Marwan Alsarraj is the Biopharma Segment manager at Bio-Rad.

Marwan Alsarraj

Although many have never heard of it, β-thalassemia is one of the most common autosomal recessive diseases globally and is a serious blood disorder. Scientists have estimated that about 1.5% of the global population carries mutations associated with β-thalassemia, with more than 60,000 new cases diagnosed every year (1). This blood disorder reduces the production of hemoglobin, the iron-containing protein in red blood cells that plays an essential role in delivering oxygen throughout the body. Scientists and researchers, therefore, seek to produce a gene therapy to treat β-thalassemia and correct the underlying genetic imbalance that causes the condition.

One promising approach involves using clustered regularly interspaced short palindromic repeats (CRISPR) to correct the mutated gene. However, scientists are still working to improve and validate CRISPR’s editing efficiency. Furthermore, CRISPR’s ability to successfully correct mutation associated with -thalassemia is still uncertain. Researchers, therefore, need to pair CRISPR editing protocols with a quality control tool, such as droplet digital polymerase chain reaction (ddPCR), that accurately detects the presence of successful CRISPR edits.

When correcting the mutations with β-thalassemia, scientists do not have a straightforward path toward addressing this condition at the genetic level. Adult hemoglobin is composed of two globin subunits, α-globin and β-globin, which must be expressed in equal numbers for hemoglobin to develop correctly. People with β-thalassemia have a genetic mutation in the gene for β-globin, 

, leading to the gene’s downregulation. With unequal quantities of α-globin and β-globin circulating, the free α-globin forms toxic precipitates that impair the development of red blood cells and kill mature red blood cells. As a result, patients can experience a wide range of severe symptoms, including an increased risk of developing blood clots, weakness, and fatigue, which often lead to early death.

Some research suggests that deleting the α-globin gene, 

, may improve outcomes. Alternatively, introducing a healthy 

 gene via a lentiviral vector improves patients’ clinical outcomes, but only if these patients already express some β-globin (2). A research group based in France and Italy recently combined these two approaches: they used CRISPR to delete one copy of 

 in hopes of restoring the balance between the two hemoglobin subunits (3).

Performing this kind of dual edit is a complex task. Researchers must first design a guide RNA (gRNA) to locate the gene to be edited. Once the gRNA identifies the correct site containing the 

 gene, Cas9 must perform the cut and facilitate the insertion of the 

 gene previously occupied. A dual edit approach only works if CRISPR correctly removes 

 in the same spot. Assessing the success of this technique requires rigorous quality control to ensure CRISPR performs the correct edits. This is where ddPCR technology comes into the equation.

While researchers commonly use quantitative polymerase chain reaction (qPCR) to assess gene-editing success, this technique can have drawbacks, especially with regards to quantification. qPCR can only estimate the transgene copy numbers by relying on a standard curve of serial dilutions to interpret samples. Therefore, the results are less sensitive and can not measure down to one gene per cell. In contrast, ddPCR is well-suited to the task, as it delivers absolute nucleic acid quantification without the aid of a standard curve. ddPCR technology is a highly sensitive tool designed to detect and quantify rare genetic variants, and it can be used to detect outcomes of CRISPR editing. For example, researchers have used ddPCR assays to detect CRISPR edits via homology-directed repair and nonhomologous end-joining (4).

ddPCR technology works by partitioning a sample into approximately 20,000-nL-sized droplets and running a separate PCR reaction in each one. Each droplet contains one or a few nucleic acid strands. If a droplet includes a strand with the target genetic sequence, that DNA will amplify, and the droplet will emit a strong fluorescent signal. On the other hand, if a droplet does not contain the target sequence, the droplet will only emit a weak fluorescence signal. By counting the strongly vs. weakly fluorescent droplets, scientists can detect specific sequences with exacting sensitivity and measure the concentration of the target sequence in the original sample.

Compared to next-generation sequencing, ddPCR technology is fast, inexpensive, and not labor-intensive, and it can detect rare events without being limited by reading depth (5). ddPCR technology is also more sensitive and accurate than qPCR. The France-Italy research group used ddPCR assays to assess the success of their dual editing approach for treating β-thalassemia (3). The researchers first used ddPCR technology to quantify 

 copy number, which correlated with -thalassemia severity. They also used the technology to detect the successful insertion of 

. In human umbilical cord blood-derived erythroid progenitor cells, the team showed robust insertion of the 

 gene, confirming on-target integration of the gene at 0.8 copies per cell.

