Clinical CAR-T Cell and Oncolytic Virotherapy for Cancer Treatment

Abstract

Immunotherapy has recently garnered success with the induction of clinical responses in tumors, which are traditionally associated with poor outcomes. Chimeric antigen receptor T (CAR-T) cells and oncolytic viruses (OVs) have emerged as promising cancer immunotherapy agents. Herein, we provide an overview of the current clinical status of CAR-T cell and OV therapies. While preclinical studies have demonstrated curative potential, the benefit of CAR-T cells and OVs as single-agent treatments remains limited to a subset of patients. Combinations of different targeted therapies may be required to achieve efficient, durable responses against heterogeneous tumors, as well as the microenvironment. Using a combinatorial approach to take advantage of the unique features of CAR-T cells and OVs with other treatments can produce additive therapeutic effects. This review also discusses ongoing clinical evaluations of these combination strategies for improved outcomes in treatment of resistant malignancies.

Chimeric antigen receptor modified T (CAR-T) cell and oncolytic viruses (OVs) are promising cancer immunotherapy platforms. Herein, Watanabe and colleagues review how these cancer gene therapy agents achieve clinical responses in tumors traditionally associated with poor outcomes and discusses emerging combinatorial strategies with radiotherapy/chemotherapy and other immunotherapies to improve clinical outcomes.

Main Text

In 2018, more than 18 million people were diagnosed with cancer, and 9.5 million died from cancer-related causes globally.¹ Tumors are complex, with dynamic and mutualistic interactions between cancer cells and their surrounding cells, which contribute to progression, metastasis, and treatment resistance. The immune system can recognize and respond to tumors both naturally and following therapeutic intervention with either pro-tumor or anti-tumor functions.^2,3 The discovery of the immunoinhibitory pathways upregulated in parallel with T lymphocyte activation uncovered powerful mechanisms by which both solid and hematologic tumors evade immune attack. However, to date, immune checkpoint therapies benefit only a subset of patients, with overall clinical response rates varying widely across cancer types.^4,5 Given the heterogeneity of tumor types, it seems likely that even more complex combinations of immunomodulatory agents may be required to obtain consistent, durable therapeutic responses against a broad spectrum of cancers.

The term “gene therapy” encompasses a wide range of treatment types that all use genetic material to modify cells, whether in vitro or in vivo. Cancer gene therapy is a method to deliver genetic materials into target cells or tissues and to express certain genes for therapeutic effect. Two types of cancer gene therapy have emerged as particularly effective during the past decades: adoptive cell therapy and oncolytic virotherapy. At least in principle, a single treatment of these agents might achieve durable and potentially curative clinical responses. In this review, we discuss current clinical progress using both types of therapy (for adoptive cell therapy, we specifically discuss chimeric antigen receptor (CAR)-modified T cells [CAR-T cells]) and how each might be particularly beneficial for different types of cancer. We also highlight the potential therapeutic relevance of combining regimens.

Introduction to CAR-T Cells

CARs confer antigen specificity to immune cells and have been tested in multiple clinical trials for the treatment of various cancers. Unlike traditional T cell receptors, CARs can target surface antigens expressed on cancer cells in a major histocompatibility complex (MHC)-independent manner. Currently, most CAR-modified cell therapies use αβT cells as a platform. These T cells are stimulated and expanded from peripheral blood mononuclear cells obtained from patients to create an autologous T cell product. In most cases, CAR transgenes are introduced into activated T cells by retroviral or lentiviral transduction.

Structurally, CARs can be divided into four regions. The first is the antigen binding domain, which is usually derived from a single-chain variable fragment (scFv) of an antibody. This region is followed by a spacer or hinge region, which is then linked to a transmembrane domain and, finally, an intracellular domain that transmits signals into CAR-modified cells. The intracellular domain could be further divided into costimulatory domains (e.g., CD28, 4-1BB) and a signaling domain (CD3ζ). Clinical and preclinical evidence demonstrates that each of these regions contributes to overall CAR-T cell function (Table 1); thus, the structure of CARs must be well conceived in order to maximize the function of CAR-expressing effector cells.^6,7

Table 1.

Impact of CAR Design on CAR-T Cell Function

Domain		Property	Consequence	Ref.
scFv		affinity	a high-affinity scFv against FRβ showed higher efficacy against AML	8
			a low-affinity scFv against ErbB2 or EGFR targeted tumor cells while sparing normal tissues with physiological target expression levels	9
		tonic signaling	accelerate exhaustion	10
Spacer		length	affect efficacy of antigen recognition depending on scFv and the location of epitope
			IgG1 hinge > hinge-CH2-CH3 (target CD22: clone HA22)	11
			IgG4 hinge > hinge-CH3 > hinge-CH2-CH3 (ROR1: 2A2)	12
			IgG4 hinge-CH2-CH3 > hinge-CH3 = hinge (ROR1: R11)	13
			IgD hinge-CD28 > CD28 (MUC1: SM3)	14
			IgG2 hinge-CH2-CH3 = hinge-CH3 > hinge (PSCA: 2B3)	15
		Fc-FcγR interaction	Fc portion (especially CH2) present in IgG-derived spacers interacts with FcγR-expressing cells, resulting in CAR antigen-independent activation	13,15, 16, 17, 18
		tonic signaling	accelerate differentiation	15
Spacer + TM		AICD	reduced AICD and cytokine release in CD8-derived spacer/TM compared to CD28-derived spacer/TM	19
		threshold for CAR reactivity	CD19CAR 4-1BB with CD28-hinge/TM recognized target cells with lower antigen density compared to the one with CD8-hinge/TM	20
TM		stability	CD8 and CD28 TMs increase surface expression of CAR compared to CD3 TM	21
			ICOS costimulatory domain requires ICOS TM domain to maximize its benefit	22
ICD	costimulatory domain	killing/persistence	CD28 costimulatory domain induces rapid expansion and robust cytotoxicity while 4-1BB contributes to sustained T cell persistence for the CD19CAR	23
		metabolism	CD28 costimulatory domain induces the effector memory phenotype and preferentially utilizes glycolysis while 4-1BB skews to the central memory phenotype with enhanced mitochondrial biogenesis	24
	signaling domain	persistence	mutation into two membrane-distal ITAM domains in CD3ζ enhanced 19–28z CAR-T cell persistence, maintaining a stem-like memory T cell phenotype	25
		threshold for CAR reactivity	CD19CAR 4-1BB with duplication of CD3ζ enhanced CAR reactivity against antigen low density while CD19CAR CD28 with two membrane-distal ITAM deletion in CD3ζ increased the antigen density threshold	20

scFv, single-chain variable fragment; TM, transmembrane; ICD, intracellular domain; AICD, activation-induced cell death.

