The Promise of Combining Radiation Therapy with Immunotherapy

Abstract

The development of immunotherapy in oncology builds upon many years of scientific investigation into the cellular mechanics underlying interactions between tumor cells and immune cell populations. The past decade has brought an accelerating pace to the clinical investigation of new immunotherapy agents, particularly in the setting of metastatic disease. The integration of immunotherapy into phase III clinical trial design has lagged in settings of advanced loco-regional disease where combination with radiotherapy may be critical. Yet such may be the settings where immunotherapies have their greatest potential to affect cancer patient survival and achieve curative outcomes. In this review, we discuss the interaction of radiation with the immune system and the potential to augment anti-tumor immunity through combined modality approaches that integrate radiation and immunotherapies. The dynamics of cellular and tumor response to radiation offer unique opportunities for beneficial interplay with immunotherapy that may go unrecognized with conventional screening and monotherapy clinical testing of novel pharmaceutical agents. Using immune checkpoint blockade as a primary example, we discuss recent preclinical and clinical studies that illustrate the potential synergy of such therapies in combination with radiation and we highlight the potential clinical value of such interactions. For various immunotherapy agents, their greatest clinical impact may rest in combination with radiation and efforts to facilitate systematic investigation of this approach are highly warranted.

Introduction

Radiotherapy is a mainstay of cancer therapy with more than 60% of patients receiving radiation in either the form of definitive, adjuvant, or palliative treatment. Growing evidence dating to the 1970s demonstrates that the immune system contributes to the anti-tumor effect generated by radiation[1]. While classically thought of as a loco-regional therapy, radiation has the potential to generate out of field “abscopal” anti-tumor responses[2], with current evidence suggesting that immunologic mechanisms underscore this effect[3]. Although the abscopal effect is exceedingly rare with radiotherapy alone[4], these observations of immune-mediated effects of radiation have led to growing enthusiasm for the potential of immunotherapies to augment the loco-regional efficacy of radiotherapy and conversely for radiotherapy to help prime a more effective systemic anti-tumor response to immunotherapies.

Immunotherapies are cancer treatments that seek to engage the patient’s own immune system to eradicate tumor cells. The historical development of immunotherapies shares many parallels with that of radiotherapy. As with radiation, initial clinical application of an immunotherapy was reported in the late 19^th century. William Coley pioneered the use of a bacterial preparation termed Coley’s toxin in the 1890s. While the clinical effect was modest, Coley’s toxin provided an early demonstration of the potential to generate an anti-tumor response by harnessing the immune system. Mirroring radiation, immunotherapies gained prominence as a component of standard cancer treatment in the mid- to late- 20^th century, albeit with considerable toxicities. This included the origins of cell therapies with the development of bone marrow transplant by Fritz Bach and others in the 1960’s[5] and the production, testing, and clinical approval of high dose interleukin 2 (IL-2) for metastatic renal cell carcinoma[6] and melanoma in the 1990’s[7]. In the 21^st century, more selective targeting significantly reduced the toxicities of both radiation and immunotherapies. In radiation oncology, this resulted largely from technological advances including highly conformal intensity modulated and stereotactic techniques combined with high-precision image guidance[8, 9]. Prominent advances with immunotherapies include the development of antibodies directly targeting tumor cells, immune checkpoint inhibitor antibodies, and chimeric antigen receptor T cell therapies[10]. Immune checkpoint inhibitor antibodies have revolutionized the approach to treating metastatic disease in several cancers including melanoma, non-small cell lung cancer, and renal cell carcinoma with some patients experiencing complete responses durable to over five years [11, 12]. Currently there are approved therapies targeting two immune checkpoints: CTLA-4 and PD-1/PD-L1. The increasing specificity of modern radiotherapy and immunotherapies has reduced toxicity profiles and facilitated their increasing roles in clinical oncology. This historical convergence in increased clinical safety and advancement of radiation and immunotherapy now makes feasible their inclusion as part of combined modality treatment approaches (Figure 1).

Fig 1.

Historical convergence of radiotherapy and immunotherapy.

