over time. There is no definition for variability. Bar16 has recently proposed that process capability index (Cpk), be
used as a measure of variability and that a value of Cpk > 2.5 indicates little or no data variability.

Postapproval (Marketing Phase)

At least one lot of drug substance and one lot of each packaging type for drug product produced each year should be
placed on long-term stability. Additional stability testing may be required to support process changes for drug substance
and/or drug product. The filing requirements for changes are covered in multiple FDA guidance documents addressing drug
product changes (SUPAC) and drug substance changes (BACPAC). This is typically an area that requires substantial
regulatory understanding and experience to know how to proceed and is beyond the scope of this article.

Conclusion

Stability is interwoven through the entire fabric of the drug product lifecycle. A detailed understanding of this
area is needed to properly design and evaluate stability studies in order to ensure minimal delays and minimize
costs in developing a new drug product. This article presents an overview of stability testing and provides the
reader with references to obtain a detailed understanding of the subject. Finally, detailed knowledge of the

http://www.southerntesting.com/StabilityStudies.htm (4 of 5)5/5/2011 3:14:20 PM

 Stability Studies: Preclinical, Clinical and Postapproval

stability requirements and the impact on other areas (e.g., container closure, process changes) is crucial for
selecting the appropriate contract lab to perform stability studies.

References

1. Grimm, W., “Stability Testing of Clinical Samples”, Drug Development and Industrial Pharmacy, 22, No. 9&10, 851-
871 (1996).
2. Berglund, M., Bystrom, K., Persson, B., “Screening Chemical and Physical Stability of Drug Substances”, Journal of
Pharmaceutical & Biomedical Analysis, 8, No. 8-12, 639-643 (1990).
3. Krummen, K., “Stability Testing During Development”, Paperback APV, 16, 209-225 (1987).
4. Witthaus, G. “Drug Stability. Accelerated Storage Tests: Predictive Value”, Top. Pharm. Sci., Proc. Int. Congr. Phar.,
275-290 (1981).
5. Federal Register, “Guideline for the Photostability Testing of New Drug Substances and Products”, May 16, 1997, Vol.
62, No. 95, 27116.
6. ICH Q1A (R2), Stability Testing of New Drug Substances and Products, 6 February 2003.
7. ICH Q6A, Specifications Test Procedures and Acceptance Criteria for New Drug Substances and New
Drug Products: Chemical Substances, 6 October, 1999.
8. ICH Q3A(R), Impurities in New Drug Substances, Step 4, 7 February 2002.
9. ICH Q3B(R), Impurities in New Drug Products, Step 4, 5 February 2003.
10. FDA Guidance, “Container Closure Systems for Packaging Human Drugs and Biologics”, May 1999.
11. ICH Q1F, Stability Data Package for Registration Applications in Climatic Zones III and IV, Step 4, 6
February 2003.
12. Futscher, N., Schumacher, P., Phar. Ind., 34, 479-483 (1972).
13. Beaumont, T., Paperback APV, 32, 177-190 (1993).
14. FDA Guidance, “Q1D Bracketing and Matrixing Designs for Stability Testing of New Drug Substances and Products”,
January 2003.
15. ICH Q1E, Evaluation for Stability Data, Step 4, 6 February 2003.
16. Bar, R., “Statistical Evaluation of Stability”, PDA J. Pharm. Sci. Technol., 57, No. 5, 369-377 (2003).

http://www.southerntesting.com/StabilityStudies.htm (5 of 5)5/5/2011 3:14:20 PM