Abstract
Purpose: How best to sequence and integrate immunotherapy into standard of care is currently unknown. Clinical protocols with accelerated nonablative hypofractionated radiation followed by surgery could provide an opportunity to implement immune checkpoint blockade.
Experimental Design: We therefore assessed the impact of nonablative hypofractionated radiation on the immune system in combination with surgery in a mouse mesothelioma model. Blunt surgery (R1 resection) was used to analyze the short-term effect, and radical surgery (R0 resection) was used to analyze the long-term effect of this radiation protocol before surgery.
Results: Nonablative hypofractionated radiation led to a specific immune activation against the tumor associated with significant upregulation of CD8+ T cells, limiting the negative effect of an incomplete resection. The same radiation protocol performed 7 days before radical surgery led to a long-term antitumor immune protection that was primarily driven by CD4+ T cells. Radical surgery alone or with a short course of nonablative radiation completed 24 hours before radical surgery did not provide this vaccination effect. Combining this radiation protocol with CTLA-4 blockade provided better results than radiation alone. The effect of PD-1 or PD-L1 blockade with this radiation protocol was less effective than the combination with CTLA-4 blockade.
Conclusions: A specific activation of the immune system against the tumor contributes to the benefit of accelerated, hypofractionated radiation before surgery. Nonablative hypofractionated radiation combined with surgery provides an opportunity to introduce immune checkpoint blockades in the clinical setting. Clin Cancer Res; 23(18); 5502–13. ©2017 AACR.
Protocols using an accelerated course of hypofractionated nonablative radiation followed by surgery might be an ideal approach to implement immunotherapy in clinical practice. In this series of experiments, we therefore analyzed the effect of surgery in combination with accelerated hypofractionated nonablative radiation in a murine mesothelioma model and demonstrated that surgery did not preclude the potential beneficial effect of high-dose radiation on the immune system as long as adequate time was provided between the end of the radiation and surgery to activate the immune system. To the best of our knowledge, this is the first study providing experimental evidence supporting the use of induction radiation before surgery and demonstrating that surgery would not be detrimental on the immune system if the timing between the different therapeutic modalities was adequate.
Introduction
Recent advances in immunotherapy for solid tumors have opened the door to a new field of therapy in oncology. One open question is how best to integrate immunotherapy into the current standard of care (including chemotherapy, radiotherapy, and surgery). Several publications have shown the potential synergistic effect of combining immune checkpoint blockade with ablative radiation in mice models, but clinical trials using this combination have so far been limited (1, 2). Part of the limitation is the lack of knowledge about the impact of standard therapy on the immune system and the risk of toxicity when immunotherapy is added to these treatments.
The development of new highly conformal radiation techniques over the past 20 years allows precise targeting of the tumor and enables safe delivery of higher radiation doses per fraction (3). Traditionally, radiation is believed to work by direct and indirect damage to the DNA from ionizing radiation and related reactive oxygen species, causing cell death and tumor destruction. However, activation of the immune responses through cell death, tumor antigen release, and modification of the tumor microenvironment may contribute to the benefit of radiation (4, 5). Ablative radiation (hypofractionated radiation doses of 8 Gy or higher) delivered in a short course of 5 or fewer fractions appears to augment antitumor immune responses, but this immune benefit is complex and a nonlinear function of dose (6, 7).
In the clinical setting, an ablative dose of radiation is not always achievable due to the risk of toxicity to the surrounding tissue (8). Therefore, surgery could be an important adjunct to the combination of radiation and immunotherapy to achieve local tumor control with an acceptable risk of toxicity from the radiation. Protocols using accelerated nonablative hypofractionated radiation followed by radical surgery have been used clinically for rectal carcinoma and malignant pleural mesothelioma (9–12).
