Five critical mistakes when using Biomagnetic Separation in CLIA IVD-kits manufacturing. Learn their cause and how can we avoid them

Transcription

1 Five critical mistakes when using Biomagnetic Separation in CLIA IVD-kits manufacturing Learn their cause and how can we avoid them A B C D E F G H

2 2 Five critical mistakes when using Biomagnetic Separation in CLIA IVD-kit manufacturing Learn their causes and how to avoid them Using a high throughput method such as chemiluminescence is very advantageous in settings such as hospital labs that need to analyze, for example, thousands of patient samples a day. If you want to look at 5-10 analyses per patient, you will be performing tens of thousands of tests daily. This becomes an impossible task using ELISAs or other older techniques, but with high throughput techniques such as automated chemiluminescent immunoassays (CLIAs), this task becomes quite doable. Dr. Fabrice Sultan The basic guide for using magnetic beads in chemiluminescent immunoassays (CLIA) Chemiluminescence immunoassay (CLIA) is a sensitive, high throughput, economical alternative to conventional colorimetric methodologies such as enzyme-linked immunosorbent assays (ELISA). As a result of their success, the number of IVD-tests based on this technique has exploded. What s more, whenever a CLIA analyzer is installed in a Hospital or Diagnostic laboratory, demand increases by tens of thousands of kits a day. As a consequence, the volume of production of CLIA- IVD kits has sky-rocketed and the small volume, manual handling and quality control methods used during development and early production stages have become inadequate. Producing more kits to cope with demand means ensuring consistency between batches and raises new concerns regarding in-batch consistency due to the different scales manufactured (liters) and final application (less than a few hundred of microliters in the analyzer).

3 3 SEPMAG has been working with leading IVD companies Production Managers, R&D scientists, Lab Technicians and Quality Assurance personnel to guarantee high performance, well parameterized biomagnetic separation processes since Over the last ten years, during which we have been developing, manufacturing and marketing advanced biomagnetic systems, we have identified five recurrent, critical mistakes. They are mistakes that delay customer projects, cause significant economic losses and in some cases, even put the production in risk. Appropriate technical information and training in the Biomagnetic Separation processes is a critical factor for a good outcome. The better the potential users understand how to avoid mistakes, the easier it will be for them to benefit from this technology. This book starts by explaining the basics of biomagnetic separation processes before using these concepts as the basis of a detailed explanation of the five critical mistakes we have identified and how to avoid them. For more resources on the subject, see our website ( and our technical Blog ( Figure 1. Advanced Biomagnetic Separation System

4 4 How to specify a Biomagnetic Separation Process? When a new CLIA-IVD kit is transferred from R&D to production, all manufacturing protocols need to be adapted to the new throughput and volume. Biomarker specifications, buffers and coating protocols all benefit from the cumulative experience in nonmagnetic kits. Coupling an antibody to magnetic beads is quite similar to either colloidal gold or latex particles, but washing protocols using biomagnetic separation is new. The use of classical (and dirty) centrifugation method makes little sense given the magnetic properties of beads. This also applies to the use of lateral flow filtration and other complex and time-consuming non-magnetic separation techniques. Although using biomagnetic separation seems the obvious choice, there are some problems in practice. The first is specifying the process itself. When we talk to IVD-kit manufacturers about their biomagnetic separation needs, they usually mention: Separation time: the time it takes to separate the solid phase from the buffer. Magnetic beads losses: the maximum number of beads (and the coupled biomarkers) lost during the process. Batch volume: the required batch size and, sometimes, process scaling compatibility issues (development, batch validation, full production). The need to avoid irreversible aggregation: irreversible bead aggregation during the separation requires a numerous amount of resources for resuspension and to check that everything has been done correctly. Because each kit (ml) must have the same characteristics, incorrect bead re-suspension increases variability. However, these are all functional parameters. They are consequences of the biomagnetic separation rather than the process itself. What it is missing here, is there is no mention of the key parameter that defines a biomagnetic separation process. So, what defines the behavior of the magnetic beads during the process? The answer to this question is quite simple, but it is usually overlooked. The key parameter in biomagnetic separation processes is THE MAGNETIC FORCE. Magnetic beads move at a specific speed, which is determined by the net force resulting from competition between the magnetic force and the drag force. The latter is caused by buffer viscosity.

