2.5 Pharmaceutical Quality: GMP, Stability Testing, and Bioequivalence
Key Takeaways
- Good manufacturing practices under Division 2 of the Food and Drug Regulations require a Drug Establishment Licence, and every marketed product carries a Drug Identification Number.
- First-order degradation gives a shelf life t90 of 0.105 divided by the degradation rate constant, the time at which 10% of the labelled potency has been lost.
- The Arrhenius relationship allows accelerated stability data at elevated temperature to predict shelf life at normal storage conditions.
- Canadian bioequivalence standards require the 90% confidence interval of the test-to-reference ratio for area under the curve to fall within 80 to 125 percent.
- Compounded preparations are assigned a beyond-use date based on published stability data and USP-based defaults, which is not the same as a manufacturer's expiry date.
2.5 Pharmaceutical Quality: GMP, Stability Testing, and Bioequivalence
Exam Focus: The syllabus lists "principles of formulation and good pharmaceutical manufacturing practice, including aspects of product development and assessment, stability testing, and quality control" and "bioequivalence testing of generic drugs" as explicit bullets under Pharmaceutics and Drug Delivery Systems.
The Canadian Regulatory Chain
A drug product reaches a Canadian pharmacy through a defined chain of approvals under the Food and Drugs Act and the Food and Drug Regulations.
| Element | What it establishes |
|---|---|
| New Drug Submission | Evidence of safety, efficacy, and quality reviewed by Health Canada |
| Notice of Compliance | Authorisation to market the product in Canada |
| Drug Identification Number (DIN) | An eight-digit number identifying manufacturer, product name, active ingredients, strength, dosage form, and route |
| Drug Establishment Licence | Authorisation for a site to fabricate, package, label, test, distribute, import, or wholesale |
| Abbreviated New Drug Submission | The generic pathway, relying on demonstrated bioequivalence to a Canadian reference product rather than repeating clinical trials |
| Notice of Compliance with conditions | Market authorisation on promising early evidence, conditional on confirmatory studies |
Good manufacturing practices (GMP) are set out in Division 2 of the Food and Drug Regulations and cover premises, equipment, personnel, sanitation, raw material testing, manufacturing control, quality control, packaging, labelling, records, samples, stability, sterile products, and recall procedures. The governing principle is that quality is built into the process, not tested into the product at the end. Finished-product testing samples a fraction of a batch and could never, by itself, assure the quality of the whole.
Recall classification follows the health risk:
- Type I — a reasonable probability that use will cause serious adverse health consequences or death.
- Type II — may cause temporary or medically reversible harm, or the probability of serious harm is remote.
- Type III — unlikely to cause any adverse health consequence, such as a minor labelling defect.
A pharmacy receiving a recall must quarantine affected stock, check dispensing records, notify affected patients for a Type I recall, and document the actions taken.
Degradation Kinetics and Shelf Life
Most pharmaceutical degradation follows first-order kinetics, where the rate depends on the remaining concentration of drug.
The concentration at time t is C = C0 multiplied by e raised to the power of negative k times t, where k is the degradation rate constant. Two derived quantities matter:
- Half-life of degradation: t1/2 = 0.693 / k — the time for potency to fall to 50%.
- Shelf life: t90 = 0.105 / k — the time for potency to fall to 90% of label claim. This is the quantity that defines an expiry date, because most products are required to retain 90 to 110 percent of labelled potency. The value 0.105 is the natural logarithm of 100/90.
The t90 formula appears on the PEBC reference sheet provided during the examination, so it does not have to be memorised, but the ability to recognise which formula the question requires does.
Worked example: a solution degrades by first-order kinetics with k = 0.0021 per month. Shelf life t90 = 0.105 / 0.0021 = 50 months.
Zero-order degradation, where rate is independent of concentration, applies to suspensions in which dissolved drug is continuously replenished from undissolved solid. Second-order kinetics are uncommon in pharmaceutical stability.