Theoretical basis for combining radiation therapy and immunotherapies

The Steel hypothesis, first conceptualized in the 1970s, describes mechanisms whereby combined modality drug/radiation approaches could improve treatment outcomes[13]. A modernization of the Steel hypothesis has been described highlighting exploitable interactions of radiation and cancer drugs in the molecular era[14] . Under this revised framework, radiation and immunotherapy agents may interact to improve clinical outcomes by five distinct mechanisms: 1) spatial cooperation, 2) temporal modulation, 3) biological cooperation, 4) cytotoxic enhancement, and 5) normal tissue protection.

Through mechanisms elaborated below, radiation has the potential to increase susceptibility of tumor cells to immune-mediated killing[15]. These radiated tumor cells also upregulate negative feedback elements (e.g. checkpoint proteins), which can dampen the immune response[16, 17]. Immunotherapy agents blocking this negative feedback may reinvigorate an immune response that was primed by radiation (Figure 2). This biological cooperation resulting from intersecting cellular and signaling mechanisms has potential to enable spatial cooperation through generation of systemic immune responses mounted against distant, out-of-field tumors, and cytotoxic enhancement via increased immune killing of radiated tumor cells[2, 18].

Figure 2. Schematic illustration of a modernized Steel hypothesis in the era of immunotherapy.

The interaction of radiation and immunotherapy is multifaceted with each component contributing to improved clinical outcomes in the treatment of malignancy. Originally described by Steel in the 1970s, the growing complexity of such interactions prompts revision of this original framework. The potentially exploitable interactions of radiation and immunotherapy include spatial cooperation, temporal modulation, biological cooperation, cytotoxic enhancement, and normal tissue protection. The interaction between anti-PD-L1 therapy and radiation therapy is diagramed as an illustrative example of biological cooperation. Abbreviations: SF, surviving fraction of cells; IMT, immunotherapy; IMM, immunogenic; RT, radiation therapy. Adapted from Morris et al. [84] and Bentzen et al. [14] with permission.

Responses to immunotherapy often are delayed compared to other forms of cancer treatment and may follow a transient increase in tumor burden. This has prompted development of new criteria for evaluation of response to immunotherapies[19]. This raises concern that, in rapidly proliferating tumors, patients who would otherwise have mounted an effective immune response may succumb to sequela of transient progression (e.g. airway obstruction). Radiation can reduce the growth of such lesions, allowing a greater window of opportunity for response to immunotherapy, thereby eliciting temporal modulation. Immunotherapy also has the potential to promote normal tissue protection, and current investigational strategies include antibody-mediated blocking of radiation-induced fibrosis by targeting TGFβ.

As these examples illustrate, the Steel hypothesis provides a framework for conceptualizing the potential cooperative therapeutic interactions between radiation therapy and immunotherapies. In this article, we review these interactions through a discussion of illustrative pre-clinical and clinical studies that investigate combinations of radiation and immunotherapy.

Bench to bedside: Combining radiation with immunotherapy to generate in situ vaccination

A growing body of eloquent preclinical work describes the immunogenic effects of radiation on the tumor microenvironment. Radiation can induce immunogenic tumor cell death and release of tumor-specific antigens[20, 21]. Tumor cells surviving radiation may not escape unscathed and undergo phenotypic changes in the expression of immune susceptibility markers[15]. Effects on the microenvironment include temporary local eradication of radiation-sensitive immune lineages including suppressor and effector lymphocytes, and local release of inflammatory cytokines and damage-associated molecular patterns resulting in local effects on endothelial cell expression of adhesion receptors, immune cell trafficking, and immune cell activation[22, 23]. On the other hand, radiation also triggers effects in the tumor microenvironment that are potentially detrimental to the development of anti-tumor immunity. These include delayed increases in tumor infiltration by suppressive regulatory T cells as well as increased infiltration and activation of inhibitory macrophage and myeloid-derived suppressor cell lineages[24–28]. In addition, certain pathways influenced by radiation can have both positive and negative effects on anti-tumor immunity and the tumor microenvironment. For example production of type 1 interferon can induce recruitment of effector T cells and antigen presenting cells, but can also drive recruitment of myeloid-derived suppressor cells[29]. Additionally, prolonged activation of type 1 and 2 interferon can drive expression of ligands for multiple T cell inhibitory receptors[30]. Targeting such detrimental immunologic effects is one approach whereby immunotherapies may be used to augment the efficacy of radiotherapy.