Clinical Results and Current Trends of CAR-T Cell Therapy for B Cell Malignancies

To date, CD19 CAR-T cells to treat B cell malignancies have been the most widely used adoptive CAR-T cell therapy and have produced remarkable clinical responses. These successes resulted in US Food and Drug Administration (FDA) approval of two CD19 CAR-T cell products, Kymriah and Yescarta. Results from multiple clinical trials with these therapies demonstrated the emergence of antigen-negative relapse after CD19 CAR-T cell treatment, which occurs with particular frequency in B cell acute lymphoblastic leukemia (B-ALL).

Park et al.²⁶ reported complete remission (CR) in 83% (44/53; 32 minimal residual disease [MRD]-negative, 9 MRD-positive, 3 unknown MRD status) of adult B-ALL patients treated with second-generation CD19 CAR-T cells (19–28z). All 9 patients who had MRD-positive CR after CAR-T cell therapy relapsed with CD19⁺ blasts, and 16 of the 32 patients with MRD-negative CR relapsed. Of this latter group of patients, four had CD19-negative relapses. In addition, Maude et al.²⁷ treated pediatric B-ALL patients with second-generation CD19 CAR-T cells (19BBz) and reported 81% (61/75) of this group had MRD-negative CR. Of note, 15 out of 22 patients who relapsed after CR had CD19 antigen-negative disease. Importantly, the former trial included the patients in first CR with <5% blast in bone marrow in addition to refractory, relapsed, or MRD patients, while the latter trial only included relapsed patients with ≥5% blasts in bone marrow. We have summarized results from these and other clinical CD19 CAR-T cell studies for B-ALL in Table 2.

Table 2.

Selected CAR-T Cell Clinical Trials

CD19 CAR Targeting B-ALL
Institution	Patient Population	Vector	scFv	Spacer	TM	ICD	No. Patients	CR No. (%)	MRD− No.	Relapse after CR No. (%)	CD19− relapse No (%)	Ref.
MSKCC	adult	retro	SJ25C1	CD28	CD28	CD28	53	44 (83)	32	25/44 (57)	4/25 (16)	26
U. Penn	pediatric	lenti	FMC63	CD8	CD8	4-1BB	75	61 (81)	61	22/61 (36)	15/22 (68)	27
Seattle CRI/UW	pediatric	lenti	FMC63	IgG4 short	CD28	4-1BB	45	40 (93)	40	18/40 (45)	7/18 (39)	28
NCI	pediatric	retro	FMC63	CD28	CD28	CD28	20	14 (70)	12	2/14 (14)	2/2 (100)	29
U. Penn	adult	lenti	FMC63	CD8	CD8	4-1BB	35	24 (69)	19	n.d	n.d	30

CD19 CAR Targeting B-Lymphoma
Drug Name (Clinical Trial)	Patient Population	Vector	scFv	Spacer	TM	ICD	No. Patients	ORR No. (%)	CR No. (%)	Ref.
Axicabtagene ciloleucel (ZUMA-1)	adult	retro	FMC63	CD28	CD28	CD28	101	84 (76)	59 (54)	31
Tisagenlecleucel (JULIET)	adult	lenti	FMC63	CD8	CD8	4-1BB	93	75 (82)	37 (40)	32
Lisocabtagene maraleucel (TRANSCEND)	adult	lenti	FMC63	IgG4 short	CD28	4-1BB	255	186 (73)	135 (53)	33

BCMA CAR Targeting MM
Institution	Vector	BCMA Targeting Domain	Spacer	TM	ICD	No. Patients	ORR No. (%)	Stringent CR No. (%)	CR No. (%)	VGPR No. (%)	PR No. (%)	Ref.
NCI	retro	scFv 11D5-3 (murine)	CD8	CD8	CD28	16 (high dose)	13 (81)	2 (13)	n.d.	8 (50)	3 (19)	34
Multicenter	lenti	scFv 11D5-3 (murine)	CD8	CD8	4-1BB	33	28 (85)	12 (36)	3 (9)	9 (27)	4 (12)	35
U. Penn	lenti	scFv clone 10 (human)	CD8	CD8	4-1BB	25	12 (48)	1 (4)	1 (4)	5 (20)	5 (20)	36
Xi’an Jiaotong University	lenti	two single domain antibodies VHH1 + VHH2	CD8	CD8	4-1BB	57	50 (88)	n.d.	39 (68)	3 (5)	8 (14)	37

ALL, acute lymphoblastic leukemia; MM, multiple myeloma; scFv, single-chain variable fragment; TM, transmembrane; ICD, intracellular domain; CR, complete response; MSKCC, Memorial Sloan Kettering Cancer Center; retro, retroviral; lenti, lentiviral; U. Penn, University of Pennsylvania; n.d., not determined; MRD, minimum residual disease; ORR, overall response rate; VGPR, very good partial response; PR, partial response.

CD19 CAR-T cells also have produced clinical benefit against B-lymphoma (Table 2). However, sample collection for this disease is more challenging than for B-ALL, and thus the status of CD19 antigen expression at the time of relapse has been less clear. Nevertheless, loss of CD19 expression has been reported in some B-lymphoma patients following successful treatment with CD19 CAR-T cells similar to that seen in B-ALL patients.38, 39, 40

The mechanisms of antigen-negative relapse vary, including (1) outgrowth of a CD19-negative population;^41,42 (2) genetic mutation;43, 44, 45 (3) lineage switch of leukemic cells from a lymphoid to myeloid lineage;46, 47, 48 (4) contamination of CAR-expressing B cells in the infused products;⁴⁹ and (5) trogocytosis.⁵⁰ Therefore, multiple current efforts are underway to prevent target antigen-negative relapse and improve CAR-T cell therapy against B-ALL and B-lymphoma.

