Enhancing Dendritic Cell–based Immunotherapy with IL-2/Monoclonal Antibody Complexes for Control of Established Tumors

Chronic illnesses have increased dramatically over the last century (1), of which malignancy remains a top threat and target for many new vaccine candidates (1). Moving away from the broad-based chemotherapy of the past, current efforts focus on activating NK and CTL for their ability to kill tumor cells directly (2, 3). Initially, the nonspecific immunomodulator IL-2 was used to enhance NK and T cell–mediated immunity to tumors (4, 5), at the expense of severe toxicity to the patient. More recently, well-tolerated dendritic cell (DC) therapy has been evaluated as a way to induce tumor Ag (TA)–specific CD8 T cells (6), but with modest potency, likely due to the relatively low CD8 T cell responses observed (7). Combinations of these two existing therapies are currently being tested to further increase CD8 T cell numbers (8), but without modifications to limit the toxicity or short in vivo t_1/2 of IL-2 that requires long duration of therapy in specialized treatment centers.

In recent years, a more precise understanding of the success and limitations of high-dose (HD) IL-2 therapy, approved for renal cell carcinoma and metastatic melanoma (9, 10), have been highlighted. HD IL-2 therapy offers greater longevity for 16% of the patient population, at the risk of 2% mortality from treatment toxicity (11). The low efficacy of HD IL-2 in patients has been suggested to stem from poor induction of NK cell proliferation (12) and the stimulation of suppressive regulatory T (TReg) cells (13). Several investigators have since demonstrated in murine models that complexing free IL-2 with the IL-2–specific mAb S4B6 greatly decreases signaling to CD4⁺CD25⁺ TReg cells as well as CD25⁺ endothelial cells (14). The S4B6 mAb serves to redirect the bioactivity of IL-2 to CD122^hi cells by competitively binding to its CD25 binding region. This unique quality decreases vascular leak syndrome, a serious side effect commonly associated with HD IL-2 therapy (14). Complexing to the IL-2–specific mAb S4B6 (IL-2c) (15) also increases its in vivo t_1/2 because IL-2c is too large to excrete from the kidneys (15–17); this results in the in vivo proliferation of NK cells and memory phenotype CD8 T cells (15). Additional studies suggest that IL-2c can influence the differentiation of effector CD8 T cells responding to soluble peptide immunization (18, 19). To overcome issues with HD IL-2–associated toxicity and low CD8 T cell responses after DC vaccination, we evaluated a short immunization approach coupling DC immunization to stabilized IL-2c infusion to amplify numbers and increase function of both NK cells and endogenous TA-specific effector CD8 T cells.

C57BL/6 (B6) mice were from the National Cancer Institute (Frederick, MD). BALB/c mice were from The Jackson Laboratory (Bar Harbor, ME). Mice with TCR-transgenic (tg) OT-I cells and SMARTA cells have been described (20, 21). The University of Iowa Animal Care and Use Committee approved animal experiments. Class I peptides used for DC pulses were Ova_257–264 (SIINFEKL), AH1_6–14 (SPSYVYHQF), and tyrosinase-related protein 2 (TRP2)_180–188 (SVYDFFVWL) peptide at a concentration of 2 μmol. Class II peptides used were Ova_323–339 (ISQAVHAAHAEINEAGR), respiratory syncytial virus protein M2_26–39 (NYFEWPPHALLVRQ), and lymphocytic choriomeningitis virus protein gp_61–80 (GLKGPDIYKGVYQFKSVEFD) at the same concentration. LPS-matured peptide-coated DCs were prepared as described (22) and injected i.v. (5 × 10⁵).