Dose, fractionation, and volume of radiation influence immunologic effects in the tumor microenvironment. Fractionation of radiation generally enables the relative sparing of normal tissues while achieving therapeutic dose delivery to cancer cells. Differences in the capacity and kinetics of DNA damage repair in normal tissues versus tumor cells underlie the rationale for this approach. However, fractionation does not spare adaptive immune cell populations, specifically lymphocytes, which have little capacity for DNA damage repair and undergo apoptosis within hours of exposure to single fraction doses of just 1-3 Gy[31]. Radiation induced lymphopenia is a negative prognostic factor and multiple studies indicate it is positively correlated with field size, dose per fraction, and fraction number[32, 33]. Additional clinical data suggest that absolute lymphocyte count is predictive of response to checkpoint blockade and is positively correlated with response rate and duration of response[34]. On the other hand, preclinical studies suggest that despite an initial local depletion of lymphocytes, hypofractionated regimens of radiation may be immune activating[35]. Additionally, recent work suggests that standard fractionation and hypofractionation induce expansion of unique immune populations with standard fractionation favoring a myeloid response and hypofractionation driving a lymphoid response that may be more favorable to adaptive anti-tumor immunity[36]. Such analyses of fractionation are challenging to control, however, due to the confounding effects of time and the dynamic nature of changes in tumor infiltrating immune cells.

Immunogenic tumor cell death increases as a function of increasing dose[37]. High dose radiation also leads to dose-dependent increases in the expression of MHC-1 and death receptors such as Fas, which are critical for T cell killing of tumor cells[38, 39]. In contrast, moderate fractional doses of 8-12 Gy may be optimal for activating a type I interferon response in tumor cells via a dose-dependent increase in the cytoplasmic leakage of DNA from micronuclei, which activates the cGAS/STING pathway[17, 40]. At higher doses, radiation-induced STING activation may decline in part due to induced expression of Trex1 exonuclease, reducing the accumulation of cytoplasmic DNA resulting in negative feedback inhibition[17]. In preclinical studies, activation of the cGAS/STING pathway has been essential for generating radiation-induced adaptive immune responses[17, 41], and the complexity of this interaction extends beyond tumor intrinsic signaling. For example, immune recognition of radiated tumors requires dendritic cell intrinsic STING activation via cytoplasmic sensing of tumor derived DNA[41], which may be mediated in part by uptake of tumor derived exosomes containing tumor cell DNA fragments[42]. At low doses (2-5 Gy), radiation modulates the tumor microenvironment by inducing release of cytokines that influence immune cell trafficking and activation[22, 43]. Also at low doses (1-3 Gy) radiation may modulate the tumor microenvironment by ablating radiation-sensitive immune populations including suppressive and effector lymphocytes[44–47]. This may create a window of opportunity by locally and temporarily depleting exhausted and suppressive T cells from the tumor microenvironment and allowing reconstitution with a more favorable infiltrate using immunotherapies.

In preclinical and clinical studies, several groups have taken advantage of the favorable immunomodulatory effects of radiation to prime a more effective systemic anti-tumor immune response[48–50]. This treatment strategy, termed in situ vaccination, utilizes a patient’s own tumor as a source of tumor-specific antigen to stimulate and diversify an effective anti-tumor T cell response. This approach takes advantage of “private antigens” which are induced by random, patient-specific, mutations and differentiation markers in tumor cells. Recent evidence suggests that these mutated proteins are the most important tumor antigens recognized by T cells[51]. Through the capacity to immunomodulate the tumor microenvironment and generate an in situ vaccination effect, radiation may play a role in rendering tumors more responsive to immunotherapies.

Pre-clinical evidence provides a clear rationale for combination of radiation with immune checkpoint blockade. Radiation can promote adaptive resistance through upregulation of PD-L1 on tumor cells[52] and the addition of checkpoint blockade can overcome this resistance mechanism and enhance the generation of abscopal responses[53]. Combination with radiation may be particularly valuable in the treatment of immunologically “cold” tumors, which are characterized by low levels of T cell infiltrate and low mutation burden resulting in few mutation-created neo-antigens[54]. Such “cold” tumors do not typically respond to immunotherapies such as immune checkpoint inhibitors[16, 18]. Even in tumors that are responsive to immune checkpoint blockade or other immunotherapies, radiation may allow for increased depth and duration of response by priming a more diversified adaptive anti-tumor immune response. For example, in the B16 murine model of melanoma radiation and checkpoint blockade activated separate immunologic mechanisms, diversification of the repertoire of T cell receptors among tumor infiltrating lymphocytes and increased clonal expansion of these cells, respectively[16]. These observations have stimulated multiple clinical studies testing combinations of radiation and immune checkpoint inhibitors. Next-generation approaches combining additional classes of immunotherapies with radiation are being developed now in preclinical studies to improve upon and further leverage the in situ vaccine effect of radiation to enhance development of anti-tumor immunity[55–58].