One approach to overcome this problem is to target multiple antigens expressed on cancer cells. In a phase I dose-escalation clinical trial reported by Fry et al.,⁵¹ CAR-T cells targeting CD22, which is widely expressed on B lineage cells, were infused into 21 B-ALL patients with (17/21) or without (4/21) prior CD19-directed immunotherapy. Of the group who received prior immunotherapies, 15 had been treated with CD19 CAR-T cells, and 8 of these patients were negative for CD19 expression at the time of CD22 CAR-T cell infusion. Seven of nine patients with prior CD19 CAR-T cell treatment who were infused with ≥1 × 10⁶ CD22-CARs/kg achieved CR, including four out of four patients with CD19-negative B-ALL who received this dose level. The group recently updated results from their clinical trial⁵² to show that, overall, CD22 CAR-T cells induced CR in 40 out of 57 evaluable patients with 35 MRD-negative CRs. Of note, 51 of these patients had prior CD19-targeted therapy, and CD19 was negative in 33 patients at the time of CD22 CAR-T cell infusion. Thirty patients experienced relapse, and the majority of these relapses were CD22-negative/dim. In another early phase clinical trial, Dai et al.⁵³ reported the infusion of CAR-T cells targeting both CD19 and CD22 with a tandem CAR into B-ALL patients. All six patients treated with these bispecific CAR-T cells achieved MRD-negative CR. Although long-term follow-up studies are needed, three of these patients relapsed, including one patient with CD19-negative/CD22-dim disease. Currently, several clinical trials targeting multiple antigens are ongoing for the treatment of B-ALL and B-lymphoma (https://clinicaltrials.gov/).

CAR-T cell therapy has also been successful in treating multiple myeloma (MM). Although MM is a B cell lineage malignancy, most CAR-T cell studies for MM have targeted B cell maturation antigen (BCMA), not CD19, since MM cells are often CD19-negative (Table 2). However, similar to CD19 CAR-T cell trials, BCMA-negative/dim relapses have been observed in some cases.^34,36 Therefore, several ongoing clinical trials for MM are simultaneously targeting multiple antigens. Despite the frequency of CD19-negative MM cells, since CD19 expression was found in an MM stem cell subset,⁵⁴ most of these trials are additionally targeting CD19. Yan et al.⁵⁵ reported results of a phase II clinical trial of CAR-T cell therapy for MM targeting CD19 and BCMA. In this trial, 21 patients were co-infused with both CD19 CAR (41BBz) and BCMA CAR (41BBz) T cells. The overall response rate (ORR) was 95% (20/22), including 9/22 with stringent CR (43%), 3/22 with CR (14%), 5/22 very good partial remission (PR) (24%), and 3/22 with PR (14%). Although one patient relapsed, the status of antigen expression was not reported. In addition to co-targeting CD19, CARs co-targeting alternative plasma cell antigens including CD38 (ClinicalTrials.gov: NCT03767751) and CD138 (ClinicalTrials.gov: NCT03196414) also are underway.

While CAR-T cell clinical trials for B-ALL, B-lymphoma, and MM showed benefit, ORRs in trials against B cell chronic lymphocytic leukemia (B-CLL) have been unsatisfactory.56, 57, 58, 59 The lack of responses in B-CLL patients could be explained by intrinsic characteristics of their T cells, such as an exhausted phenotype of CD8⁺ T cells,⁶⁰ less expansion of naive-like T cells,⁶¹ and aberrant metabolic reprogramming of T cells.⁶² These phenomena led to the initiation of combinatorial therapy for B-CLL with both CAR-T cells and immune checkpoint inhibitors (ICIs), as described in CAR-T Cells and Other Immunotherapies.

Challenges of CAR-T Cells Targeting Other Malignancies

Besides targeting lymphocytic malignancies, CAR-T cells also have been developed to target myeloid malignancies such as acute myeloid leukemia (AML) and myelodysplastic syndromes (MDSs). Numerous clinical trials targeting myeloid markers such as CD33 and CD123 are ongoing (https://clinicaltrials.gov/). To date, limited data from these trials have been published.63, 64, 65 Since most target antigens for myeloid diseases are also expressed on normal myeloid and hematopoietic stem cells, the potential incidence of on-target/off-tumor toxicity is one common concern for CAR-T cell therapy for these malignancies, especially therapies targeting CD33 and CD123. While B cell aplasia caused by CD19 CAR-T cells is manageable, the potential toxicities associated with these target antigens are significantly riskier. In fact, Gill et al.⁶⁶ and Kenderian et al.⁶⁷ showed that while CD123 and CD33 CAR-T cells have potent cytotoxicity, they also cause significant myelotoxicity in humanized mice. Therefore, research is focused on increasing the safety of these CAR-T cells by utilizing either (1) a safety switch to deplete CAR-T cells in vivo, if needed, by co-expressing with truncated epidermal growth factor receptor (EGFR) (ClinicalTrials.gov: NCT02159495 and NCT03114670), herpes simplex virus thymidine kinase (HSV-TK)⁶⁸ (ClinicalTrials.gov: NCT04097301), or the inducible caspase-9 (iCasp9);⁶⁹ or (2) mRNA-encoded CARs that would allow T cells to only transiently express the CAR to prevent long-term myelotoxicity.⁷⁰

Despite the clinical success of CAR-T cells targeting hematological malignancies, clinical outcomes of CAR-T cell therapy targeting solid tumors remain poor.⁷¹ Solid tumors are more challenging to target with CAR-T cells in part due to tumor antigen heterogeneity, with a wide range of expression levels that leads to the outgrowth of antigen-negative populations under pressure from single-antigen targeting CAR-T cell attack. This antigen escape was observed in clinical trials targeting EGFRvIII⁷² and interleukin (IL)-13Rα2⁷³ for glioblastoma. As mentioned before, targeting two or more tumor antigens simultaneously using multiple CARs74, 75, 76 or tandem CARs⁷⁷ is one strategy to overcome this problem. Recently, other groups have targeted multiple antigens for solid tumor using CAR-T cells secreting bispecific T cell engager (BiTE)⁷⁸ or antibody mimic receptors (amRs)⁷⁹ in preclinical models. None of these strategies for solid tumors has yet been tested in the clinic.

In addition to heterogeneity of tumor antigen expression, the hostile solid tumor microenvironment (TME) is another barrier to successful CAR-T cell therapy.⁸⁰ The TME is made up of multiple immune inhibitory cells as well as extracellular components such as immune inhibitory cytokines and extracellular matrix (ECM). A number of strategies have been explored to enhance CAR-T cell function by targeting each component of the TME. For instance, to target the ECM, Caruana et al.⁸¹ engineered CAR-T cells to secrete heparanase, which degrades heparan sulfate proteoglycans (HSPGs), a component of the ECM. The authors demonstrated increased infiltration and enhanced anti-tumor effect of their heparanase-secreting CAR-T cells compared to control CAR-T cells in a preclinical neuroblastoma and melanoma xenograft mouse model. In addition, to maintain CAR-T cell activity within the TME, multiple groups are targeting inhibitory molecules including membrane proteins (largely focused on the interaction between PD-1 and PD-L1) and soluble factors. Some of these strategies include modifying CAR-T cells to (1) express dominant negative receptor of PD-1⁸² or transforming growth factor β (TGF-β)⁸³ to neutralize their inhibitory effect, (2) secrete anti-PD-1 scFv to block the PD-1/PD-L1 pathway,^84,85 or (3) knock out inhibitory receptors.86, 87, 88, 89 Moreover, to further enhance the function of effector cells in the TME, CAR-T cells have been engineered to express a switch receptor, PD-1/CD28, which converts negative signals from PD-1 to immunostimulatory signals through the CD28 endodomain.⁹⁰ Similarly, inhibitory signaling through immunosuppressive cytokines such as IL-4 and TGF-β1 can be inverted into immunostimulatory signals by introducing chimeric receptor with extracellular cytokine receptor domains fused to immunostimulatory endodomains.91, 92, 93, 94

