Justine Bentley

We are a pharmaceutical tablet manufacturer that is increasingly concerned by counterfeit drugs. We have received tablets that we suspect are counterfeit, but examination of the bulk composition has shown no significant differences. Could surface analysis techniques be utilized to determine the difference between genuine products and copies? 

The answer is yes if the route of manufacture is different. However, before going into specifics we should define counterfeit drugs, explain the background and why they are such a large problem. 

Counterfeiting was first recognized by the World Health Organization in 1985 and the agency estimates that the sales of fraudulent drugs will reach $75 billion (€47.9 billion) by 2010.

				 Currently, counterfeit pharmaceuticals are believed to account for 10% of the global supply chain and this percentage rises to 70% in some developing countries. Even in the US, a country renowned for having some of the most stringent regulations for the pharmaceutical industry, the sale of counterfeit drugs was estimated at $32 billion (€20.4 billion) in 2006. 

While the direct effect on pharmaceutical company profits is important, the most concerning aspect is the health risks. In the best case scenario the counterfeit drug performs as the genuine article with no side-effects, in the worst case scenario the counterfeit drug results in death. 

Currently, counterfeit drugs can be split into several categories: 					

				 products without active ingredients 

 products with incorrect quantities of active ingredients 

 products with the correct quantities of active ingredients but with fake packaging 

 products with high levels of impurities and contaminants 

Counterfeit drugs with the wrong chemical composition or contaminants can be investigated by a range of techniques such as nuclear magnetic resonance, near infrared spectrometry, thin layer chromatography, desorption electrostatic ionization, Raman spectroscopy, X-ray diffraction and liquid chromatography–mass spectrometry. However, those that contain the correct ingredients in the correct amounts (i.e., copies of the original) are more difficult to identify while potentially being as deadly as those with the incorrect ingredients.      

Regarding the initial question, we need to understand why drugs with the correct ingredients in the correct amounts are a problem. 

At first, these drugs may not appear to have the same health concerns as other types of counterfeit drugs, but it is not just the profits of pharmaceutical companies that suffer because of the practice of copying originals. Just because a drug contains the same ingredients as a genuine drug doesn't mean it will react in the same way or, more precisely, have the same effects on the person taking it. Drugs are specially designed to release APIs at set times if the composition is correct, but if the distribution of the counterfeit is different then it will produce undesired effects. 

In relation to the question, we will focus solely on solid dosage forms — tablets. 

There are two main routes of manufacture for pharmaceutical tablets: 					

 Granulation — a process by which particles within a formulation are enlarged to facilitate the flow of properties in a particular drug. This can be done by using either a wet or dry granulation technique that uses mechanical pressure to create the desired effect. 

 Direct compression — a process that does not require agglomeration of particles, instead active ingredients and excipients are pressed together in a specific ratio to give the correct dosage. 

The main factors to be considered when determining a tablet manufacturing route are: 					

 necessity for controlled release of drug 

 handling conditions during manufacture and intake 

 degradation of raw materials relating to time of manufacture 

Counterfeiters usually use the cheapest and easiest route to manufacture drugs rather than the most effective and accurate one. Determining the route that has been used can prove which is a genuine drug and which is a counterfeit. This is where surface analysis can help. 

Figure 1 ToFSIMS spectra of a pharmaceutical tablet.			

So how do we distinguish the route of manufacture when the bulk composition is the same? The distribution of lubricant can be different depending on the route of manufacture. In granulation, the lubricant is added to granules prior to compression and coats the surface of the granule, whereas in direct compression the lubricant is more uniformly distributed. This difference enables surface analysis to determine the route of manufacture. 

 Tablets with identical ingredients were formed using granulation and direct compression. A series of fracture surfaces were exposed for analysis as the surfaces of many pharmaceutical tablets are treated, uniform and nonrepresentative of the bulk distribution. A scalpel was used in a controlled atmosphere to 'crack' the tablets, exposing a fresh noncontaminated bulk surface that was immediately analysed. 

Step 2: identify the species present at the fracture surface.

 Spectral Time-of-Flight Secondary Ion Mass Spectrometry (ToFSIMS) was used to identify the diluent (i.e., is the tablet a calcium phosphate or cellulosic-based formulation), the API and the lubricant at the fracture surface for each tablet. 

Figure 1 shows part of a general ToFSIMS spectrum of a cellulose-based diluent. 

Step 3: quantify the species present at the fracture surface. 

Standard X-ray photoelectron spectroscopy was used to quantify the elements, as well as their oxidation states (i.e., the percentage of carbon as C-C bonds and C-O bonds), present at the fracture surface. 

Step 4: determine the route of manufacture.

