Cancer remains a leading cause of death in developed countries. One therapeutic technique that has yet to be optimized is the combination of local radiation therapy and immunotherapy, which is thought to have a synergistic effect (1–4). Since 1953, radiation therapy has been found to aid antitumor activity outside of the irradiated site in a phenomenon known as the abscopal effect (5–8). The exact mechanism behind the abscopal effect is unknown, but current research suggests that local radiation generates tumor-specific antigens that are processed and presented to T cells for systemic antitumor activity (2, 6, 8, 9). However, the inability to predict when the abscopal effect will happen had led to clinical limitations. In addition, the posited idea of antigen generation by radiation may be limited by modulators of the immune system including program cell death protein-1 (PD-1) and program cell death protein ligand-1 (PD-L1) (6, 8–10). For instance, Park et al. (10) showed that the abscopal effect induced by stereotactic ablative radiotherapy (SABR) was minimized by PD-1 in preclinical mouse models of melanoma and renal cell carcinoma. New immunotherapy medications that target immune modulators such as PD-1 and PD-L1 have led to a new interest in the abscopal effect with radiation. For instance, Smilowitz et al. (11) showed in 2016 that combining radiation therapy with immune checkpoint inhibitors improved outcomes in a mouse model of intracerebral melanoma. As such, it is thought that immune checkpoint inhibitors present a way to circumvent the immunosuppressive components that limit the influence of abscopal effects.

There remain a lot of unknowns about the relationship between immunotherapy and the abscopal effect. In particular, the optimal sequence of immunotherapy with radiation remains unknown (12). Specifically, is immunotherapy best to give before, concurrently, or after localized radiation? Various studies have indicated that the optimal sequence remains unknown (12).

To analyze the optimal sequence, we conducted a retrospective review at our institution of the best time to administer immunotherapy targeting PD-1 and PD-L1 with localized radiation. A preclinical study in 2014 suggested that concurrent delivery of radiation with PD-L1 antagonism was better than subsequent administration of immunotherapy (13). Initial experience at our hospital also suggested similar results with the use of PD-1 inhibitors and stereotactic body radiotherapy (SBRT) or stereotactic radiosurgery (SRS) (14). Our hypothesis was that administering immunotherapy concurrently with SBRT (i.e., a sandwiched approach) would produce the best improvement in overall survival (OS) compared to administering immunotherapy before SBRT/SRS or after SBRT/SRS.

We conducted a single-center retrospective chart review for this study. The study was conducted with institutional review board (IRB) approval under UMC IRB 15-001726.

Based on our analysis, we see that the OS was significantly greater for those patients receiving immunotherapy concurrently or after radiation compared to those who received immunotherapy prior to radiation. This contrasts with previous preliminary data we had that suggested that immunotherapy sandwiched between SBRT/SRS was more beneficial, but the data were more limited in that abstract (14). A recently published phase 2 trial also showed improvement for patients who received pembrolizumab after SBRT for NSCLC (16). However, results did not meet significance as defined by the trial and were influenced by patients with PD-L1-positive tumors. Therefore, the data on how to combine therapies are still inconclusive. With respect to OS based on the decision to start immunotherapy, our results were not significant. We suspect that this may also be confounded due to differing times between the decision to start immunotherapy and the patient actually receiving immunotherapy, but this is a valuable clinical decision point. As such, the applicability of our results to clinical decision-making remains limited.

Our data expand on existing knowledge about the optimal timing of radiation therapy and immunotherapy. Secondary review of the KEYNOTE-001 trial suggested that receiving pembrolizumab after radiation promoted greater progression-free survival and OS for NSCLC. This was for any amount of time prior to pembrolizumab treatment (17). A year later, a phase 1 trial that treated urothelial carcinoma with radiotherapy and pembrolizumab demonstrated that the sandwiched approach was more effective than immunotherapy after radiation (18). Interestingly, our results also show that either a sandwiched approach of immunotherapy and radiation or providing immunotherapy after radiation shows benefit.

Of note, there are several limitations to our study. As mentioned above, external application is limited clinically because there is no significant difference in OS in the timing of immunotherapy/radiation when analyzed from the decision to start immunotherapy. While the end of radiation is a usable point in a retrospective review, its use as a clinical decision point is not feasible.

Additionally, as a retrospective review, we attempt to limit inherent differences in the groups of selections. However, the OS of those patients who received immunotherapy prior to SBRT/SRS was significantly lower than those who received immunotherapy concurrently or after. This suggests a difference in the groups outside of the parameters chosen for inclusion and exclusion in the study.

Furthermore, our study has analyzed several different types of cancer in addition to combining SBRT and SRS, and this leads to concerns over its applicability. Overall, our results show the need for clinical trials with robust methodology that specifies sequencing of immunotherapy and radiation and does so for a particular cancer. Chemotherapy regimens are currently delineated for specific cancers, and the same approach should be applied when using immunotherapy and radiation.

Our results raise more questions about determining the optimal sequence for radiation and immunotherapy. There needs to be a wider collection of retrospective data combined with randomized trials to determine what sequence is best for what cancer. In addition to the timing of SBRT/SRS with immunotherapy, there is a need to optimize other portions of SBRT/SRS and immunotherapy administration including the optimal dose of radiation and fractionation. Overall, more data are needed to elucidate the optimal timing of immunotherapy with SBRT/SRS. There are several studies currently exploring this topic (2–4), and it is important that the full benefit of this combination is understood to improve clinical care.

The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

The studies involving human participants were reviewed and approved by UMC IRB 15-001726. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AH, AJ, HA, PW, and NS aided in designing the study. SW and AH collected the data for the retrospective review. SW, AJ, and MP contributed to the statistical analysis and wrote the first draft of the article. All authors contributed to article revision and read and approved the submitted version.

Author PW was employed by Circulogene, Birmingham, AL, United States.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.