Medical Device Link.

PROCESSING TECHNOLOGIES

Figure 1. (click to enlarge) Advantages of lateral-flow immunoassays as a point-of-care or field-use application.

Lateral-flow assays represent a well-established and appropriate technology for a variety of point-of-care (POC) and field-use applications. However, this technology has not been widely applied when very sensitive, highly reproducible, or quantitative results are required. In recent years, interest in this technology has increased in various non-traditional market areas. A combination of factors has driven this renaissance in lateral-flow technologies: patent pressures on existing technologies; other new technologies with better sensitivity, reproducibility, and quantitative or objective result recording; and market demand for appropriate POC or field-use technologies that can be brought to market in a short time frame and for a reasonable investment.

The advantages of lateral-flow immunoassay (LFIA) systems are well known (see Figure 1). Critical among such advantages are that LFIAs represent an appropriate POC and field-use technology that can be brought to market quickly for a small investment, and can be applied to a broad range of applications. Such advantages cannot be claimed by other putative POC technologies currently being developed, including sensor and array-based technologies.

Figure 2. (click to enlarge) Market segments for lateral-flow immunoassays in point-of-care and field-use technologies.

Lateral-flow assays are already being produced or developed in several market segments (see Figure 2). However, manufacturing lateral-flow assays that meet the requirements of these segments continues to be a challenge. As the applications expand, demands on the technologies increase, requiring improvements in sensitivity, reproducibility, and manufacturability. For example, quantitation and objective read/record technologies that can be linked to laboratory information systems are being demanded. In addition to improved materials, assay technologies, reader technologies, and manufacturing processes, there is also a growing need for a more multidisciplinary approach to lateral-flow assay development.

During the past few years, new approaches to manufacturing lateral-flow assays have emerged to meet such demands. Each new approach has resulted in taking a fresh look at the fundamentals of how IVD companies manufacture such technologies.

This first article in a two-part series addresses the issues that remain challenging in developing and manufacturing sensitive, highly reproducible LFIAs, and it outlines the innovations taking place to overcome the limitations of traditional technologies. This article looks at traditional manufacturing technologies and materials, and it discusses not only the limitations of each but also the steps that are being taken to overcome such limitations.

Nitrocellulose Membranes

Figure 3. (click to enlarge) Outline of a lateral-flow manufacturing process.

Figure 3 outlines a general manufacturing process for traditional lateral-flow test strips. The materials and processes used for manufacturing such systems and the ways in which IVD manufacturers use the materials have remained largely unchanged.

Nitrocellulose has been the material of choice in a majority of lateral-flow assay systems. While several attempts have been made to introduce other materials into the market (e.g., nylon and PVDF membranes), such attempts have had limited success. The reasons for this limited success include cost, limited utility, the education needed regarding new chemistry and processing requirements, and inertia from the large bank of existing experience in using nitrocellulose.

Even though nitrocellulose is not an ideal matrix for analytical membranes in LFIAs, it does have certain characteristics that make it useful in this application. Such characteristics are low cost, true capillary flow, high protein binding capacity, relative ease of handling (with Mylar-backed membranes), and available products with varying wicking rates and surfactant contents.

Despite these positive characteristics, nitrocellulose has a number of drawbacks. Such drawbacks include imperfect reproducibility of performance within and between lots, shelf-life issues, flammability and breakage issues (with unbacked membranes), and variable characteristics due to environmental conditions such as relative humidity.

As a result of these issues with nitrocellulose, assay developers and IVD manufacturers have spent a considerable amount of time and effort on optimizing chemistries that overcome the inherent material issues. They have also developed manufacturing processes that guarantee adequate performance during the product’s entire shelf life. Carefully controlling key dispensing, dipping, and drying processes, and paying attention to chemical and biological treatments of the membranes to prevent introducing additional variation into the finished products are critical to success.

Reagent Dispensing for Membranes

For standard LFIAs, at least one test line and one control line are laid down on a nitrocellulose membrane. This process uses a dispenser to lay down a constant volume of fluid per unit length of the membrane. IVD manufacturers use two dispensing methods: contact and noncontact.

In a contact dispensing system, a tip dispenser is dragged along the surface of a membrane, and a pump releases a constant volume of fluid out of the tip and onto the surface. This system is simple and low in cost. However, this process involves contact with the membrane’s surface, which can lead to scoring. In addition, since the process of getting fluid into a membrane relies on not only the contact between the membrane’s surface and the tip, but also the membrane’s absorption characteristics to soak up the fluid into the matrix consistently, the line widths can vary. This variability can be a major issue in reader-analyzed assays, since the reader will generate results that are related to the line widths. This translates into inconsistent test results and an inability to quantify the system completely.

Figure 4. The XYZ 3050 platform with BioJet Quanti 3000 and AirJet Quanti 3000 dispensers by BioDot Inc.