Malignant pleural mesothelioma is an aggressive malignancy with a median survival of less than 18 months despite aggressive treatment with chemotherapy, surgery, and radiation (13, 14). Immunotherapy has shown some encouraging results, but randomized trials have yet to confirm its potential benefit (15, 16). The protocol of accelerated hypofractionated radiation followed by surgery could be an ideal setting to implement immunotherapy (17). We have thus developed a mouse model to better understand the impact of this radiation protocol on the immune system. We observed that a short course of hypofractionated radiation in combination with CTLA-4 blockade was able to improve local control and induce an abscopal effect through an activation of the immune system with upregulation of activated CD8+ T cells and downregulation of regulatory CD4+CD25+Foxp3+ T cells (Treg) in the tumor (18). In the series of experiments reported herein, we therefore aimed to determine the short- and long-term effects of nonablative hypofractionated radiation on the immune system when combined with surgery.
Materials and Methods
Tumor cell lines and mice
AB12 and AE17 malignant pleural mesothelioma cell lines were both derived from an asbestos-induced tumor in a BALB/c and C57BL/6 mouse, respectively. AB12 was kindly donated by Dr. Jay Kolls, University of Pittsburgh (Pittsburgh, PA). AE17 was obtained from the European Collection of Cell Cultures. AE17-OVA was developed by stably transfecting the parental cell line (AE17) with secretory ovalbumin (sOVA). The cell line was kindly provided by Dr. Steven Albelda, University of Pennsylvania (Philadelphia, PA), and Dr. Delia Nelson (University of Western Australia, Crawley, Australia).
AB12 and AE17 were grown in RPMI1640 culture media (Life Technologies Inc.) supplemented with 10% heat-inactivated FBS (Life Technologies Inc.), 2 mmol/L l-glutamine, 100 U/mL penicillin, 100 μg/mL streptomycin and nonessential amino acids. The transfected cell line AE17-OVA was maintained in the same medium supplemented with 400 μg/L neomycin analogue G418 (geneticin; Invitrogen). Cells were plated in tissue-culture coated flasks (BD Biosciences Canada), grown in a 37°C and 5% CO2 environment, and passaged when 70% confluent.
Eight- to 12-week-old BALB/c and C57BL/6 syngeneic mice were purchased from Jackson Laboratories, and acclimatized in the animal colony for 1 week before experimentation. The animals were housed in microisolator cages, 5 per cage, in a 12-hourlight/dark cycle. Sterile water and rodent food were given ad libitum. Animal care and experiments were performed in accordance with institutional and Canadian Institute of Health guidelines. All animal experiments were approved by the Animal Research Ethics Board at the Toronto General Research Institute (University of Toronto, Toronto, CA).
In vivo tumor growth experiments
Mice were injected subcutaneously in the right flank with 1 × 106 AB12 cells, AE17 or AE17-OVA cells in 100 μL of PBS at day 0. For rechallenge experiments, cells were injected subcutaneously in the left flank with the same conditions. After removing fur and cleaning the skin, injections were made with a syringe and 25–27G needles. Tumor growth was monitored every 3 days. Tumor dimensions were measured using microcalipers. Tumor size is expressed as tumor area in squared millimeters using the longest length and the perpendicular width (length × width). Mice were sacrificed when tumor dimension reached 150 mm2 or showed signs of ulceration as per institutional ethics protocols.
Local radiotherapy
Radiation was given using the X-Rad 225Cx small-animal image-guided irradiator (Precision X-Ray). The irradiator has a 225 kVp X-ray tube (Varian Associates) and a flat-panel silicon detector mounted on a 360° rotation C-arm gantry. The automated stage is movable on the x, y, and z axis. It is all housed in a self-shielded cabinet and is remotely controlled by a computer (Dell Precision 690, Intel Xeon CPU running Windows XP). The mean targeting displacement error is ≤0.1 mm in the x-y-z planes. Radiation was given to mice under isoflurane anesthesia. To initially visualize the animal, the tumor fluoroscopic mode was used. To target precisely the tumor, a scout cone-beam CT was created at a 40 kVp tube potential and 0.5 mA current. The tomography was then reconstructed at a 0.4-mm voxel size. The beam source was collimated to either a 1.5-cm or 2-cm diameter circular field. To confirm the area to be irradiated, the tumor was then visualized under fluoroscopic imaging with the collimator in place, immediately prior to delivery of treatment. Radiation was delivered at a tube potential of 225-kVP and a 13-mA current for a dose rate of 3.02 Gy/minute. The daily dose was given from 2 angles, half from above (180 degrees) and half from below (0 degrees). Total dose was given in divided fractions over 3 days according to treatment protocols. After radiation, mice were placed back in their cages and housing facilities.