5 5 It is IMPORTANT to remember that we are talking about the magnetic force, not the magnetic field. Remember this basic concept: the application of a perfectly spatially-uniform magnetic field to a magnetic bead does not generate a magnetic force. It merely generates torque, which aligns the magnetic moment of the bead with the field applied to it. To generate a magnetic force (and consequently the movement of the magnetic beads), a Magnetic Field that is not uniform in space must be applied. A simple permanent magnet separates magnetic beads in a vial because the magnetic field applied is not uniform. Therefore, there is no point in trying to define the process by the value of the magnetic field (usually expressed in Tesla or Gauss), because what is important is how it changes with distance. Figure 2. Spatial uniform magnetic field generates torque (left). A spatiallynon uniform magnetic field is necessary to generate a magnetic force (right). As shown by the formula expressing the value of the force, this depends on the change of the scalar product of two factors: the magnetic moment of the beads and the magnetic field. The spatial variation in the magnetic field is defined by our magnetic separation rack, which is fixed if permanent magnets are used as the field source. The magnetic moment of the beads changes, depending on the field applied. If the beads are superparamagnetic, the magnetic moment varies linearly with the magnetic field when this is low (the materials have constant magnetic susceptibility). At high values of the applied magnetic field, the magnetic moment of the beads becomes saturated and is approximately constant. In the linear magnetic response region of the bead, it is difficult to have a magnetic field profile leading to a well-defined homogenous force because a constant gradient of the square of the applied field is required. However, when magnetic beads are saturated, having a constant magnetic field gradient leads to a well-defined homogenous magnetic force.

6 6 When both conditions are met - magnetic saturation of the field and constant magnetic field gradient, the MAGNETIC FORCE governing the Biomagnetic Separation processes is defined. Magnetic saturation With these concepts in mind, it is easy to understand how to avoid the most common, critical mistakes when using this technology in CLIA-IVD kit production. Figure 3. Magnetic force expression for the different regions of the superparamagnetic bead magnetization curve. Mistake #1 Blaming the magnetic beads Product development is a time-consuming, expensive process for CLIA-IVD kit manufacturers. There are several steps involved: Selecting the biomarker Choosing the right coupling Selecting the right magnetic bead Low kit-to-kit variability. Batch aliquots (typically less than a milliliter) of production batches (liters scale) must be consistent. If not, variability causes problems when interpreting the results in the analyzer. You are well versed with the first two points but what is the right bead? Assuming you have the right biomarker and a perfect coupling, the ideal magnetic bead should have the following properties: High recovery/fast separation, compatible with the timing of the analyzer step. It needs to be fast enough during large-scale production processes without high bead and coupled biomarker losses. No aggregation problems. Beads should be easy to re-suspend. It makes no sense to separate quickly if several additional sonication steps are required, which are difficult processes to control/implement in large volumes. Figure 4. CLIA IVD kit test

7 7 What happens when these requirements are not met? The most common response to defective magnetic separation is that the magnetic beads is to blame. Users then contact the magnetic bead supplier (or alternative supplier) for protracted discussions about the product, revision of the coating protocols a lengthy process that consumes significant resources and generally delays the product development and/or launch. If a biomarker couples well with the magnetic bead, changing the magnetic beads is extremely expensive, delays the project and in most cases DOES NOT solve the problem. So what s missing from the equation? Remember that magnetic force depends, not only on the magnetic moment of the beads, but also on the magnetic field profile of the separation device. How does the device affect biomagnetic separation? Let s look at an example. The diagram shows how the same suspension of magnetic beads behaves in two different biomagnetic separation systems (link to video in Youtube?). The magnetic rack on the left (blue), separates slowly. This means waiting for a long time to complete the separation or accepting high losses of beads and biomarkers. However, if we wait for a long time in order to avoid losses, some beads are affected by irreversible aggregation problems as the force over the earlier separated beads (the closest to the bottom part) are subjected to an extremely high local force for protracted periods. Figure 5. Separation of the same magnetic bead-suspension in two different magnetic racks