The alternative is a noncontact dispensing system such as the BioJet Quanti 3000 system by BioDot Inc. (Irvine, CA) (see Figure 4). The BioJet uses a microsolenoid valve coupled to a positive-displacement pump to project drops of fluid onto the membrane and overlap the drops at a tight pitch in order to create a continuous line. With the reagents absorbed into the membrane at the point of impact, issues concerning the interface between the surface and the tip dispenser are removed. As a result, the line widths can be more consistent than with a contact tip dispenser. Comparisons can be made between developed line widths that are dispensed using contact and noncontact dispensers (see Figure 5).

Figure 5. (click to enlarge) Comparison of developed line width of protein lines dispensed on membranes using contact and noncontact dispensers.

In-line processing methods can give a greater degree of control, because the dispensing process is linked to camera systems that can examine the quality of the dispensed line and mark out any bad parts (see Figure 6). The in-line dispensing equipment can generate consistent line widths, while the camera systems can assess the line quality in real time (see Figure 7).

Membrane Blocking

Figure 7. (click to enlarge) Output from a quality control camera system on an RTR 4500. Dispensing run over a 50-m dispense length, 50-mm/sec, human IgG at 1 mg/ml, 0.8 µl/cm on nitrocellulose.

Membrane blocking serves multiple functions in lateral-flow systems, including preventing or reducing nonspecific binding of samples or conjugates to the capture lines and the entire membrane. Blocking is also used to control flow rates and stabilize test and control-line proteins. The blocking process involves immersion of the striped membrane in an aqueous solution of proteins, surfactants, and polymers. The membrane is then removed, blotted, and dried.

While the blocking process can be performed in either a batch or continuous mode, using continuous, in-line processes achieves the greatest consistency of dipping and drying. Both steps are critical to ensuring adequate product performance and reproducibility. Any variation in product performance caused by inhomogeneous drying conditions in batch ovens can be overcome by using in-line drying systems. For example, drying towers process each part equally, are simple to validate compared with batch ovens, are not subject to variations in temperature or humidity, and can be highly efficient.

Figure 6. In-line camera and bad-part marking system on an RTR 4500 inline dispensing system by BioDot Inc.

In-line dipping and drying results in a more homogeneous treatment of materials because of the following factors: consistent dwell time of each unit length of material in the dip tank; consistent drying of each unit length of pad in the drying towers; and reduced solute drag issues in batch tanks, as the in-line dip tank is continuously refilling and mixing, which helps to keep the solution concentration consistent.

Requiring blocking of nitrocellulose membranes is an unfortunate corollary to using this material. While the blocking process is necessary to prevent product variation during the product’s lifetime, it can add cost and complexity to manufacturing.

Figure 8. (click to enlarge) (a) Comparison of coefficients of variation and line intensities, blocked versus unblocked membranes. (b) Effects of blocking in stabilizing the flow rate of a membrane over time. Bars represent the time for assay reagents to flow laterally through 25 mm of membrane. DB = dip blocked, UB = unblocked.

Figures 8a and 8b demonstrate the effects of blocking on product performance initially and over time. The graphs compare the performance of a quantitative, visually read gold-based assay with and without dip blocking. At first, when the assays were blocked and run, it appeared as if blocking resulted in lower signal strength and a higher coefficient of variation (CV), leading to the conclusion that blocking should not be performed on this assay. However, when the effects of removing the blocking step on the system’s stability were examined, it was clear that not blocking the membranes would have a significant effect on the assay due to changes in assay run time upon storage in a desiccated pouch. IVD manufacturers should always carefully assess the requirements for blocking due to such effects.

Conjugate Pad: Glass Fibers, Polyesters, and Rayons

In a typical assay system, the conjugate pad accepts the conjugate, keeps it stable during the product’s shelf life, and releases it efficiently and reproducibly. In reality, variations in conjugate application, drying, and release from the material are the primary sources of variation in assay performance, as measured within and between lot CVs. Assay sensitivity can also be adversely affected by poor conjugate mixing and release in the conjugate pad. Depending on the system, while it may be more important to achieve fast or slow release of the conjugate, the release must always be consistent.

Because of the nature of the materials used, conjugate pads should be pretreated to ensure the appropriate release and stability characteristics. Pad pretreatment is performed by immersion in an aqueous solution containing proteins, surfactants, and polymers, followed by drying. Similar to membrane dipping and drying, this process can be performed in either a manual batch mode or a continuous in-line mode, with the latter giving the best opportunity for homogeneous processing of entire batches of materials.

The method used for adding conjugate to the treated pad is critical to the final performance of the test. The following two methods are used to add conjugate: immersion, in which a treated conjugate pad is immersed in a conjugate suspension; and dispensing using quantitative noncontact dispensers such as the AirJet Quanti 3000 by BioDot (see Figure 4).

Quantitative dispensing of conjugates onto conjugate pads results in improved consistency of conjugate uptake compared with an immersion process. Dispensing conjugates also affects the consistency and efficiency of the release of conjugate from a conjugate pad. In addition, dispensing conjugates reduces any washing away of the pretreatment reagent, which occurs during immersion.

Table I. (click to enlarge) Percent recovery of paramagnetic particles from a variety of conjugate pads upon running of the assay (n = 10).