Surgical resection of subcutaneous tumors
Under general anesthesia with isoflurane, mice with flank tumors were shaved and cleaned with isopropanol. Tear gel was applied on both eyes and a heating lamp was used to prevent hypothermia. Skin around the tumor was infiltrated with Marcaine (bupivacaine 0.25%) prior to incision. Two different approaches were used depending on the experiment. Blunt surgery was performed by blunt dissection removing all macroscopic tumor, but no skin or surrounding healthy-looking tissue (R1 resection). Radical surgery was performed by removing the skin on top of the tumor and 0.5-cm margin of healthy-looking subcutaneous tissue around the tumor (R0 resection). Sterile prolene sutures (5-0 or 6-0) were used to close the wound. Marcaine was administered immediately after closing the wound for postoperative analgesia and the mice were observed until complete recovery. Mice were then monitored at 6, 24, and 48 hours after surgery and meloxicam 1 mg/kg was given subcutaneously for postoperative analgesia.
In vivo depletion of CD4+ and CD8+ specific T cells
Anti-CD4 MAb from rat GK 1.5 hybridoma or anti-CD8 Mab from 2.43 rat hybridoma (Bio X Cell) were diluted to a final concentration of 1 mg/mL with PBS or 2 mg/mL for double depletion. Intraperitoneal injections for 3 consecutive days with 0.2 mL (0.2 mg) of purified Mab were performed. For double depletion the total volume injected was 0.2 mL, consisting of 0.1 mL of each mAb at a concentration of 2 mg/mL. Examination of peripheral lymphoid organs at day 6 revealed that >95% of the cells were depleted. The depleted condition was maintained with 0.2 mg injections of mAb every 3 days.
Immune checkpoint blockade
Mouse mAb (9D9) to CTLA-4, PD-1 or PD-L1 (Bio X Cell) was diluted to a final concentration of 1 mg/mL with PBS and kept at 4°C until further use. Intraperitoneal injections were made with 0.2 mL (0.2 mg) of the purified antibody every 3 days for the length of the specified treatment.
Tumor digestion
Tumors were removed and placed in 15-mL conical tubes filled RPMI1640 culture media and stored on ice until further use. Tissue was chopped into 2-mm pieces and transferred to 15-mL conical tubes containing digestion media consisting in RPMI1640, DNAse (Roche 10104159001) and Liberase TM (Roche Diagnostics). Tubes were placed in a shaking water bath for 30 minutes and when the pieces were soft and malleable the solution was filtered and mashed through a 70-μm cell strainer. Cells were then washed with PBS and remaining cells were counted and viability was assessed.
Flow cytometry
Cells were resuspended in FACS buffer, and stained for 30 minutes at 4°C with α-CD16/CD32 Fc block (BD, Pharmingen), and a combination of the following mouse-specific antibodies: CD3, CD4, CD8, CD44, CD45, CD69, CD137 (4-1BB), TIM3, PD-1, ICOS (BD, Pharmingen). Cells stained with tetramer were incubated for 30 minutes with the Class I H-2Kb SIINFEKL tetramer prior to surface staining. All samples were then washed twice with FACS buffer and analyzed immediately using a BD LSR II flow cytometer (BD Biosciences) and FlowJo V10 software (FlowJo LLC). Tumor samples were pooled and analyzed as a single sample to have sufficient cells.
