Test Method Selection and Specification Setting in Stability Testing for Combination Medical Device Products


Stability testing for combination medical devices is relatively new in comparison to traditional pharmaceutical dosage products and has recently gained more attention for manufacturers and regulators.  Test methods and specifications are integral parts of stability testing for combination products and play important roles in the success of a stability program. Combination medical devices have special considerations due to the complexity of the device design and integration of drug components.  Additionally, with new regulations and less experience in industry this results in challenges for the developers of combination medical devices.  This paper provides details and descriptions about how to select test methods and setting specification in stability testing for combination medical device products and also provides general practice guidelines about how to implement the changes of test methods and specification in the course of stability testing. 

1. Introduction

Combination products are comprised of two or more regulated components such as drugs, devices, or biologicals that are physically, chemically, or otherwise combined or mixed and produced as a single entity to produce a therapeutic effect.  They also may be comprised of two or more separate products packaged together in a single package or finally as products packaged separately that according to its labeling is for use only with another individually specified drug, device, or biological product where both are required to achieve the intended use, indication, or effect (1,2). Single entity drug-device combination products are perhaps the most common and some examples include pre-filled insulin syringes, metered inhalers, catheters or contact lenses coated with antibiotics, transdermal patches, drug eluting stents, and steroid eluting pacemaker leads.   

We recently published an overview of the determination of shelf life for combination devices (3) and this follow up article focuses on the selection of test methods and establishment of specifications explicit to the drug component of combination products.  While test methods and specifications have been extensively discussed in the pharmaceutical industry and ICH guidelines can be widely applied towards combination products, stability testing programs for combination medical devices are still relatively new for both industry and regulators and some areas are not clear for combination products developers.  Since there are many flavors of combination devices and testing and specification settings will be highly contingent upon the device type, this paper uses drug eluting medical devices as a representative model for a reasonably-complex combination product that can provide general approaches to the reader to apply towards their own products.

Before specific discussion about methods and specifications, it would be helpful to understand the following considerations which illustrate potential differences in perspective from the pharmaceutical industry compared to the combination product point of view and this may help lay out the rationale for the method selection and specification development.  Extensively, since the drug in a combination device is typically coated onto, contained inside of, or materialistically incorporated into the device, the approaches may be very different compared to standard pharmaceutical dosage forms.  Additionally, the primary function or intention of a drug in combination devices may not be intended to cure a disease or prevent a medical condition.  Instead, the drug purpose could be to improve biocompatibility or prolong the lifetime of a device when implanted or inserted into body.  For instance, drug eluting stents and pacemaker leads both release drug over time to locally reduce scar tissue overgrowth at the implant site.  Drug analysis considerations for combination products must also consider sample preparation factors such as: finished goods may require difficult disassembly techniques or destruction to remove or extract drug compounds, the drug may be degraded or lost during sample preparation, drugs may be contained within an excipient matrix such as complex polymers which could directly impact testing, the interaction between device and drug or leachable compounds may impact testing results, and finally the amount of absolute drug in a device is typically very low compared to pharmaceuticals and therefore quantification of drug and impurities can be challenging,.  All of these factors may directly impact the consistency of test results and great detail should be placed into developing highly sensitive robust analytical procedures that specify detailed instructions in regard to equipment, temperature conditions, times of and allotted for sample preparation, and procedures of removal of components.  Lastly, combination devices are very expensive on a per unit basis in comparison to pills, tablets, or liquid formulations and regulatory agencies may also have strong opinions and become directly involved with test parameter selection and method development.  

Note:  Combination products come under 21 CFR 4 CGMPs which outline which of the existing CGMPs primarily are to regulate the development and manufacture of the combination product (1).  The decision basically depends upon ultimate risk to the patient / end user posed by the combination product’s components, and often with drug / device combination products, the highest risk component is the drug, hence the driving CGMPs would be 21 CFR 210, 211, drugs. However, in all cases, even where the device is the lower risk component, 21 CFR 820.30, design control for devices, must be employed during development of the combination product.  