The data in Table I demonstrate the results of a comparison between dip coating and dispensing conjugates onto two different conjugate pads. In this experiment, paramagnetic particles were applied to each pad and normalized, such that each pad received the same total amount whether dip coated or dispensed. The dried pads were then assembled into LFIA strips, which were run with 150 µl of buffer. The recovery of conjugate was determined by total iron analysis via inductively coupled plasma mass spectrometry.

The recovery of conjugate from the pad was significantly different between the dipped and dispensed conjugate systems. Since dispensed conjugates are concentrated onto a smaller area than those applied through immersion, contact with the conjugate pad material is limited, resulting in improved particle release from the material.

Figure 9. (click to enlarge) (a) Comparison of sample pad release of analyte. (b) Comparison of analyte recovery from two sample pad materials.

The data further demonstrate an improvement in assay CVs resulting from quantitative application of conjugate via dispensing (see Table II). The test consisted of a sandwich immunoassay for a marker in serum. The latex particle conjugate was applied via a saturation method versus dispensing using the AirJet Quanti 3000 by BioDot. The conjugate was normalized such that each test received the identical amount. The results were measured using an optical reader. The CVs in the critical clinical range for this analyte were improved in the pad with dispensed conjugates.

In addition to the conjugate system, the choices of labels and conjugation methods are critical. Covalent coupling can be important to the ability to perform quantitative assays due to the more stable bonds between the ligand and particle versus passive adsorption methods. New labels and reading methodologies will be discussed further in the second article of this series.

Sample Pad: Celluloses, Glass Fibers, Polyesters, and Other Filtration Media

In LFIAs, the sample pad accepts the sample and treats it to make it compatible with the assay. This process must be done without destabilizing the analyte and by delivering it with high efficiency and reproducibility to the rest of the assay. The materials chosen can affect assay performance due to the inhomogeneity of many available materials and the type of binders that they contain. If required, the method of pad pretreatment must be carefully designed to avoid buffer concentration gradients and edge effects upon drying. In-line methods are best used for homogeneous treatment of the pads.

Table II. (click to enlarge) Comparison of data generated using a variety of conjugate application methods in a serum marker assay.

The importance of choosing appropriate sample pad materials is demonstrated by the effects of changing materials on assay performance as measured by CV and linearity in a competitive-format quantitative assay for progesterone (see Figures 9a and 9b). Two different sample pad materials were evaluated in this system. Both pads were made of cellulose material and were otherwise indistinguishable. Each pad was treated with a buffer to maintain pH and ensure that the analyte was not bound, and with a polymer to stabilize the flow rate. The results were obtained by reading the developed strips on a test-strip reader by BioDot, which utilized a CCD camera to capture and calculate line density.

Sample pad A demonstrated better recovery of the analyte, resulting in improved CVs, particularly on the lower end of the curve, and greater linearity over the entire analysis range. Correct choice of the sample pad materials in this system affects the ability to manufacture an acceptable quantitative system.

Lamination, Cutting, and Assembly

Using multiple materials for different functions in lateral-flow assays means that laminating the individual components onto a backing material is required. In traditional, non-reader-based lateral-flow assays that use manual lamination methods, the lamination process can have large tolerances built in. Such tolerances can result in variations in overlap and final line placement in a cassette from test to test. Overlap variations can result in variations in the run quality of a strip, although such variation is acceptable in many less-demanding applications.

Figure 10. LM6000 in-line lamination system by BioDot Inc.

However, in some of the more demanding applications, variations in run time and fluid front conformation can be detrimental to assay performance. Particularly in reader-based systems, the evenness of line development across the entire width of a strip, the speed of running to completion, and the position of a developed line in the assembled cassette can be critical to the test’s success. Such factors place a higher demand on the lamination process and the cutting and cassette assembly processes. Automation is key to the success of such processes at tight tolerances. Using in-line lamination equipment with camera systems and material edge sensing, as well as sensing in cutting and assembly processes, can produce parts at acceptable tolerances for such applications (see Figure 10).

Conclusion

As they exist today, standard lateral-flow assay systems can be sensitive, reproducible, and economically manufactured, and can be appropriately used in a range of application areas that few other assay formats can match. With appropriate choices of materials, manufacturing equipment, and manufacturing process technologies, IVD manufacturers can produce assays that meet stringent performance criteria, although significant care in those choices is required.

In order to reach the next level of utility, innovation in and better application of methodologies are required. A fresh approach to materials design, manufacturing process design, assay design, and detection and sample-handling technologies has the capacity to lead to significant improvements in the performance of this technology. The key will be in implementing such changes while maintaining the essence of what makes lateral-flow assays a truly appropriate and applicable point-of-care and field-use technology.

The second article in this series will discuss alternatives to the standard approaches to materials and manufacturing processes, highlight the improvements in assay performance that such novel approaches can generate, and discuss the processing challenges that they bring.

Brendan O’Farrell, PhD, is a principal at Diagnostic Consulting Network (Irvine, CA) and vice president of technology development at BioDot Inc. Jeff Bauer is also a principal at Diagnostic Consulting Network and is principal research scientist at BioDot Inc. The authors can be reached at bofarrell@dcndx.com and jbauer@dcndx.com, respectively.