Results from the first FDA-approved in-human trial of CRISPR-edited T cells for cancer treatment suggest that such therapies can be used safely in patients, allaying some of the concerns about potential AEs and advancing progress toward the development of more effective cell-based immunotherapies.
Carl H. June, MD
Results from the first FDA-approved in-human trial of CRISPR-edited T cells for cancer treatment suggest that such therapies can be used safely in patients, allaying some of the concerns about potential adverse effects (AEs) from the gene-editing tool and advancing progress toward the development of more effective cell-based immunotherapies.1
Investigators who led the trial at the University of Pennsylvania’s Abramson Cancer Center in Philadelphia (UPenn) for patients with refractory cancers are planning a follow-up safety study that will increase the dose of engineered T cells and target a different protein with the goal of boosting efficacy. Several other academic and industry groups in the United States are preparing similar studies and several such trials are under way in China.
The technology is also being used to build a map of cancer gene targets and biologists are creating tissue models to test for AEs associated with CRISPR editing.
Additionally, investigators are employing the technique to explore treatments for sickle cell anemia and other diseases, including coronavirus disease 2019 (COVID-19). In mid-March, as the pandemic worsened in the United States, scientists from the New York Genome Center in New York, New York, reported that CRISPR-based genetic screens can be used to identify potential predictive and therapeutic targets for COVID-19.2
CRISPR, which refers to clustered regularly interspaced short palindromic repeats, are genomic sequences bacteria use to recognize and destroy foreign DNA through an enzyme, frequently CRISPR-associated (Cas) protein Cas9, guided by an RNA molecule (Figure). This approach enables gene editing that is much faster, less expensive, and more specific than prior methods.3
The positive findings from the UPenn team were announced in February amid heightened concern in the global research community over potential misuse of CRISPR. In December 2019, a Chinese court sentenced scientist He Jiankui to 3 years in prison for misconduct related to a CRISPR experiment that modified the germline in human embryos to confer resistance to HIV.4 His work, which reportedly resulted in the births of 3 genetically engineered babies, was widely condemned for violating ethical norms and taking unnecessary risks.
The UPenn trial and other proposed studies of CRISPR-modified T cells differ significantly from He’s experiment: the oncology research does not involve embryos, it targets somatic rather than germline cells, and the CRISPR/Cas9 editing tool is currently used ex vivo to modify cells in a lab, rather than being injected into patients to change DNA in the body. The UPenn study also underwent an ethics review and a lengthy FDA approval process.
Concerns for patients in the T-cell study include immune reactions to the Cas9 protein and off-target edits that may create unsafe chromosomal rearrangements, such as changes that cause T-cell lymphoma. Although every T-cell therapy with a different set of edits will have to be checked again, the UPenn investigators have so far found no such AEs, said Jennifer R. Hamilton, PhD, a postdoctoral fellow at the Doudna Lab at the University of California Berkeley.
“Continuing to look at safety will be important. It’s really early days in being able to do these studies,” said Hamilton, who coauthored a Science perspective on the trial with CRISPR pioneer Jennifer Doudna, PhD.5 That said, she added, “It’s a really exciting time for gene editing and engineered cells. A lot of what we figured out in the gene editing field can enable better anticancer engineered cell therapeutics.”
Engineered T cells created with editing tools like CRISPR may eventually be able to treat patients with metastatic cancers that cannot be managed with existing immunotherapies alone, said Carl H. June, MD, a coauthor of the UPenn T-cell study. June is the Richard W. Vague Professor in Immunotherapy in the Department of Pathology and Laboratory Medicine and director of the Center for Cellular Immunotherapies, both at Penn Medicine. In 2015, he was named a Giants of Cancer Care® award winner for his work with chimeric antigen receptor (CAR) T-cell therapy in patients with leukemia and lymphoma.
“Synthetic biology offers the ability to make cells much more potent than the way we’re born with. That’s what our goal was here,” June said. “It will work hand in hand, I think, with things like checkpoint therapies and vaccines.”