2. Selection of Test Methods 

Test method and attribute selection in stability testing is based on ICH Q1A (4), stating that “Stability studies should include testing of those attributes of the drug products that are susceptible to change during storage and are likely to influence quality, safety, and/or efficacy.  The testing should cover, as appropriate, the physical, chemical, biological, and microbiological attributes, preservative content (e.g., antioxidant, and antimicrobial preservative), and functionality tests (e.g., for a dose delivery system).” The test attributes that are considered generally applicable to all combination products include appearance, drug content (assay), impurities/ degradation products, in-vitro drug release/elution, particulate matter and antioxidant preservative(s) content. In addition, some other tests attributes, such as sterility and package integrity, are performed at release, annually, and at expiry. Depending on the nature of the combination product and its manufacturing process, additional methods using cutting-edge technologies to solve analytical problems should be considered as appropriate to ensure the stability profile of the combination product is fully characterized.  All analytical test methods should be clearly described in the internal analytical procedure and fully validated following regulatory guidance prior to using in stability testing (5,6).   These methods should be documented in a quality system following current cGMP practice unless in the very early stages of product development.  The following sections address common tests performed for stability analysis using drug-eluting devices as a model for combination products.


Appearance testing is the inspection of drug components before and or after removal from the device by microscopy, visually inspection, or use of other machine vision techniques to document any abnormal observations.   Any removal processes used and items to be inspected along with acceptance criteria need to be detailed in the procedure.  Inspection methodology could include: scanning device components and packaging to make sure they are intact and properly positioned, checking for signs of damage, looking for presence of foreign debris, and checking for voids, tears, flashing, or any other abnormalities in controlled release matrices.  Color of the components or drug substance may be representative of oxidative or hydrolytic stress and may also be a part of appearance testing. Furthermore, any aberrant observations made by an inspector that aren’t explicitly covered in the procedures should be recorded an investigated further.  

Assay (Total Drug Content): 

The assay method should be able to quantitatively determine the total active pharmaceutical ingredient in the device and ensure that the commercial product is consistently made and that the drug remains within specification during the entire shelf life.  Assay testing should be specific and capable of overcoming interferences from background matrix.  The method should be stability indicating and ideally would also be capable of simultaneously detecting impurities.  HPLC is very often used for its analytical and high throughput capabilities and it is a common practice to use the same testing method for both stability and lot release. For combination devices, the assay testing should be performed on an individual unit basis with the reported assay result comprised of the average measurement from each individual unit.  The common pharmaceutical testing practice of combining multiple units may not be applicable for combination device testing. In general, 10 units per lot should be sufficient for drug content and it is generally acceptable to use this same data for determining content uniformity at lot release / T0 as evaluated per USP chapter <905> following a transdermal systems approach (7). 

Impurities or Degradation Products 

The impurity test method needs to be specific and stability-indicating and may be completed as part of the assay test method or as a separate method.  The testing may be performed on individual devices or it may be necessary to use a pooled or composite sample preparation from multiple devices to boost sensitivity as degradation products should only be a small percentage of the API.  In general, HPLC is often used in the analysis of impurities with detection by either single or multiple wavelength UV to aid identification.  The reporting of degradation products follows ICH Q3B (8). Normally, total degradation products, specified degradation products (i.e. impurities created by known degradation pathways), and unidentified degradation products are reported. The quantitation limit for the analytical procedure should be not more than (≤) the reporting threshold. Degradation product levels can be measured by a variety of techniques, including those that compare an analytical response for a degradation product to that of an appropriate reference standard or to the response of the drug substance itself. 

In Vitro drug release (drug elution)

Drug release or elution testing is required for both lot release and stability testing throughout the product shelf life.  In vitro drug release testing is not intended to mimic in vivo performance.  Instead the purpose is to ensure that API is released from the device in a controlled, consistent, and reproducible manner that can be measured at a few selected time points. The in vitro drug release method is typically much faster than in vivo and this shortens the lot release testing time and captures any potential issues earlier in the evaluation.

For some combination products, the drug release can be significantly different than a standard or controlled release pharmaceutical. A chief distinction for combination devices is that the API may be sub contained in a background matrix inside of or integrated into the inner workings of a device and could have limited amount of contact with biological fluids or tissues to dissolve or transport drug.  Therefore, the traditional fluid dynamics and mass transfer assumptions and behavior could be dramatically different and special considerations (e.g. degassing solutions, mechanical agitation of the device) for method development may be required.  Additionally, the small amount of starting material and complicated test processes can generate large variability which creates a great deal of difficulty in setting specifications at multiple time points.  Since there isn’t a compendial test method available, very often, a combination product developer needs to work closely with regulatory agencies for method development and specification setting before implementation of testing occurs.