Pathways of degradation and how formulation counters them
| Pathway | Susceptible drugs | Countermeasure |
|---|---|---|
| Hydrolysis | Esters, amides, beta-lactams, aspirin | Buffer to the pH of maximum stability; supply as a dry powder for reconstitution; use non-aqueous or water-miscible cosolvents |
| Oxidation | Epinephrine, ascorbic acid, phenothiazines, morphine | Antioxidants such as sodium metabisulfite; chelating agents such as edetate disodium; nitrogen headspace; light protection |
| Photolysis | Nifedipine, amiodarone, riboflavin, nitroprusside | Amber glass, opaque overwrap, protective cartons |
| Racemisation and isomerisation | Adrenaline, tetracycline | pH control and temperature control |
| Physical instability | Emulsions, suspensions, proteins | Control particle size and viscosity; avoid freeze-thaw cycles and agitation |
The Arrhenius relationship links the degradation rate constant to temperature, so a product stored at elevated temperature degrades faster in a predictable way. Accelerated stability testing exploits this by storing samples at conditions such as 40 degrees Celsius and 75% relative humidity and extrapolating back to the shelf life at the long-term condition of 25 degrees Celsius and 60% relative humidity. Accelerated data support a provisional expiry date; real-time data must confirm it. Note that the Arrhenius approach does not hold for products whose failure mode is physical rather than chemical, which is why suspensions, emulsions, and biologics require real-time data.
Stability testing evaluates chemical potency, degradation products, physical appearance, microbiological quality, and the performance of the container closure system, since a light-sensitive product in a clear vial and a moisture-sensitive tablet in a poorly sealed bottle both fail regardless of the molecule's intrinsic stability.
Beyond-Use Dates in Compounding
A manufacturer's expiry date is supported by full stability data on the product in its original unopened container. A beyond-use date (BUD) applies to a compounded or repackaged preparation and is assigned by the pharmacist. It is derived from published stability-indicating data for the specific formulation where such data exist, and from conservative default rules based on the formulation type and the risk level of the compounding process where they do not.
A beyond-use date can never exceed the earliest expiry of any ingredient used, and it must account for the new container, the new environment, and the fact that opening the original package changes the conditions the manufacturer validated.
Bioequivalence
A generic product contains the same active ingredient in the same dosage form, strength, and route as a Canadian reference product. It need not repeat clinical efficacy trials; instead it must demonstrate bioequivalence in a pharmacokinetic study, usually a randomised crossover in healthy volunteers with adequate washout.
The measured parameters are:
- AUC — area under the plasma concentration-time curve, representing total extent of absorption.
- Cmax — maximum concentration, representing rate of absorption.
- Tmax — time to maximum concentration.
The Canadian standard requires the 90% confidence interval of the ratio of the test product to the reference product for AUC to fall entirely within 80 to 125 percent, with a corresponding requirement on Cmax. Two points are commonly misunderstood and are worth stating precisely:
- The criterion applies to the confidence interval, not merely to the point estimate. A product whose mean ratio is 100% can still fail if its variability is high.
- The 80 to 125 percent range is not a licence for a 45-percentage-point difference between two generics in an individual patient; it is a statistical equivalence boundary on average population exposure, and the interval is asymmetric because it is calculated on a logarithmic scale, where 1/1.25 equals 0.80.
Narrow-therapeutic-index drugs — such as levothyroxine, warfarin, digoxin, lithium, cyclosporine, tacrolimus, and several antiepileptics — face tighter bioequivalence requirements, and the practical recommendation is to keep a patient on a consistent product and to monitor if the manufacturer changes.
Provincial interchangeability is a separate, jurisdictional decision. Health Canada determines bioequivalence; each province decides which products its drug plan will treat as interchangeable and therefore substitutable at the counter. Biologics add a further distinction: a biosimilar is not a generic, is not declared interchangeable through the same process, and substitution is governed by provincial policy rather than by pharmacist discretion.
A solution degrades by first-order kinetics with a rate constant of 0.0035 per month. What is its shelf life?
Which requirement must a Canadian generic product meet to be considered bioequivalent to its reference product?
Accelerated stability testing at 40 degrees Celsius and 75% relative humidity is used to predict shelf life at normal storage. Which principle underlies this approach?
A pharmacist compounds an oral suspension from commercially available tablets. Which statement about dating the preparation is correct?