Clinical investigation of immunotherapy agents in combination with radiation

Retrospective studies analyzing combinations of radiation and checkpoint blockade

Clinical safety is a central concern in translating combination therapies to patients. Early clinical data describing safety of radiation and immunotherapy combinations stems from retrospective analyses. For example, in two separate series of metastatic melanoma patients who received non-brain radiotherapy during their course of checkpoint blockade, researchers found that this combination was not associated with higher than expected rates of adverse events[59, 60]. Shaverdian et al conducted a secondary analysis of the KEYNOTE-001 trial and found that patients treated with both radiotherapy and pembrolizumab experienced longer progression-free survival and overall survival than patients who did not have previous radiotherapy, with an acceptable safety profile[61]. To determine safety of combining stereotactic radiosurgery with immunotherapy Martin et al analyzed 480 patients with newly diagnosed brain metastases secondary to non-small cell lung cancer, melanoma, and renal cell carcinoma treated with stereotactic radiation. Addition of immunotherapy was associated with symptomatic radiation necrosis and this association was especially strong in patients with melanoma (p = .03)[62]. Together, these studies demonstrate safety with combining immune checkpoint blockade and radiotherapy, however, caution is warranted with stereotactic radiosurgery.

Randomized prospective trials

Several prospective trials have investigated the addition of immune checkpoint blockade to radiotherapy. Based on strong preclinical data in a spontaneous murine model of prostate cancer[63], Kwon et al conducted a multi-center phase 3 clinical trial, which included men with at least one bone metastasis from castration-resistant prostate cancer that had progressed after docetaxel treatment. Patients were randomly assigned to receive radiotherapy to a bone metastasis (8 Gy in one fraction) followed by either ipilimumab or placebo. Ipilimumab did not increase overall survival compared to placebo (p=0.053). However, on subgroup analysis, patients with favorable prognostic features (no visceral metastasis, no anemia, normal alkaline phosphatase) who received ipilimumab experienced a statistically significant improvement in survival compared to placebo[64].

In a phase 1 trial Tang and colleagues enrolled patients with solid tumors having at least one metastatic lesion in the liver or lung to receive stereotactic ablative radiation therapy (50-60 Gy in 4-10 fractions) and ipilimumab either sequentially or concurrently. Response outside the radiation field was assessable in 31 patients with 3 exhibiting a partial response and 10 experiencing clinical benefit[65].

Blockade of the checkpoint PD-1/PD-L1 in combination with radiation has demonstrated similar efficacy in prospective trials. The PACIFIC trial randomized patients with locally advanced, unresectable, non-small-cell lung cancer who have previously received platinum-based chemoradiotherapy to receive either durvalumab or placebo. This study demonstrated that progression-free survival was significantly longer among patients receiving durvalumab compared with placebo, and secondary end points of 12-month and 18-month progression-free survival rates, objective response rate, and duration of response also were improved with durvalumab[66].

FDA approved indications for checkpoint blockade are largely restricted to the metastatic setting. However, emerging evidence suggests a potential role for immunotherapy in non-metastatic settings. Preclinical work demonstrates the potential for immune checkpoint blockade to prevent metastatic progression from localized disease[67]. These findings have implications for patients with high-risk locally advanced tumors. The PACIFIC trial provides strong supporting clinical evidence, and its success represents a clear opportunity and need to investigate the addition of immunotherapy to potentially curative combined modality therapies in non-metastatic disease settings.