Alternatively, CAR-T cells have been modified to secrete immunostimulatory cytokines to overcome the hostile TME and augmenting their function. For example, IL-12-secreting CAR-T cells improved anti-tumor responses by modifying the TME in preclinical models of colorectal adenocarcinoma,^95,96 melanoma,⁹⁶ sarcoma,⁹⁶ and ovarian cancer.^97,98 CAR-T cells modified to secrete other cytokines, including IL-15,99, 100, 101 IL-18,^102,103 and IL-23,¹⁰⁴ have also been evaluated in preclinical solid tumor models in which they augmented CAR-T cell function. Some of these strategies have been tested in the clinic (ClinicalTrials.gov: NCT02498912 and NCT03721068), although the results have not been reported.

Combinatorial Treatment with CAR-T Cells

As discussed, CD19 CAR-T cells for the treatment of hematological malignancies have demonstrated remarkable responses in many patients with B-ALL. However, those same achievements have not translated well to other types of lymphomas and leukemias or solid tumors. In addition to antigen escape, lack of T cell persistence and T cell dysfunction or exhaustion contribute to mechanisms of CAR-T cell failure. Therefore, many strategies have been proposed to enhance the efficacy of CAR-T cells by improving tumor infiltration, persistence, or overall function.¹⁰⁵ Chemotherapy and radiotherapy are the mainstays of cancer treatment and in combination with CAR-T cells can help modulate the TME to eliminate immunosuppressive cells and improve T cell persistence.¹⁰⁶ With the objective of improving CAR-T cell efficacy in the hostile TME, recent clinical efforts have also combined CAR-T cells with immune checkpoint blockade (ICB) to reduce T cell exhaustion.

CAR-T Cells and Chemotherapy

The first-in-human study to evaluate chemotherapy in combination with adoptive cell therapy occurred more than a decade ago. This study showed that lymphodepletion with fludarabine prior to antigen-specific T cell infusion improved T cell persistence and increased plasma levels of homeostatic cytokines IL-7 and IL-15.¹⁰⁷ This study and other preclinical evaluations of chemotherapy in combination with CAR-T cell infusion led to clinical investigation of lymphodepleting regimens before CAR-T cell treatment.¹⁰⁸ Importantly, two clinical trials investigating the effects of CAR-T cells targeting CD30 in lymphoma patients compared response rates with and without lymphodepletion (ClinicalTrials.gov: NCT01316146 and NCT02917083). No dose-limiting toxicities were observed in the CD30 CAR-T trial, but responses were limited with three CR, three stable disease (SD), and three progressive disease (PD) patients.¹⁰⁹ Subsequently, the RELY-30 trial added cyclophosphamide to fludarabine (Cy/Flu) for 3 days prior to CD30 CAR-T cell infusion, resulting in a peak expansion of CAR-T cells 1–2 weeks after treatment at 1–2 logs higher than without lymphodepletion. Current reports of RELY-30 show six of nine patient CRs with grade 1 cytokine release syndrome (CRS), which is suggested to be a result of lymphodepletion, but overall combination therapies appear safe.^110,111

Turtle et al.¹¹² reported improved response in relapsed and/or refractory CD19⁺ non-Hodgkin’s lymphoma (NHL) patients who received Cy/Flu (50% CR, 72% ORR) versus fludarabine only (8% CR, 50% ORR). Patients who received Cy/Flu had improved T cell expansion and persistence along with a reduced immune response to murine scFv. Investigators also reported severe cytokine release syndrome in 13% of patients receiving chemotherapy (ClincialTrials.gov: NCT01865617). Patient’s with aggressive NHL showed an ORR of 51% with 40% CR with CD19 CAR-T cells when receiving high-intensity cyclophosphamide and fludarabine, and these positive outcomes were related to a favorable cytokine profile.¹¹³

With current efforts to improve CAR-T cell efficacy in solid tumors, studies targeting HER2 (human epidermal growth factor receptor 2)-positive sarcomas reported enhanced T cell expansion and persistence with a second-generation HER2-specific CAR-T cell after 2 days of Cy/Flu followed by an additional 3 days of fludarabine.¹¹⁴ Preliminary results in a 10-patient trial (ClinicalTrials.gov: NCT00902044) reported 60% overall survival 1 year after HER2.CAR-T cell infusion, with eight patients experiencing grade 1/2 cytokine release syndrome that was resolved after supportive care.¹¹⁵ Historically, 3-year survival for pediatric patients with metastatic osteosarcoma is 20%.¹¹⁶

Currently, lymphodepleting chemotherapy is generally not used to target malignant cells in combination with CAR-T cells, but rather the studies discussed above have confirmed that lymphodepletion prior to CAR-T cell infusion improves adoptively transferred T cell expansion and persistence, leading to the majority of trials adopting this regimen. Chemotherapy before CAR-T cell infusion has led to indications of severe cytokine release syndrome in a few patients, but most have been controlled. Additionally, chemotherapy has reduced immunosuppressive cells in the TME and improved tumor antigen expression, which has contributed to synergistic benefits between this conventional therapy and CAR-T cells.¹¹⁷

CAR-T Cells and Radiotherapy

Radiation therapy (RT) is a standard of care treatment used to stimulate tumor-specific immunity for many solid tumors. The benefits of RT stem from direct tumor cell apoptosis and release of antigen and damage-associated molecular patterns (DAMPs) that prime the immune system to target both local and distant tumor sites. Clinical studies have combined RT with CAR-T cells to help de-bulk the primary tumor and improve T cell trafficking.¹¹⁷ In a phase II trial for relapsed/refractory diffuse large B cell lymphoma (DLBCL) investigators compared CD19 CAR-T cells with either chemotherapy (C-CAR) or radiotherapy (R-CAR) (ClinicalTrials.gov: NCT03196830). Patients receiving R-CAR showed a higher ORR at 100% compared to 25% in the C-CAR cohort and no indications of neurotoxicity, and less severe cytokine release syndrome was observed in R-CAR recipients while 75% of C-CAR recipients exhibited neurotoxicity and 100% showed signs of severe cytokine release syndrome.¹¹⁸ In another clinical study, no patients treated with radiotherapy 30 days prior to CD19 CAR-T cell infusion experienced any grade 3 or higher cytokine release syndrome or toxicities attributed to CAR-T cell treatment.¹¹⁹ While there was no significant improvement in the ORR in patients who received induction RT prior to CAR-T cells, results suggest that this combination can potentially improve the safety and reduce toxicities. Five of 19 patients who did not receive RT experienced grade 3 or higher cytokine release syndrome, suggesting a benefit of RT reduced toxicity, but this is a limited finding and further studies in different tumor settings are required to confirm this combination strategy.¹²⁰

