In Vivo Effects of Lattice Radiation Therapy on Local and Distant Lung Cancer: Potential Role of Immunomodulation

Abstract

Radiation is a potent immune-modulator that elicits cell death upon tumor, stromal and angiogenic compartments of tumor microenvironment. Here, we test a novel approach of high-dose radiation delivery using three dimensional volume based lattice radiation therapy (LRT) to understand the impact of different volume irradiation in eliciting both local and metastatic/distant tumor control through modulation of tumor immune micro-environment. To study such effects of LRT, tumors were implanted in both hind legs of C57BL/6 mice using Lewis lung carcinoma 1 (LLC1) cells. Mice were divided into five groups: untreated; partial tumor volume groups included two 10% vertices, one 20% vertex and one 50% vertex of the total tumor volume; and 100% open-field irradiation. Tumors implanted in the left flank were irradiated with a single dose of 20 Gy while the tumors in the right flank were unirradiated. Tumor growth and regression as well as immune responses (such as Th1 and Th2; T-cell infiltration) were determined after radiation treatment. Results demonstrated that both 100% open-field irradiation and 20% volume irradiation (in two 10% volumes) resulted in significant growth delay in the irradiated tumor. Further, all types of radiation exposures, partial or 100% volume, demonstrated distal effectiveness, however, 20% volume irradiation (in two 10% volumes) and 50% tumor volume irradiation led to maximum growth delay. Mice treated with partial tumor volume radiation induced a robust IFN-γ and Th1 response when compared to whole-tumor irradiation and down-modulated Th2 functions. The presence of increased CD3⁺ cells and TRAIL in partially irradiated tumor volumes correlated well with tumor growth delay. Further, serum obtained from any of the LRT treated mice caused growth inhibition of endothelial cells when compared to serum obtained from either untreated or open-field irradiated groups. These results indicate that high-dose partial volume irradiation can cause an improved distant effect than the total tumor volume irradiation through activating the host immune system.

INTRODUCTION

Radiotherapy (RT) is the most commonly used therapeutic modality for most cancers in combination with/without chemotherapy and surgery. However, in certain tumors, conventional RT effectively targets only local tumors while leaving the metastatic tumors intact, with high chances of tumor recurrence and causes radiation-induced injury to normal tissue surroundings. The profound effects of ionizing radiation (IR) on tumors can be significantly improved by increasing the radiation dose in a hypofractionation setting, yet without affecting the surrounding normal tissues, however, this is compromised by small tumor volume. In recent years, several new RT strategies have been reported to deliver high dose of radiation, which includes two-dimensional (2D) high-dose GRID called Spatially Fractionated Grid Radiation (SFGRT) (1–4); the standard of care approaches such as three-dimensional (3D) conformal radiation (5, 6); intensity-modulated radiation (6); and newly introduced modern 3D high-dose lattice radiotherapy (LRT) (7).

Hypofractionated radiation therapy is a categorical approach that uses 3–6 Gy fractions whereby the α/β ratio, biologically effective dose (BED) and four R’s of radiotherapy are applicable. Hypofractionation schedules that use doses above 8 Gy per fraction whereby the biological changes can potentially differ from the “classical” four R’s is generally known as the high-dose hypofractionation radiation therapy (HDHRT). There are data to suggest that the use of HDHRT is effective as an alternative means of dose escalation with conventional fractionation treatment schedule. Biologically, new mechanistic insights suggest that HDHRT may cause four unique effects that can be further exploited for sensitization. HDHRT at doses above 15 Gy can (a) cause nontargeted pharmacodynamics intra-tumoral bystander effects as well as abscopal effects potentially mediated by TNF-α, TRAIL, PAR-4 and ceramide (1, 2, 8); (b) tumor endothelial cell death at doses above 8–11 Gy (9); (c) increased recognition of radiation-induced enhanced antigen presentation, such that a single fraction may incite an immune response that enhances the effects of radiation (10); and (d) result in a better response of those tumors that are heterogeneous with different cell populations and different clonal radiosensitivity (11).

The abscopal effect is potentially important for tumor control and is mediated through cytokines and/or the immune system (2, 8, 12–15). In this context, inducing antitumor immunity, mainly cellular-mediated immune responses, has become a major focus of tumor regression studies (15, 16). Increased activation of tumor specific cytotoxic T cells can directly target tumors irrespective of their proliferation efficiency, location and chemo-sensitivity (17). In addition, maintaining strong memory T-cell responses against tumors may give long-term protection against recurring tumors. However, inducing strong immune responses to tumor cells is hampered by poor immunogenicity and lack of co-stimulatory molecules to activate naive T cells. Although, ionizing radiation exposure can trigger anticancer immune responses, to mount efficient immune responses a very high dose of ionizing radiation is required suggesting that a massive tumor necrosis is essential to induce specific immune responses. This high-dose IR can be easily delivered using 3D LRT that was developed based on our previous radiobiological data of tumor growth using GRID (7). In the current study, we investigated the effects of high-dose LRT delivered to partial tumor volumes in eliciting both local and metastatic/distant tumor control through modulation of tumor immune microenvironment. Findings of this data has the potential to translate in clinic as well as further understand biologic basis for combining immune modulators with HDHRT.

