Eric B. Kmiec, PhD
- The development of clustered regularly interspaced short palindromic repeats (CRISPR) and its CRISPR-associated protein (Cas9) has revolutionized the field of gene editing by enabling genetic spellchecking or genetic disruption at a frequency and degree of precision never believed possible.
- The potential for toxic responses or chemical interactions resulting from combination therapy has formed the foundation for the advancement of gene editing in cancer immunotherapy
- Although partial or complete ablation of CTLA-4, PD-1, and other genes associated with checkpoint blockade would help to alleviate the problem of variable responses encountered by traditional drug treatments in combination immunotherapy regimens, other challenges remain.
The capacity to modify the human genome as a way to develop therapeutic strategies for cancer has held a certain fascination for geneticists, molecular biologists, and molecular oncologists for quite some time. Although the reengineering of eukaryotic genomes has been possible with single-stranded modified oligonucleotides1,2 and complex meganucleases from yeast,3 it has only been within the last 4 years that the process of gene editing in real time has begun to emerge as a legitimate treatment option. The development of clustered regularly interspaced short palindromic repeats (CRISPR) and its CRISPR-associated protein (Cas9) has revolutionized the field of gene editing.4-6 These genetic tools enable genetic spellchecking or genetic disruption at a frequency and degree of precision never believed possible.
CRISPR (and Cas9) is derived from a natural activity found in microbes, which use these systems as a specific response pathway, generically known as adaptive immunity, to defend themselves against viral infection and re-infection.7-10 Technologic advances and reengineering have enabled scientists to capture, redesign, and implement this natural activity in human cells. The CRISPR component, as used in eukaryotic cells, consists of a specific guide RNA (sgRNA) component that directs a ribonucleoprotein complex to stably bind at a specific site on the DNA where a double-strand break is executed through the action of protein Cas9, the other component of the ribonucleoprotein complex. The induction of a double-strand break in human cells activates the DNA damage response pathway(s), and, as part of that response, the broken chromosomal DNA is rejoined, albeit with some loss of DNA at the cleavage site. This DNA fraying is caused most often by the process of nonhomologous end joining, which is a natural survival response in mammalian cells to DNA damage in order to avoid aberrant chromosomal segregation at mitosis.
Although these processes can be mutagenic, and they often are in the natural response of cells to anticancer drugs such as camptothecin or cisplatin, this nonspecific molecular process can lead to cell death. However, if such DNA breakage and resection activity could be harnessed and precisely directed at targeted genes, as opposed to acting at random on chromosomal DNA, important lessons about the roles of those genes and the biological pathways in which they function can be ascertained. That is exactly what has been developing during the past 3 to 4 years; what was once a complex and lengthy process for the generation of a knockout cell line, a knockout mouse, etc., has now become an almost routine laboratory exercise, in large part, because of the domestication of the CRISPR/Cas9 system.
CRISPR/Cas9 can also be used to create a double-strand break at a site within the genome that harbors a mutation, often in the form of a single-base error or a multiple-base pair deletion, by adding a donor DNA repair template to the CRISPR/Cas9 cleavage toolkit.11-14 For the purposes of this commentary, however, we will not discuss homology-directed repair because a number of reviews have recently been published.15 Furthermore, molecule-directed repair is not presently part of the design for the major clinical applications of gene editing in oncogenesis.
Gene Editing and Cancer Immunotherapy
One of the most dramatically expanding fields in cancer research is cancer immunotherapy, in which some focus has centered on the generation of CAR-T cells. These cells express tumor-targeting receptors and, when tested in model systems, have shown promise for the treatment of various leukemias and lymphomas.16 CAR cells often bear a single-chain variable fragment on their surface that recognizes a highly expressed antigen located on the surface of tumor cells; a signaling domain that induces T-cell killing is activated upon binding to the receptor on the tumor cell. T-cell–mediated killing of tumor cells is now an appealing strategy for personalized medicine, and trials are underway (NCT01601313 and NCT02030832).
Although the idea has scientific merit and initial results are promising, the generation of the appropriate number of CAR-T cells is an expensive and time-consuming process. This, in large part, is because the T cell used as the therapeutic agent must be isolated from the patient, modified outside the body, and reintroduced into the patient. Although this process and a route of administration has been clinically achievable, the extent of T-cell expansion required to generate sufficient numbers of activated cells for clinical impact remains quite challenging.
As such, work is underway to generate a compiled population of CAR-T cells that can be used in a wide variety of patients. Such an advance would be viewed as an important breakthrough, but it is believed that such a universal population could still pose problems, such as graft-versus-host reactions/disease. CRISPR/Cas9 gene editing tools designed to have a higher degree of precision and efficiency have enormous potential in CAR-T cell therapeutics and cancer immunotherapy.
By taking a step back and viewing the field in totality, we see that cancer immunotherapy is beginning to be introduced into the clinic and is gaining significant traction through two fundamental approaches. The first is passive immunotherapy, in which antitumor antibodies or cytotoxic T cells (and natural killer cells, often modified in some fashion) are introduced into the patient. The second is more of an active immunotherapy that develops through the stimulation of the immune cells via checkpoint blockade, most often through the use of antibodies directed against immune regulatory checkpoint molecules expressed on T cells. By averting the checkpoint blockade, T cells are able to overcome negative signals that block their function and expansion.