Further prospective studies have attempted to identify mechanisms underlying clinical responses observed with combination therapy. In a clinical trial investigating combining radiation therapy with anti-CTLA-4 in non-small cell lung cancer patients, objective responses were observed in 18% of enrolled patients, and 31% had disease control[68]. The strongest response predictors were increased serum interferon-β after radiation therapy and early dynamic changes of blood T cell clones, which is in agreement with preclinical mechanistic data. Interrogation of the T cell receptor repertoire in responding patients from this study demonstrated detection and expansion of T cell clones not present at baseline following radiation therapy, consistent with an in situ vaccine effect. Additionally, in one patient, investigators were able to identify a tumor neoantigen recognized by a population of neoantigen specific T cells that were not identified prior to radiotherapy[68].

In a randomized trial specifically designed to measure abscopal response, McBride et al randomized patients with metastatic head and neck squamous cell carcinoma to receive either nivolumab alone or nivolumab with stereotactic body radiotherapy to a single lesion (9 Gy × 3) between the 1st and 2nd doses of nivolumab. The primary end-point of objective response rate in non-irradiated lesions was not improved with combination therapy[69]. In a phase 2 trial, Theelen et al randomized patients with metastatic non-small-cell lung cancer to receive pembrolizumab alone or in combination with radiotherapy (8 Gy x3). There was a doubling in overall response rate between combination therapy and pembrolizumab monotherapy, however this was not statistically significant (p=0.07)[70]. Caution is warranted in interpreting this response rate, however, as PD-L1 status was not balanced between groups and may confound this outcome. In subgroup analysis, patients with PD-L1 negative tumors experienced significantly improved progression-free survival and overall survival [70].

Ongoing clinical trials

Collectively, the current body of clinical data suggest that combining radiation with checkpoint blockade is safe and demonstrate that an in situ vaccine effect can be achieved with radiation. However, randomized, prospective studies have not yet shown a capacity for radiation to augment clinical response in the metastatic setting. With locally advanced disease, the PACIFIC trial provides strong evidence for upfront combination therapy. Combining radiation with immunotherapy is an active and growing area of investigation that extends beyond combination with immune checkpoint blockade to include preclinical and clinical testing of radiation with every class of immunotherapeutic[71–74]. The spectrum of immunotherapy agents currently being tested clinically is diverse and is summarized in Table 1.

Table 1.

Immunotherapy agents currently under clinical investigation in combination with radiation therapy.

Category	Examples of Immune Rx	Disease Site	Phase	Number of Current Studies (N)	Fold Change in N from 2018 [56] to Present
Checkpoint Inhibitors	Anti-CTLA-4	Cervix, Melanoma, Head & Neck, Pancreas, Liver, Lung	I/II/III	98	5
	PD-1/PD-L-1	Esophageal, NSCLC, Malignant Glioma, Melanoma (Brain metastases), Invasive Bladder, Oligometastatic Breast, Head & Neck, Pancreas, Gastric, Colorectal, Follicular Lymphoma	I/II/III	451	5
Cytokines	IL-2, IFN, GM-CSF, and TGF-beta blockade	Metastatic Breast, NSCLC, Glioblastoma, Follicular Lymphoma, and Pancreas	I/II	149	16
Cell Therapy	CAR T cells (Anti BCMA, CD19, CD-30, TAI-meso, EGFRvIII, mesothelin, CD22)	B-cell Lymphomas, Pancreas, Glioblastoma, Follicular Lymphoma, Pancreas	I/II	18	0.67
Vaccines/Oncolytic Viruses	AdV-tk, Sipluleucel-T, G207, ADV/HSV-tk, Oncolytic Adenovirus Ad5-yCD/mutTKSR39rep-hIL12 and Ad5-yCD/mutTKSR39rep-AD	Prostate, Pancreas, Malignant supratentorial neoplasms, NSCLC, Triple Negative Breast, Prostate, Glioma, Ovarian, NSCLC, Sarcoma, Glioblastoma, Neuroblastoma,	I/II/III	23	0.66
Other Ta rgeted Immune Rx	OX40 antibody, CDX-301, GITR, and TLR-4,7,9 Agonists	Melanoma, Renal Cell Carcinoma, NSCLC, Breast, Sarcoma, Cutaneous T-cell and Recurrent Lymphoma	I/II	22	0.76