MATERIALS AND METHODS

Mice, Cells and Injection

Female C57/BL6 mice, 6 weeks of age, were purchased from Jackson Laboratory (Bar Harbor, MA) and allowed to acclimate for at least a week before use. The mice were housed in a pathogen free environment and cared for according to the guidelines stipulated in the “Guide for the Care and Use of Laboratory Animals”, DHHS publication NIH-85–23 and approved by Institutional Animal Care and Use Committee (IACUC) protocol. The Lewis lung carcinoma 1 (LLC1), a mouse cancer cell line that can develop syngenic tumors in C57BL/6 mice, and HUVEC cells were purchased from the American Type Culture Collection (Manassas, VA). LLC1 cells were cultured in Dulbecco’s modified essential medium (DMEM; GIBCO) supplemented with penicillin-streptomycin (1%; GIBCO), sodium pyruvate (0.1%; GIBCO) and fetal bovine serum (FBS, 10%; Atlanta Biologicals, Lawrenceville, GA) at 378C and 5% CO2. HUVEC cells were cultured in ATCC-formulated F-12K medium supplemented with 0.1 mg/mL heparin and 0.03–0.05mg/mL endothelial cell growth supplement (ECGS) and FBS (10%).

For experiments in which tumor-bearing mice were used, tumors were initiated by subcutaneous injection of 2 × 10⁶ LLC1 cells into the flanks of both left and right hind legs of each mouse. Tumors were allowed to grow to an area of 5 × 5 mm before irradiation and sorted within 10% differences in the intra- and inter-tumor volumes.

Irradiation of Tumors

Mice tumors were irradiated using a novel method called LRT and compared with open-field radiation treatment. LRT was developed based on a long-standing technique GRID, or spatially fractionated radiation (SFGRT) that delivers radiation in a pattern of high-dose “pipes” surrounded by lower dose regions within a radiation field, allowing for normal tissue sparing that is not possible when the entire tumor is treated with high dose (4). LRT is a dosimetrically superior three-dimensional technique for delivery of spatially fractionated radiation, delivering high-dose radiation islands within the tumor, but allowing for maximal normal tissue sparing compared to the GRID technique (7).

A 100 kV industrial X-ray machine (Phillips, Netherlands) was used for the radiation treatments. The dose rate with a 2 mm aluminum plus 1 mm beryllium filter was ~1.85 Gy/min at a focus-surface distance of 10 cm.

Open-Field (Conventional IR) Irradiation

For open-field irradiation, whole tumors were irradiated at a dose of 20 Gy.

Rotational Delivery Platform for High-Dose LRT

A rotating platform was fabricated to generate multiple focused high-dose regions (lattice) in a mouse tumor using 100 kV X rays. As shown in Fig. 1A, the mice-rotating platform was designed to have two rotating axis, one for motorized continuous rotation, second for manual adjustment to set the angle θ, between the beam axis and the rotation axis, allowing multiple positioning of high-dose points for rotating-focus delivery.

FIG. 1.

Rotational delivery device for high-dose LRT delivery in mice. Panel A: GAFCHROMIC Film-Cube Phantom LATTICE Irradiation Setup. Panel B: Sequential Film Analysis showing LATTICE formation. Panel C: Dose profiles of lattice. Panel D: Maximum dose at a specific depth normalized to the dose at the lattice vertex. Panel E: Scheme of different types of radiation treatments given to the left side tumor in mice bearing subcutaneous tumors in both right and left flanks. LATTICE radiotherapy is a novel three-dimensional technique that delivers radiation high-dose islands to large tumors with minimal damage to surrounding normal tissue. To deliver a total dose of 20 Gy, we used three different LRT set ups: (1) lattice two (10%) vertices, (2) lattice one (20%) vertex and (3) lattice one (50%) vertex. The other two groups were untreated control group and open-field irradiation to the whole tumor.

To test the applicability of the mice-rotating platform, a stack of GAFCHROMIC EBT films were assembled into a 2.5 × 2.5× 2.5 cm film cube phantom. The cube was then affixed to the rotating platform. A special collimator was fabricated to form a single circular beam of 1 mm size. The cube is positioned such that the rotating axis and the beam intersect at the depth of 9 mm. The platform was then rotated to 25° to initiate cone rotation for 10 min beam-on time. The film cube was offset four times to repeat the same irradiation forming five focused high-dose islands (Fig. 1A). The dose rate was 80 cGy/min (free-space) at the rotating intersection. The cone-dose distribution is demonstrated in Fig. 1B. The test shows that the device was capable of delivering high-dose lattice to small tumor size in animal settings with high precision and steep dose gradient (Fig. 1C). The dose-sparing effect by the rotating cone technique is shown in Fig. 1D, where dose at the surface is only 20% of the maximum dose delivered to the lattice dose vertex.

