Female 7–8-week-old C57BL/6 mice were obtained from Taconic Farms (TAC, Germantown, NY). Mice were housed at the University of Wisconsin Biomedical Research Model Services vivarium in accordance with the Guide for Care and Use of Laboratory Mice. All experiments were performed under a protocol approved by the institutional Animal Care and Use Committee.

GD2⁺ B78-D14 (B78) melanoma cells were obtained from Ralph Reisfeld (Scripps Research Institute) in 2002. They were derived from B78-H1 cells, which were originally derived from B16 melanoma. B78 tumor cells were grown in RPMI-1640 (Corning) supplemented with 10% heat inactivated FBS (Gibco), 2mmol/L L-Glutamine (Corning), 100 U/mL penicillin (Corning), and 100ug/mL streptomycin (Corning), with periodic supplementation with 400 μg/mL G418 and 50 μg/mL Hygromycin B. Mycoplasma testing by PCR was done routinely.

B78 cells (2x10⁶) were injected intradermally (i.d.) as we previously reported (14) into the shaved right flank of syngeneic C57BL/6 mice (Taconic Farms). In the two-tumor model, mice were injected with B78 cells into both right and left flanks. Tumors were measured twice per week using electronic calipers. Tumor volumes were calculated using the formula: 0.5 x(small diameter)² x large diameter. Mice were euthanized once their largest tumor diameter reached 20mm. Treatment began once tumors reached ~100-300mm³. Mice were dosed using an X-Rad 320 Animal Irradiator (Precision X-Ray). Mice received one fraction of radiation totaling 12Gy to their right flank tumors. The day of RT was day 0 of treatment for all experiments. Mice were injected intraperitoneally with CY at a dose of 100mg/kg on the day of RT. Mice were injected IT into the right tumor with hu14.18-IL-2 immunocytokine (IC, 50 mcg) (AnYxis Immuno-Oncology GmbH of Vienna, Austria) on days 5-9 following RT. In previous experiments (5), IT injection of rat IgG as a control for hu14.18-IL-2 IC did not cause tumor necrosis or antitumor effects; therefore, in this current study, control mice did not receive IT injections of IgG or vehicle (PBS). In rechallenge experiments, tumor-free or naïve mice were injected i.d. in the middle of the abdomen with 2x10⁶ B78 cells. We used the same dose of B78 tumor cell rechallenge as we used for initial tumor implantation, because we have established in previous experiments that mice cured by a combination of RT and IT-IC reject this dose of B78 tumor rechallenge while naïve mice do not (5).

B78 tumor-bearing mice were randomized into control or treatment groups. On the indicated day after the treatments, mice were euthanized, and tumors were excised, minced, and digested into single cell suspensions using a Miltenyi gentleMACS Octo Dissociator (2.5 mL of complete RPMI media, 2.5 mg of Collagenase IV, and 250 g of DNase I per tumor). After dissociation, the resulting suspensions were passed through a 70m filter. Cells were then washed in PBS and incubated with Tonbo Ghost Dye Violet 510 for 30 min at 4°C, followed by washing in flow buffer (PBS-1% FBS) and a 10-minute incubation with anti-CD16/32 FC Shield (Tonbo, Clone 2.4G2) at 4°C. Cells were surface stained in Brilliant Stain Buffer (BD Biosciences) for 30 minutes at 4°C with CD45 FITC (30-F11), CD25 BB700 (PC61), CD8a APC-Fire 750 (53-6.7), and CD4 BV785 (GK1.5); the antibodies were obtained from BD Biosciences, BioLegend, or Invitrogen. Samples were washed, then fixed and permeabilized for 30 minutes at 4°C with the FoxP3/Transcription Factor Staining Buffer Set from eBioscience following kit instructions. Samples were then stained with FoxP3 PE-Cy7 (FJK-16s) from eBioscience for 30 minutes at 4°C. All samples were then washed in a permeabilization buffer and resuspended in flow buffer for acquisition on an Attune flow cytometer (ThermoFisher). Data were analyzed with FlowJo v10 software (BD). Tregs were defined as live, single, CD45+ CD4+ CD25+ FoxP3+ cells.

Tumor growth results from each mouse were summarized by the time-weighted average (area under the volume-time curve, calculated using trapezoidal method). Time-weighted averages were compared between treatment groups overall by Kruskal-Wallis tests. If significance was found using the Kruskal-Wallis test, then pairwise comparisons were conducted using Mann-Whitney tests. No p-value corrections were applied to the pairwise tests. Survival curves were compared with log rank tests.