Approximately 3 × 10⁴ naive Thy1.1 OT-I CD8 T cells or 2 × 10⁴ naive Thy1.1 SMARTA CD4 T cells were transferred into naive Thy1.2⁺ B6 mice i.v (23) at day −1. A total of 5 × 10⁵ LPS-matured/peptide-coated DCs was injected i.v. at day 0, followed by 1.5 μg rat Ig or murine IL-2/anti–IL-2c at days 4–6. Complexes were made by incubating murine IL-2 (PeproTech) with S4B6 anti–IL-2 mAb at a 2:1 molar ratio (7.5 μg/ml IL-2/250 μg/ml S4B6) for 15 min at 37°C.

Splenocytes from naive B6 hosts were labeled with 0.08 μmol CFSE (no peptide) or 1 μmol CFSE (1 μmol OVA_257–264). Targets (5 × 10⁴ each) were mixed 1:1 ratio of CFSE^lo/CFSE^hi with titrated numbers of effector OT-I CD8 T cell numbers from the spleens of rat Ig or IL-2c–treated mice. Killing was measured at E:T ratios of 0.3 and 1 over 4 h at 37°C. Percent specific lysis was calculated as 100 − (100 × [(% CFSE^lo in control well/% CFSE^hi in control well)/(% CFSE^lo in experimental well/% CFSE^hi in experimental well]).

B16, B16-fLUC, and 4T1 tumor cells expressing firefly luciferase (4T1-fLUC) were provided by Lyse Norian (University of Iowa). B6 mice were injected s.c. in the left hind flank with 2 × 10⁴ B16-fLUC cells in a 1:1 volume of PBS and Matrigel (BD Biosciences; catalog number 356234) or 5 × 10⁴ B16 cells i.v. A total of 1 × 10⁵ 4T1-fLUC cells was orthotopically inoculated in mammary pad #9 of female BALB/c mice. Bioluminescence imaging was performed on anesthetized mice on an IVIS Spectrum (Caliper) system ∼10 min after i.p. injection of 1 mg luciferin (Gold Biotech, St. Louis, MO). Imaging on IVIS was performed according to the manufacturer’s protocols, and bioluminescence was quantified in radiance.

Human CD8⁺ T cells were purified with the RosetteSep CD8⁺ T cell enrichment mixture (StemCell Technologies). A total of 2 × 10⁵ CD8⁺CD3⁺ T cells was plated in 96-well plates coated with 0.1 μg/ml OKT3/anti-CD28 at day 0. On day 4, additional RP10 1640 media or human IL-2 (huIL-2; 10 ng/ml; PeproTech)/MAB602 (R&D Systems) complexes were added to each well. huIL-2/MAB602 complexes were prepared by incubating huIL-2 (10 ng/ml) with MAB602 (100 ng/ml) for 15 min at 37°C. Ki67 positivity, total numbers of CD3⁺CD8⁺ T cells, and granzyme B expression were determined by flow cytometry on day 7.

Abs were used with the indicated specificity and the appropriate combinations of fluorochromes. Intracellular stains were performed following manufacturer’s protocols. Foxp3/transcription factor staining buffer kit (eBioscience, catalog number 00-5523-00) was purchased for transcription factor labeling. The following Abs were purchased from BioLegend: Thy1.1 (OX-7), CD8α (53-6.7), 41BB (17B5), programmed cell death-1 (PD-1; 29F.1A12), and LAG3 (C9B7W). The following Abs were purchased from eBioscience: GITR (DTA-1) and CTLA-4 (UC10-4B9). Anti–IFN-γ was purchased from BD Biosciences. Anti-human granzyme B and murine IgG1 Abs were purchased from Invitrogen. PK136 was purchased from eBioscience; 2.43 Ab was purchased from Santa Cruz Biotechnology. Anti–IL-2 mAb were purified from S4B6 hybridomas and purchased from American Type Culture Collection.

Intravenous and s.c. B16 tumor studies were performed using B16 and B16-fLUC cells at indicated concentrations. Lungs were perfused with PBS and harvested from mice at indicated day postinoculation. Visible tumor nodules were quantified from IL-2c and rat Ig control groups. Tumor area for B16-fLUC and 4T1-fLUC (1 × 10⁵ cells/mouse) studies were determined by two digital caliper measurements taken on anesthetized mice.