CAR-T Cells and Other Immunotherapies

The discovery of ICIs to improve T cell antitumor activity revolutionized the field of immunotherapy. Immune checkpoints can occur through CTLA-4/CD80/86, PD-1/PD-L1, TIM-3/HMGB1, and LAG-3/MHC class II binding.¹²¹ Studies have emerged combining ICIs with CAR-T cells to enhance the functionality of both adaptive and adoptive T cells and promote a more robust immune attack (Table 3). Indeed, CAR-T cell-treated patients often relapse with CD19⁺ tumors due to reduced function of therapeutic T cells.¹²² To attenuate T cell exhaustion, investigators designed a study to deliver ICIs for PD-1 blockade 14 days after CAR-T cell infusion. Three patients re-established B cell aplasia after checkpoint blockade, indicating CAR-T cell function.¹²³ A clinical study from the same institution is also testing pembrolizumab (PD-1 ICI) in lymphoma patients who failed to respond or relapsed after CD19 CAR-T cell treatment (ClinicalTrials.gov: NCT02650999). Additional trials also are investigating PD-1/PD-L1 blockade in combination with CD19-specific CAR-T cells (ClinicalTrials.gov: NCT02706405, NTC02926833, NCT03310619, and NCT00586391).

Table 3.

Selected Clinical Trials of CAR-T Cells and ICIs

Sponsor	Phase	Target	Trial Title	CAR	ICI	NCT No.: ClinicalTrials.gov	Status
City of Hope Medical Center	I	glioma	IL13Ralpha2-targeted chimeric antigen receptor (CAR) T cells with or without nivolumab and ipilimumab in treating patients with recurrent or refractory glioblastoma	IL13Ralpha2	nivolumab, ipilimumab	NCT04003649	recruiting
University of Pennsylvania	I	glioblastoma	phase I study of EGFRvIII-directed CAR T cells combined with PD-1 inhibition in patients with newly diagnosed, MGMT-unmethylated glioblastoma	EGFRvIII	pembrolizumab	NCT03726515	active
University of Pennsylvania	I/II	DLBCL; FL; MCL	pembrolizumab in patients failing to respond to or relapsing after CAR T cell therapy for relapsed or refractory lymphomas	CD19	pembrolizumab	NCT02650999	active
Fred Hutchinson Cancer Research Center	I	DLBCL; mediastinal (thymic) large B cell lymphoma	JCAR014 and durvalumab in treating patients with relapsed or refractory B cell non-Hodgkin lymphoma	CD19	durvalumab	NCT02706405	recruiting
Baylor College of Medicine	I	neuroblastoma	3rd generation GD-2 chimeric antigen receptor and iCaspase suicide safety switch, neuroblastoma (GRAIN)	GD2	pembrolizumab	NCT01822652	active, not recruiting

ICI, immune checkpoint inhibitor, DLBCL, diffuse large B cell lymphoma; FL, follicular lymphoma; MCL, mantle cell lymphoma.

Treatment with checkpoint blockade can promote T cell functionality, yet poor T cell persistence can still limit effective anti-tumor responses. In a phase I clinical trial of GD2 targeting CAR-T cells for the treatment of melanoma (ANZCTR: ACTRN12613000198729, CARPETS), investigators found an increase in PD-1 and Lag-3 expression on stimulated CAR-T cells. PD-L1 blockade reduced activation-induced cell death of these CAR-T cells.¹²⁴ Based on these findings, pembrolizumab was added to a phase I trial of GD2-specific CAR-T cells for neuroblastoma (ClinicalTrials.gov: NCT01822652). Ultimately, PD-1 blockade did not significantly enhance the persistence of T cells compared to lymphodepletion only, but this was limited to a small number of patients in each cohort. Additionally, the authors suggest that beneficial effects of checkpoint blockade on CAR-T cell persistence may depend on the timing of the treatment and tumor target sensitivity to ICIs.¹²⁵

As discussed above, solid tumors attenuate CAR-T cell anti-tumor effects via multiple mechanisms, and blocking a single pathway (e.g., PD-1/PD-L1 interaction) may be insufficient to overcome CAR inhibition. Additionally, different solid tumors have different TME.¹²⁶ Thus, further investigations for each tumor type are required to improve the anti-tumor effect of CAR-T cells.

Oncolytic Virotherapy

Oncolytic viruses (OVs) selectively replicate in and kill cancer cells either due to natural tumor tropism or genetic manipulation of the wild-type (WT) virus. Recently, viral-based gene therapy approaches have garnered a lot of scientific and clinical interest as effective cancer immunotherapies.¹²⁷ OVs have proven safe, with fewer treatment-related severe adverse events (SAEs, grade 3–4) than many other treatments, including chemotherapy, small molecule inhibitors, cytokines, monoclonal antibodies (including ICI), and CAR-T cell therapies.¹²⁸ However, to date, OVs as monotherapies have provided only modest anti-tumor effects in patients. Thus, it is now generally accepted that OV-mediated tumor cell lysis alone is insufficient to eradicate patient tumors due to complex tumor heterogeneity, physical barriers to viral spread, and the immunosuppressive microenvironment.

OVs can mediate anti-tumor effects through a variety of mechanisms, including direct tumor cell lysis and activation of the anti-tumor immune response through the release of tumor-associated antigens (TAAs), intracellular DAMPs, viral pathogen-associated molecular patters (PAMPs), as well as release of stimulatory cytokines and chemokines to reverse the immunosuppressive TME.¹²⁹ This activation promotes the infiltration of immune cells into the TME, turning immunologically “cold” tumors into “hot” tumors.¹³⁰ OVs can also induce immunologic cell death (ICD).^131,132 These revelations shifted the OV field from designing OVs with higher lytic potential to OVs that stimulate more robust immune responses. In addition to their lytic potential, many OVs can also act as gene therapy vectors. Thus, in order to increase immune stimulation, OVs are being designed to deliver therapeutic transgenes (“armed” OVs).