8 8 If the effect of the separation rack is ignored and the only results considered are those of magnetic bead performance, we will reach the wrong conclusion that alternative beads are required. However, the behavior of exactly the same suspension in an Advanced Biomagnetic Separation System (right, orange) paints a very different picture of the situation. The beads separate quickly, all moving at approximately the same speed. Irreversible aggregation problems disappear, because local retention force and exposure times are much shorter. As stated in the previous chapter, you need to look at the biomagnetic separation process as a whole, remembering that the force depends on both the magnetic beads and the separation device. This will mean you avoid making Mistake #1 (Blaming the magnetic beads). What is our recommendation for CLIA-IVD kits manufacturers? We suggest that rather than using different magnetic beads, you will find it cheaper and quicker to check whether the problem lies with the magnetic separation rack. If your suspension works well with a different Biomagnetic Separation Systems, you will see that the problem is not the magnetic beads and that there is no need to rebuild the coating protocol. Mistake #2 Using bigger magnets to avoid losses In section one we discussed why blaming the magnetic beads for faulty biomagnetic separation can be a big mistake. The main reason is that the problem may be the magnetic rack rather than the beads. Being aware of the significance of the biomagnetic separation system is not enough. Another common mistake is to assume that a bigger magnet will increase the magnetic force. Magnetic force, however, does not depend on the magnetic field but on the magnetic field gradient. If the prism shape is too big with respect to the working volume, the magnetic field becomes homogenous and the force is greatly weakened, the result of which is: excessively long separation times and/or higher losses. If the prism is small, both the magnetic field and the magnetic field gradient decay very quickly with distance, and the force is unable to attract the beads which are farthest away. Even worse, the force close to the magnet is too high, causing irreversible aggregation. Figure 6. Permanent magnet prism (left) and advanced biomagnetic system (right) for separation up to 1 liter of magnetic bead suspension. Both devices use the same number of permanent magnets.

9 9 This is best illustrated with an example. Let s compare what happens with a large magnet and with an advanced biomagnetic separation system, both using the same amount of permanent magnet. To compare these systems we need to look at the magnetic force. An easy way to do this is by looking at the magnetic field gradient maps 1. The simple permanent magnet generates a very low gradient just a few millimeters away from the magnet while the force farther away from the magnet is very low. This means high bead and biomarker losses. Even with very long separation times, complete separation of the solid phase is almost impossible. The optimal magnetic force is the value that simultaneously offers both quick separation and low losses; a retention force high enough to retain the beads when the supernatant is extracted, which is also mild enough to avoid irreversible aggregation. In this example, the cyan area represents the optimal value 2 for the magnetic force. The region where the suspension works under the optimal conditions is just a tiny fraction of the batch volume. Nearer the prism, the force increases quickly and, consequently, there is a very high risk of irreversible bead aggregation. This risk is exacerbated by the long separation time required by this kind of magnetic separation rack. Figure 7. Magnetic Force maps for a Permanent Magnet prism (left) and an Advanced Biomagnetic System (right). 1 Using only the magnetic field gradient overestimates the magnetic force in the low field region (where the beads magnetic response is linear). For the purposes of this chapter, this approach overestimates the values for the prism magnet, as most of the working volume is in the lower field region. 2 As the magnetic force is the product of the magnetic moment of the beads and the magnetic field gradient, the exact optimal value depends on the magnetic bead. The example uses the optimal value for most of the commercial magnetic beads.

10 10 In contrast, advanced biomagnetic separation systems use magnetic field profiles by generating homogenous magnetic force. This means the force is the same throughout the working volume. A higher force is generated over the beads farther away from the vessel walls. The magnetic beads move faster, reducing separation times and guaranteeing complete recovery of the solid phase and coupled biomarkers. In the retention area, the force is still the same. If the system is designed to obtain optimal values, the beads experience enough force to be retained during buffer extraction, but this force is still low enough so as not to generate irreversible processes. Feedback from companies confirms these claims. A European company testing a SEP- MAG Q1L reported the following results : 200 nm magnetic beads, suspension at 3%: Recovery of over 99% at the end of the cycle. 300 nm magnetic beads, suspension at 1%: Recovery of over 99.5% after six minutes. 500 nm magnetic beads, suspension at 1%: Recovery of over 99.5% after three minutes, both in water and organic buffer. Figure 8. Testimonials by Resyn Bioscience, a magnetic bead producer, regarding the use of Advanced Biomagnetic separation systems. Magnetic bead manufacturer ReSyn Biosciences was even more enthusiastic and shared its satisfaction with the Biomagnetic Separation process improvements when using advanced systems, even when the magnetic beads were in a viscous liquid suspension on Facebook and Twitter.