Novel Therapy is Feasible
The phase I trial (NCT03399448) was designed to assess the safety and feasibility of infusing autologous cancer/testis antigen 1 (NY-ESO-1) T-cell receptor (TCR)—engineered T cells, called NYCE cells, in patients after CRISPR/Cas9 editing of the TRAC-α (TRAC), TCR-β (TRBC), and PDCD1 loci. The trial used a TCR rather than a CAR to reduce the incidence of cytokine release syndrome associated with CAR T-cell therapy and allow better assessment of potential AEs from CRISPR/Cas9 editing.1
“The ‘feasible’ aspect comes in because it’s the most complex cell manufacturing we’ve ever done, and the patients were really beat up—in the case of 1 patient, 8 different lines of chemotherapy and 3 different bone marrow transplants,” June said. “So you don’t necessarily know if their cells will still grow.”
The trial team enrolled 6 patients, harvested T cells from 4 of them, and used CRISPR to knock out PDCD1, which encodes the PD-1 immune checkpoint, and 2 endogenous TCR domain genes, TRAC and TRBC. The endogenous TCRs were removed to prevent mispairing and competition for expression with the synthetic TCR, increasing the effectiveness of the therapy and preventing autoimmune effects that have been seen in animal studies, June said. Disrupting PD-1 has been shown to improve T-cell killing of tumor cells. The cells were then transduced with a lentiviral vector to express a transgenic TCR for NY-ESO-1, a cancer/ testis antigen.
One patient progressed during the cell-preparation process and entered hospice. The NYCE cells were reinfused in the 3 other patients: 2 with refractory advanced myeloma and 1 with a refractory metastatic sarcoma not responding to multiple prior therapies. The patients were given lymphodepleting chemotherapy with cyclophosphamide and fludarabine on days 3 and 5 prior to T-cell administration and a single infusion of 1 × 108 manufactured engineered T cells per kilogram.1
Figure. Mechanisms of CRISPR Gene Editing (Click to Enlarge)
The best clinical responses were stable disease in 2 patients. One had a mixed response with an approximately 50% decrease in a large abdominal mass that persisted for 4 months, although other lesions progressed. As of December 2019, all patients had experienced disease progression, with 2 receiving other therapies and 1 dying from myeloma.
Engineered T cells taken from the blood at 9 months continued to exhibit anticancer activity in lab tests. The NYCE cells were also present in tumor and blood marrow biopsies but their level of functioning there is unknown, June said.
One potential pitfall of gene editing is that it may result in unintended chromosomal deletions and rearrangements, investigators have noted in preclinical findings.6 CRISPR/Cas9 works by causing DNA double-strand breaks, and the Cas9 “molecular scissors” protein can end up cutting in the wrong place. After an on-target cut, the natural DNA repair process that completes an edit can also result in the end of a chromosome becoming attached to another chromosome, creating a translocation.1
UPenn investigators assayed for 12 potential translocations, finding them in progressively declining numbers during the T-cell manufacturing process and after reinfusion. The falling frequency of these rearrangements suggests that “they conferred no evidence of a growth advantage over many generations of expansion in the patients” and thus no sign of oncogenicity, study authors wrote.
“T cells are pretty fault-tolerant. You can kind of have mayhem happen on them and they don’t turn into cancer. If you do it to your bone marrow stem cells, you’re much more likely to get a leukemia or some other kind of cancer,” June said.
Cas9 is derived from Streptococcus pyogenes, which causes strep throat and other infections. The protein provokes an immune response in healthy people that could destroy engineered T cells. However, testing of the 3 patients did not show any such responses, which the investigators said could be a result of the small amount of Cas9 in the engineered T-cell product or immunodeficiency in patients with extensive previous treatment histories.1
Hamilton attributed the absence of unwanted responses to the method by which the editing complex was inserted into the T cells. Investigators used a 2-step process, first knocking out the 3 genes with electroporation of ribonucleoprotein (RNP) complexes comprising recombinant Cas9 and then transducing the cells with a viral vector to express TCR, which in turn, recognizes NY-ESO-1.5 The Cas9 was not integrated into the cell DNA and was not replicated, and the protein degraded to below-detectable levels before reinfusion, she said.