Unless indicated otherwise, significance of bar graphs were calculated by Student t test using GraphPad Prism 5 for Macintosh (GraphPad). The p values <0.05 were considered significant. Survival curves were analyzed by the Mantel–Cox test.

Prior studies reveal that treatment of mice with IL-2c can enhance tumor control over soluble IL-2, potentially through activation of NK and NKT cells expressing the low-affinity IL-2R (CD25, CD122, and CD132) (15, 24). To determine the relative potency of IL-2 versus IL-2c treatment in tumor control, we used a lung metastasis model in which B6 mice were injected i.v. with B16 melanoma, and daily IL-2c or soluble IL-2 therapy was initiated 3 d later. IL-2c treatment enhanced survival of tumor-bearing mice (Fig. 1A) and reduced tumor size (Fig. 1B), although no significant decreases in lung tumor numbers were observed (Fig. 1C). As previously described (15, 24), treatment with IL-2c enhanced NK and NKT cell numbers (Fig. 1D) as well as activation state (CD16 and granzyme B expression (Fig. 1E). Thus, IL-2c treatment is superior to soluble IL-2 and can enhance NK cell activation and limit tumor burden to prolong survival of mice with B16 lung metastasis.

The TRP2 is a defined endogenous TA for melanoma in both humans and mice. The TRP2 epitope is presented by either H-2K^b (mouse) or HLA-A201 (human) (25, 26). In addition, TRP2-specific CD8 T cells were capable of limiting B16 melanoma in various vaccination regimens or immunotherapy (27). Of note, analyses of TRP2-specific CD8 T cells in the blood of tumor-bearing mice at day 3 after completion of IL-2 therapy revealed ∼10-fold higher numbers in mice receiving IL-2c treatment compared with soluble IL-2 treatment (Fig. 1F, 1G). To date, no preclinical studies of IL-2c have offered comprehensive data on its potential to target and expand nonmutated self-peptide TA, a trait essential to an effective tumor therapy (19, 24, 28). Although the TRP2-specific CD8 T cell response was modest in both groups, these data suggest the potential of IL-2c to amplify endogenous TA-specific CD8 T cell responses that, in addition to NK cells, could contribute to IL-2c–enhanced control of B16 metastases.

To initially determine if IL-2c treatment could enhance the CD8 T cell response to a relevant Ag-specific vaccination, we combined a DC immunization with Ova_257–264 peptide and IL-2c treatment. IL-2c has been shown to stimulate proliferation in memory phenotype CD8 T cells and NK cells expressing the low-affinity IL-2R (29). Because expression of the high-affinity IL-2R (CD25, CD122, and CD132) on newly activated CD8 T cells is largely extinguished by 4 d post–DC immunization (30), rat Ig control or IL-2c were administered on day 4–6 post–DC priming. Strikingly, IL-2c treatment markedly enhanced the number of DC-primed endogenous Ova_257–264-specific CD8 T cells by >50-fold at day 7 after DC immunization (Fig. 2A). Previously, Pipkin et al. (18) demonstrated that IL-2 signaling induced a transcriptional program promoting effector differentiation and cytolytic potential of activated CD8 T cells. Thus, we monitored expression of costimulatory molecules that have previously been shown to enhance differentiation of effector T cells (31–33). IL-2c treatment upregulated 41BB and GITR expression on Ova_257–264-specific CD8 T cells, but failed to upregulate OX40 (Fig. 2B). These results demonstrate the capacity of IL-2c to massively enhance accumulation of DC-primed effector CD8 T cells and selectively upregulate costimulatory molecules.