Implanted murine LLC1 syngenic tumor of the left hind leg of C57/BL6 mice were treated with single 20 Gy lattice high-dose vertices that included Group I with no treatment; Group II: left tumors were treated with two 10% vertices (with total irradiated tumor volume of 20%); Group III: One 20% vertex (with total irradiated tumor volume of 20%); Group IV: One 50% vertex (with total irradiated tumor volume of 50%); and Group V: 100% of the tumor volume was irradiated (Fig. 1E). Each group had 12 mice that received radiation treatment when tumors reached a volume of 25–35 mm3. For each mouse irradiation, exact tumor volume was calculated to determine total volume of radiation based on the LRT groups. These experiments were performed twice.

Tumor Growth Regression and Delay

Tumor volumes were measured with calipers before irradiation and every alternate day after irradiation until the tumor size reached 10% of the mouse’s body weight. Tumor volume was calculated according to Feldman et al. (18) using the formula: π/6 × 1.58 (length × width)3/2. Tumor volume at each time point (V_t) was normalized to the initial volume (V₀) to calculate the fold change in tumor volume. Treatment enhancement ratios were calculated as V_t/V₀ of the control untreated group divided by V_t/V₀ of the treated group. Tumor growth delay (TGD) was calculated as the time required for increase in the tumor volume to five times the initial tumor volume.

ELISpot Assay

Three mice from each group were sacrificed on days 3 and 7 after irradiation to collect the spleen. Enzyme-linked immunosorbent spot (ELISPOT) assays were performed to determine LLC1 tumor specific cytokine secretion; IL-2, IFN-γ and IL-6 from untreated and irradiated mice splenocytes. ELISPOTS were carried out as per the manufacturer’s protocols (R&D Systems) using 96-well plates (Millipore). The plates were coated with 10 μg/ml rat anti-mouse IFN-γ or IL-2 or IL-6 antibodies in PBS at 4°C overnight. The plates were washed three times with PBS and blocked with complete growth media for 2 h. Frozen mice splenocytes (1.5 × 10⁶ cells/well) were thawed and added to each well of the ELISPOT plate and stimulated for 24 h at 37°C, 5% CO₂, in the presence of LLC1 cell lysate (60 μg/well). After 24 h, spots were developed with biotinylated anti-IFN-γ, or IL-2 or IL-6 antibodies (R&D Systems), followed by Streptavidin-horse radish peroxidase (HRP) (Vector Laboratories, Burlingame, CA) and an AEC, 3-amino-9-ethylcarbazole, substrate kit (Vector Laboratories, Burlingame, CA) according to the manufacturer’s instructions. The membrane was read by automated reader (CTL Immunospot, Shaker Heights, OH) for quantitative analyses of the number of IL-2, IL-6 and IFN-γ spots forming cells (SFC) per 1.5 million cells plated.

ELISA

Three mice from each group were sacrificed on days 3 and 7 after irradiation to collect blood from heart-puncture and 150–200 μl of serum was collected. Presence of IL-2, TNF-α and IFN-γ in mice serum was detected by multi-analyte ELISArray kit (SABiosciences, Valencia, CA) as per the manufacture’s protocol. Presence of IL-4, IL-10 and CXCL1/KC in the serum was determined using DuoSet ELISA development kit (R&D Systems, Minneapolis, MN).

Real-Time Cell Electronic Sensing (RTCES) Method

HUVEC cells were pre-plated on the surfaces of microelectronic sensors that are composed of circle-on-line electrode arrays and are integrated into the bottom surfaces of the 16-well plate. HUVEC cells were then either incubated with the serum obtained from mice sacrificed 3 days after lattice or open-field irradiation or from the untreated group. Controls were incubated with normal FBS. Changes in cell number were monitored and quantified by detecting sensor electrical impedance. The dynamic response of the cells to these treatments was continuously monitored by real-time cell electronic sensing (RTCES; xCELLigence) system (Roche Applied Science) by obtaining the data points at every 12 h. Cell numbers at each time point (N_t) were normalized to the cell numbers at the time of treatment (N₀).

Acid Sphingomyelinase (ASMase) Assay

Three mice from each group were sacrificed on day 7 after irradiation to collect blood from heart-puncture and 150–200 μl of serum was collected. ASMase assay was performed as per the manufacturer’s protocols (Amplex Red Sphingomyelinase Assay Kit, Invitrogen) using 96-well plates. Serum samples (10 μl) from each mouse were used for this assay and were done in triplicates.

Immunohistochemistry

Tumor samples resected from control and treated mice were fixed in formalin. Paraffin embedding steps were performed with Thermo Shandon Pathcentre Tissue Processor (Thermo Scientific). Slides with 4–5 μm thick tissue sections were prepared for immunohistochemical staining.