Barriers to effective translation of preclinical findings to patients

Preclinical barriers

One challenge that confronts the immuno-oncology research community is the difficulty of designing and choosing regimens with sufficient justification and likelihood of benefit, to be prioritized for rapid initiation of clinical testing. The lab mouse can be studied in large numbers, in reproducible circumstances, with an immune system that is similar to that of humans, using clinically relevant therapeutics or their murine surrogates. However, implantable tumor models rely on murine cancer cell lines that are immortalized and generally more immunogenic than human tumors, which makes translation of immunotherapy regimens difficult. Actual human cancer cell lines or tumor fragments can be implanted in immunodeficient mice; however, these studies in mice lacking a functional immune system cannot be used to study the capacity of immunotherapy to act on these tumors via endogenous immune elements. While grafting immunodeficient mice with human hematopoietic stem cells can create humanized mice, interpretation of data from such models is complicated by the xenogeneic mismatch between the immune cells and normal tissues of the host as well as allogeneic mismatch between the immune cell donor and the tumor cells.

To partially circumvent these challenges, we and others have begun utilizing companion canines with cancer to validate observations made in immunocompetent murine tumor models. Each year approximately 6 million dogs will develop cancer, many of whose owners are unable to afford treatment but are willing to participate in clinical trials. Canines develop a wide variety of spontaneously occurring cancers that share many characteristics with human cancers. In addition, many pets receive state-of-the-art medical care that can include image-guided radiotherapy and experimental and proven immunotherapeutics, much like human patients, which provides a unique translational opportunity to test combination therapies in a heterogeneous population with spontaneously developing tumors[75].

Several key cancer biology similarities exist between dogs and humans including patterns of response or resistance to conventional therapy, as well as metastasis and recurrence[76]. Many specific cancers are functionally identical in dogs and humans at the histological level, including osteosarcoma, mucosal melanoma, mammary tumors, soft tissue sarcomas, non-Hodgkin lymphoma, bladder cancer, and others[76–80]. Genome-wide studies have identified similarities in gene dosage between corresponding cancers in dogs and humans, offering insights into potential conserved pathogenesis mechanisms involving key driver genes. Dog and human tumors also have many key similarities at the transcriptional level, and several tumor types are considered to be indistinguishable between species[81].

In an effort to improve the translational drug development process, the National Cancer Institute created the Comparative Oncology program, which includes a clinical trial network of 20 academic veterinary teaching hospitals across the United States and Canada. Since its creation in 2004, the program has completed 12 multicenter clinical trials in pet dogs with spontaneous cancers, and its 24^th trial concept is currently open for enroNment[82]. Several canine immunotherapies are available and these include cytokines such as IL-2, IL-12, and IL-15 as well as a USDA approved cancer vaccine Oncept^®, which targets the tyrosinase protein often expressed in melanoma. A canine anti-PD-1 has also been tested and is pending USDA approval[81]. Additionally, it is possible to administer clinical quality radiation therapy to dogs and the spatial separation of tumors and normal tissues in canines more closely reflects that encountered in humans, when compared to mice[83].

Clinical barriers

A wealth of preclinical data describes inflammatory changes both in the tumor and surrounding microenvironment in response to varying doses and fractionation schedules of radiation. However, dynamic changes in murine tumors and immune systems are inherently different from what may occur in human patients. Therefore, it is imperative to confirm preclinical findings in humans to allow for effective translation of preclinical success to patients. While it is a reasonable task to secure funding for phase 3 trials whose endpoints are survival, it is very difficult to convince funding agencies to sponsor phase 1 trials designed to ask fundamental questions about dose and fractionation whose endpoints are biologic correlates of immune activation in response to radiation. In contrast to pharmaceuticals, no single private entity has ownership of radiation as an intervention. This limits the funding potential for mechanistic research in human subjects. While difficult to undertake, results of such mechanistic studies will prove to be invaluable when attempting to rationally combine radiation with immunotherapy.