Similar to NK cells, Ova_257–264-specific CD8 T cells responding to IL-2c also expressed more granzyme B (Fig. 2B), but did not further upregulate expression of the inhibitory receptors CTLA-4, PD-1, or LAG3 compared with DC primed CD8 T cells in the absence of IL-2c treatment (Fig. 2C). To address additional functions important in tumor control, we used an adoptive transfer model of OT-I TCR-tg cells. This TCR-tg approach with DC priming allowed us to measure CD8 T cell number and function, independent of changes in TCR repertoire in the translational context of DC immunization. Similar to the endogenous CD8 T cell response, IL-2c treatment amplified DC-Ova_257–264-primed OT-I CD8 T cell numbers (Supplemental Fig. 1A), 41BB, GITR, and granzyme B expression (Supplemental Fig. 1B), and did not alter CTLA-4, PD-1, or LAG3 expression (Supplemental Fig. 1C). As many human tumor-associated Ags are derived from nonmutated self-proteins expressed at low levels, Ag sensitivity (also termed functional avidity) is an important consideration when developing T cell–mediated tumor immunotherapies (34). Indeed, high T cell avidity was shown to correlate with T cell recognition of tumors and their therapeutic efficacy (35). IL-2c treatment enhanced Ag sensitivity of monoclonal OT-I cells for IFN-γ production by ∼2-fold (Supplemental Fig. 1D), indicated by a decreased effective concentration 50 (EC₅₀) (Supplemental Fig. 1E). Consistent with the sustained expression of granzyme B and enhanced sensitivity to Ag, OT-I cells generated by DC prime and IL-2c treatment exhibited enhanced in vitro cytolytic activity (Supplemental Fig. 1F, 1G). These data demonstrate an increase in both Ag sensitivity for cytokine production and per-cell killing capacity by CD8 T cells following DC + IL-2c treatment without increasing inhibitory receptor expression.

HD IL-2 therapy has been shown to induce TReg cells that contribute to poor disease outcomes (13, 36). IL-2c treatment did result in a modest increase in the frequency of Foxp3 expressing TReg cells (Fig. 2D, top panel); however, it has been suggested that in the context of immunotherapy, the ratio of effector CD8 T cells/TReg may be the critical parameter determining the success of treatment (37). Our results show that amplification of the Ova_257–264-specific CD8 T cell response by IL-2c was much greater than the increase in TReg cells, and the ratio of effector Ova_257–264-specific CD8 T cells/TReg therefore increased substantially during the treatment (Fig. 2D, bottom panel). To examine whether a complex mixture of class I and II peptide epitopes on DCs might induce a greater TReg response, we peptide-pulsed DCs with Ova_257–264 and class II peptide epitopes Ova_323–339, gp_61–80, and M2_26–39. We subsequently treated with IL-2c days 4–6 post–DC immunization and quantified CD4⁺Foxp3⁺ T cells, which yielded a similar TReg induction and enhanced CTL/TReg ratio as seen in IL-2c–treated mice that received DC pulsed with Ova_257–264 alone (Fig. 2E). Furthermore, we tested whether DC + IL-2c would directly polarize naive CD4 T cells to an effector Th1 or suppressive TReg phenotype. We adoptively transferred 2 × 10⁴ tg SMARTA CD4 T cells into naive B6 recipients that were subsequently treated with DC-gp_61–80 + IL-2c from days 4–6. Activating CD4 T cells directly using DC + IL-2c enhanced the quantity of CD4 T cells by 10-fold and polarized them to a Th1 phenotype, without inducing a discernable TReg population (Fig. 2F). These data demonstrate that IL-2c treatment in the context of DC immunization results in preferential expansion of effector CD8 T cells over TReg cells and can also be used to enhance Th1 CD4 T cells against a defined epitope.