In drug release testing, UPS dissolution apparatus 4 or others as appropriate are commonly used for combination products.  Like the assay method, HPLC with a UV detector is often used for quantifying the drug released. During the method development, selection of drug release medium, control of test conditions, and sampling practices is critical. Factors such as adequate solubility of the drug in the in-vitro release media to ensure ‘sink’ conditions under experimental conditions throughout the period of study, stability of the drug in the media and the media components during the entire period of study and the compatibility of the media with the detection technique are considered important aspects of the development phase of the method  A minimum of three-time points for drug release profile to cover initial burst, middle and final drug release should be evaluated against specifications; however, additional time points may be used for information purposes only and longer release times, even up to a few days, could be needed to measure a minimum release according to the specification.  It is imperative that the drug release test method should demonstrate discriminatory power, i.e. the method can detect the difference when drug formulation or manufacturing processes changes by different drug release profiles.  When drug release profiles from different processes, formulation or different test methods need to be compared, statistical approach or FDA f1 (difference factor) and f2 (similarity factor) approach may be necessary to demonstrate equivalency of different profiles (9).  


Particulates on a combination device may come from the manufacturing environment or different sources within the device, such as shedding from drug coating / surface or device delivery system from mechanical perturbations.  Any particulates on a cardiovascular device poses serious risks to patient safety including embolization and stroke and could have other significant risks based on implantation site location.  Particulate matter testing can also evaluate the amount of residual material that may leach off a medical device into the bloodstream throughout the life of the implant.  The test method for particulates largely depends on the products and drug components evaluated and USP <788> may be referenced for general particulate testing principals (10).  Before particulate testing occurs, a combination product may also need to be mechanically actuated or put through a procedure to mimic the twists and turns of implantation (e.g. coronary arterial stents delivered through the groin).  

Packaging integrity (vs. sterility) 

Packaging integrity testing is to ensure that the sterile barrier will be maintained during the shelf life and should be tested as part of stability testing.  It can be used in lieu of sterility per 2008 FDA guidance (11). Physical or chemical test methods may be used, such as leak testing, dye penetration, seal stretch, etc.  Testing can be done annually or at the last time point.


Bacteria shed endotoxins in large amounts while actively growing and dividing and upon cell death.  Therefore, it may be evaluated during stability testing to ensure levels remain within specification during shelf life.  The testing follows USP <85> (12) and may not be required globally and testing annually or at the last time point should be sufficient.

Polymer Molar Mass

Polymers have properties that are highly dependent on molar mass and are often used as a matrix to confine or apply the drug during manufacturing (e.g. creation of a monolithic controlled drug component or spray on coatings).  Therefore, depending on the products and role of polymers in the products, polymer testing is optional for stability testing.  Gel Permeation Chromatograph (GPC), which is essentially HPLC with specialized size exclusive based columns, is used in tandem with refractive index and or multi-angle laser light scattering detectors.  The column(s) and mobile phase are the main consideration during method development.  Number average molar mass (Mn), weight average molar mass (Mw) and polydispersity index (PDI), are characteristics of interest along with FTIR or NMR measurements for copolymer composition analysis.

Antioxidant or drug stabilizer:  

Additives such as antioxidants like BHT (Butylated Hydroxytoluene) may be added to stabilize the drug substance.  Depending on the device, testing for the drug stabilizer may be required for both lot release or stability testing per agency request. Since the concentration of drug stabilizer is very low within the device, the detection and quantification of stabilizer can be a challenge.  HPLC or more sensitive instrument such as GC/MS etc. may be employed.  Excipients in the product matrix may interfere with the analysis.