11 11 So, how can you avoid Mistake #2 (Use bigger magnets to avoid losses)? Think in terms of magnetic force rather than magnetic field, because there is no sense asking how many Tesla (or Gauss) are generated by a magnetic separation rack. Remember that the force needs to be balanced correctly because excessive force can lead to irreversible aggregation problems. Choose the right force for the farthest areas, as well as the retention area. As a general rule, we recommend you seek external expert advice, because IVD companies do not tend to be experts in magnetic design. Mistake #3 Defining the process based purely on the separation time Not all mistakes made in CLIA-IVD kit manufacturing involve the magnetic rack itself. The third mistake we have detected involves process validation. Biomagnetic separation processes are often validated solely by specifying a separation time. The problem with this approach is that validation is then linked to a specific magnetic field profile and vessel size. What s more, the separation time is merely a consequence of the speed (directly proportional to the magnetic force) and the distance travelled by the beads. The separation time does not describe the working conditions of the magnetic beads but is a consequence of the force they experience and the specific dimensions of the vessel. In short, this is not a good way to validate a process. Additional information is required to characterize the biomagnetic separation process. One option is optical monitoring of changes in opacity (the suspension is dark at the beginning of the process and transparent once the beads are separated). This makes it possible to characterize the process not only the end (separation time), but for the entire time that the vessel is inside the magnetic separation rack. Figure 9. Magnetic bead suspension opacity before (left) and after (right) the biomagnetic separation process

12 12 If the Biomagnetic Separation System has well-defined conditions (i.e. homogenous force), the opacity versus time curve typically has a sigmoidal shape. The two parameters that define this curve are the exponent p and the time t50.the first reflects the steepness of the curve and the second the time it takes to reach 50% of the difference between the maximum and minimal opacity. These two parameters depend on a number of magnetic bead characteristics (diameter, % magnetic content, magnetic material) and the suspension (buffer viscosity, beads concentration). As the shape of the measured curve is affected by all these process parameters, monitoring makes it possible to establish detailed references for the process. When well-defined magnetic force conditions are used, the process should follow the reference curve. When the entire process is monitored rather than separation time alone, Quality Problems can be identified more quickly. Deviations from the reference curves reveal numerous production problems (aggregation, incorrect bead characteristics and incorrect concentration) that can be detected during the Biomagnetic Separation step. This means corrective actions can be taken sooner, thus reducing costs. So, how can we avoid Mistake #3 (Defining the process based purely on the separation time)? It s simple! Just characterize the Biomagnetic Process according to the Magnetic Force (or Magnetic Field Gradient), use homogenous Magnetic Force so all the beads are in the same condition and monitor the process. As well as reducing the kit variability, these steps improve Quality Control of the process and are useful for early detection of production problems. Figure 10. Set-up for optical monitoring of a Biomagnetic Separation process and the typical shape (sigmoidal) of the resultant curve

13 13 Mistake #4 Neglecting process scalability When developing a CLIA IVD-kit, the initial focus is on the biomarker and how to coat the magnetic beads. Biomagnetic separation conditions usually get swept to one side. During these early stages, separation processes are usually developed on a small scale using existing laboratory magnetic racks. Once the kit has been developed, the batch volume is increased and a different magnetic separator is used. If the working conditions are not well defined, the magnetic force over the beads is completely different with the new system, which is when losses and irreversible aggregation problems occur. A costly re-engineering process is then needed to resolve these issues and to keep losses and clump formation to levels low enough to provide a cost-effective and efficient production process. But, what problems are encountered when scaling up inhomogeneous magnetic separation racks? Figure 11. Schematic representation of magnetic force on a small scale (left) and a large scale (right) inhomogeneous magnetic separation racks

14 14 On a small scale, it is easy to have a high magnetic field gradient. Even if all the beads are not magnetically saturated the separation time is short because the distances travelled are also short. But inhomogeneous conditions on a larger scale involves greater losses (lower force at larger distances) and exponentially longer separation times. Over short distances, forces can be higher, increasing the risk of irreversible aggregation. The latter issue is aggravated by longer separation times. In contrast, homogenous Biomagnetic Separation conditions are easy to scale up. When using homogenous gradient the force can be kept constant even at large volumes. As a result, a biomagnetic separation process without magnetic beads losses, and without irreversible aggregation of the beads can be reproduced at different volumes. To do this, beads need to be magnetically saturated for optimal performance (constant force). In advanced biomagnetic systems like SEPMAG systems, the device always maintains the same suboptimal volume at instant t=0 (about 7% of the volume), guaranteeing that the whole batch volume is subject to constant force whatever the scale. Figure 12. Advanced Biomagnetic Separation systems at different scales. The key to avoiding Mistake #4 (Neglecting process scalability) is to correctly validate the magnetic separation conditions early on in the development process, preferably when working on a small scale. Working with homogenous Biomagnetic Separation from the initial small volume, will give you a welldefined process condition, which makes scaling-up straightforward. It also drastically reduces the length of time and the resources needed to move from R&D to production, and to scale up larger volumes when the demand for kits so requires. Figure 13. Use of small tubes in homogenous biomagnetic separation systems. The magnetic force over all beads is the same when a large bottle is placed in the system.