We hypothesized that CY will be beneficial when combined with IC or RT+IC because it was reported that CY can improve immunotherapy by reducing Tregs (12, 13). Therefore, first we wanted to confirm that CY reduces Tregs within the tumors in our B78 melanoma model. Supplemental Figure 1 shows that indeed, Tregs were significantly reduced 5 days after a single injection of CY. Next, we tested if CY would augment the antitumor effect when combined with IC. The treatment schema is depicted in Figure 1A. When CY was combined with IT-IC, a significant reduction of tumor growth (Figure 1B) and prolongation of survival compared to untreated mice (Figure 1C) were achieved, whereas treatment with IT-IC alone or CY alone did not show statistically significant tumor reduction. In addition, the combination of CY and IT-IC had significantly greater effect than CY or IT-IC in both suppressing tumor growth and extending the survival. The combination of IC and RT, in contrast to RT or IC alone, significantly reduced growth of the tumors that were ~150mm³ when treatment began (Figure 1D). This combination of RT and IC also prolonged survival compared to untreated mice, although RT alone prolonged survival as well (Figure 1E). These findings confirm the beneficial effect of combining RT and IT-IC, as we reported previously (5).

Next, we asked if combining all three treatment modalities, RT, immunotherapy (IT-IC) and chemotherapy (CY), would be more effective than IT-IC combined with RT or with CY. Figure 2 shows the combined data of four separate experiments. When CY was combined with RT and IT-IC, this triple modality combination induced a statistically significant reduction of tumor growth compared with untreated mice and with mice treated with RT + IC, whereas this effect for the triple modality was not statistically significant from IC + CY (Figure 2A). However, this combination of RT, IC and CY induced statistically significant prolongation of survival of mice compared to RT + IC or IC + CY (Figure 2B). No treatment-related toxicities (ruffled fur, lethargy or diarrhea) were observed. In addition, the combined treatment with RT, IC and CY showed a greater number and percentage of cured mice (10/20) than in RT + IC (2/20) or IC + CY (4/15) treated groups (Figure 2C), although the difference between the triple group and IC+CY was not statistically significant. Overall, the results in Figure 2 show the advantage of combining CY with RT and IT-IC. Flow cytometry analyses of tumor-infiltrating lymphocytes on day 10 (10 days after RT and CY injection and 1 day after the last treatment with IT-IC) showed that the mice receiving the triple combination of RT, CY and IC, as well as mice treated with CY+IC, had a reduced percentage of CD4⁺ T cells (Figure 3A) whereas the relative number of CD8⁺ cells was not changed (Figure 3B). Importantly, the mice receiving the triple combination of RT, CY and IC, as well as mice treated with CY+IC, had significantly reduced percentages of Tregs (Figure 3C) and increased ratios of CD8⁺ T cells/Tregs (Figure 3D) compared to untreated mice, suggesting greater activation of effector T cells. There were no reproducible statistical differences of macrophages and myeloid cells between the groups (data not shown).

Next we tested if the triple combination of CY, RT and IT-IC induced a systemic antitumor immune response. B78 tumor cells were injected simultaneously in mice on both the right and left flanks. When the tumors reached ~ 100mm³, the tumors on the right were treated with RT and IT-IC (Figure 4A), while the CY was given i.p. When these mice were treated with the RT+IC+CY combination, antitumor effects were noted against the left (untreated) tumors (Figure 4B). When we compared the volumes of the untreated tumors of mice across two experiments, we found that 61 days following RT, mice treated with RT+ IC+ CY had significantly more tumor regression compared to mice treated with RT, RT+IC, or RT+CY (Figure 4C). Survival following the triple treatment was significantly improved as compared to RT and RT+IC (Figure 4D). The overall statistical differences between the groups, and between tumor volumes on day 61, are presented in the Supplemental Table 1. There was no statistical differences on day 61 between tumor volumes in untreated (Figure 4C; Mean +/- SD = 267 +/- 378, n=9) and treated (Supplemental Figure 2; Mean +/- SD=364 +/- 330, n=9) tumors following RT+IC+CY therapy, although the untreated tumor group had 2/9 mice with complete tumor regression, whereas the treated tumor group had 4/9 mice with complete tumor regression. Overall, the results with a two-tumor model show that a combination of RT+IC+CY can elicit a systemic antitumor response against a distant tumor.

To test if our treatments resulted in long term immunological memory, cured mice were rechallenged with B78 tumor cells 60 days after the primary tumor rejection. The results show that the mice that were cured by a combination of RT+IC or RT+ IC+ CY, in contrast to naïve mice, were able to reject rechallenge of B78 tumors (Figures 5A, B), indicating that they developed immunological memory to B78 melanoma.