Tumor sections were probed with CD3 antibody for T-cell infiltration and with TRAIL antibody. Paraffin embedded tissue sections were deparaffinized by xylene and rehydrated in 100%, 75% and 50% ethanol then distilled water. Blocking and antibody staining steps were performed with Biocare autostainer, Nemesis 3600 (Biocare Medical). Antigen retrieval step was accomplished by heating the slides in citrate buffer (Target Retrieval Solution, Citrate pH 6, Dako), followed by peroxidase (Peroxidazed-1, Biocare Medical) and protein block (Background Sniper, Biocare-Medical) for 10 min each. Tumor sections were incubated with primary antibody (Cell Signaling Tech.) for 30 min followed by secondary antibody (4 plus Biotinylated Universal Goat Link; Biocare-Medical) for 15 min and then by Streptavidin-HRP (4plus Streptavidin HRP Label; Biocare-Medical) for 15 min. Sections were then stained with 3, 3′ diaminobenzidine (DAB) (Betazoid DAB Chromogen Kit; Biocare-Medical) to detect antibody positive regions. The slides were counter stained with hematoxylin, dehydrated and sealed with a cover slip and sealant. Entire slides were digitally scanned by an AperioScanScope CS scanning system (Aperio Technologies Inc., Vista, CA) and analyzed by the Aperio ImageScope Viewer software. The positive pixel count v9 algorithm was used to measure positive tumor regions for CD3.

Statistical Analysis

For the syngenic tumor studies, the associations between mean differences among groups were tested by use of the two-tailed paired Student’s t test. For ELISPOT, ELISA and RTCES, the results of test of significance are reported as P values calculated from Student’s t test using two-tailed distribution.

RESULTS

Single Fraction, High-Dose LRT Significantly Delayed Growth of Both Local and Distant Tumors

To study the effects of LRT on tumor growth (both local and distant tumors), we implanted LLC1 cells in both right and left hind legs of mice. When tumor size reached around 5 × 5 mm, animals were randomized based on their tumor size for irradiation using two lattice 10% volume vertices or one lattice 20% volume vertex or one lattice 50% volume vertex or open-field irradiation (100% tumor volume) or left unirradiated (Fig. 1E). Local irradiation with a single dose of 20 Gy was delivered to the implanted left-side tumors as described in the Materials and Methods. Since LLC1 is a very aggressive tumor, tumor growth was measured at alternate days after radiation treatment. The tumors (both left- and right-side tumors) of unirradiated control mice grew progressively and reached the maximum tumor growth within 12 days (Fig. 2). Mice treated with two lattice 10% vertices had reduced tumor growth both locally and distantly suggesting that 20% irradiated tumor volume has the potential to cause delay in the growth of the primary tumor potentially through an intra-tumoral bystander event on the surrounding unirradiated tumor cells and of the distant unirradiated tumor (located at the right hind leg) through abscopal effects (Fig. 2). However, when 20% of the tumor volume was irradiated in a single vertex (one 20% lattice vertex) the effects on tumor growth were less than two 10% vertices group. This suggests that both intra-tumoral bystander and extra-tumoral abscopal effects are important for tumor growth inhibitory and delay effects. On the contrary, the conventional open-field irradiation to the whole tumor was more effective in the directly irradiated left tumor compared to the unirradiated right tumor. Interestingly, lattice single (50%) vertex did not have any significant effect on the growth of irradiated tumor but had abscopal effect on distant unirradiated tumor (Fig. 2). Overall, two lattice 10% vertices treatment was significantly effective in inducing tumor growth delay of both left and right tumors (P < 0.025 and P < 0.05, respectively) while one lattice 50% vertex treatment was more effective in right tumor (P = 0.034). This experiment was independently performed twice and similar results were observed.

FIG. 2.

LLC cells were injected in the backside (near both the back flanks) of C57BL/6 mice to implant two tumors. Tumor in the left flank was either left untreated or irradiated (20 Gy) with either LRT (two 10% vertices, one 20% vertex or one 50% vertex) or open-field irradiation (100% tumor volume). Tumor diameters were measured using digital calipers and tumor volume was calculated. Tumor volume at each time point (V_t) was normalized to the initial volume (V₀). Standard error is shown as the error bars. Panel A: Left-side tumor and (panel B) right-side tumor. Panel C: Treatment enhancement ratios were determined by dividing V_t/V₀ of untreated control with Vt/V0 of different treatments. P values for LRT groups compared with open-field irradiation and ≤0.05 are indicated by *. Panel D: Tumor growth delay is shown in the form of time required (in days) for increase in the tumor volume to 5 times the initial tumor volume (Δt_5x).