Immunofluorescence
Frozen tissue samples on slides were fixed with cold acetone for 10 minutes. Paraffin embedded samples were deparaffinized with Xylene, 100% ethanol, 95% ethanol, and 70% ethanol respectively and antigen retrieval was performed by immersing samples in 100°C citrate buffer for 20 minutes. Samples were blocked with 5% BSA in Tris-buffered saline for one hour before the addition of primary antibody. After incubating overnight at 4°C, sections were washed in TBS+0.2% Tween 20. Slides were subsequently incubated for 1 hour at room temperature with the appropriate fluorescently labeled secondary antibody. Slides were further washed with TBS+0.2% Tween 20 before adding mounting media with DAPI nuclear stain. Coverslips were placed on top and sealed with nail polish.
Fluorescently labeled cells or tissues were visualized with the WaveFX (Quorum Technologies Inc) confocal microscope system. Pictures were analyzed using ImageJ V1.47 (NIH, Bethesda, MD). Corrected total cell fluorescence (CTCF) was calculated by the formula CTCF = integrated density – (area of cell × mean fluorescence of background reading).
Ovalbumin ELISA
AE17-OVA and AE17 cell culture supernatants were collected 3 days after seeding cells. Cells were then trypsinized and washed twice with PBS. For cell lysates, cells were collected by centrifugation, 5 minutes at 1,000 × g. Cells were then subjected to ultrasonication for 4 cycles on ice. Cell lysates were collected by centrifugation at 1,500 × g for 10 minutes at 4°C to remove cellular debris. Cell lysate or media was placed in wells coated with a biotin-conjugated antibody specific to OVA from an ELISA kit (Biomatik corporation). Samples were left for 2 hours to bind anti-Ova antibodies. Avidin conjugated to horseradish peroxidase (HRP) was then added to each well and incubated for 1 hour. Finally TMB substrate solution was added and those wells containing OVA, biotin-conjugated antibody and enzyme-conjugated avidin exhibited a change in color. Reaction was terminated by the addition of sulphuric acid. Concentrations were determined by four-parameter logistic test using a standard curve. Samples were measured in duplicate.
Statistical analysis
Statistical analysis was performed with GraphPad Prism 5 (GraphPad Inc). More than two groups were compared using one-way ANOVA analysis. Unpaired two-tailed Student t test was used to analyze two groups. A P value of less than 0.05 was considered statistically significant. Results have been presented as mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001 in all figures.
Results
Development of a model of nonablative hypofractionated radiation
Our initial goal was to develop a mesothelioma model that mimicked the clinical setting in which mice received accelerated nonablative hypofractionated radiation. To optimize the model, we compared 4 different doses of local radiotherapy (LRT) targeting the tumor. These experiments were done in BALB/c mice with AB12 cell line following previous in vitro and in vivo experiments performed by our group (18). A total of 1 × 106 AB12 tumor cells were inoculated subcutaneously in the right flank and LRT was started 7 days after tumor inoculation. On the first day of LRT treatment, mice were randomized into the following groups: (i) no treatment, (ii) 15 Gy over 3 days (5 Gy/fraction ×3), (iii) 22.5 Gy over 3 days (7.5 Gy/fractions ×3), (iv) 30 Gy over 3 days (10 Gy/fraction ×3), and (v) 22.5 Gy in a single dose.
Untreated mice showed rapid tumor growth up to 22 days when animals were sacrificed. All four groups treated with LRT achieved tumor growth stabilization for at least 7 days before tumor growth resumed. No mice were cured by LRT alone. There was no significant difference in tumor size among the four LRT treatment groups, but mice irradiated with 30 Gy in 3 fractions and 22.5 Gy in one fraction showed signs of distress and lost 10% to 15% of total body weight during the first 2 weeks after treatment. These results confirmed that the tumor model was sensitive to LRT and that these doses of radiation were not ablative. We chose a radiation dose of 15 Gy in 3 fractions (5 Gy × 3) for the following experiments.