Direct translation of this approach to humans would require generation of IL-2c with human IL-2 and mAb. To address this possibility, we generated IL-2c with human IL-2 and the MAB602 anti-human IL-2–specific Ab, complexes of which have been shown to stimulate proliferation of memory phenotype CD8 T cells in mice (16). IL-2c MAB602 treatment enhanced Ki67 expression, as a measure of proliferation (Supplemental Fig. 2A, B), cell recovery (Supplemental Fig. 2C), and granzyme B expression (Supplemental Fig. 2D) of anti-CD3/CD28–stimulated human CD8 T cells. Thus, IL-2c treatment is a potentially viable means to amplify human CD8 T cells responding to Ag stimulation.

The prior results (Fig. 2) were obtained with a foreign Ag in non–tumor-bearing mice. To address the potency of IL-2c to amplify a DC-primed TA-specific CD8 T cell response in tumor bearing mice, B16 cells were injected i.v., and DC-TRP2 was initiated 3 d later followed by IL-2c treatment of some mice at days 4–6 post–DC immunization (Fig. 3A). IL-2c dramatically enhanced the frequency (Fig. 3B) and total numbers (Fig. 3C) of TRP2-specific CD8 T cells over DC-TRP2 priming alone or in mice with no treatment. Of note, all mice that received DC + IL-2c survived the 32-d study period, whereas 100% of mice receiving no therapy and 50% of mice receiving DC-TRP2 without IL-2c succumbed to tumor (Fig. 3D). Mice that received DC-TRP2 + IL-2c exhibited reduced tumor size (Fig. 3E) and substantially reduced tumor numbers at day 30 compared with surviving mice that received DC-TRP2 without IL-2c (Fig. 3F). The reduction of tumor numbers was a specific improvement of the DC + IL-2c treatment over IL-2c treatment alone (Fig. 1). These data suggest that IL-2c treatment, in conjunction with DC-TRP2 priming, can enhance control of pre-existing lung metastasis.

Importantly, late intervention with DC + IL-2c was capable of inducing an equally potent TA-specific CD8 T cell response. To ensure a high tumor burden, 2 × 10⁵ B16 cells were injected i.v. in naive B6 mice, followed by DC-TRP2 + IL-2c 21 d later (Supplemental Fig. 3A). TRP2-specific CD8 T cells were measured in the blood at day 7 post–DC immunization with a 30-fold increase in frequency (Supplemental Fig. 3B) and 40-fold increase in total numbers (Supplemental Fig. 3C). DC + IL-2c treatment initiated as late as 21 d after tumor injection was able to substantially reduce the number of tumor nodules in the lung (Supplemental Fig. 3D, 3E). These data suggest that DC + IL-2c treatment has the potential to effectively amplify immune responses and control tumor load in patients with advanced cancer.

To address the relative contribution of CD8 T cells and NK cells in tumor control, mice were injected i.v. with B16 melanoma, followed by CD8-α or NK cell–specific depleting Abs (Fig. 4A). One week after B16 injection, some mice were immunized with DC-TRP2, and some mice in each group received IL-2c treatment at day 4–6 post–DC immunization. The presence or absence of NK cells had no impact on the number of TRP2-specific CD8 T cells, whereas CD8-depleted mice contained no detectable TRP2-specific CD8 T cells (Fig. 4B, 4C and data not shown). Again, all mice that received DC-TRP2 + IL-2c treatment survived the experimental period, whereas 60% of untreated mice succumbed (Fig. 4D). Some mice in DC-TRP2 + IL-2c groups depleted of either NK cells or CD8 T cells also succumbed during the experimental period (Fig. 4D), suggesting that both effector populations contributed to enhanced survival. Depletion of CD8 T cells or NK cells in DC-TRP2 + IL-2c–treated mice resulted in higher numbers of tumors in the surviving mice compared with mice containing both cell populations (Fig. 4E). Thus, DC-TRP2 + IL-2c treatment enhances survival and control of pre-existing lung metastases through a combination of amplified tumor-specific CD8 T cells and activated NK cells.