Immune Responses in LLC Tumor Model: Lattice High-Dose Radiation Induces Increased Secretion of Inflammatory Cytokines

Having ascertained that there are local and distant tumor growth delay effects of partial tumor volume irradiation, we then analyzed if the activation of immune responses play a role in such observed LRT effects using ELISA for cytokine secretion (IFN-γ, IL-2, IL-10 and IL-4), ELISPOT for lymphocytes responses (IL-2, IL-6 and IFN-γ), and immunohistochemistry (IHC) to quantify intra-tumoral CD3 T-cell infiltration. Analysis was mainly focused on cytokines that were implicated in tumor regression and/or involved in T-cell mediated immune responses, particularly the Th1 mediated inflammatory cytokines. Cytokine IFN-γ is implicated in tumor growth suppression (19). Expression of IFN-γ in the early phase of the tumor growth is important as it induces specific antigenic immune responses to irradiated tumors (20). Radiation treatment including all types of LRT and open-field moderately induced IFN-γ secretion in the serum obtained 3 days postirradiation (Fig. 3A). However, IFN-γ levels reduced in 7 days postirradiation serum.

FIG. 3.

Secretion of cytokines in the serum of tumor bearing mice collected after days 3 and 7 of different treatments was assessed by ELISA. Serum samples (15 μl) were used for each cytokine. Levels of TH1 mediated inflammatory cytokines: IFN-γ (panel A); IL-2 (panel B) and levels of TH2 mediated cytokines important for inducing humoral responses, (panel C) IL-4 and (panel D) IL-10 were determined. In each experiment, 3 animals per group were analyzed.

IL-2 is produced by macrophages and T cells that play a crucial role in T-cell responses by maintaining memory, proliferation and differentiation of T cells (21, 22). IL-2 was significantly induced in both days 3 and 7 serum samples obtained after treatment with one 50% lattice vertex (Fig. 3B). Although open-field radiation treatment did not induce IL-2 secretion on day 3, there was a significant increase on day 7 (P = 0.07) (Fig. 3B). We did not observe any changes in the other two LRT groups compared to untreated control groups. Since LRT showed a strong effect on the cytokines secretion involved in Th1-mediated responses, we studied the effect of various radiation treatments on cytokines involved in humoral responses such as IL-4 and IL-10. As shown in Fig. 3C and D, there was a significant decrease in both IL-4 (P = 0.16 at day 3 and 0.03 at day 7) and IL-10 (P = 0.0015 at day 3 and 0.006 at day 7) secretion in serum obtained from one 50% lattice vertex at both time points. Treatment with two 10% lattice vertices significantly reduced IL-4 (P = 0.09 at day 3 and 0.038 at day 7) and IL-10 (P = 0.00096 at day 3 and 0.575 at day 7) secretion in serum although IL-10 levels returned to normal on day 7. These results clearly demonstrate that LRT induces robust systemic cellular mediated immune responses.

LATTICE Induced LLC1 Tumor Antigen-Specific IFN-γ, IL-6 and IL-2 Secretion by Splenocytes

Since we observed an increase in Th1 mediated cytokines in serum, next we analyzed the immune response of IL-2, IL-6 and IFN-γ secreting splenocytes using ELISPOT in samples collected after day 3 of radiation treatment. Using the ELISPOT assay, we quantified the LLC1 tumor antigen specific IL-2, IL-6 and IFN-γ secreting splenocytes stimulated with LLC1 tumor lysate for 24 h. As shown in Fig. 4, two 10% lattice vertices (P = 0.015 for IL-2 and 0.15 for IFN) and one 50% lattice vertex (P = 0.001 for IL-2 and 0.008 for IFN) significantly increased LLC1 tumor specific IL-2 and IFN-γ secreting splenocytes compared to one 20% lattice vertex as well as open-field radiation treatment. However, in samples collected after day 7 of radiation treatment, LLC1 tumor specific IL-2 secreting splenocytes levels returned to normal (data not shown). In contrast to IL-2 and IFN-γ, both LRT and open-field irradiation resulted in a reduction in LLC1 tumor specific IL-6 secreting splenocytes with maximum reduction observation following two 10% lattice vertices and one 50% lattice vertex.

FIG. 4.

To determine the radiation-induced tumor antigen-specific IL-2, IL-6 and IFN-γ secretion by splenocytes, LLC1 lysate reactive splenocytes were quantified using ELISPOT assay. Splenocytes from untreated, lattice two (10%) vertices, lattice one (50%) vertex, lattice one (20%) vertex, and open-field IR were cultured with LLC1 lysate (60 μg/well) for 24 h. Data are shown as the number of spots or reactive splenocytes from the total cells obtained 3 days postirradiation.

Increased CD3 Infiltration in Tumors from Mice Irradiated with High-Dose Radiation

Next, we analyzed for T-cell lymphocyte infiltration into the tumor in response to different volumetric radiation treatments. Tumor infiltrating T-cell lymphocytes are crucial for tumor regression mediated by immunogenic cell death. Both ionizing radiation treated and untreated tumors were collected after days 3 and 7 and stained for CD3⁺ lymphocytes as described in the Materials and Methods. As shown in Fig. 5A and B, one 50% lattice vertex and one 20% lattice vertex significantly enhanced CD3 infiltration in both irradiated and unirradiated tumors. One 50% lattice vertex treatment consistently enhanced CD3⁺ T-cell infiltration compared to other LRT volumetric treatments as well as open-field radiation treatment. Increased CD3⁺ tumor infiltrating lymphocytes correlated with reduced tumor growth in response to LRT treatments indicating that the abscopal response can be attributed to cellular mediated immunity.