The team could not have detected this integration using qPCR. Because of the inherent variability in how qPCR results are measured, the technique cannot detect gene copies at concentrations lower than two or three copies per cell. Without ddPCR technology, these researchers would not have shown that their CRISPR strategy has potential for future clinical testing.

Nearly 40 clinical trials are currently investigating whether CRISPR can be used to treat genetic diseases, and regulatory agencies might approve the first CRISPR-based gene therapy in less than a decade (6). But given the continued challenge of developing a reliable CRISPR editing protocol, biopharmaceutical companies developing CRISPR therapies must take care to ensure their therapies are safe and effective by using tools like ddPCR technology to provide the confidence they need. For example, ddPCR assays can be designed to detect any CRISPR edits by using probes that span the junction between the native genome and the donor sequence. Researchers and biopharmaceutical manufacturers can screen out cell lines containing unsuccessful edits before they even reach patients by analyzing cells lines for specific CRISPR edits. This, in turn, will increase the chance of clinical success for CRISPR-based gene therapies and open the door to a new generation of treatments for difficult-to-treat genetic diseases, such as -thalassemia.

 5 (11) (2010).
2. S. Mettananda, et al., 

 125 (24) 3694–3701 (2015).
3. G. Pavani, et al., 

 5 (5) 1137–1153 (2021).
4. Y. Miyaoka, et al., 

 1768, 349–362 (2018).
5. C. Peng, et al., 

 11 (2020).
6. US National Library of Medicine, “NIH,” 

Marwan Alsarraj is the Biopharma Segment manager, Digital Biology Group, at Bio-Rad.

When referring to this article, please cite it as M. Alsarraj, “Assessing the Success of CRISPR Gene Therapies Using ddPCR,” 

Articles

Using more exacting analytical tools can give a clearer assessment of gene-editing outcomes.

Assessing the Success of CRISPR Gene Therapies Using ddPCR

Chimeric antigen receptor (CAR)-T cells have already changed how certain cancers are treated. As of February 2022, FDA had approved five CAR-T cell therapies to treat liquid tumors in patients with various types of leukemia and lymphoma (1). The biopharmaceutical industry has invested heavily in the future of this novel treatment modality, and CAR-T cells now account for 56% of cell immunotherapies in development (2). By 2030, one forecast predicts that the CAR-T cell market will grow to $21.7 billion (3).

With this growth, however, comes the realization across the industry that CAR-T cell therapy development is a complex process that requires standardization, regulation, and meticulous quality control to reliably generate safe and effective therapies. In addition, when developing CAR-T cells, biomanufacturers can have their choice of cell sources and bioprocesses, which can lead to variability in success as well as openings for inefficiency and waste. Analytical tools such as droplet digital polymerase chain reaction (ddPCR), a technology platform used to quantify nucleic acids, can support CAR-T cell manufacturers in developing successful CAR-T cell products.

The importance of quality control in CAR-T cell development is evident in the development process itself. First, a physician draws blood from a patient and a technician extracts the T cells. Next, a technician delivers the 

 gene to the T cells, typically using viral vectors. The cells take up the gene and express it, generating the CAR protein. This protein, in turn, gets shuttled to the cell surface,` thereby transforming the T cell into a CAR-T cell. Finally, the technician amplifies the CAR-T cells in a bioreactor and delivers the batch to the clinic, where a physician transfuses the cells back into the patient.

Each step in this process depends on nature: technicians rely on viruses and T cell machinery to generate CAR-T cells as well as on the proper environmental conditions to expand the cells, while physicians rely on the patients’ bodies to tolerate the cells and degrade them at an appropriate rate. Because developers and physicians cannot fully control all of these factors, they must monitor the creation and administration of CAR-T cells to ensure they perform as expected and will not induce adverse reactions in patients. Below are several sources of variability in CAR-T development and how ddPCR technology can help resolve some of the uncertainty inherent in the process.