15 15 Mistake #5. Inappropriate safety precautions when working with magnetic fields The first four mistakes we described in this document are related with the production process. However, even if you get a perfect reproducible, high performant process, it is a last mistake you should avoid. We have frequently see IVD-manufacturers to adopt solutions implying high safety risk for the operators and the equipment. You should be aware that magnets can generate a high risk of accident, attracting ferromagnetic objects. These objects can be other magnets, scissors, screwdrivers or any magnetizable object. The risk of having parts of the body are trapped between the two objects increases very quickly with the size of the magnets used. For a fridge magnet the worst scenario is that you pinch your finger. For large Rare Earth magnets, careless users can suffer severe injuries, including multiple bone fractures. For people with pacemakers, the risk is still higher: magnetic fields can interfere with the device, causing malfunctions. That is why you find big danger signs near the MRI areas in hospitals. But not only people can be harm. Laboratory equipment is also be affected by magnetic fields, especially the ones including magnetic recording media. Many unexperienced people has seen how they Credit Cards and/or they company badges (if magnetic) have been erased. Computers and hard disk drivers can also lost their information, and electronics of many lab systems can also be affected. The need of keeping safety distance has a big repercussion on Clean Room occupation. You need to keep clear a DANGER AREA (Field > 3 mt, 30 Gauss) were it is a risk of accidents by the mechanical attraction between magnets and magnetizable objects. You need also clear a larger CAUTION AREA (Field >0.5 mt or 5 Gauss) where it is a risk of erasing magnetic recording media and pacemakers malfunction. These safety measures have a large repercussion in Clean Room occupation. As the figures shows, if the magnetic separation rack has been designed without specific attention to safety, the large stray fields will imply that computer and equipment placed inside the CAUTION area are in danger, then a big space should be cleared and no additional equipment can be placed there. If you need to put several magnetic separation devices in the same room, the problem would still be worse. Figure 14. Repercussion of the CAUTION (grey) and DANGER (orange) areas on the Clean Room occupation.

16 16 By contrast, if the Biomagnetic Separation System has been designed keeping the stray fields small, it would be plenty of free space for placing safely other equipment in the CLEAN ROOM/FLOW BOX. Several Advanced Biomagnetic Separation Systems can be placed in an area smaller than the needed for a single classical magnetic separation rack. Then, how to avoid Mistake #5 (Inappropriate safety precautions when working with magnetic fields)? First of all, you should ALWAYS comply with the safety measures: respect the recommendations for the Caution and Danger areas. Then, in addition of the performance as magnetic separator, you should also pay attention to the Stray Magnetic Fields generated by the Magnetic Separation racks. Whenever possible, you should choose Biomagnetic Separation Systems with small Stray fields. By doing this, you would reduce the risk of accidents, by minimizing the danger area, but also you will saves space on the laboratory/clean room/flow box. How SEPMAG may help? Since 2004, we are helping our customers to avoid these 5 mistakes. Our SEPMAG Systems have homogenous Biomagnetic Separation conditions. Moreover, the magnetic force values are adapted to optimal performance for most commercial magnetic beads. As we use homogenous conditions, process validation and scale-up are straightforward. When needed, special force values (gentler or stronger force), can be provided under request. We may help you to experimentally determine the optimal force value for your magnetic beads suspension. Our QCR hardware and software allows monitor each single biomagnetic separation step, and using the data to ensure the quality of the process. All these benefits, in an extremely safe environment, with no maintenance nor running cost. Do you need additional information? Please, check our Resources for more FREE literature on the subject and check our BLOG for the latest updates on Biomagnetic Separation. If you wish, you may always CONTACT us. About the author Lluís M. Martínez, Chief Scientific Officer at SEPMAG Founder of SEPMAG, Lluis holds a PhD in Magnetic Materials by the UAB. He has conducted research in German and Spanish academic institutions. Having worked in companies in Ireland, USA and Spain, he cumulates more than 20 years of experience, applying magnetic materials and sensors to industrial products and processes. He has filed several international patents on the field and co-authored more than 20 scientific papers, most of them related with the movement of magnetic particles. martinez@

17 www. SEPMAG SYSTEMS Parc Tecnològic del Vallès E Cerdanyola del Valles. Barcelona