 Once the information from steps 1–3 is known, the route of manufacture can be determined. It is not as straightforward as comparing the composition, which we already know is the same. This method uses a more complex equation that takes into account the composition of the diluent, as well as the quantity of the lubricant at the fracture surface. 

 If further confirmation is necessary, ToFSIMS can be used. ToFSIMS imaging provides colour-coded molecular maps of the fracture surface. As previously mentioned, the lubricant is distributed differently depending on the route of manufacture. By mapping the lubricant at the surface, differences between genuine and counterfeit drugs can be identified. 

Figure 2 shows a ToFSIMS image of a generic pharmaceutical tablet where the outer case is coloured blue, the inner matrix is shown in green with the central core containing the active pharmaceutical ingredient in red. 

Figure 2 ToFSIMS image of a cross section of a pharmaceutical tablet.			

In conclusion, whilst surface analysis techniques can not directly identify a counterfeit tablet they can investigate the route of manufacture, which can lead to the identification of a counterfeit tablet when compared with a genuine product.  

1 www.who.int/mediacentre/news/releases/2006/pr09/en/ 

Articles

As counterfeiters become more cunning and technologically advanced, spotting their handiwork is increasingly difficult. Can surface analysis techniques be used to outwit them?

The counterfeit detective

Our oral pharmaceutical product is packaged by a range of contract packaging companies. Adverse properties of the final packaged product from one particular company indicate that the product shelf life is shorter than it should be. It has been suggested that this could be related to gas filling issues. Is this anything that surface analysis could clarify?

This could certainly be the case — gas contamination and compositional errors are an increasingly common issue, particularly in the pharmaceutical industry where gas-filled packaging is frequently used. However, 'could' is no good when your product and, ultimately, your profits are suffering. Clarification is required. 

The particular value of surface analysis in this scenario is in identifying the correct course of further investigation to avoid costly and unnecessary testing procedures.  

Pharmaceutical packaging can be for powder, liquids, tablets, capsules, creams and gels, and is generally more sophisticated than packaging for other industry sectors. 

High-barrier packaging, such as gas-filled packaging,

maintains the atmosphere within the packaging, but this atmosphere must be modified to provide a long product shelf life. This is achieved in one of two ways: vacuum packing or gas flush packaging. It is the second method that concerns us.  

Gas flush packaging reduces the amount of oxygen surrounding the drug to slow or eliminate the growth of aerobic life forms, which increase the rate of oxidation reactions. Often, the displaced oxygen is replaced with nitrogen, carbon dioxide or sometimes argon. The gas composition of an average blister-packed pharmaceutical product, for example, would contain less than 1% oxygen compared with the 21% typically found in air. 

Pharmaceutical products that show signs of oxygen or moisture attack often indicate atmospheric issues in the packaging itself. Unfortunately, spotting these symptoms is not always straightforward, and without definitively identifying the root cause as being atmospherically related, the resolution remains unclear.  

This is where surface analysis techniques can be invaluable. While some harmful effects caused by atmospheric issues can be observed visually, such as discolouration, most are much less obvious and can only be confirmed using extremely sensitive analytical equipment. As an example, one of the most common side-effects of atmospheric interference is a change in surface morphology. Differences in the surface contact area can interfere with engineered dissolution rates and affect the speed of absorption 

. Topographical changes at the surface of the pharmaceutical, or simply a change in roughness, can easily be demonstrated using noncontact surface profiling (3DP), where changes at the submicron level can be monitored.  

Sometimes chemical changes at the drug surface occur because an unprotected pharmaceutical product undergoes surface chemistry oxidation or hydration. Once again, surface analysis techniques are the ideal identification tools; for example, x-ray photoelectron spectroscopy (XPS) offers quantified data on elemental and oxidation state changes within the top 5–8 nm of a surface. The technique not only enables the overall level of oxygen (as combined in oxygen-containing groups) at the surface to be measured, but can also differentiate between the form of the oxygen — such as whether it is present as an acid, carbonyl, ester or alcohol. Figure 1 shows XPS technology in action. 

Whereas XPS provides quantified data, time-of-flight secondary ion mass spectrometry (ToFSIMS) provides detailed molecular information from the outer 1–2 nm, enabling unequivocal fingerprinting of drug molecular ions, excipients and any modified contaminants or modified species. One demonstrable example of the technique in practice is 'blooming'. Oxygen attack, caused by atmospheric issues in packaging, often results in 'blooms' occurring at the drug surface. By taking repeated measurements from the drug surface to the sample subsurface, ToFSIMS can differentiate between blooming caused by oxidation and a straightforward contamination of the external surface. 

Additionally, scanning electron microscopy can be used to analyse the physical form of the 'bloom' crystals to identify how they formed. In the case of atmospheric problems, material at the drug surface is likely to have been dissolved and then slowly recrystalized because of water vapour ingress. 