Gene editing has been used to knock out a series of endogenous genes in T cells that are known to activate or participate in graft-versus-host disease.17 In some cases, this work was deemed successful and has led to a variation of the use of gene editing in cancer immunotherapy. A CRISPR/Cas9 system now can be designed to knock out genes encoding T-cell inhibitory receptors and even ligand signal molecules that act to suppress the capacity of T cells to kill the tumor cells.
Two of the genes central to the control of such expansion encode CTLA-4 and PD-1.18 PD-1 is often chosen as a universal target for immune checkpoint inhibitors that block PD-1 interaction (on lymphocytes) with antigen-presenting cells and tumor cells. In non–small cell lung cancer, in which PD-1 has been inhibited, significant clinical benefit has been observed.19
A number of PD-1– and PD-L1–blocking agents have been generated and even have been used in combination with CTLA-4 inhibitors for the treatment of melanoma.20,21 Although the combination of PD-1 and CTLA-4 antagonists has improved response rates, an increase in the associated toxicity with this particular combination has been concurrently observed.22 As reported by Mahoney et al.,22 the rationale for combining CTLA-4 and PD-1 blockers remains strong because both CTLA-4 and PD-1 are expressed on lymphocytes and because these two pathways have different mechanisms for inhibiting the functions of these cells.23-25 This type of therapeutic strategy requires critical evaluation because multiplexing specific chemotherapeutic agents, including targeted therapies, can lead to an elevated and complex toxicity profile.
However, it is important to note that combination therapy with immune checkpoint inhibitors can lead to a synergistic effect of each chemotherapeutic agent. Recent data suggest that certain kinds of chemotherapy protocols actually may induce a priming of the immune system and may enable cells to more efficiently respond to immune checkpoint inhibitors.26 Thus, combination therapy may act in a way that essentially sets up or potentiates the microenvironment(s), thereby providing a more fertile ground for the activity of a chemotherapeutic agent—in this case, an immune checkpoint inhibitor. Under these circumstances, the inhibitor can act more efficiently, which can lead to a direct reduction in tumor burden.
Therefore, combination therapy may require an exquisite balance between achievement of a significant degree of efficacy and minimization of a significant degree of toxicity. This type of balancing act is particularly important when two traditional drugs or immunotherapy-based treatments are used, and caution clearly is warranted in the establishment of trials for combination therapies. It is also important to remember that PD-1 and CTLA-4 act at different positions within the T-cell differentiation cascade, and complete ablation of each may need to be temporally regulated to achieve maximum efficacy. A new review by Hoos27 provides a methodical evaluation of next-generation immuno-oncologic drugs.
A New Generation of Combination Therapy and a Word of Caution
The potential for toxic responses or chemical interactions has formed the foundation for the advancement of gene editing in cancer immunotherapy. By using a genetic reengineering strategy, toxicity or drug interactions could be avoided. Toward this end, the National Institutes of Health Recombinant DNA Advisory Committee approved a clinical trial at the University of Pennsylvania in which CRISPR/Cas9 will be used to knock out the PD-1 gene and the endogenous T-cell receptor gene for the treatment of melanoma (Protocol #1604-1524). PD-1 also is the target of a clinical trial in China focused on metastatic non–small cell lung cancer. There, it has been reported that patients have received initial treatments, which have gone smoothly, and that one patient will receive a second injection. Details of safety and efficacy have not been fully reported. The clinical plan, however, aims to treat 10 to 12 patients, and each will receive two, three, or four injections; monitored evaluation will last 6 months or longer.28
As of this writing, additional trials in China are planned: Investigators at Peking University, in Beijing, plan to initiate three trials that use CRISPR-based strategies against bladder, prostate, or renal cell cancers.28 Could these gene-editing tools also lead to the development of the universal T cell described above that not only retains aggressive killing activity but also fails to induce significant anti-immune responses? Because of the efficiency with which these tools work, it is likely that possibility will become a reality.
Although partial or complete ablation of CTLA-4, PD-1, and other genes associated with checkpoint blockade would help to alleviate the problem of variable responses encountered by traditional drug treatments in combination immunotherapy regimens, other challenges do remain.
For example, a large number of new sources of Cas9 may act less aggressively when carrying out DNA cleavage. These analogs are sought because worries of off-site mutagenesis directed by indiscriminate DNA breaks by Cas9 still remain vexing. It should be noted, however, that many of the viral gene therapy approaches, particularly those that use retroviruses, also produce insertional mutagenesis at indiscriminate sites. Many of these trials have been approved by the U.S. Food and Drug Administration and have had varying levels of success or failure.
With regard to the ablation of the PD-1 gene and others, this issue must be considered because the genetic surgery required to disrupt genes that interfere with T-cell killer activity must be solely targeted (Figure). If the targeting is not precise and if other genes are disrupted to any measurable extent, then any positive effect that gene editing could have on cancer immunotherapy could quickly become a process of diminishing returns.
Finally, it is well recognized that many of the patients who receive a transfer of genetically modified T cells will, in fact, develop resistance, succeeded by immune escape. Thus, a cycle of informed treatment could be required even when genetic disruption or total ablation of gene function is successfully attained. However, the incorporation of this remarkable genetic tool to the innovative world of cancer immunotherapy is likely to lead to the emergence of clinical strategies that either will yield effective cures or at least will contribute to the transformation of cancer into a manageable chronic disease.
About the Author: Dr. Kmiec is the director of the Gene Editing Institute, which is part of the Center for Translational Cancer Research at the Helen F. Graham Cancer Center & Research Institute, positioned within the Christiana Care Health System.