Prussian blue nanoparticle-based photothermal therapy combined with checkpoint inhibition for photothermal immunotherapy of neuroblastoma

Abstract

We describe “photothermal immunotherapy,” which combines Prussian blue nanoparticle (PBNP)-based photothermal therapy (PTT) with anti-CTLA-4 checkpoint inhibition for treating neuroblastoma, a common, hard-to-treat pediatric cancer. PBNPs exhibit pH-dependent stability, which makes them suitable for intratumorally-administered PTT. PBNP-based PTT is able to lower tumor burden and prime an immune response, specifically an increased infiltration of lymphocytes and T cells to the tumor area, which is complemented by the antitumor effects of anti-CTLA-4 immunotherapy, providing a more durable treatment against neuroblastoma in an animal model. We observe 55.5% survival in photothermal immunotherapy-treated mice at 100 days compared to 12.5%, 0%, 0%, and 0% survival in mice receiving: anti-CTLA-4 alone, PBNPs alone, PTT alone, and no treatment, respectively. Additionally, long-term surviving, photothermal immunotherapy-treated mice exhibit protection against neuroblastoma rechallenge, suggesting the development of immunity against these tumors. Our findings suggest the potential of photothermal immunotherapy in improving treatments for neuroblastoma.

1. Introduction

Advances in the field of nanomedicine have resulted in multiple classes of nanoparticles for treating cancer that have either received FDA approval or are currently undergoing clinical evaluation.^1,2 An emerging area of interest in cancer nanomedicine is the use of nanoparticles in combination with immunotherapies (such as antibodies, immunoadjuvants, or immune cells), which specifically target and/or activate the immune system.^3–7 These combination “nanoimmunotherapies” offer the potential for improved and durable responses over conventional cancer therapies (e.g. surgery, chemotherapy, radiation therapy) by conferring immunity to treated subjects, which provides long-term protection against cancer recurrence. Motivated by these advantages of nanoimmunotherapies, we describe “photothermal immunotherapy”, a nanoimmunotherapy that combines Prussian blue nanoparticle (PBNP)-based photothermal therapy (PTT) with anti-CTLA-4 checkpoint inhibition for treating neuroblastoma. Neuroblastoma is the most common extracranial solid tumor of childhood, accounting for 15% of cancer-related deaths in children.^8,9 Approximately 50% of neuroblastoma patients are initially diagnosed with aggressive disease.¹⁰ Various multimodal treatment options have resulted in incremental, but limited progress in treating patients within this “high-risk” group. Therefore, there is an urgent need for more effective therapies for this resistant tumor.

Nanoparticle-based PTT functions as a rapid and minimally invasive method for reducing tumor burden using near infrared (NIR) light-absorbing nanoparticles and a low power NIR laser.^11,12 Several reports have demonstrated the efficacy of PTT using diverse nanoparticles including gold nanoshells,^13,14 gold nanorods,^11,15 gold nanocages,^16,17 and carbon nanotubes^18,19 in animal cancer models such as breast cancer,²⁰ squamous cell carcinoma,²¹ and prostate cancer.²² Here, we utilize PBNPs, whose photothermal properties have only recently been described,^23,24 for PTT. As compared to common alternative nanoparticles used for PTT, PBNPs offer several advantages: they are easily synthesized in a single and scalable step using a one-pot synthesis scheme, do not require costly synthesis materials,^23,25–28 and are already FDA-approved for human oral use to treat radioactive poisoning.^29,30

Additionally, we describe in this study, that PBNPs can be synthesized with pH-dependent properties, which makes them suitable for intratumoral (i.t.) administration of PTT because tumor environments exhibit mildly acidic pHs whereas the blood and lymph exhibit mildly alkaline pHs.^31,32 Given that surgery is one of the mainstays of neuroblastoma treatment, we anticipate that PBNP-based PTT can be administered intraoperatively in conjunction with surgery or in a minimal residual disease setting. An additional advantage of the pH-dependent stability of PBNPs is that it mitigates concerns associated with the long-term fate and potential toxicity of the nanoparticles within the body, an advantage over non-degradable nanoparticles used for PTT.

An important feature of nanoparticle-based PTT in cancer therapy is its immunostimulatory properties,^3–5 which have been leveraged along with the tumor-debulking effects of PTT to induce long-term tumor remission in several animal tumor models.^20,33 However, these innate effects of PTT have also been observed to be insufficient in inducing long-term remission in more aggressive cancer models including neuroblastoma.^21,22,24 These findings create the opportunity for testing PTT in combination with complementary immunotherapies to obtain improved and durable responses in treating aggressive cancers. To this end, we investigate anti-CTLA-4 checkpoint inhibition immunotherapy in combination with PBNP-based PTT. Checkpoint inhibition uses monoclonal antibodies to target key immune checkpoints such as CTLA-4 and PD-1³⁴ in order to reverse immune suppression, unleashing potent antitumor responses by activating endogenous immune cells (e.g. T cells).^35,36 Checkpoint inhibitors including anti-CTLA-4 (e.g. ipilimumab) and anti-PD-1 (e.g. nivolumab) have received FDA approval for the treatment of advanced cancers such as metastatic melanoma.^37,38 However, clinical responses have been restricted to modest subsets of patients.³⁹ We hypothesize that combining PBNP-based PTT with anti-CTLA-4 checkpoint inhibition will improve outcomes over those obtained with either modality alone for treating neuroblastoma.