Combination therapy with LRT and surgery
We next analyzed the role of radiotherapy in combination with blunt surgery (R1 resection). Mice were randomized the first day of LRT into the following groups: (i) no treatment; (ii) LRT (5 Gy × 3); (iii) surgery; and (iv) LRT (5 Gy × 3) and surgery. In the LRT and surgery group, blunt surgery was performed 5 days after completion of LRT.
Mice treated with blunt surgery alone had rapid tumor recurrence and tumor growth rate was faster after resection compared with untreated tumors. In the groups treated with LRT alone and with the combination LRT surgery, tumor growth was significantly slower than in untreated mice or mice treated with surgery alone. There was no difference between the treatment group LRT alone and LRT surgery (Fig. 1).
Upregulation of tumor-infiltrating CD8+ T cells after local radiation
To evaluate the involvement of the immune system on tumor growth after LRT, tumor samples from 15 Gy (5 Gy × 3) radiated mice were compared with untreated mice on day 2, 7, and 12 after LRT. Immunofluorescent staining revealed that the number of CD3+CD8+ double positive T cells 2 days after LRT was not significantly different between radiated and untreated mice. However, on day 7 and 12 after LRT, the number of CD3+CD8+ T cells was significantly higher in the LRT group compared with untreated tumor (Fig. 2). Flow cytometry showed that the frequency of tumor infiltrating CD45+ CD3+CD8+ T cells were 5× higher in irradiated than in untreated tumors on day 7 (Fig. 2).
Tumor-infiltrating CD8+ T cells are OVA-specific
To determine whether tumor-infiltrating CD8+ T cells induced by radiation were tumor antigen–specific, we inoculated C57BL/6 mice with the AE17-OVA cell line. LRT-treated tumors were excised and analyzed 10 days after the end of radiation. Tumor-infiltrating cells were stained with H-2Kb tetramers containing the OVA protein–derived peptide SIINFEKL.
After gating for live cells, the proportion of CD44+ and SIINFEKL tetramer cells were identified among the CD3+CD8+ double positive population. Radiated tumors showed a trend toward greater proportion of tetramer-specific CD8+ T cells compared with untreated tumors (Fig. 3). The result of this experiment gives further evidence that LRT promotes recruitment of lymphocytes into the tumor using a different tumor cell line and mice strain. About 30% of the recruited lymphocytes are specific for the OVA-derived peptide SIINFEKL in the treated tumor as compared with only 15% in the untreated group.
Expression of 4-1BB and PD-1 by tumor-infiltrating lymphocytes
To assess the change in phenotype of tumor-infiltrating lymphocytes after a short course of nonablative radiation, we analyzed the expression of the inhibitory receptor PD-1 and the activation marker 4-1BB in radiated tumor 3 and 9 days after LRT in C57BL/6 mice. The early activation marker 4-1BB was significantly upregulated in the tumors 3 days after radiation compared with untreated controls, but this difference decreased over time and was not statistically significant on day 9 after radiation (Fig. 3). In parallel, radiation led to a significant reduction of CD8+ T cells expressing PD-1 on day 3 and 9 after radiation compared with untreated controls (Fig. 3). These findings suggest that accelerated hypofractionated radiation may transform the immunosuppressive microenvironment of the tumor even though the radiation was not ablative.
Depletion of CD4+ T cells and CD8+ T cells partially abrogates the effect of LRT on tumor growth
To examine whether CD4+ T cells and CD8+ T cells played a role in the tumor response to radiotherapy, animals were depleted of CD4+ T cells, CD8+ T cells, or both starting 1 day before LRT and throughout the length of the experiment. Mice were randomized to the following groups: (i) no treatment, (ii) LRT only, (iii) LRT and CD4+ T-cells depletion, (iv) LRT and CD8+ T-cell depletion, and (v) LRT and double depletion.
Double depleted mice treated with LRT showed significantly greater tumor size than animals treated with LRT only (Fig. 4). However, tumor size in double depleted mice remained significantly smaller than untreated mice demonstrating that LRT still had an impact on tumor growth independently of CD4+ and CD8+ cells. Mice depleted in CD8+ T cells alone showed significantly greater tumor size than mice depleted in CD4+ T cells alone, suggesting that the benefit of LRT on the immune system was predominantly mediated by CD8+ T cells.
