January 1, 2016 (Vol. 36, No. 1)

Angie Breinin R&D Scientist SAFC
Ian M. Taylor Ph.D. Commercial Director Solentim

Increasing Workflow Productivity and Assuring Clonality for IND and BLA Submissons to Regulators

New therapeutic monoclonal antibodies (mAbs) and biosimilars are commonly produced by clonal, mammalian cells. Selection and optimization of new clones are key in bioprocess development. If clonality cannot be assured, the FDA has stated it will require additional controls on manufacturing and other aspects of the product quality control strategy.

If biopharma companies cannot provide regulators with detailed evidence of clonality in Investigational New Drug (IND) or Biological License Application (BLA) submissions, the additional manufacturing steps required could cause costly delays to a clinical trial or drug launch.

The traditional method for selecting single clones for the Master Cell Bank (MCB) and manufacturing involves seeding via limiting dilution or with fluorescence-activated cell sorting (FACS). Since some colonies may not subsequently divide, or the cells may not produce sufficient protein, scientists may need to identify and track hundreds of such wells to identify a high-producing clone.

The traditional approach has other disadvantages. Researchers must identify small colonies at early time points in the growth of the well; this is a highly subjective process that varies based on a scientist’s skill level. Using a microscope also requires scientists to constantly adjust the focus around each of the 96 wells to obtain a precise view of its contents. This method of identifying single cells is time-consuming and can even cause scientists to feel dizzy or nauseated after prolonged periods at the microscope.

The result: a possible decrease in the quality of data gathered in a day. Lack of data traceability is another issue with this method. Wells are marked manually, and regulators must rely on statistical data generated by researchers. This means that there is no photo documentary evidence to prove monoclonality.

These issues and the desire for traceable data have resulted in the development of automated cell imaging systems for biopharma companies and CMOs. Early generations of such systems have included their own set of disadvantages that impair the ability to identify single cell clones, such as poor illumination around the well edges, loss of data along tiled image stitch lines, the inability to see a single cell on the day of seeding, and lack of focus that lead to indistinct colonies.

Early identification techniques used by some researchers involve fluorescent screening. Ideally, the image data should be achieved without using fluorescence in clone screening. Fluorescence should generally be avoided, as labelled antibodies represent a potential contamination risk and cell surface stains have led to some toxicity, which has inhibited colony growth.

Cell Metric Platform

To address these issues, Solentim offers a family of Cell Metric™ imaging workstations dedicated to cell-line development. This platform includes a proprietary camera and illumination technology integrated with software that monitors and analyzes 96 or 384 microplates. Each well is captured using brightfield imaging, and cells are clearly visualized as dark spheres with white halos, making them easily distinguishable from debris.

The lighting generates an even illumination across the well, so there are no dark shadows around the well edges, and using brightfield means fluorophores are not required to detect cells. The combination of a high-resolution camera and precise stage movements means there are no stitching effects; the system can accurately see a single cell on the day of seeding and importantly also resolve two cells stuck together.

A key attribute of this imaging technology is its focus assurance across the whole microtiter plate. Plates are naturally distorted during the manufacturing process, typically by 200 to 300 µm (Figure 1) and this varies among manufacturers and from plate to plate and batch to batch. The imager maps the contours of the plate’s bottom surface and adjusts so the well bottoms are always in focus. This focus assurance is essential in order to qualify a cloning method and should also reduce the incidence of wells with “ghost” colonies (where a colony forms, but no cell is detected at Day 0).

Scientists will typically perform a time course of microplate images: the day of seeding Day 0, Day 1, Day 2, Day 3, Day 7 and Day 14, for example. The imager’s post-run analysis software allows researchers to toggle from colony outgrowth images back to Day 0 images of one or two cells. Scientists need only to go back and look at images for the wells that formed a colony. The imager can also measure confluence of the colony outgrowths as an indicator of relative growth rates.