Next Stop: CARs
June said his group’s next trial will aim for greater efficacy by using a CAR rather than a TCR, switching to CD19 as the target, and giving multiple infusions of engineered T cells at higher doses. It will also use more advanced cell culture reagents than were available when the NY-ESO-1 study protocols were established several years ago, which will enable higher levels of editing. Although the completed trial knocked out 15% to 45% of the target genes at the 3 loci, newer methods allow greater than 90% on-target editing of T-cell DNA, June said.
Intensifying the therapy may increase the number of chromosomal translocations, but that is not a major concern if the gene editing still avoids creating highly proliferative mutant cells, Hamilton noted.
Investigators at Memorial Sloan Kettering Cancer Center in New York have proposed a similar trial that could further boost the effectiveness of engineered T cells by more precisely inserting a CD19-directed CAR. Rather than using a lentivirus to infect the T cell and place the CAR randomly into the DNA, the alternative method uses CRISPR to knock out the native TRAC gene and an adeno-associated virus to place the CAR there. In a mouse study, the TRAC-CAR cells were more durable and demonstrated superior antitumor activity than conventional CAR T cells.7 In another preclinical study, investigators dispensed with viruses entirely, electroporating a linear piece of DNA into cells along with Cas9 RNPs.8
Several other in-human trials of CRISPRengineered T cells to treat cancer are under way. CRISPR Therapeutics, a Cambridge, Massachusetts—based firm cofounded by CRISPR pioneer Emmanuelle Charpentier, PhD, is treating patients in 2 trials evaluating allogeneic CRISPR/Cas9 gene-edited CAR T-cell therapies.9 A phase I/II study (NCT04035434) of the CD19-directed therapy CTX110 aims to recruit 95 patients with relapsed or refractory B-cell malignancies and non-Hodgkin lymphoma (NHL). A phase I study (NCT04244656) of the anti-BCMA agent CTX120 is enrolling up to 80 patients with relapsed or refractory multiple myeloma.
The company is also planning a trial of CTX130, a CAR T-cell therapy targeting CD70 for the treatment of solid tumors and hematologic malignancies, and has active trials of a hematopoietic stem cell therapy for sickle cell disease (NCT03745287) and transfusion-dependent β-thalassemia (NCT03655678). Its research areas include a treatment for type I diabetes and in vivo CRISPR-based therapies for cystic fibrosis and other rare genetic diseases.
Intellia Therapeutics, also based in Cambridge, is planning a trial next year of NTLA-5001, a CRISPR-engineered autologous T-cell therapy directed toward Wilms tumor 1 antigen, an overexpressed protein associated with acute myeloid leukemia.10 The company is working on a sickle cell treatment as well as in vivo therapies for hereditary angioedema and for transthyretin amyloidosis, a protein misfolding disorder that affects the heart and nerves.
Other pending trials include one recently posted by the Center for Cell and Gene Therapy at Baylor College of Medicine in Houston, Texas. The phase I study (NCT03690011), which is not yet recruiting, will evaluate a CRISPR-edited, CD7-directed CAR therapy with added CD28 to treat patients with T-cell acute lymphoblastic leukemia or lymphoma or NHL, according to the ClinicalTrials.gov website. Several other US companies and academic institutions are researching CRISPR-based cancer therapies but have not yet reached the clinical trial stage.
Uses for CRISPR Expand
Although the UPenn findings mark a milestone for CRISPR research in the United States, Chinese investigators have been testing the technology for several years. The first-ever in-human use of a CRISPRbased therapy occurred in 2016, when Chinese researchers knocked out the PD-1 gene on T cells to treat a patient with metastatic non—small cell lung cancer.11,12 There are currently at least 10 trials of CRISPR-based therapies in China, targeting mesothelin-positive solid tumors, CD19-positive leukemia or lymphoma, esophageal cancer, Epstein Barr virus—associated malignancies, HIV in patients with cancer, and other conditions such as pulmonary tuberculosis and sepsis, according to a search of ClinicalTrials.gov.