Long-term immunologic protective memory after LRT and surgery
We next investigated the role of LRT before radical surgical resection of the tumors (R0 resection) to assess the ability of accelerated nonablative hypofractionated radiation to generate an effective immunologic memory response. C57BL/6 mice were inoculated with AE17-OVA cells and after 9 days were randomized into the following treatment groups (i) surgery only, (ii) LRT and surgery 24 hours later (LRT-Surg 24 hrs), (iii) LRT and surgery 7 days later (LRT-Surg 7d). A total of 7 days was chosen based on the time course of tumor infiltrating CD8+ T-cell upregulation (Fig. 2). Cured mice were rechallenged at 90 days with the same tumor delivered to the opposite flank.
All 10 mice in the surgery alone group were tumor free 90 days after treatment, and 9 mice in both groups treated with LRT and surgery were tumor free. One mouse in each group treated with LRT and surgery was lost during surgery due to tumor infiltration of the chest wall. After tumor rechallenge, tumor growth was significantly smaller in the LRT-Surg 7d group compared with the other 2 groups (Fig. 5). In the LRT-Surg 7d group, 3 of 9 mice completely rejected the tumor, whereas no tumor rejection occurred in the other two groups. This finding suggests that accelerated nonablative hypofractionated radiation 7 days before surgical removal of the tumors promotes a protective immunologic memory response.
Role of T cells in the long-term protection of radiated mice
To determine the role of CD4+ and CD8+ T cells in the long-term protection of radiated mice, we first confirmed that cured mice were protected by rechallenging them a second time with AE17-OVA; 20 of 21 mice rejected their tumor. These 20 mice were then rechallenged again after randomization into the following groups: (i) CD4 depletion (n = 6), (ii) CD8 depletion (n = 7), (iii) Double depletion (n = 7). Depletion of lymphocytes started one week before AE17-OVA cells inoculation.
Double depleted mice for CD4+ and CD8+ cells lost their antitumor memory response and displayed rapid tumor growth rate, similar to untreated mice challenged for the first time. In contrast, all CD8+-depleted mice completely rejected the tumor (Fig. 5). CD4+-depleted mice rejected 1 of 6 tumors and growth rate of the remaining 5 tumors was significantly slower than double depleted mice. These results suggest that memory CD4+ T cells are critical to provide long-term benefit from radiation, while memory CD8+ T cells contribute to rejection of the tumor in the long-term but are not sufficient to provide long-term immune protection against the tumor. Hence, CD4+ T cells are sufficient to mount an effective immune response against the tumor in the long-term despite the absence of CD8+ T cells.
CTLA-4 blockade improves the beneficial effect of LRT
In this experiment, the goal was to determine the role of immunotherapy combined with a short course of nonablative radiation in improving the therapeutic effect of radiation on tumor growth. We therefore compared the effect of immune check point blockade in combination with hypofractionated radiation. Anti-CTLA-4, anti-PD-1, and anti-PD-L1 have been successfully used in combination with radiation previously, but have not been compared to each other (19–22). Mesothelioma tumor–bearing mice were therefore randomized to: (i) no treatment, (ii) anti-PD1, (iii) anti-PDL1, (iv) anti-CTLA-4, (v) LRT, (vi) LRT and anti-PD1, (vii) LRT and anti-PDL1, and (viii) LRT and anti-CTLA-4. Treatment started on day 9 in all groups, with LRT and/or mAb injection. The mAb injection was repeated every 3 days for a total of 3 doses. The AE17 and AB12 cell lines do not express PD-L1 (Supplementary Fig. S1). However, as shown in Fig. 3, a large proportion of tumor-infiltrating lymphocytes express PD-1, suggesting that targeting the PD1/PD-L1 pathway may be of value.