Using the techniques outlined previously provides a clear picture of the form of attack the drug surface has suffered. The amount of tests available to troubleshoot potential product lifespan problems is immense, but with our data a likely candidate can be identified and the correct form of further analysis can be selected. In the case of atmospheric related symptoms, this further analysis usually takes the form of residual gas analysis (RGA). This technique has recently been refined for pharmaceutical packaging gas analysis (PPGA) to allow intact pouches or blister packs to be studied. Similarly, modified RGA analysis can also be used to study entrained bubbles of contaminated gas to identify a likely contamination source.  

Taking its routes from secondary ion mass spectrometry, PPGA enables the residual gas from sealed packaging or glass bubbles to be leaked in a controlled manner into the residual gas analyser (Figure 2). 

Figure 2: Detail of fracture chamber and pumping arrangement.

 fracturing of samples is based on a modified ultra high vacuum (UHV)

valve assembly where the central sealing spindle is replaced by a diamond-tipped shaft. This shaft is mechanically driven by rotation of the valve assembly, bringing the diamond tip into contact with the surface of the sample (Figure 3). Increasing the pressure on the tip causes rupturing of the sample with resultant release of the internal atmosphere into the evacuated chamber. An UHV leak valve is then used to bleed the gas into the main instrument chamber for mass spectral gas analysis. 

Figure 3: Glass bubble trial sample mounted for fracturing. 

Returning to the specific query outlined at the beginning, the benefits of combining analytical techniques are two fold. By using the initial surface analysis techniques to identify the specific symptoms demonstrated by the drug, the field of potential causes is significantly narrowed.  

In turn, appropriate further action can be taken more quickly and with less risk of wasted effort. From a business perspective, this means that lifespan issues are likely to be resolved faster, without the expenses of unnecessary extensive testing. 

Figure 4: Mass spectrum from gas-filled packaging. 

In effect, surface analysis provides the key to unlocking the correct door to a solution, and can make a big difference to profit lines and continued production. 

 joined CSMA in 2005 as a Technical Sales Consultant. She promotes the features and benefits of surface analysis techniques to a range of sectors and disciplines, providing valuable advice on troubleshooting and on delivering analytical solutions for commercial applications, including successful product development and reverse engineering. 

If you have any questions for the experts at Surface Surgery then please send an email to 

Gas-filled packaging problems

Our company is involved in developing and manufacturing APIs that can be utilized with drug-eluting stents (DES). Despite ensuring constancy in pharmaceutical composition, we are experiencing issues with variations in drug release during 

 studies. We are working closely with a stent manufacturer to develop the system, but could surface analysis techniques investigate the problem further? 

In short the answer is yes — particularly as the composition constancy is closely monitored. However, before going into the types of analysis that may shed light on your problem, it is useful to understand more about the development of DES. 

Stents are metal mesh tubes that have been used since the 1990s. They are inserted post-angioplasty to treat blocked coronary arteries. Initially, they were simply bare metal tubes that had been passivated and deemed fit to be implanted. At first, these metal scaffolds provided positive results with arteries remaining open after insertion, but, with time, it became apparent there was an issue with artery reclosure (restenosis). This issue led to the development of drug-coated stents that release specially designed drugs targeted at preventing restenosis. Figure 1 shows a stent approximately 3 cm in length prior to expansion.  

The coating used for drug-eluting stents facilitates the controlled release of pharmaceuticals to ensure a specific dosage is released during a set period of time. 

Controlling drug release in this way delivers drug therapy to the point at which it is most required, improving efficacy and avoiding the issue of 'human error' when it comes to self-medication. For such benefits to apply, DES coatings must release drugs 

 according to a predictable, therapeutically rational programmed rate.  

The release processes used to achieve this 'rational programmed rate' are varied; to name but a few:  

diffusion (from matrix and membrane systems)

Increasingly important applications of controlled release technology also include patterned, targeted, triggered and closed-loop delivery. The question is: what can affect these systems and, subsequently, cause release variation problems? The following have all proven to be common culprits:  

variation in coating composition (i.e. drug distribution)

It is essential to characterize and monitor each of these issues during initial studies to ensure the drug elution profile is correct before

 studies begin. This is where surface and interface analysis techniques can provide valuable analytical data. The most appropriate techniques are dynamic secondary ion mass spectrometry (DSIMS) and 3D non-contact surface profiling (3DP). 

DSIMS is suitable for this particular issue because the depth profiling mode can closely monitor composition with depth. As well as identifying foreign contaminants that may be present at the very surface of a drug-eluting stent, this type of profiling also analyses the chemical composition of each layer (down to the metal substrate itself) once the stent is treated with its specific coating. 