Meanwhile, in vivo use of CRISPR in humans, in which a full gene editing construct is injected into the body to modify somatic DNA, is just getting started. The first in vivo CRISPR-based clinical trial (NCT03872479), launched last year by Allergan and Editas Medicine, seeks to evaluate AGN-151587 (EDIT-101) in patients with Leber congenital amaurosis 10, a rare type of blindness caused by a genetic mutation. Labs around the world are conducting preclinical work on in vivo CRISPR therapies for cancer and many other diseases. June said it may take another decade of research before such therapies are feasible.
“The issue there is the efficiency of it and making sure it’s safe and doesn’t go germline, meaning it doesn’t go into sperm or eggs,” he said. “There are a lot of technical challenges, but the field’s been advancing very rapidly.”
CRISPR is also being used to facilitate development of cancer therapies and highly targeted precision medicine. The Broad Institute of MIT and Harvard is using CRISPR and other technologies to create a Cancer Dependency Map that profiles cancer cell line models for genomic information and sensitivity to genetic and small molecule perturbations.
CRISPR loss-of-function screens are used to knock out thousands of different genes in cell lines and measure which deletions lead to cell death, allowing investigators to determine which genes are essential for each cancer type and link them to tumor genetic or molecular features, said Jesse Boehm, PhD, scientific director of the Cancer Dependency Map and associate director of the Broad Cancer Model Development Center.
New data sets are regularly published on an open-access website, Depmap.org. Broad scientists have identified about 500 strong cancer targets, including 150 that may be targetable with existing drugs, and they have so far attempted to confirm 25 of those links through preclinical analyses, Boehm said.
In a study published last year, investigators used CRISPR and other technologies to screen for potential targets against malignant rhabdoid tumors (MRTs), which are aggressive cancers of the kidney, brain, and soft tissues that usually affect young children. They identified MDM2 and MDM4, canonical negative regulators of the tumor suppressor p53, as valid targets. They then demonstrated that 2 investigational drugs, idasanutlin (MDM2-specific) and ATSP-7041 (MDM2/4-dual) caused MRT regression in mice.13
Investigators and colleagues at the Broad Institute also used CRISPR screening data to identify WRN, which encodes a DNA helicase, as a synthetic lethal target in microsatellite unstable cancers, setting the stage for drug company research on compounds that can be combined with checkpoint inhibitors to treat colorectal tumors and other cancers.14
Boehm said his group would like to see their findings eventually inform therapeutic decision making in a manner similar to the way research has elucidated clinical pathways. “Ten years from now, we do aspire for the Dependency Map to be something that oncologists use to prescribe the right drug to the right patient, essentially achieving the promise of precision medicine,” he said. In the longer term, it may become possible to screen individual patient tumor tissue samples using CRISPR-based methods to determine the most effective therapy, Boehm added.
Another research effort seeks to better understand the variety of potential AEs from gene editing. With funding from the National Institutes of Health (NIH) Common Fund’s Somatic Cell Genome Editing program, investigators are studying human tissue models, delivery methods, reporter systems, and other technologies that facilitate testing of the effects of CRISPR/Cas9. Different groups in the NIH consortium are working with animal models, brain organoids, and T cells, among other systems.
Todd McDevitt, PhD, a senior investigator at Gladstone Institutes in San Francisco, California, is working on creating tissues from human stem cells to use as a substrate for testing CRISPR and other tools. McDevitt said in an interview that he is interested in knowing whether a certain volume of Cas9 proteins will cause cells to malfunction, for example, by disrupting ion regulation in cardiac tissue and affecting the heartbeat. Some of the potential effects cannot be tested in animal models, he said.
“Those are too different, so that’s why we’re trying to use some of the human models for those kinds of questions,” said McDevitt, who also is a professor of bioengineering at the University of California San Francisco. “Ultimately, a new product would have to go through several of these kinds of things in different systems to really test the effects well.”
The transduced cells persisted in the blood for up to 9 months after infusion, far longer than previously reported trials with NY-ESO-1—engineered T cells, which had half-lives in the blood of about 1 week, the investigators said. No patient experienced cytokine release syndrome or overt AEs attributed to the cell infusion.