In the development of CAR-T cells, one key initial step is using viruses, such as adeno-associated virus (AAV) or lentivirus, 

 gene. But once the initial conditions are set, it is difficult to control the behavior of these viruses and the genetic sequences they contain. As a result, the number of 

 transgenes that enter each cell and integrate into the genome can vary. Depending on the success of this transfection step, the treatment may not work at all, or worse, it may elicit a systemic inflammatory response that can cause further harm to patients (4). To protect patients, FDA recommends that CAR-T cell manufacturers performing viral transduction screen out cells containing less than one or more than four transgene copies per cell (5). To do this screening process, technicians need to quantify the 

 transgene copy number accurately and precisely.

Quantitative polymerase chain reaction (qPCR) is the standard method for quantifying nucleic acids, but it is not sensitive enough to quantify 


transgenes with the accuracy and precision needed to satisfy regulators. The qPCR technique quantifies nucleic acids indirectly, using a standard curve to estimate copy number. Scientists must generate this standard curve using serial dilutions, a process that is time-consuming and prone to human error. This error creates variability that makes it challenging to quantify small copy numbers. If developers cannot tell with certainty whether or not a CAR-T cell batch contains at least one transgene copy per cell, they cannot be sure whether their transfection method was successful.

ddPCR technology, however, can offer precise measurements of transgene copy number because of its ability to quantify gene sequences directly without using a standard curve. The technique involves partitioning a sample into approximately 20,000 nanoliter-sized droplets that contain one or a few nucleic acid strands each, with some droplets containing the target gene and some not. A separate PCR reaction takes place in each one. Starting with a sample of nucleic acids from CAR-T cells and using primers targeted to the 

 transgene, DNA amplification will only occur in the droplets containing that gene. As the gene amplifies, a reporter gets cleaved from the probe and emits a fluorescent signal. The fluorescent droplets can be counted to derive the transgene concentration in the original sample. Unlike qPCR, ddPCR does not measure the degree of amplification to estimate copy number in the original sample; instead, it counts whether or not amplification took place, providing a digital measure of copy number. This ability makes ddPCR technology more accurate and precise and, therefore, more reliable for assessing CAR copy number in CAR-T cells.

These attributes were quantified in a study performed by 

 at the Huazhong University of Science and Technology in Wuhan, China (6). The researchers directly compared qPCR and ddPCR instruments in their ability to quantify DNA standards and CAR transgene copies in clinical peripheral blood samples from patients on CAR-T therapy. They found that ddPCR demonstrated greater sensitivity than qPCR. Specifically, among a series of DNA standards, ddPCR assays detected as few as 3.2 copies/mL, while qPCR showed a negative result at that concentration. Furthermore, their ddPCR assay could reliably detect as few as five copies per reaction among clinical blood samples, while qPCR could not detect concentrations lower than 20 copies per reaction. Additionally, ddPCR technology showed lower intra-assay and inter-assay coefficients of variance for the series of diluted standards than qPCR, suggesting that ddPCR is more repeatable and reproducible.

Meanwhile, researchers at the National Institutes of Health Clinical Center wanted to see if this method of copy number quantification was robust. To do this, 

 tested the consistency of ddPCR assay results following several changes to the development workflow (7). Specifically, the researchers transfected T cells using either lentiviral or retroviral vectors. They then assessed transfection success right away or after the cells had been frozen for three or six weeks, using three different technicians across two different laboratories. When they quantified 

 copy number amidst all of these protocol variations, their ddPCR assay delivered consistent results.

ddPCR technology was also central to optimizing 

’s transduction protocol. The researchers varied the viral vector-to-T cell ratio as well as varied their centrifugation protocol (i.e., either centrifuging their samples or not) following transduction to identify the best protocol for maximizing transduction efficiency and copy number per cell. They used flow cytometry to measure the effect of these variables on transfection efficiency while using ddPCR technology to measure copy number. Both of these measures were correlated throughout the experiments. Transfection efficiency and copy number per cell increased as the vector-to-target cell ratio increased, but only when the samples were not centrifuged. Conversely, centrifuging the samples increased transfection efficiency at lower vector-to-target ratios. Overall, these data demonstrated that ddPCR technology can reliably quantify 

 transgene copy numbers across different conditions and can be used to assess the impact of various protocol deviations on the success of CAR-T cell production.

 gene can lead to immune reactions, but this scenario is not the only case in which a CAR-T cell could theoretically pose a risk to patients. For example, CAR-T cell production using lentivirus could potentially create replication-competent lentiviruses (RCL) that could infect patients and induce the rapid growth of T-cell neoplasms. This scenario is purely theoretical; however, to be safe, FDA recommends that manufacturers test their clinical vector lots and manufactured cell products for RCL. In addition, the agency recommends that doctors test their patients’ blood after receiving CAR-T cell therapy (8).

ddPCR technology can be used to detect RCL. In one study, researchers 

 at the Mayo Clinic in Rochester, Minn., developed a ddPCR assay that detects the envelope sequence of the virus (9). The limit of detection for their assay was 10 copies/μL. The technique, overall, provided sensitive, specific, and reproducible RCL detection, expediting quality-control assessments compared to traditional cell-based or qPCR-based approaches. 