Melanoma initiates as a localized primary tumor in the skin. To address the potential that DC-TRP2 + IL-2c treatment could control a solid tumor, B16 melanoma cells expressing firefly luciferase were injected s.c. into the flank of B6 mice and tumor size was evaluated by bioluminescence imaging. Mice were randomized into groups with similar sized tumors (Fig. 5A, 5B) at day 7 after tumor inoculation and subsequently immunized with DC-TRP2 ± IL-2c treatment. IL-2c treatment amplified DC-primed TRP2-specific CD8 T cells in solid tumor-bearing mice (Fig. 5C, 5D) and substantially slowed the rate of tumor growth (Fig. 5E) compared with mice given DC-TRP2 alone. Thus, DC-TRP2 immunization can be enhanced by IL-2c treatment for control of a solid tumor.

The results so far evaluated the impact of IL-2c treatment on DC-primed CD8 T cells and tumor control in B6 mice. The 4T1 breast cancer cell line establishes a solid tumor after orthotopic transplant in BALB/c mice and spontaneously metastasizes in a manner similar to human breast cancer, leading to host mortality (38, 39). The AH1 peptide, derived from the endogenous retroviral gp100 Ag, serves as a tumor-specific Ag for the 4T1 tumor (40). To extend our analyses to an additional mouse strain, epitope, and tumor model, we inoculated female BALB/c mice with 4T1-fLUC and monitored tumor growth by bioluminescence. Mice were randomized into three groups with equivalent tumor burdens (Fig. 6A, 6B) at day 7 post–tumor inoculation. Some mice were then immunized with DC-AH1 ± IL-2c. IL-2c treatment enhanced the AH1-specific CD8 T cell response (Fig. 6C, 6D) as observed previously for response to Ova_257–264 or TRP2 peptide. In contrast to the prior study with B16 melanoma, DC or DC + IL-2c treatment had little impact on the growth rate of the primary 4T1-fLUC tumor (Fig. 6E). However, DC + IL-2c treatment prevented the development of metastases visible by whole mouse bioluminescence imaging and thus reduced the overall live tumor burden (Fig. 6F, 6G). Furthermore, none of the mice in the DC + IL-2c group succumbed during the study period, whereas 60% of control mice and 50% of DC-AH1 alone mice succumbed (Fig. 6H). Metastases were readily apparent in untreated mice or mice given DC-AH1 without IL-2c. Importantly, 100% of mice in the untreated control group and 75% in the DC-AH1 group that succumbed to the 4T1 tumor exhibited detectable metastases (Fig. 6H). Strikingly, there was no visible metastasis in the DC-AH1 group that received IL-2c, consistent with the lack of mortality from this group (Fig. 6H). Of note, IL-2c alone did not prevent metastases in the experimental time frame (Supplemental Fig. 4A, 4B). Thus, DC-AH1 + IL-2c treatment had specific efficacy against metastases in the 4T1 model even with minimal control of the primary tumor site.

The immune system is naturally equipped with the means to eliminate tumor cells; however, unmanipulated endogenous responses appear insufficient for full tumor control. To this end, multiple approaches have been tested as immunotherapies against cancer. To be effective, immunotherapies would ideally activate highly functional cytotoxic NK and TA-specific CD8 T cells with the capability to eradicate tumor cells without concurrent activation of regulatory cells that could dampen immunity. Although currently approved HD IL-2 infusion can modestly increase CD8 T cell numbers, induction of TReg cells and high toxicity following therapy remain important limitations of this treatment (13, 41). DC-based immunotherapy can additionally stimulate TA-specific CD8 T cells, but the relatively low-magnitude T cell responses after DC immunization may underlie the modest impact on clinical outcomes in malignancy (7). Thus, single therapeutic modalities appear to lack potency, and the field has turned to combination immunotherapies with the potential to enhance clinical outcome (32, 42). However, combination DC with soluble IL-2 treatment does not address the toxicity or long duration of IL-2 therapy (8). In this study, we evaluate a strategy combining IL-2 with a specific mAb to generate relatively stable, nontoxic IL-2c that can be used to augment the ability of DC immunization to enhance the magnitude and function of TA-specific effector CD8 T cells, while also amplifying NK cell function and numbers.