FIG. 5.

Panel A: Immunohistochemistry staining for CD3 infiltration in tumors obtained from mice irradiated with high-dose radiation at days 3 and 7 post treatment. Antibody positive regions are brown. Panel B: Entire slides were digitally scanned and analyzed by the AperioImageScope Viewer software. The positive pixel count v9 algorithm was used to measure positive tumor regions for CD3 and shown as percent T-cell infiltration.

Effect of Ionizing Radiation on Keratinocyte Chemoattractant (KC/GROα/CXCL1)

Increased secretion of KC leads to angiogenesis, tumorigenesis and inflammation (23, 24). Above results clearly demonstrated that LRT suppresses both local and distant tumor growth, induces cytokines responsible for cellular mediated immune response leading to increased CD3⁺ T-cell infiltration in tumor. In this experiment, we further examined KC expression by ELISA using serum from 3 and 7 days post IR, to understand the contribution of KC immune response towards prosurvival. Interestingly, KC expression level decreased in response to one 50% lattice vertex (P = 0.004 at day 7), one 20% lattice vertex (P = 0.007 at day 7) and open-field IR (P = 0.001 at day 7) treatments compared to two 10% lattice vertices (P = 0.845 at day 7) (Fig. 6). These results indicate there is some degree of prosurvival immune response events coupled with immunogenic cell death events in two 10% lattice vertices group.

FIG. 6.

Secretion of keratinocyte chemo attractant (KC/GROα/CXCL1) in the serum of tumor bearing mice collected after days 3 and 7 of different treatments as assessed by ELISA. Serum samples (15 μm) were used. In each experiment, 3 animals per group were analyzed.

Real-Time Analysis of HUVEC Cell Growth, ASMase Levels, and TNF-α Secretion Using Serum Obtained from Different Radiation Treatments and TRAIL Expression in the Tumors

High-dose radiation greater than 12 Gy induces endothelial cell death (9) and we reported that post-SFGRT treatment serum samples caused induction of apoptosis in HUVECs (1). In addition, vascular endothelial cells has been reported to play an integral part in the antigen presentation event during cell-mediated immune phenomena (25). Since there was a robust systemic cell-mediated immune response for LRT volumetric single dose of 20 Gy, effect of serum obtained from groups I to V was tested on endothelial cell growth as assessed by RTCES. As shown in Fig. 7A, when the HUVEC cells were grown in the serum obtained from any of the LRT groups, the cell growth was significantly reduced compared to cells that were grown in the serum obtained from either the untreated or open-field IR groups. Maximal effect on HUVEC cell growth inhibition was observed with serum from one 50% lattice vertex group (P = 0.006 at 120 h).

FIG. 7.

Panel A: HUVEC cells were either incubated with the serum obtained from irradiated mice 3 days after lattice or open-field irradiation or with FBS. Changes in cell number were monitored continuously by obtaining the data points at every 12 h and quantified by detecting sensor electrical impedance. Cell numbers at each time point (N_t) were normalized to the cell numbers at the time of treatment (N0). *P < 0.01; (panel B) ASMase assay was carried out using 10 μl of serum samples obtained from mice 7 days after irradiation; (panel C) secretion of TNF-α in the serum of tumor bearing mice collected after days 3 and 7 of different treatments was assessed by ELISA. Serum samples (15 μl) were used, and (panel D) Immunohistochemistry staining for TRAIL in tumors obtained from mice irradiated with high-dose radiation at days 3 and 7 post-treatment. Antibody positive regions are brown.

Previously, we reported that in patients treated with SFGRT, an increase in ASMase activity with LDL-enriched ceramide was observed in post-treatment serum samples (1). In this study, we analyzed ASMase activity in the mouse serum after 7 days and correlated with effect on HUVECs. There was a significant increase in ASMase activity levels after LRT and open-field irradiation, however, effects were maximal with LRT two 10% vertices and one 50% vertex at day 7 (Fig. 7B). This increase correlated with HUVECs growth inhibition.

We previously reported that increased levels of serum TNF-α correlated with overall survival to SFGRT (2). Further, we reported that A549 cells secreted increased TNF-α and TRAIL in response to 10 Gy dose and these factors were responsible for induction of apoptosis and clonogenic death in untreated A549 cells (8). Based on this observation, these studies analyzed the levels of TNF-α in serum and TRAIL expression in tumor tissue and correlated with tumor growth delay. Compared to the untreated control group, increased secretion of TNF-α in the serum (Fig. 7C) was observed at day 3. Expression of TRAIL was elevated in irradiated left-side tumors plus unirradiated right-side tumors of all the treatment groups (Fig. 7D) compared to controls with higher elevations observed in 50% as well as 20% tumor volume irradiation. These findings demonstrated that partial tumor volume treatment with high-dose irradiation of 20 Gy can mitigate the growth of vasculature and simultaneously can increase apoptosis in the epithelial compartment.