Currently, there are several types of OVs undergoing clinical investigation, including HSV, adenovirus (Ad), vaccinia virus (VV), measles virus (MV), and reovirus, among others (recently reviewed elsewhere¹³³) (Figure 1), either as single-agent treatments or in combination with standard chemotherapy/radiation treatments or various immunotherapies. As of May 27, 2020, there were 89 active clinical trials investigating OVs (Tables S1 and S2). Here, we summarize key aspects of the most widely used viruses and discuss some recent clinical results and their implications for the future of OVs.

Figure 1.

Characteristics of Selected Clinically Evaluated OVs

Graphical representations of the most commonly utilized OVs in clinical trials are shown. General information about each virus, including genomic makeup, virion size, and receptor usage, is provided along with examples of genomic modifications made for clinically evaluated OVs. Notations of some advantages and disadvantages of each type of OV in the context of vector generation and clinical applicability are also given.^127,133,134

HSV

The most significant milestone reached in the OV field in recent years has been the FDA and European Medicine Agency’s approval of an oncolytic herpes simplex virus (oHSV) expressing granulocyte-macrophage colony-stimulating factor (GM-CSF), talimogene laherparepvec (T-VEC), for the treatment of advanced melanoma. HSV has a large genome (152 kb) that can be easily manipulated, as well as broad tropism including epithelial cells through HSPGs, immune cells via herpesvirus entry mediator (HVEM), and neurons through nectins.¹³⁴ However, wild-type HSV can cause significant pathogenicity. Thus, to overcome pathogenesis and ensure tumor cell selectivity, oHSVs are genetically modified to delete the neurovirulence gene ICP34.5, allowing replication only in cancer cells with a disrupted protein kinase R (PKR)-eukaryotic initiation factor 2 (eIF2) pathway.^135,136

T-VEC, in addition to removal of ICP34.5, also contains a deletion of the ICP47 gene to promote an anti-viral response in normal cells while stimulating MHC class I expression in cancer cells, allowing for increased TAA presentation.^137,138 This stimulation functions synergistically with T-VEC-expressed GM-CSF to stimulate antigen-presenting cells within the tumor mass. In a phase III trial in patients with unresectable advanced melanoma, T-VEC virotherapy resulted in a 26.4% ORR compared to 5.7% with recombinant GM-CSF alone.¹³⁹ This study demonstrated the synergistic potential of OVs with gene therapy for immune stimulation.

Another oHSV, HSV1716 (ICP34.5 deleted), has shown some efficacy and increased survival in patients with glioblastoma multiforme (GBM).¹⁴⁰ Recently, HSV1716 was evaluated in a phase I clinical trial in pediatric and young adult patients with relapsed/refractory solid non-CNS tumors. A single intratumoral (i.t.) dose of HSV1716 was shown to be safe in this patient population. While no objective responses were observed, evidence of viral replication and inflammation were seen and patients receiving higher doses of HSV1716 had increased survival.¹⁴¹ Due to the success of T-VEC and demonstrated safety and potential efficacy of other oHSVs, numerous clinical trials investigating herpes-based OVs are currently underway.

Ad

Ads are nonenveloped double-stranded DNA (dsDNA) viruses with a 36-kb genome. For decades, Ads have been studied as gene therapy vectors, as they efficiently infect both dividing and non-dividing cells, have broad tropism, and can be readily modified. Ads are highly immunogenic but cause only mild, self-limiting infections in healthy individuals. In 2005, an E1B-deleted oncolytic Ad (OAd), Oncorine (H101), was approved by the Chinese Food and Drug Administration for the treatment of squamous cell cancer of the head and neck or esophagus in combination with chemotherapy. In a phase III trial, i.t. H101 with chemotherapy had a 78.8% response rate compared to just 39.6% to only chemotherapy.¹⁴² H101 combined with transarterial chemoembolization (TACE) has subsequently exhibited overall survival benefit in treating advanced hepatocellular carcinoma (HCC)¹⁴³ and it is currently in a phase III trial in China (ClinicalTrials.gov: NCT03780049). A similar OAd developed in the United States, ONYX-015, demonstrated only mild efficacy in two patients with pancreatic carcinoma¹⁴⁴ and is not being evaluated clinically.

The more commonly used OAds contain a 24-bp deletion (Δ24) in the E1A gene, which confers cancer cell replication selectivity based on the aberrant free E2F transcription factor in many malignant cells. DNX-2401 is a promising Δ24 OAd containing an additional capsid modification for enhanced cancer cell transduction (RGD-motif insertion) currently being evaluated in patients with high-grade gliomas (ClinicalTrials.gov: NCT03896568) and in pediatric patients with diffuse intrinsic pontine glioma (DIPG) (EudraCT: 2016-001577-33). In a recent phase I trial for gliomas, a single i.t. injection of DNX-2401 was safe with no AEs greater than grade 3. Reduction in tumor mass was observed in 18 of 25 patients, with 3 patients achieving a CR with progression0free survival (PFS) ≥3 years. Importantly, analysis of resected tumors revealed macrophage and CD4⁺ and CD8⁺ T cell infiltration in response to DNX-2401 treatment. Additionally, patients’ resected tumors had increased expression of DAMPS, high-mobility group proteins B1 (HMGB1), heat shock proteins (HSP), and ATP, suggesting DNX-2401 infection induces ICD of tumor cells.¹⁴⁵ These data demonstrate the synergy between direct tumor cell lysis and immune stimulation by OAds.

Enadenotucirev, a group B-based chimeric OAd, has been delivered intravenously (i.v.) to patients with advanced solid tumors in a phase I study in which 5 of 61 patients achieved SD for ≥12 weeks. The presence of neutralizing antibody (NAb) to enadenotucirev was low prior to treatment, although most patients did develop an antibody response that reduced OAd activity in subsequent treatment cycles.¹⁴⁶ These data highlight the need to consider the route of administration of particular OVs. OVs, such as OAd and oHSV, can achieve significant responses when locally delivered but have decreased potency when delivered i.v. and will therefore require multiple local administrations when treating patients with metastasized disease or be combined with other systemic treatments.

