Integrated lab-on-a-chip

In 2010, we celebrated the 20^th anniversary of a technology concept that has had significant impact on how analytical tasks are currently performed in chemistry, biology, and medicine. The idea of a ‘miniaturized total analysis system’ (μTAS)¹ triggered significant scientific activity that has led to a broad range of applications and also finds increasing use in commercial lab-on-a-chip systems.² In the early years, most miniaturized devices performed simple operations (such as electrophoretic separation or mixing of liquids³), but functional integration has increasingly become possible. This will ultimately lead to a fully integrated ‘sample-in, answer-out’ system.

Figure 1 shows a typical set of biochemical-assay operations. The sample is first introduced onto the chip, where it needs processing for further use on the device. This particular task has frequently proven challenging because there are generally a large number of required steps and different media and volumes are frequently used. For example, in the identification of rare cells in whole blood—such as when detecting tumor cells or in prenatal diagnostics—one first has to remove as much ‘background’ noise (white and red blood cells) as possible. The remaining cells must then be lysed, ideally selectively, to access their genetic material. The latter is usually concentrated (using magnetic beads), before amplification using, for instance, a polymerase chain reaction (PCR) or equivalent methods. This is done to generate enough signal for detection. Separation of the desired molecular species, perhaps in the form of an electrophoretic or chromatographic step, often occurs before the sample is transported to the detection area. A suitable detection technique (such as using laser-induced fluorescence or an electrochemical sensor) is then employed to extract the signal. Finally, the sample has to be discarded properly, which—for biologically active material—means on-chip waste storage to minimize contamination risks.

Figure 1. Typical process flow for a biochemical assay in a microfluidic device. PCR: Polymerase chain reaction. RCA: Rolling-circle amplification. DEP: Di-electrophoresis.

We have developed a microfluidic system that includes all or most of these steps. One has to keep in mind that there is a fundamental difference in the design process and the underlying physics compared to other fields of engineering, especially micro-electronics, to which microfluidics is often compared. In the latter fields, development of individual modules is facilitated by the fact that their mutual interactions are limited and often calculable using simple constraints. This allows assembly of module libraries that can be simply transferred from one development case to another. However, this is not easily feasible in microfluidics, since the performance of such a functional module is influenced by the complete system and all functions before and after the specific module. Therefore, following emerging best practice, we combined the theoretical (or modeling) approach with experimental data from module prototypes that—coupled together—match the complexity of the fully integrated device. This stepwise approach also simplifies the search for and correction of possible errors. Figure 2 shows one of our module test platforms, which combines modules for plasma generation from whole blood, a free-flow separation unit, PCR, and a sample-collection channel in a plug-and-play-like setup. One important property that facilitates this modular configuration is our use of standard sizes and interfaces, enabling module use in a toolbox-like fashion.⁴

Figure 2. Modular test platform, combining modules for plasma generation from whole blood, a free-flow separation unit, PCR, and a sample-collection channel.

When the microfluidic functionality has been validated and the required assays have been performed, integration of these steps onto a single device can be done. Figure 3 shows such a highly integrated microfluidic cartridge for isolation of cancer cells from a whole-blood sample.⁵ Note that this functional integration requires combining elements made from different materials (such as filtration membranes, septa, or silicon-based sensors). This area requires further research, since many existing assembly technologies are not applicable to devices incorporating biomolecules because of their sensitivity to heat or other chemicals, such as solvents or glues. This represents one of the directions of our ongoing research.

Figure 3. Fully integrated test chip for development of a cancer-diagnostic system. (Courtesy: Project ‘SmartHealth,’ European Union contract FP6-2004-IST-NMP-2-016817.)

In summary, current technological and scientific advances indicate that one of the truly important potentials in analytical-device miniaturization—i.e., integration of all analytical/diagnostic process steps—is coming within reach of commercial product development. This will enable providers of such applications with a means to extend them to novel fields or create new systems. For the microfluidic community, this could be an important step toward finding the still missing ‘killer application.’