3DP complements data obtained using DSIMS and provides important information on surface parameters. These data can prove particularly useful in identifying drug release issues caused by base material roughness or coating thickness. 

Common surface parameters used to analyse the surface roughness/coating thickness of drug-eluting stents are:  

: arithmetic mean of the deviations from the mean.  This parameter represents the mean surface roughness, measured over the whole analysed area.

: height of the 10 points of the surface.  Mean of distance between the five highest peaks and five deepest valleys. A matrix of 3×3 pixels is taken into account to identify the peaks and valleys. 

: peak to peak. This parameter is a measure of the height difference between the highest pixel and the lowest pixel. 

To illustrate the way these techniques and calculations are used more effectively, it is beneficial to look at a sample analysis process. 

The first port of call in analytical procedures designed to discover potential surface issues affecting the controlled release of pharmaceuticals from drug-eluting stents, is to examine an uncoated metal stent that has been treated only with an oxide layer. Figure 2 shows the results of this initial step using DSIMS depth profiling. Depicted in the high-depth resolution DSIMS profile are the outer 15 nm of a bare metal stent. While this particular profile reveals that the oxidization layer used to treat the bare metal stent (prior to coating application) is uniform and measures approximately 2 nm in thickness, it also reveals a distinct carbon presence (black line marked 'C').  

Figure 2 Results using DSIMS depth profiling. 

This reading levels out, indicating that carbon is used in the composition of the metal stent itself. However, the initial peaked reading suggests that carbon is also present at the oxide/metal interface. This could represent a contamination — one that could have an undesirable effect on subsequent pharmaceutical treatments and result in release inconsistencies.  

Figure 3 shows a 2D image of an uncoated metal stent. The imaging technology can identify issues in surface topography, which may cause drug/coating distribution problems. Using data taken from various points along the sample, a mean surface roughness can be calculated. These data are particularly useful when the analysis moves on to stage two: investigating the coated stent.  

Again, the depth profiling of DSIMS provides valuable data relating to the now multilayered stent surface — from the outer nanometres of the topcoat through to the bulk of the metal substrate.  

Figure 4 shows a DSIMS depth profile of a drug-coated stent. Represented in the diagram are the polymer matrix (12C 21H), the drug (12C 14N) and the metal substrate (56Fe 16O

. The profile reveals several key pieces of information when assessing potential causes of drug release issues. First, coating layer thickness can be determined by examining the concentration differences at different sample depths. This is particularly important information as erratic drug release is often caused by inconsistencies in coating thickness — thicker patches result in a slower release rate while thinner ones release the product too soon. Alternatively, patches of thinner coating could also indicate a reduced level of drug presence all together. Either way, using surface analysis of this nature quickly identifies such issues.  

Returning to the information depicted in Figure 4, we see that a high concentration of drug is not present until a depth of approximately 3μm — a feature you would expect in any time-delay release mechanism. A high concentration of the drug nearer the surface may indicate an issue with drug distribution, while an unusually low drug concentration at a lower depth may reflect a problem with undesirable drug migration during storage.  

No foreign contaminants are shown in this profile, but organic and inorganic contaminants (i.e., from manual handling or cleaning solvents) would also be easily identified by the depth at which they occur, giving a strong indication of where they may have been introduced in the DES or pharmaceutical coating manufacturing process.

Figure 3 3DP 2D image of an uncoated metal stent with a colour-coded thickness map.  

Once again, data produced using the DSIMS depth profiling is complemented by 3DP analysis. This technique is particularly useful in analysing issues relating to drug release timings as it is the only nondestructive method of obtaining a detailed breakdown of layer thicknesses and corresponding coating distribution. 

Figure 5 shows a 3DP 2D image of a drug-coated stent with a colour-coded thickness map. As polymer coatings are transparent to light, the 3DP technique can be switched from analysing surface topography to providing coating thickness values. It is possible to visually identify coating thickness and areas of uneven depth; the shallower areas perhaps indicating patches of poor drug concentration. The line shown in the diagram depicts where a line scan was taken across the stent to obtain data to calculate coating thickness.  

Figure 4 DSIMS depth profile of a drug-coated stent. 

Similarly, mean surface roughness can be calculated (by repeating measurements at several points along the stent) and contrasted with that of a stent prior to coating with the drug/polymer matrix. If the mean surface roughness for the coated stent is higher, it is likely to indicate a coating issue and potential inequality in drug dispersal.  

Figure 5 3DP 2-dimensional image of a drug-coated stent with a colour-coded thickness map. 

Our company is involved in developing and manufacturing APIs that can be utilized with drug-eluting stents (DES). Despite ensuring constancy in pharmaceutical composition, we are experiencing issues with variations in drug release during in vitro studies. We are working closely with a stent manufacturer to develop the system, but could surface analysis techniques investigate the problem further?