DISCUSSION

A well-known phenomenon of radiation treatment, apart from direct effects, are the nontargeted effects and this has been extensively reviewed in ref. (26). Recently, nontargeted events in the form of bystander (within the irradiated tumor) and abscopal (distant unirradiated tumor) effects were demonstrated in mice bearing contra-lateral lung adenocarcinoma xenograft tumors after treatment with SFGRT (27). Maximal abscopal effect was observed in unirradiated right tumor when mice were exposed to 15 Gy SFGRT followed by 5 fractions of 2 Gy irradiation to the left tumor suggesting that the abscopal effect can be amplified by sequential combination of SFGRT with conventional fractionation.

The concept of LRT was developed based on SFGRT that is discussed at length in a recently published review (28) as this approach is used for the delivery of HDHRT. Clinically, we defined the design of 3D GRID as LRT that takes advantage of SABR systems more safely and efficiently (29). In this study, LRT was developed for the small animals and this platform can be used for any biological irradiators. Animal LRT can be precisely planned to treat as small as 10% of tumor target volume to 100% (Fig. 1 A–D) and this system provided an opportunity to compare the effects of full tumor irradiation versus irradiation to several partial tumor volumes. Results from this study demonstrated that 20% irradiation (in two 10% volumes) showed significant response in both left and right hind leg tumors and further 50% tumor volume irradiation demonstrated effectiveness in the distal unirradiated right tumor (Fig. 2). This is the first experimental evidence to demonstrate that partial tumor volume radiation can be as effective as whole tumor irradiation. This phenomenon can be exploited in situations where whole tumor irradiation is not possible due to toxicity to critical surrounding normal tissue structures.

The importance of radiation-induced distant effect in tumor control in mice has been demonstrated (30–35). It is also well recognized that several factors including the dose of radiation per fraction, number of fractions and total dose of radiation influence the immune response against tumors (36). In addition, these effects are mediated by cytokines thereby implicating immune modulation in the control of both local and distant/metastatic tumors (37, 38).

Macrophage activation, cell-mediated immune response and promoting tumoricidal activity are functions of Th1 cells secreting cytokines IFN-γ and IL-2. In the murine adenocarcinoma model, IFN-γ has been demonstrated to modulate the antitumor effects of radiotherapy (39). Recently, it was reported that in mice bearing colon adenocarcinoma, all the tumors that responded to 15 Gy irradiation exhibited significant increase in intratumoral IFN-γ compared to nonresponders (40). Consistent with this report, this study observed an induced secretion of LLC1 tumor-specific IFN-γ by splenocytes in treated tumors (Fig. 4). Further, a similar increase in IFN-γ levels was observed in the serum obtained after day 3 of LRT and open-field irradiation (Fig. 3A). These findings are consistent with several other reports where T-cell dependent radiation-induced abscopal effect (41) correlated with the induction of IFN-γ producing T cells (42) and this phenomenon could not be observed in immune-compromised mice (43). Thus, partial irradiation of tumor volume using LRT can equally cause induction of IFN-γ levels when compared to complete tumor volume irradiation and hence, this induction of IFN-γ can play a role in significant growth delay of both irradiated and unirradiated tumors (Fig. 2). Shigematsu et al. reported that the Th1 cytokine IL-2 together with IFN-γ and IL-12 were induced in irradiated dendritic cells (DCs). This set of cytokine expression profile in DCs correlated with increased T-cell proliferation when irradiated DCs were co-cultured with T cells, whereas, proliferation of T cells was absent with unirradiated DCs co-culture. (44). In this study, there was an increased secretion of IL-2 in the serum obtained after days 3 and 7 of irradiation when 50% tumor volume was irradiated (Fig. 3B), however, such increase was seen only 7 days after open-field irradiation. Further, an induced secretion of LLC1 tumor specific IL-2 (Fig. 4) was observed after days 3 and 7 after two 10% and one 50% tumor volume irradiation and such increase was absent in open-field irradiation. These findings demonstrate that such IL-2 inductions in partial tumor volume irradiations can be more effective in promoting the proliferation of T cells than in full-tumor volume irradiations. Overall, it is evident that partial-tumor volume irradiation induces a robust Th1 response when compared to whole-tumor irradiation (Table 1).

TABLE 1.

Secretion or Levels of Several Factors after LRT or Open-Field Irradiation Compared to Untreated Controls in the Serum Obtained at Days 3 or 7 after Irradiation

	LRT						Open field
	Two 10% vertices		One 20% vertex		One 50% vertex		100%
Factors	Day 3	Day 7	Day 3	Day 7	Day 3	Day 7	Day 3	Day 7
IFN-γ	↑	NC	↑	NC	↑	NC	↑	NC
IL-2	NC	NC	NC	NC	↑	↑↑	NC	↑↑
IL-4	↓	↓↓	NC	↓	↓	↓↓	NC	↑
IL-10	↓↓	NC	↓	↑	↓↓	↓↓	↓	NC
KC	↓	NC	↓	↓	↓	↓↓	↓	↓↓
ASMase	ND	↑↑	ND	↑	ND	↑↑	ND	↑
TNF-α	↑↑	↑	↑↑	↑	↑↑	↓	↑↑	↑

Notes. ↑ Indicates upregulation over controls and ↓ indicates downregulation over controls. ND = not done; NC = no change.