Scientists can spend more time analyzing and scrutinizing these selected wells to ensure that they truly are clonal and generate an automated Clonality Report. The format of this report is designed around the recent FDA requirements regarding the establishment of clonal cell lines. To generate this report, scientists select the wells identified as clonal, annotate the image feature (as single cell or debris) and then select the imaging time points to include for these wells.

The single cell image feature is shown within the context of the whole well image, and the report has a full audit trail of clone identity, time, and date stamps. These reports can be printed or exported as PDF or PowerPoint documents, and image quality is maintained in the exported formats, allowing for easy regulatory review. They can also form part of an IND application, BLA, or be included in a client dossier for a CMO. These reports can be rapidly generated, are consistent between users, contain a secure audit trail, and represent proof that can travel.

Figure 1. The heights of well bottoms measured on-the-fly for all the wells in the microplate during imaging. Focus is dynamically adjusted for these plate deformities so that every well is in focus. This adjustment happens simultaneously with image acquisition.

Proof of Concept

SAFC currently uses the Cell Metric system to determine monoclonality in CHOGS cell lines expressing mAbs for biotech and pharma clients. SAFC researchers use an automated whole well imager, as customers are increasingly asking for documentary evidence that cells in the MCB are derived from a single clone. The Cell Metric platform is available as a stand-alone system or, for increased throughput, as the Cell Metric CLD (Figure 2), which includes a built-in incubated microplate loader and cassettes.

The imager was assessed in proof-of-concept experiments using CHOZN®
GS-/-ZFN-modified CHO cells (Sigma-Aldrich) expressing a human recombinant IgG cultured in chemically defined, animal component-free EX-CELL® CD CHO Fusion medium (Sigma-Aldrich). Single cell cloning was performed using limiting dilution in CellBIND® 96 well clear flat-bottom polystyrene microtiter plates (Corning) in 80% EX-CELL CHO Cloning Medium (Sigma-Aldrich) and 20% conditioned EX-CELL CD CHO Fusion medium.

The cells were cultured until confluent (37°C, 80% humidity, 5% CO²) and for 14 days they were imaged at Day 0 (when cells were seeded) and also on Days 1, 3, 7, and 14. The imager was used to determine well confluency. When the wells were 80–100% confluent the plates were consolidated and replica plated for a static titer assay. Supernatant was sampled from the wells on Day 7, and the supernatant was analyzed for protein titer to identify the most productive clones.

Figure 2. Cell Metric CLD imager for cell-line development. Image courtesy of Horizon Discovery

The results (Figure 3) show that on Day 0, the system generated a clear image of a single cell; on Day 1, this has divided into two cells; by Day 3, a loci of growth from the nascent colony has started.

Since the success of these proof-of-concept assessments, the automated  cell imaging system has been seamlessly integrated into the existing bioprocess workflow and has increased throughput of microplate imaging from 10–20 to 30–50 plates per day. Throughput has increased, too, because at the beginning of the experimental run, SAFC researchers can effectively image every single well on the day of seeding. Since the imager captured the data from Day 0, they can examine the most promising wells on Day 7 and track back to Day 0 to determine the wells that originated from a single cell.

The case study illustrates the significant workflow benefits resulting from the use of a dedicated automated imager to establish clonality in cell-line development. The research at SAFC has illustrated that it is now possible to demonstrate ease of clone screening and the ability to measure confluence with the Cell Metric imager.

Additionally, the Clonality Report includes high-quality, raw data images that show the choice of the top clones in a format appropriate for a client dossier document or for an IND submission delivered directly to a regulator.

Figure 3. Cell metric images of CHOGS cells (from left to right) on Day 0, Day 1, and Day 3 (magnified in the wells)

Angie Breinin is an R&D scientist at SAFC and Ian Taylor, Ph.D. ([email protected]), is sales & marketing director at Solentim.

Previous articleUPDATE: Endo Terminates Agreements with VIVUS, Acerus
Next articleScientists Use CRISPR to Treat Duchenne Muscular Dystrophy