Treatment with mAbs alone had no significant impact on tumor growth. Mice treated with LRT and anti-CTLA-4 antibody had significantly smaller tumors compared with mice treated with LRT only. There was limited difference in tumor size between LRT alone and the combination group LRT and anti-PD1 or LRT and anti-PDL1 (Fig. 6). In an additional study, we observed that mice treated with LRT and anti-CTLA-4 had three distinct response patterns, no response, partial response, and good response (Fig. 6). IFNγ-producing CD8+ T cells infiltrating the tumor were significantly higher in the combination group LRT and anti-CTLA-4 compared with the untreated group (Fig. 6).
Discussion
Previous work had shown that radiation could enhance the abscopal effect in a tumor bearing host, improve the capacity of adoptively transferred T cell to infiltrate the tumor, and synergize with immune checkpoint blockade to provide better local and distant control of the tumor (23–25). These previous experiments, however, focused on treating the primary tumor with radiation alone. Although surgery could be an important adjunct to radiation to provide optimal control of the primary tumor, the potential impact of surgery in the context of hypofractionated radiation on the immune system has, to the best of our knowledge, never been experimentally analyzed. Surgery could potentially limit the beneficial effect of radiation on the immune system by removing the source of tumor neoantigen release or by creating a nonspecific anti-inflammatory state in the vicinity of the tumor.
In this study, we combined a nonablative dose of radiation with surgery to treat the primary tumor and analyzed the impact of both treatments on the immune system in the short and long term. We observed that surgery alone led to a faster tumor growth compared with untreated mice suggesting that the nonspecific inflammatory reaction created by the surgical dissection was detrimental in this mesothelioma model. Accelerated, hypofractionated radiation prior to surgery, however, prevented the negative impact of surgery. We also found that a short course of nonablative radiation before complete resection of the tumor could provide an in situ vaccination with long-term protection if there was sufficient time for the immune system to generate a specific immune response against the tumor before surgery.
Radiation is traditionally delivered with normofractionated doses of 2 Gy per fraction, generally for 5 to 6 weeks to obtain a radical dose. However, over the past decade hypofractionated treatment with higher dose delivered in a shorter time frame has been increasingly used in the adjuvant setting after surgery to limit the inconvenience and cost related to this long treatment (26, 27). The data presented in these series of experiments suggest that hypofractionated radiation treatment switched from the adjuvant to the neoadjuvant setting may enhance the immunoprotective effect, thus keeping the total dose to a minimum.
T cells are extremely sensitive to radiation and therefore tumor-infiltrating lymphocytes are damaged after each dose of radiation (5, 28, 29). This toxic effect of radiation on T cells therefore masks the potential benefit of radiation on the immune system in conventional radiation due to the daily delivery of radiation over the course of several weeks. A short course of radiation in contrast takes full advantage of the activation of the immune system by allowing the lymphocytes from the systemic pool to reinfiltrate the tumor not long after the end of radiation and continue to rise once the radiation is stopped. In accord with previous studies, we observed that after 7 days the number of infiltrating T cells in the tumor was at least 5 times higher after radiation than in untreated tumors and that the rise in tumor-infiltrating lymphocytes was tumor antigen specific (30, 31).
Lymphocytes infiltrating solid tumors after radiation have been examined in different murine models, including previous publications from our group (4, 18). Radiation can reverse the immunosuppressive state of the tumor by decreasing the number of myeloid-derived suppressor cells, switching tumor-infiltrating macrophage from an M2 to an M1 phenotype and creating a proinflammatory environment where danger signals and cytokines generate the migration of lymphocytes toward the tumor (30–32). Accordingly in our model, we observed that this short radiation course rapidly upregulated the expression of 4-1BB and decreased the expression of PD-1 on lymphocytes. Although the expression of 4-1BB was only temporary, its impact on the response to radiation may be significant as expression of 4-1BB on lymphocytes can stimulate production of IL2 in a CD28-independent way, a critical step for activation of T cells and for prevention of an anergic state (33). In addition, 4-1BB also enhances CTL cytolytic activity (34). The decreased expression of PD-1 on the other hand was prolonged suggesting the absence of T cells' exhaustion over time (35, 36).
