Designing a globally acceptable registration stability protocol

Initially, the International Conference on Harmonisation (ICH) guideline Q1A described harmonized stability requirements for the ICH tripartite regions (EC, Japan and the US), which are in Climatic Zones I and II.¹ Later, postissuance of the World Health Organization report of the 37th meeting of the WHO Expert Committee on Specifications for Pharmaceutical Preparations in 2001,² ICH guideline Q1F was issued.³ To minimize required storage conditions, ICH Q1A was revised so that the Zone I and II intermediate condition aligned with the WHO proposals of 2001 on the proposed long-term storage condition for Zone IV countries (i.e., 30 °C/65% relative humidity [RH]). In 2003, it appeared that global harmonization had been achieved.

However, since that time, a further set of opinions and regional guidelines have developed leading to a diversity of opinion.^4–7 A number of countries in Zones III and IV, led by the Association of South East Asian Nations (ASEAN) region, decided to review guidelines in light of new information on climatic conditions in Zone IV.⁵ Although many are along the lines of the original ICH guidelines, there are a number of differences and positions that have evolved, particularly away from ICH Q1F, which was subsequently withdrawn on 8 June 2006.

The ASEAN countries and Brazil, among other regions/countries, are now adopting a position of 30 °C/75% RH as the most appropriate long-term storage condition.^5–7 These are now classified as Zone IVB countries. In September 2006, the WHO Eastern Mediterranean Region Office stability guideline was finalized, which describes:

• 30 °C/35% RH as the Zone III condition (now including Egypt)

• 30 °C/65% RH for the Zone IVA condition (now including Iran)

• 30 °C/75% RH for the Zone IVB condition.⁷

This regional guideline was initially accepted at a WHO expert meeting in Geneva (Switzerland) in October 2006, as a draft for a potential WHO global stability guideline. There is a possibility that the zones into which all countries fit could be gathered and a guideline released at the next meeting of WHO experts in October 2007.

It should be noted that in Annex 1 of the Eastern Mediterranean Region (EMR) guideline, countries classified previously as being in Zone III have frequently selected 30 °C/65% RH as the recommended long-term storage condition. Thus, if a company wanted to maximize the chances of success of a stability study in as many regions as possible, one could see a situation arising for storage/testing at 25 °C/60% RH, 30 °C/65% RH and 30 °C/75% RH out to end of shelf-life.

Large stability studies for drug products and devices with multiple strengths, packaging configurations and orientations can cost millions of dollars for the analytical testing alone. Depending on the cost model, adding an extra storage condition can increase costs by up to 20%. A goal of the pharmaceutical industry must be to have a globally acceptable registration stability package without unnecessary testing, saving on manufacturing requirements, cost and time.

Considerations when designing a reduced stability protocol

It is highly likely that, because of the length of time products are in the development pipeline, they have been initially developed considering stability at 25 °C/60% RH and, at least for some period of time, 30 °C/65% RH, with a knowledge of stability characteristics at 40 °C/75% RH for up to 6 months.

It is not necessarily easy to extrapolate stability behaviour at these conditions to behaviour at 30 °C/75% RH for long-term storage. The Arrhenius or van't Hoff equations are often used to model temperature effects, however humidity effects can be more challenging to predict, particularly for drug products. Table 1 shows variation in vapour pressure at different temperatures/relative humidities. For example, from this table it can be seen that going from 25 °C/60% RH to 30 °C/75% RH, vapour pressure increases by 67%. Depending on the dosage form, its susceptibility to moisture related degradation (chemical and physical) and the packaging permeability, this could have a significant effect on a product's shelf-life. Generally, for a given temperature, the higher the RH, the more prone products become to both chemical and physical degradation.⁹

Table 1 Pressure of water vapour at different temperatures and relative humidities.8

However, one must be careful when assuming that a higher relative humidity will be a worst-case scenario. This is particularly true for the performance of hard gelatine capsules, for which capsule brittleness is inversely related to moisture content.¹⁰ Under ambient conditions, a minimum quantity of sorbed moisture is necessary to act as a plasticizer and maintain the capsules in a pliable state, although too much water can cause the capsules to become sticky and may lead to gelatine crosslinking, frequently leading to a drop in in vitro dissolution performance.¹¹ Thus, an intermediate moisture content range is desirable.

