Minimize incubation times with flow-through immunoassays
Matrix interference can be a major challenge in the development of immunoassays to study a number of disease states. Kathi Williams and her colleagues at Genentech Inc., USA, experienced considerable problems with matrix interference when developing an ELISA for Phase III studies of a humanized monoclonal antibody when a new patient group was introduced. The solution was to transfer the assay to the Gyrolab platform that, thanks to its flow-through technology helped minimizing matrix effects by reducing the incubation times with the capture reagent antibody.
Adding a new patient group brings matrix interference
Etrolizumab is a very promising humanized monoclonal antibody directed against the β7 subunit of two integrin heterodimers involved in highly debilitating inflammatory bowel disease, which includes ulcerative colitis (UC) and Crohn’s disease (CD). The assay development team at Genentech had been successful in developing and validating a sandwich ELISA for pharmacokinetics analysis in Phase I/II studies on UC, but adding CD patients in the Phase III trials brought with it matrix interference. While dilutional linearity was good for samples from the healthy population and the UC and CD patient groups, the recovery of spiked etrolizumab was low in CD patient samples, despite efforts to solve the problem by changing capture and detection reagents (Figure 1). The team needed to find an effective solution fast.
Figure 1: Performance of a modified ELISA tested by spike recovery at 800 ng/mL in ten samples each of normal healthy control and disease indications and then diluted to the MRD. Low recovery of spiked etrolizumab indicated persistent matrix interference despite modifications to the ELISA. From Figure 2A, Williams et al, 2017.
In search of an effective solution
Matrix interference can be caused by substances in the disease population, such as rheumatoid factor in rheumatoid arthritis patients, or heterophilic antibodies in autoimmune diseases such as irritable bowel syndrome. Solutions include adding substances to the assay that bind or compete with the interfering molecules, incorporating sample pretreatment steps, or using reagents that lack the Fc fragment. Most of these approaches are labor intensive and expensive, making them impracticable for larger studies.
The assay development team considered diluting out the interference but this would have reduced the sensitivity to unacceptable levels, and dilution didn’t appear to solve the problem for CD samples. Increasing incubation times only exacerbated the problem, suggesting that reducing incubation times might be the answer. This observation led the team to decide to transfer the assay to the Gyrolab system, which is based on flow-through immunoassays that minimize incubation times. There were also other advantages, including automation to improve robustness, and the possibility of improving the dynamic range and thereby ensuring more samples were in range without the need for additional dilutions.
Gyrolab system delivers in two weeks
The team successfully transferred the ELISA directly to the Gyrolab system and the immunoassay gave extremely robust results in the qualifying runs within two weeks, ready to be transferred to a CRO for validation. Transferring to the Gyrolab system increased the dynamic range over ten-fold, and a direct head-to-head comparison of the validated ELISA with the Gyrolab assay showed good correlation. The assay is now being used to assess six ongoing Phase III studies.
Figure 2: Spike recovery at high (10 000 ng/mL) and low (91 ng/mL) in eight each normal healthy subjects and disease indications, before being diluted to the MRD. All spikes were recovered within ±20%. From Figure 4A, Williams et al 2017.
Find out more about Gyrolab system and how it helped to solve the problem of matrix interference at Genentech by watching the recorded webinar, ‘Overcoming Disease Specific Matrix Effects in a Clinical Pharmacokinetic Assay’