CD25 expression is maintained for only 3 d following DC immunization alone (30) and then responding CD8 T cells express only the low-affinity IL-2R (CD122 and CD132) (43). Based on these data, we combined initial DC immunization with IL-2c delivered from days 4–6 after DC immunization to amplify TA-specific CD8 T cells, while minimizing suppressive mechanisms and toxicity. This rationale was based on literature indicating that a specific IL-2c (IL-2 complexed with the S4B6 mAb) signals primarily through the low-affinity IL-2R and not the high-affinity IL-2R expressed on some TReg cells and vascular endothelium (14, 16). Our results demonstrate the superior efficacy of DC-TA + IL-2c immunization strategies for tumor control over IL-2c or DC immunization alone. The combination of DC-TA + IL-2c infusion not only activates and expands NK cell, but also enhances the number, Ag sensitivity, cytotoxicity, and CTL/TReg ratio of effector TA-specific CD8 T cells without increasing inhibitory receptor expression that could blunt activation (44). These quantitative and qualitative improvements in innate and adaptive cytotoxic cell types translate to better control of tumor burden in multiple models of pre-existing malignancy. Notably, we show that DC + IL-2c enhances number and function of endogenous TA-specific CD8 T cells for significant reduction of tumor burden. Moreover, by enhancing nonmutated self-peptide–specific CD8 T cell responses, DC+IL-2c treatment can be effectively used against tumors that do not undergo high rates of mutation and express neoantigens (45, 46); thus, providing a viable treatment option for patients with cancer carrying tumors that traditionally exhibit the highest resistance to immunotherapies (47).

In many cases, primary tumors can be limited through surgical intervention (48). However, metastatic disease is more difficult to detect and prevent (49). Perhaps the most striking findings from our data are the enhanced capacity of the DC-TA + IL-2c combination therapy to deal with pre-existing metastases (B16 i.v. model) or prevent detectable metastases (4T1 model). Whether this results solely from enhanced numbers or specific functional attributes of DC-TA + IL-2c–generated NK cells or TA-specific CD8 T cells remains a question of great interest. For example, TA-specific endogenous CD8 T cells likely emerge from a preimmune repertoire that has been limited in number and affinity to avoid autoimmunity (50, 51). The decreased affinity for TA may be partially overcome by the enhanced Ag sensitivity and increased cytolytic capacity observed in T cells stimulated by combination DC-TA + IL-2c treatment. Additionally, the enhanced expression of late costimulatory molecules (41BB and GITR) by IL-2c treatment has the potential to render responding TA-specific CD8 T cells and NK cells resistant to suppressive mechanisms within tumors (32). Furthermore, DC + IL-2c treatment dramatically enhanced TA-specific CD8 T cell/TReg ratios, a relationship that has been suggested to be critical for successful immunotherapy (37). Finally, it will be of great interest to determine if additional combination with checkpoint blockade will further increase the capacity of DC + IL-2c immunotherapy to combat malignancy. The model systems described in this study should be ideally suited to address these questions.

Our results establish the proof of principle for combining DC-TA immunization with IL-2c infusion for tumor control. Importantly, complexes of human IL-2 and anti-human mAb can also enhance the proliferation, accumulation, and function of human CD8 T cells. As DC therapy is currently U.S. Food and Drug Administration–approved (7), and IL-2c has been shown in multiple preclinical studies to address serious pitfalls with currently approved HD IL-2 therapy (9), we consider this DC + IL-2c immunization strategy to be a viable candidate for Food and Drug Administration approval as an immunotherapy for patients with cancer.

We thank members of the Harty Laboratory for helpful discussion.

The authors have no financial conflicts of interest.