As mentioned previously and demonstrated by T-VEC, although modest anti-tumor efficacy can be achieved through direct oncolysis alone, stimulation of the immune response provides synergistic effects. Similar to T-VEC, ONCOS-102 is a chimeric OAd expressing GM-CSF. Local treatment with ONCOS-102 to a patient with malignant pleural mesothelioma resulted in CD8⁺ T cell infiltration to the tumor, including induction of TAA-specific T cells in peripheral blood mononuclear cells (PBMCs).¹⁴⁷ Similar results were also seen in a patient with ovarian cancer.¹⁴⁸ Studies in a cohort of 12 patients with various solid tumors treated with repeated i.t. administration of ONCOS-102 in combination with chemotherapy confirmed T cell infiltration at the tumor sites and systemic tumor-specific T cell responses. Importantly, ONCOS-102 treatment induced tumor PD-L1 expression,¹⁴⁹ suggesting that combinatorial treatment with ICB could enhance anti-tumor efficacy. This strategy is currently undergoing clinical evaluation (ClinicalTrials.gov: NCT02963831).

VV

VV is a member of the poxvirus family with a large dsDNA genome (190 kb), which is replicated in the cytoplasm of infected cells. VV has been used in vaccine strategies, as it can infect a variety of cells, having high tropism to tumor cells, but it causes mild infection in healthy individuals and can induce a potent immune response.¹³⁴ To date, two VV vectors have been evaluated clinically as cancer vaccines: TRICOM for prostate cancer (ClincialTrials.gov: NCT02326805)^150,151 and TroVax for colorectal cancer.^152,153

As gene therapy vectors, VVs are attractive because they can accommodate large transgenes (25 kb)¹³⁴ and are usually modified for tumor cell selectivity through deletion of the TK gene.¹⁵⁴ Pexa-Vec (JX-594), which has the TK gene deleted and expresses the bacterial lacZ gene used as a marker for viral replication and human GM-CSF, is the most clinically studied oncolytic VV (oVV).¹⁵⁵ i.t. Pexa-Vec administration proved safe with some evidence of anti-tumor efficacy in phase I trials in adult¹⁵⁶ and pediatric¹⁵⁷ solid tumor patients. In a phase II trial in hepatocellular carcinoma patients, i.t. Pexa-Vec had a 62% response rate, and the high dose (1 × 10⁹ plaque-forming units [PFU]) improved overall survival compared to the low dose (1 × 10⁸ PFU).¹⁵⁸ Although a phase III trial of Pexa-Vec was initiated for patients with HCC (ClincialTrials.gov: NCT02562755), it was discontinued after failing to meet interim efficacy goals.¹³³ A clinical trial is currently ongoing to investigate the use of Pexa-Vec in combination with checkpoint inhibitors with other advanced and metastatic solid tumors (ClinicalTrials.gov: NCT02977156).

MV

MV contains a negative stranded RNA genome of approximately 15-kb. MV infects cells via CD46 and signaling lymphocytic activation molecule (SLAM), where it replicates within the cytoplasm and causes cell death via fusogenic syncytia formation.^127,134 Wild-type MV is a serious human pathogen. As such, according to the World Health Organization (WHO), 86% of children throughout the world received the measles vaccine in 2018,¹⁵⁹ and thus pre-existing immunity to MV must be addressed by any therapeutic MV strategy.

The first clinical trial of an oncolytic MV investigated i.t. administration of an unmodified attenuated MV Edmonston-Zagreb strain (MV-EZ) to cutaneous T cell lymphoma.¹⁶⁰ The treatment was safe, and some response was seen in four of five injected tumors. Multiple clinical trials have since evaluated modified Edmonston strain MVs. The first to be tested was MV-CEA, which is an oncolytic MV that expresses the carcinoembryonic antigen (CEA), an inert peptide used to monitor gene expression kinetics.¹⁶¹ MV-CEA was safely delivered intraperitoneally (i.p.) to patients with recurrent ovarian cancer with no dose-limiting toxicities (DLTs).¹⁶² Responses were dose-dependent, and 14 of 21 patients achieved SD. Treatment with MV-CEA increased overall survival over expected median survival. A subsequent study delivering MV-CEA intracranially to treat GBM was completed recently (ClinicalTrials.gov: NCT00390299).

MV-NIS was the second modified MV tested. The virus expresses the human thyroidal sodium-iodide symporter (NIS), used to monitor virus spread by radioiodine imaging, and it has the potential to be combined with ¹³¹I radiotherapy. When given i.p. to recurrent ovarian cancer patients, MV-NIS again elicited SD in 13 of 16 patients and increased overall survival compared to expected survival.¹⁶³ More recently, in a phase I study in which MV-NIS was delivered i.v. to patients with myeloma, of 11 patients treated at the higher dose level, 3 had transient responses and 1 achieved a durable hematological response.¹⁶⁴

Reovirus

Reovirus infection is common but usually produces subclinical infections.¹²⁷ Reovirus, which is naturally selective for cancer cells with an inactive PKR signaling pathway,¹³⁴ is a double-stranded RNA virus, approximately 23 kb, which replicates in the cytoplasm of infected cells. Clinical studies of unmodified reovirus, including i.t. administration for gliomas^165,166 and i.v. to patients with advanced tumors¹⁶⁷ and relapsed myeloma,¹⁶⁸ have demonstrated safety. This unmodified reovirus, marketed as Reolysin (pelareorep), given i.v. over multiple cycles also demonstrated safety in a phase II trial for metastatic melanoma. Although objective responses were not observed in this trial, viral replication within tumors was confirmed in 2 of 13 patients biopsied tumors 1 week after treatment.¹⁶⁹ As discussed below, Reolysin is currently being evaluated in combination with chemotherapies and ICB to enhance its therapeutic effect.

Combinatorial Treatment with OVs

OVs have demonstrated remarkable safety profiles when administered to a wide range of patients, including pediatric and young adult patients. However, significant anti-tumor efficacy of OVs given as a monotherapy has been less impressive. In addition, while reovirus and MV can be utilized without genetic modification, most must be altered to enhance cancer cell selectivity (safety) and oncolytic potency. Additionally, many OVs can be “armed” with therapeutic transgenes, the most potent of which are immunostimulatory molecules.

Modification of OVs to enhance the lytic effect is promising, but the OVs showing the greatest therapeutic efficacy are OV vectors delivering immunostimulatory transgenes, e.g., T-VEC. Although T-VEC has been approved by regulatory agencies and is currently being used clinically, the complete response rate can still be improved. The current trend in OV therapy is to combine the properties of OV (tumor de-bulking, immune stimulation) with other treatment regimens such as standard of care chemotherapy, radiotherapy, and most recently ICI to achieve synergistic therapeutic responses (Table S2).