Kontny and Mulski found that brittleness tended to occur below approximately 40% RH, which may be important, particularly when considering hot and dry countries.¹⁰ Although drug retardation is often reported in vitro because of crosslinking, Bremecker and List found drug release increased with rising RH levels for lipophilic and hydrophilic drugs from hard gelatin capsules.¹²

At a particular temperature, water sorption/desorption of tablets is generally related to RH; however, across temperatures, behaviour may not be a simple function of either relative or absolute moisture content in the atmosphere (Table 2).

Table 2 Moisture sorption/desorption for tablets at various temperatures and humidities.9

As can be seen by comparing water sorption/desorption at 25 °C/33% RH and 40 °C/32% RH, predictions of tablet stability cannot necessarily be made by examining either RH or absolute humidity. In another study, Matsuda et al. showed that ranitidine HCl active pharmaceutical ingredient (API) degraded at a maximum rate of 60–70% RH and that the degradation rate lowered above 70% RH (the critical RH); this was found to be independent of temperature.¹³ It should be noted, however, that above 70% RH the API had dissolved.

Udeala and Aly found a decrease in stability for some thiamine hydrochloride directly compressed tablet formulations with increasing tablet moisture/RH of storage, followed by an increase in stability on further moisture increase (i.e., beyond a certain point, increasing RH of storage could lead to an improvement in stability characteristics depending on the formulation).¹⁴

For formulations containing amorphous sugars (e.g., freeze-dried formulations), a decrease in storage temperature, where RH is kept constant, can result in an increase in the extent of water sorption with possible adverse effects on hydrolytic stability.¹⁵ This negative deviation from ideality (Raoult's law) was postulated to be the increasing dominance of the intermolecular hydrogen bonds between water and sugar over water–water hydrogen bonding at lower temperatures.

Humidity may also have an effect on the route of degradation. Stanisz found differences in the mechanisms of degradation with humidity; however, it should be stressed that in this case a difference was only noted at the extreme of low humidity (0%) and that between 50.0% and 76.4% RH, at fixed temperature, the same degradation mechanisms were found.¹⁶ Genton and Kesselring studied the effects of temperature and humidity on the stability of nitrazepam in solid-state and found maximum degradation rate at 75% RH, which then decreased at higher relative humidities at all temperatures.¹⁷

A proposed globally acceptable protocol

The statement on the ICH website where Q1F used to be states that the regulatory authorities in the ICH regions have agreed that the use of more stringent humidity conditions such as 30 °C/75% RH will be acceptable should the applicant decide to use them. However, when designing a globally acceptable, but reduced testing protocol, an understanding of the product gained during development, and particularly during stress testing studies, is important.

Key points

The innovator must understand the robustness of the product such that RH/moisture, and its relation to temperature, is characterized regarding a product's stability and performance. At the very least, if this is not understood then one of the consequences may be that more protective packaging than is required will be chosen to remove the risk of failure on stability; and, frequently, the more protective the packaging the higher the cost.

Currently, there is limited data available in literature comparing product performance at 25 °C/60% RH and 30 °C/65% RH to 30 °C/75% RH. Stability behaviour for a product at 40 °C/75% RH will provide an indication as to whether a product is particularly susceptible to moisture; however, if specific data are not available, or is limited in duration, from a company's product over the long-term, then there will be risks associated with a reduced registration stability study. This is particularly true if only what are considered the most stressful conditions are studied leading to a reduced or unworkable shelf-life in countries with less demanding storage requirements.