3. Specification Setting

Specifications are important in stability testing and help ensure combination products maintain quality throughout their shelf life.  The definitions of specifications / acceptance criteria and specification setting guidance can be found in the ICH Q6A (13). Sufficient data must be collected and evaluated. Specifications shall be fully justified and statistical approaches widely used in the specification setting process. Supporting data could include product development / validation, preclinical / clinical, lot release and stability testing.  When setting specifications, variability from manufacturing process and testing should both be considered.  Acceptance criteria for lot release should be more stringent than those used for stability testing to allow for some degree of change in the product characteristics over time while still remaining safe and effective. Specifications should be defined and documented in a product specification document along with test procedure and stability protocol in a quality system.  Finally, a proposed specification and justification should be provided in a regulatory submission, e.g. CMC section in Pre-Market Approval (PMA) for combination medical devices to the FDA. Aspects related to specifications for relevant stability tests are discussed below using drug-eluting devices as a model for combination products:


The acceptance criteria should include description of defined observations as discussed in the previous section for test method and acceptance criteria.  Any color testing performed should have a physical / optical reference standard for a more quantitative measurement.


A typical acceptance criterion is an average value between 90.0% and 110.0% of label claim. This is considered a conservative range that will be likely to obtain regulatory approval.    Wider ranges have been seen for some devices and may need stronger justification. Care should be taken in the use of significant figures in specification setting and data reporting.  

Impurity and Degradation Products 

The acceptance criteria should follow ICH Q3B for specified and unspecified impurities.  Due to the nature of combination products which contain a small amount of total drug, a non-standard delivery mechanism, and low amounts of impurities in the devices, the acceptance criteria for total impurities and some individual impurities can be proposed at relatively higher amounts compared to those typically in pharmaceutical products.  It can be a challenge to justify acceptance criterion and obtain approval from various agencies. Toxicology data or literature data may be useful for justification. 

In vitro drug release (Dissolution or Elution)

The acceptance criteria setting may reference USP<711> (14) and the L1, L2 and even L3 approach may be used with only the failure at the last L3 stage of testing considered to be an Out of Specification (OOS) result.  Due to the complexity of the device matrix and drug extraction or elution process and low drug content, a great deal of variability is often observed, which brings great challenges for specification setting.  Thus, a wider ranges of acceptance criteria are likely to be proposed and accepted for devices in comparison to pharmaceutical products. However, the specifications should be set in a way to ensure consistent performance from lot-lot, not allow the release of any lots with elution profiles outside those that were tested clinically, and detect any changes that may occur during the storage and that are likely to influence the quality, safety and/or efficacy of the combination product.  The plateau or 80% of drug release in general can be considered the acceptance criterion for the final time point.  It would be very helpful for specification setting and post-approval changes if IVIVC (in Vitro-in Vivo Correlations (IVIVC)) are developed.


The criteria for particulate counts may follow USP <788> (10), such that the average number of particles present in the units tested does not exceed 6000 per device for particle sizes ≥ 10 μm and does not exceed 600 per device for particles ≥ 25 μm.  However, agencies may require performance based specifications depending on the application and implantation site of devices.  Due to the large variability observed in testing for devices, a multi stage approach, such as L1, L2 may be used with agency approval.


The endotoxins limit for a medical device is dependent on the intended use of the device and what the device contacts (e.g., blood, the cardiovascular system, cerebrospinal fluid, intrathecal routes of administration, permanently implanted devices, and devices implanted subcutaneously).  For most medical devices, the acceptance criteria may follow USP <161> which limits endotoxins to less than 20 USP Endotoxin Units per device (15). 

Packaging integrity

The acceptance criteria of packaging integrity tests are dependent on the specific test performed and may directly apply technical guidelines, such as from the ASTMs.  For example, the acceptance criteria could be worded as “no bubble stream is detected upon inflation of packaging to a known failure point” for a leak test.  Quantitative criteria could also be used, for example a peel test may require a peel strength of at least 1.0 lb from each side of the packaging.

Molecular weight 

Since most polymers used in combination devices are not considered as active pharmaceutical ingredients (excluding items such as pegylated-proteins) there is no general guidance available and the criteria need to be created from internally generated data with statistical sound approaches and rational. The criteria can be a limit (e.g. not less than x Dalton) or a range depending on the type of polymer and its application.  The test may not be included in stability testing if justifications based on the stability data are proposed and approved by agencies.  For some cases, specifications may not be strictly required and data could be collected on a for information basis only.