OVs and Chemotherapy

First-line therapy for the treatment of cancer typically involves surgery, if accessible, followed by chemotherapy and radiation. The most common chemotherapies are cytotoxic agents used to target and kill rapidly dividing cancer cells. Some examples of chemotherapies include nucleotide analogs (5-fluorouracil [5-FU], gemcitabine), alkylating agents (cyclophosphamide, cisplatin), cytostatic agents (rapamycin), mitotic inhibitors (paclitaxel, docetaxel), and DNA intercalators (doxorubicin).¹⁷⁰ While these drugs can be efficacious in many settings, they do not discriminate between normal and malignant cells, often resulting in severe toxicity. OVs overcome this disadvantage with high-level specificity to tumor cells.¹⁷¹ Many studies have emerged utilizing potential synergistic combinations of these two therapeutic strategies, as chemotherapeutics provide broad cytotoxicity and tumor de-bulking while OVs provide specific oncolysis and therapeutic transgenes. Additionally, chemotherapeutics suppress the immune system, which could dampen anti-viral responses and improve the spread of OVs.¹⁷²

The addition of T-VEC to neoadjuvant chemotherapy (NAC) was recently reported in an early phase I clinical trial for patients with triple-negative breast cancer (TNBC).¹⁷³ Stage II and III patients were enrolled with two dose levels of intra-tumoral delivery T-VEC given four to five times within 12 weeks of i.v. paclitaxel (80 mg/m²) followed by dose-dense doxorubicin and cyclophosphamide prior to surgery. Five of nine patients attained pathologic CR (pCR) with only small residual foci detected in the remaining four patients. No DLTs were observed, and preliminary results suggested that combination treatment increased the number of tumor-infiltrating lymphocytes in resected tumors.

A phase I study combining HF10, an oHSV, with gemcitabine and erlotinib completed in 2018 resulted in three PRs and four SDs out of nine patients with pancreatic cancer.¹⁷⁴ Histopathological analysis from two resected tumors found CD4⁺ and CD8⁺ T cells near residual cancer cells, further suggesting that the addition of HF10 not only induced tumor cell lysis but also stimulated anti-tumor immunity.

A current clinical trial investigating a non-replicating Ad, VB-111, expressing a Fas-chimera transgene in combination with paclitaxel compared to paclitaxel alone in patients with platinum-resistant ovarian cancer recently reported an improved overall survival of 498 days compared to 173 days, respectively (p = 0.028).¹⁷⁵ Preliminary tumor specimens also observed CD8⁺ cytotoxic T cells infiltrating tumor sections.

While many of these clinical combination studies focus on the treatment of solid tumors, trials also are investigating the effect of OVs with other therapeutics for hematological malignancies. The combination of chemotherapy, dexamethasone and the proteasome inhibitor carfilzomib, with Reolysin improved viral replication in MM patients, leading to objective responses observed in 86% of patients.¹⁷⁶

OVs and Radiotherapy

Along with chemotherapy, radiotherapy is a mainstay of treatment for many solid tumors. While radiotherapy has been used to enhance viral oncolysis, viruses also have been used as radiosensitizing agents.¹⁷² A current clinical trial is investigating whether a single low dose of radiation can enhance virus replication and tumor cell killing in combination with G207, an oHSV.¹⁷⁷ Pre-clinical studies suggest that one dose of radiation (5 Gy) 24 h after virus inoculation increased virus replication, presumably by damaging the anti-viral response.¹⁷⁸

Nineteen patients enrolled in a phase II study to treat non-metastatic non-small-cell lung cancer (NSCLC) received 60 Gy during 6 weeks with three i.t. injections of OAd expressing p53 (INGN 201 [Advexin]).¹⁷⁹ Three months after treatment, no viable tumor was detected in 12 patients, leading to 1 CR and 11 PRs. In comparison, conventional radiotherapy leads to only 10%–20% locoregional control of NSCLC.¹⁸⁰ While local tumor control was observed at the primary site, results demonstrated high metastatic failure rate, leading to a current phase III clinical trial combining INGN 201 with both radiotherapy and systemic chemotherapy (ClinicalTrials.gov: NCT00041626).

OVs and Immunotherapy

As mentioned earlier, many OVs are engineered to express therapeutic transgenes to help stimulate and recruit T cells to tumor sites. This feature makes OVs attractive candidates for combination strategies with ICIs to further enhance anti-tumor activity.¹²⁹ A phase Ib clinical trial reported a 33% CR in patients with advance melanoma treated with T-VEC in combination with pembrolizumab and increased CD8⁺ T cell infiltration and interferon (IFN)-γ expression.¹⁸¹ A multicenter phase III trial is currently underway to assess efficacy in unresectable melanoma (ClinicalTrials.gov: NCT02263508). Pembrolizumab is also being used in phase II studies in combination with DNX-2401, a modified OAd that stimulates immune responses (CAPTIVE/Keynote-192; ClinicalTrials.gov: NCT02798406). Interim results show promise for GBM patients with a median overall survival of 12.3 months.¹⁸² ICI has had a limited effect on a small population of colorectal cancer patients as a monotherapy but in combination with Pexa-Vec, patients are exhibiting a favorable safety profile after durvalumab (anti-PD-L1) treatment.¹⁸³

Innovative strategies combining cancer vaccines and OVs have shown promise in preclinical non-human primate studies. Vaccination with a replication-defective Ad5 expressing cancer-testis antigen, MAGE-A3, followed by infusion of oncolytic Maraba MG1 virus encoding the same antigen results in a prime-boost effect, increasing the number of circulating MAGE-A3-specific T cells and MAGE-A3 antibodies.¹⁸⁴ Clinical investigations of this prime-boost combination with pembrolizumab are now underway in NSCLC patients (ClincialTrials.gov: NCT02879760).

Conclusions

As discussed above, CAR-T cells and OVs have the potential to target both hematologic and solid tumors; however, as monotherapies, their efficacy has been limited, especially against solid tumors. Therefore, the field of cancer gene therapy has been moving toward combinatorial treatment strategies with chemotherapy, radiotherapy, and other immunotherapies such as ICIs. Although clinical results are needed to evaluate their overall efficacy, thus far several publications have shown encouraging results with combinatorial treatments. Furthermore, taking advantage of the potential of OVs to de-bulk tumors and produce immunostimulatory molecules, combining CAR-T cells and OVs also has been tested in preclinical models as reviewed eleswhere.185, 186, 187 Although there is only one clinical trial listed to test the combination of HER2-specific CAR-T cells with an OAd expressing anti-PD-L1 and IL-12 (ClinicalTrials.gov: NCT03740256), this trial is not yet recruiting. Overall, the combination of these two gene therapies may improve outcomes for patients with multiple cancers.

Author Contributions

Conceptualization, M.S.; Writing – Original Draft, N.W.; Writing – Review and Editing, N.W., M.K.M., A.R.S., and M.S.; Supervision, M.S.; Funding Acquisition, M.S. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest

The authors declare no competing interests.