Despite radiation being a critical component to treatment for a multitude of cancer patients, there is a historical lack of clinical trials that formally explore combinations of radiation and other therapeutic agents such as immunotherapy. A search of currently registered clinical trials in the United States during the preparation of this review suggests that in the era of immunotherapy this disparity persists, yet has improved considerably compared to other therapeutics[84]. Among all current trials for cancer, 3516 (13%) investigate an intervention with radiotherapy and 5240 (22%) test an immunotherapy, with 761 (4%) trials evaluating a combination of the two interventions (Table 1). Compared to a similar Boolean search conducted in 2018, this represents approximately a fourfold increase in overall number of current combination therapy trials[85]. Focusing on phase III trials for cancer, a total of 543 (18%) investigate an intervention with radiotherapy and 567 (19%) evaluate an immunotherapy, while 78 (3%) examine a combination of these two interventions (Figure 3). This relatively strong uptake in the testing of radiotherapy in approximately 14% of phase III immunotherapy trials has no doubt been driven, at least in part, by the rationale arising from strong preclinical and early phase clinical studies outlined above. However, there continues to be a disparity in testing combinations of radiation and immunotherapy in the non-metastatic setting. Despite the success of the PACIFIC trial, <1% of early phase trials investigating a combination of radiation and immunotherapy are open to patients with non-metastatic disease. Additionally, there are only two active phase III trials testing a combination of radiation and immunotherapy in patients with localized disease: one testing chemoradiation in combination with the PD-L1 inhibitor atezolizumab in muscle invasive bladder cancer, and the other investigating combining radiation with the cancer vaccine ProstAtak®(AdV-tk) in localized prostate cancer. Effective collaboration between radiation oncology and industry investigators will be critical to redress this discrepancy, as will the attraction of proportionate federal funding, which has historically lagged in radiation oncology.

Fig 3. Distribution of current phase III clinical trials in oncology.

A search of www.clinicaltrials.gov for phase III clinical trials returned 2602 trials for condition = “cancer”. When intervention = “radiation” was added to this search 543 studies were identified. When the 2602 phase III cancer trials were sorted by intervention, 567 studies involved an immunotherapy agent as defined in this review. Of these, 78 studies examined a combination of an immunotherapy agent and radiation.

Future directions

Radiation primarily has local effects that can be strongly immunogenic however, in the context of metastatic disease it is currently unclear whether all sites of disease need to be targeted by radiation to optimally synergize with immunotherapy. Radiating multiple tumor sites may increase risk of immunosuppression and in such settings it may be critical to consider blood pool, draining lymph nodes, spleen, and/or bone marrow as organs at risk during treatment planning[86]. In patients with metastatic sites not amenable to external beam radiation or with occult disease, emerging targeted radionuclide therapies could offer an alternative approach to delivering radiation to all tumor sites[87–90]. Revisiting additional radiotherapy modalities such as brachytherapy and/or particle therapy may also prove useful in combination with immunotherapy. Preclinical studies indicate that the immunogenic effects of radiation are sensitive to the radiation dose and field size. Due to its powerful conformality and dose heterogeneity, brachytherapy may confer meaningful advantages over external beam radiation when it comes to priming an in situ vaccine effect by simultaneous engaging multiple different dose-dependent immunomodulatory mechanisms in a single tumor microenvironment[85]. As we gain further understanding of the complex interplay of tumor and immune signaling pathways, it may be beneficial to explore combinations of radiation with multiple classes of immunotherapies and molecularly targeted therapeutics simultaneously. Current combination strategies are not yet fully optimized with many unknowns including the appropriate sequence to administer immunotherapy in relation to radiation, appropriate route of delivery of immunotherapy (systemic or local), and appropriate choice of immunotherapy (or immunotherapies) to combine with radiation. Additionally, as described by preclinical and clinical studies there are likely to be considerable differences in clinical efficacy with patient-specific tumor and immune microenvironment characteristics underlying this observation. With additional breakthroughs in precision medicine, this may enable the logical design of personalized approaches to complex combined modality treatment[91].

Conclusion

In the era of immunotherapy, radiation has the potential to become a critical component of systemic cancer therapy. Combined modality approaches with immunotherapy may increase the curative capacity of radiotherapy in patients with locally advanced disease, as seen with the success of the PACIFIC trial. This therapeutic potential is supported by strong preclinical evidence providing abundant rationale for clinical testing of radiation and immunotherapy combinations. Mechanistic hypotheses originating from preclinical studies in murine models, such as the in situ vaccine effect of radiation, have been confirmed in human analyses, thus warranting further testing of next generation strategies. Retrospective data and prospective clinical trials indicate that combinations of radiation and immunotherapies are generally safe. In contrast with prior molecular targeted agents, the uptake of radiation combined with immunotherapy in clinical studies has been more robust. Driven by strong preclinical rationale, further support for preclinical investigation is needed now to achieve successful translation to proof of clinical benefit. Investment now in clinical trials that combine radiation with immunotherapy is highly warranted in oncology.