Double depletion of CD4+ and CD8+ T cells significantly abolished some of the early therapeutic benefit of radiation on tumor growth in our model. This finding suggests that both CD4+ and CD8+ T cells play an important role in the therapeutic response to radiation and contribute to the benefit of radiation in our model. These observations are in line with the literature indicating that CD8+ T cells are required for the therapeutic effects of ablative radiation (6, 37). The role of CD4+ T cells in the early response to radiation has been more variable, possibly due to the presence Treg such as CD4+CD25+Foxp3 within the tumor (18).
We and others have shown that Treg are upregulated in the tumor after a short course of radiation (18, 38). In our experience, the use of CTLA-4 blockade was able to block the upregulation of Treg, thus enhancing the impact of radiation on the primary tumor and the abscopal effect at distant tumor sites (18). The benefit of CTLA-4 blockade in combination with radiation in this series of experiments confirms our previous observation and supports the potential importance of Treg in generating the immunosuppressive microenvironment in mesothelioma (39, 40).
In contrast to the early response to radiation, this series of experiments demonstrate the critical role of CD4+ T cells in the long term. Mice previously protected against the tumor failed to reject the tumor when CD4+ and CD8+ T cells were depleted (double depletion). CD4+ depletion alone led to the development of slowly growing tumor suggesting that CD8+ T cells and other immune cells were still able to respond to the tumor but could not prevent their development. On the other hand, mice depleted in CD8+ alone did not develop the tumor, demonstrating that memory CD4+ cells are a key component in the prevention of tumor recurrences in the long term. These findings reinforce the increasing evidence that CD4+ T cells are a key component to generate long-term protective antitumor immunity, while tumor will typically escape isolated protection from CD8+ T cells alone (41, 42).
A time frame of 7 days or potentially longer between radiation and surgery is important to reach the peak of the adaptive T-cell immune response before surgery (6, 43). Mice treated with radical surgery alone or with radiation followed by radical surgery within 24 hours displayed rapid tumor regrowth after tumor rechallenge despite their initial cure, demonstrating the absence of durable immune protection without an adequate activation of the immune system. This finding correlates with clinical observation where a time gap of at least 5 days between accelerated radiation and surgery in locally advanced rectal cancer was shown to be a powerful and independent prognostic factor for better survival (44).
This study has limitations inherent to the animal models and correlative studies in patients will be important. We are thus currently planning to modify our induction radiation protocol to 15 Gy in 3 fractions. This approach would allow us to preserve the lung at the time of surgery and facilitate the concurrent implementation of immune checkpoint inhibitors.
In conclusion, we demonstrated the importance of the immune system in the benefit of clinical protocols using accelerated, hypofractionated radiation followed by surgery. This benefit is associated with CD8+ and CD4+ T cells and can lead to an in vivo vaccination with long-term protection. These findings suggest that these protocols provide an opportunity to introduce immune checkpoint blockade in clinical practice.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Authors' Contributions
Conception and design: L. De La Maza, M. Wu, L. Wu, M. de Perrot
Development of methodology: L. De La Maza, M. Wu, L. Wu, Y. Zhao, M. de Perrot
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. De La Maza, L. Wu, M. de Perrot
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. De La Maza, L. Wu, Y. Zhao, M. Cattral, M. de Perrot
Writing, review, and/or revision of the manuscript: L. De La Maza, M. Wu, L. Wu, Y. Zhao, M. Cattral, A. McCart, J. Cho, M. de Perrot
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. De La Maza, L. Wu, H. Yun, M. de Perrot
Study supervision: L. Wu, M. de Perrot
Other (provided feedback over the course of the study re design, analysis, and conclusions): A. McCart
Grant Support
This work was supported by the Mesothelioma Applied Research Foundation, the Princess Margaret Hospital Foundation, and the Canadian Mesothelioma Foundation.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.