Th1 secretions promote growth inhibition by impeding the function of Th2 cells; on the contrary, Th2 cells secretions can inhibit the functions of Th1 cells. IL-4 and IL-10 secreted by Th2 cells can abrogate the effector function of Th1 responses and mediate immunosuppressive effects (45–47). Further, Th-2 mediated immunosuppressive effects have been implicated in tumor growth. In this study, IL-4 levels were significantly downregulated in both days 3 and 7 after partial-tumor volume irradiation (two 10% LRT and one 50% LRT) when compared to whole-tumor volume irradiation (Fig. 3C and D). Similar kinetics was observed for IL-10 for day 3 (Fig. 3C and D) after partial-tumor volume irradiation (two 10% LRT and one 50% LRT). These findings indicate that partial-tumor volume irradiation can elicit a Th1 response that can down-modulate Th2 functions (Table 1).

A multifunctional cytokine IL-6 play a key role in mediating a response during the acute inflammation and IL-6 is robustly induced in response to ionizing radiation and implicated in involving STAT signaling (48). In this study, IL-6 (Fig. 4) secretion by splenocytes was reduced in treated tumors and this may be related to the dose, as most studies have reported increases in IL-6 in low-dose irradiation linked to bystander effects (49).

Increased CD3⁺ T-cell infiltration has been associated with favorable response in post-irradiated tumor specimens (50, 51). In this study, an increase in CD3⁺ T-cell migration was observed in LRT irradiated left hind leg tumor groups and this was equivalent to open-field irradiated group (Fig. 5). Interestingly, there was a surge in CD3⁺ T-cell infiltration in the right leg unirradiated tumor after 50% tumor volume irradiation. On the contrary, right leg unirradiated tumor showed a decline in CD3⁺ staining after open-field irradiation. These findings concur with the reported literature in that the presence of increased CD3⁺ cells in the two 10% volume as well as one 50% volume irradiation correlated well with tumor growth delay. These results indicate that high-dose partial volume can cause an improved distant effect than the total-tumor volume irradiation and this distant effect is mediated by activating the host immune system. Clinically, some anecdotal responses in historically radio-resistant large tumors treated with SFGRT (4, 52) may be a result of immune activation within tumor as well as distant effects. Tumor infiltrating lymphocytes are crucial for tumor regression mediated by immunogenic cell death. It is possible that high-dose open-field irradiation could kill these cells while partial irradiation of tumor will still have these infiltrating cells in the unirradiated areas of the tumor to promote stronger immunogenic cell death than the open field irradiation.

In addition, KC levels were reduced in response to both open-field irradiation as well as partial volume irradiation suggesting that LRT has the ability to down-modulate the functions of KC that has been implicated in angiogenesis (53), cell evasion (54) and proliferation (55) (Fig. 6). Together, these results suggest the role of cellular mediated immunity specifically after partial irradiation of tumor volume in reducing the growth of both irradiated and unirradiated tumors (Summarized in Table 1).

Previously, we demonstrated induction of endothelial cell apoptosis using the serum obtained from SFGRT-treated patients (1). This study confirmed this observation as the growth of endothelial cells was significantly reduced when grown in the serum obtained from the mice treated with LRT (Fig. 7A). These results corroborated with increases in TNF-α, TRAIL and acid sphingomyelinase (ASMase) in mice serum after LRT (Fig. 7B–D). These findings implicate the role of such serum factors in distant effects together with immune activation (Table 1).

In conclusion, this study for the first time in the literature attempts to compare the local and distant effects of whole-tumor irradiation versus 20% and 50% volume irradiation. The findings demonstrate that two 10% or one 20% or one 50% volume irradiation using LRT were equally effective as open-field irradiation in eliciting local and distant tumor growth delay coupled with increased immune activation, intra-tumoral immunogenic death and anti-angiogenic effect. In addition, except for two 10% volume LRT treatment, there was a decrease in prosurvival immune response. These results indicate that LRT of 20–50% tumor volume can be used in the clinics before open-field irradiation of the 100% tumor volume to activate an enhanced immune response. Further, since LRT with two 10% vertices could delay the growth of both irradiated and unirradiated distant tumor, we can speculate that increasing the number of high-dose vertices may enhance the efficacy of LRT. Together, the tumor growth and the immune response data presented here suggest that high-dose LRT if delivered in a way that directly irradiates only about 20–50% of the tumor volume either alone or followed by an adjuvant open-field radiation therapy could be a novel strategy for local and distant/metastatic tumor treatment.