’s results suggest that ddPCR could be used to screen out CAR-T cells that contain RCL so that these cells never reach patients.

Physicians infuse patients with a specific dose of CAR-T cells that is intended to deliver a therapeutic benefit without causing harm to patients. However, there is no way to fully predict CAR-T cell kinetics once the cells enter the body. Consequently, physicians must monitor CAR-T cell concentration in the blood on a serial basis to ensure the cells are present in the appropriate concentration for the right amount of time.

CAR-T cells are supposed to survive in the body for a few months, but, if they persist for too long, they can elicit a prolonged inflammatory response even after they’ve effectively treated the patient’s cancer (4). Conversely, if the cells degrade too quickly, they will not do their job. Therefore, physicians must monitor CAR-T cell kinetics to optimize clinical follow-up regimens, understand treatment failure, and understand the cause of adverse events.

There are currently no standard methods for measuring CAR-T cell kinetics, but ddPCR assays present an attractive option. Because ddPCR technology can detect small concentrations of nucleic acid sequences in liquid samples, it can be used to precisely quantify CAR-T cell concentration in the blood. Furthermore, because liquid biopsies based on ddPCR technology are minimally invasive, these tests can be run on a serial basis without causing further harm to patients.

 in Germany demonstrated the feasibility of using ddPCR technology for this purpose in patients with various types of lymphoma (10). Two CAR-T therapies, tisagenlecleucel and axicabtagene ciloleucel (axi-cel), are currently approved to treat this disease. In patients with lymphoma treated with CAR-T cell therapy, plasma CAR-T cell concentration correlates with treatment response and patient outcomes, indicating that CAR-T cell kinetics can serve as a predictive biomarker of therapeutic efficacy (11–13). To test this hypothesis, researchers developed a ddPCR assay that examines axi-cell kinetics. First, they tested it on positive control samples containing various known concentrations of CAR-T genomes. Their assay reliably detected one CAR gene per genome.

The researchers then tested 54 samples from seven patients between four weeks and nine months following treatment. They found that CAR-T cell kinetics varied greatly between patients, demonstrating the importance of serial monitoring. Looking closer at each patient, the researchers discovered that CAR-T cell persistence and peak expansion correlated with clinical responses. For example, the patient whose 

 levels dropped to undetectable levels at day 75 after treatment showed no clinical improvement, while the patients whose 

 levels were still detectable after nine months went into remission. At the other end of the spectrum, the patient with the highest peak 

 copy levels experienced severe side effects following therapy. Altogether, these data revealed the variability in CAR-T cell kinetics and the importance of serial monitoring.

ddPCR and CAR-T cell therapy go hand-in-hand

ddPCR technology has already proven useful in multiple research areas over the past decade. CAR-T cell therapy is a key current example that demonstrates the technology’s full potential to help biopharmaceutical manufacturers and physicians keep patients safe and healthy. CAR-T cell therapy development is not simple, but by quantifying 

 transgenes and other genetic elements using ddPCR assays throughout the process, CAR-T developers will be more successful in developing therapies that will achieve regulatory approval, and physicians can be more confident that they can make a meaningful difference in their patients’ lives.

1. University of Pittsburgh Medical Center Hillman Cancer Center, “FDA-approved CAR T-Cell Therapies,” 

2. Cancer Research Institute, “Cancer Cell Therapy Landscape,” 

CAR T-cell Therapy Market - Global Industry Analysis, Size, Share, Growth, Trends, and Forecast, 2021-2031

8. US Department of Health and Human Services, FDA, CBER, 

Mark White, PhD, is the associate director of Biopharma Product Marketing, and Marwan Alsarraj is the Biopharma Segment manager; both are at Bio-Rad.

Standardization is needed in CAR-T cell therapy development due to its inherent complexity.