Antioxidant or Drug Stabilizer

Since there is no compendial specification for reference, the criteria need to be developed internally.  The criterial may be limits or ranges.  Specifications may be established for information only or removed from stability testing if justified and approved by agencies.

4. How to Implement Changes of Test Method and Specification during Stability Testing

During stability testing, a new method could be developed or the current method could be improved with better accuracy and or lower variability.  This is a part of the natural development process as the product becomes more mature and the knowledge of the devices and the variability from products and testing becomes better understood. There is also likely to be more feedback from agencies regarding specifications for some attributes that may need to be adjusted to be better reflect the ranges of test parameters or fulfill agency requests. As test method and specifications need to be consistent, the stability data will need to be compared to the initial data and methodology and be evaluated with respect to specifications and the changes of test method. 

For test method changes, the newly developed or revised methods should be fully validated and an equivalency study between the new or revised method and current methods should be conducted.   The data from the study should demonstrate equal or better accuracy and equal or lower variability.  A bridging study may be used to collect stability data from new or revised and current methods prior to using the new method in stability testing.  Stability protocols need to be amended to add new or revised methods with justification and stability reports will need to describe when the method changes occurred and also justify any impacts. As needed, the new or revised methods may need to be submitted to agencies for approval prior to use.  The new method will need to be documented in the internal quality system.

For specification changes, all stability data should be fully evaluated against the new specification(s) prior to submitting to agencies as the change may cause failures or questions within ongoing stability studies. The new specification should be fully justified and appropriately documented and should be submitted to agencies for approval prior to implementation. 

5. Summary

Combination products require special considerations for selection of test methods/ parameters and specification setting in stability testing.  Appropriate method and parameter selection and specification setting in stability testing are critical for determining product shelf life and allowable shipping conditions.  Thorough evaluation and appropriate process controls should be in place when test methods and specifications are changed and the updated methods and specification can be implemented during stability testing.  Finally, stability personnel should work in a cross functional manner with those testing the mechanical aspects of the device to determine any potential impacts of physical product aging on chemical performance that may not be readily determined by chemical analytical methods alone.



  1. FDA Final Guidance for Industry, 2017 Current Good Manufacturing Practice Requirements for Combination Products
  2. FDA draft Guidance for Industry, 2008, Coronary Drug Eluting Stents-Nonclinical and Clinical Studies
  3. J. Liu, T. Chiesl, F.W. Tang and M. Kamberi, “Shelf Life Determination for Combination Medical Device Products”, Pharmaceutical Outsourcing, 32-33, July/August 2018
  4. ICH Q1A Stability Testing for New Drug Substances and Drug Products
  5. FDA Guidance for Industry, 2015, Analytical Procedures and Methods Validation for Drug and Biologics 
  6. ICH Q2B Method Validation for New Drug Products
  7. USP 41-NF 36 General Chapter <905>, “Uniformity of Dosage Units”, page 6673
  8. ICH Q3B Impurity Testing for New Drug Products
  9. FDA Guidance for Industry, 1996, Dissolution Testing of Immediate Release Solid Oral Dosage Forms
  10. USP 41-NF 36 General Chapter <788>, “Particulate Matters in Injections”, page 6537
  11. FDA Guidance for Industry, 2008, Container and Closure System Integrity Testing In lieu of Sterility Testing as a Component of the Stability Protocol for Sterile Products
  12. USP 41-NF 36 General Chapter <85>, “Bacteria Endotoxin Test”, page 6012
  13. ICH Q6A, 1999: Specifications: Test Procedures and Acceptance Criteria for New Drug Substance and New Drug Products 
  14. USP 41-NF 36 General Chapter <711>, “Dissolution”, page 6459
  15. USP 41-NF 36 General Chapter <161>, “Transfusion and Infusion Assemblies and Similar Medical Devices”, page 6085

Product Added Successfully

This product has been added to your account and you can access it from your dashboard. As a member, you are entitled to a total of 0 products.

Do you want access to more of our products? Upgrade your membership now!

Your Product count is over the limit

Do you want access to more of our products? Upgrade your membership now!

Product added to cart successfully.

You can continue shopping or proceed to checkout.

Comments (0)

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
  • Use to create page breaks.
Enter the characters shown in the image.
Validation Master Plan Download banner