At this time of changing regional requirements and limited development experience at the 'new' condition of 30 °C/75% RH, a proposed compromise between minimizing the amount of testing required and minimizing risk of failure on stability is described in Table 3.

Table 3 Proposed globally acceptable stability protocol.

For products characterized as particularly stable/robust during development, particularly if packaged in moisture impermeable packaging, and where no inverse relationship has been found during development between temperature/RH and stability, then a study could be reduced further to move to testing at 30 °C/75% RH only, perhaps with spares at 25 °C/65% RH, again reducing costs significantly.

For aqueous-based drug products packaged in semipermeable containers, calculations for deriving water loss rate as described in Q1A(R2) can be applied.

As stated previously, Q1F has now been withdrawn. It should also be noted that of the conditions that were described in ICH Q1F for testing at elevated temperature and/or extremes of humidity, the condition 25 °C/80% RH equates to a vapour pressure of around 2500 Pa compared with 5531 Pa at 40 °C/75% RH and 3182 Pa at 30 °C/75% RH. Furthermore, storage at these latter conditions is for 6 months and long-term, respectively, compared with 3 months described for 25 °C/80% RH. Therefore, as well as not now being part of ICH guidance, it could be scientifically argued that satisfactory data at the higher stress conditions removes the need to do studies at 25 °C/80% RH.

However, there is still an expectation in Japan for studies on a batch of API for 3 months at this condition. The remaining requirement for studies at 50 °C/20% RH is also only for stress testing studies on API for Japan.

Conclusions

For products well-characterized during development, significant reductions can be made to the design of registration stability protocols. Even when full characterization over long-term conditions has not occurred, there are opportunities to reduce costs when designing a globally acceptable stability study.

Jon Beaman is a senior director, global head of registration stability at Pfizer (UK).

References

1. ICH Q1A Stability Testing of New Drug Substances and Products. www.ich.org

2. Amendment from Draft of the 37th Report of The WHO Expert Committee on Specifications for Pharmaceutical Preparations (Geneva, Switzerland, 22–26 October 2001). www.who.int

3. ICH Q1F Stability Data Package for Registration in Climatic Zones III and IV. www.ich.org

4.Stability Studies in a Global Environment (Geneva, Switzerland, 13–14 December 2004). www.who.int

5. ASEAN Guideline on Stability Study of Drug Product (22 February 2005). www.aseansec.org

6. Resolution — RE No.1 of July 29 2005, Federative Republic of Brazil, Ministry of Health National Agency of Health Surveillance. www.brasil.gov.br

7. Regional Guideline for the WHO EMR: Stability Testing of Active Substances and Pharmaceutical Products, Draft 2.0 (19 April 2006). www.who.int

8. R.M. Tennent, Science Data Book (Oliver and Boyd, Edinburgh, UK, 1971).

9. A.V. Katdare and J.F. Bavittz, Drug Dev. Ind. Pharm., 10(7), 1041–1048 (1984).

10. M.J. Kontny and C.A. Mulski, Int. J. Pharmaceutics, 54, 79–85 (1989).

11. G.A. Digenis, T.B. Gold and V.P. Shah, J. Pharm. Sci., 83(7), 915–921 (1994).

12. K.D. Bremecker and P.H. List, Pharmazeutische Industrie, 43, 1026–1028 (1981).

13. R. Teraoka, M. Otsuka and Y. Matsuda, J. Am. Pharm. Sci., 82(6), 601–604 (1993).

14. O.K. Udeala and S.A.S. Aly, Drug Dev. Ind. Pharmacy, 14(12), 1735–1764 (1988).

15. B.C. Hancock and C.R. Dalton, Pharm. Dev. Technol., 4(1) 125–131 (1999).

16. B. Stanisz, Acta Poloniac Pharmaceutica, 61(2), 91–97 (2004).

17. D. Genton and U.W. Kesselring, J. Pharm. Sci., 66(5), 676–680 (1977).