To test this hypothesis, we utilize a hard-to-treat, syngeneic, Neuro2a mouse model of neuroblastoma,^36,40,41 where we locally (i.t.) administer the pH-sensitive PBNPs for PTT (i.t.), and systemically (intraperitoneally; i.p.), administer the anti-CTLA-4 checkpoint inhibitor (Figure 1). First, we describe the pH-sensitive properties of PBNPs and their suitability for i.t. PTT. Next, we study the tumor-debulking and immunostimulatory effects of PBNP-based PTT. Third, we test the efficacy of the photothermal therapy in treating neuroblastoma in the syngeneic animal model and describe the immune cells that play a key role in eliciting the observed responses. Finally, we test the ability of animals previously treated with photothermal immunotherapy to reject tumor rechallenge. The findings of this study will be the basis for further preclinical testing and eventual clinical translation of our novel therapy for treating patients with neuroblastoma.

Fig. 1.

Prussian blue nanoparticle-based photothermal therapy combined with checkpoint inhibition for photothermal immunotherapy of neuroblastoma. A) Prussian blue nanoparticles (PBNPs) that exhibit pH-dependent stability are i.t. (locally) injected into tumors. B) A low power, near infrared (NIR) laser irradiates PBNPs within the tumor effecting photothermal therapy (PTT) of the tumor, which results in tumor cell death. C) PTT also elicits an immune response resulting in the increased infiltration of lymphocytes and T cells to the tumor area. D) I.p. (systemically) administered anti-CTLA-4 reverses immunosuppression, unleashing the antitumor immune responses of endogenous immune cells, particularly T cells. The above processes combine to yield improved tumor responses and development of immunity against tumor rechallenge in a mouse model of neuroblastoma.

2. Methods

2.1. Materials, Antibodies, and Cells

Please refer to the supporting information document for details.

2.2. Animals

All animal studies were approved by the Institutional Animal Care and Use Committee of Children’s National Health System, Washington, DC (Protocol # 00,030,439). The studies were conducted to ensure humane care of the animals as per the IACUC’s guidelines. 4–6 week old female A/J mice were purchased from Jackson Laboratory (Bar Harbor, ME). The animals were acclimated for 3–4 days prior to tumor inoculation.

2.3. PBNP Synthesis

PBNPs were synthesized using a previously described scheme.²⁴ Briefly, an aqueous solution of 6.8 mg FeCl₃·6H₂O (2.5 × 10⁻⁵ mol) in 5 mL of Milli-Q water was added under vigorous stirring to an aqueous solution containing 10.6 mg of K₄Fe(CN)₆·3H₂O (2.5 × 10⁻⁵ mol) in 5 mL of Milli-Q water. After stirring for 15 min, the precipitate was isolated by centrifugation (20,000 ×g for 5 min) and rinsed by sonication (5 s, high power) in Milli-Q water. The isolation and rinsing steps were repeated 3 × before the particles were resuspended by sonication in Milli-Q water.

2.4. pH-Dependent Properties of PBNP

PBNPs at pH 5.5, 7.0 or 7.4 were prepared by adding the appropriate amounts of mild acid or base to the PBNP suspensions (0.8 mg/mL) in Milli-Q water until the desired, stable pH was obtained. The Vis-NIR absorbance spectra of the PBNPs at the different pHs over 7 days were measured on the Genesys 10S spectrophotometer using the VISIONlite software (ThermoFisher Scientific). Dynamic light scattering (DLS) analyses of the size distributions of the PBNPs at the different pHs over 7 days were measured using a Zetasizer Nano ZS (Malvern Instruments, Worcestershire, UK) as per the manufacturer’s specifications. Transmission electron microscopy (TEM) images of the PBNPs at the different pHs on day 7 were prepared by loading 5 μL of the PBNP suspensions on to a copper grid (Electron Microscopy Sciences, Hatfield, PA), and imaging using a JEM 2100 FEG TEM/STEM.

2.5. Mouse Neuroblastoma Model

For establishing the syngeneic neuroblastoma mouse model, 10⁶ Neuro2a cells transfected with luciferase were suspended in PBS and subcutaneously injected into the shaved backs of 4–6 week old female A/J mice as previously described.^24,36,40,41 Tumor growth in these mice was monitored following tumor inoculation by imaging tumor bioluminescence using an IVIS Lumina III imaging system (PerkinElmer, Waltham, MA), which allows for quantitative analysis of tumor volume over time. Tumor volumes were calculated using this imaging system as previously described.⁴²

2.6. In Vivo PTT

For the in vivo PTT studies, neuroblastoma (Neuro2a) tumor-bearing mice were treated when their tumors reached a diameter of at least 5 mm (~60 mm³ volume) measured using bioluminescence measurements. Mice were anesthetized prior to and during treatment using 2–5% isoflurane. The mice were i.t. injected with 50 μL of PBNPs resuspended in Milli-Q water (1 mg/mL), and the tumor area was irradiated with an 808 nm NIR laser from Laserglow Technologies (Toronto, ON, Canada) at 1.875 W/cm² for 10 min. The temperatures reached during PTT were measured using a FLIR thermal camera (Arlington, VA).

2.7. Anti-CLTA-4 Injections

Anti-CTLA-4 antibody (150 μg per mouse) was administered i.p. on days 1, 4, and 7 following tumor establishment for the combination (PTT + anti-CTLA-4) group, and on days 0, 3, and 6 following tumor establishment for the anti-CTLA-4 only group.

2.8. Tumor Infiltration Studies

Whole tumors were extracted from PTT-treated and untreated tumor-bearing mice, and were minced and processed through a 70 μm filter to obtain single-cell suspensions, which were analyzed for leukocyte (CD45+) and T cell (CD3+) infiltrates by flow cytometry. Cells were stained with FITC-conjugated CD45 and CD3 antibodies, and the samples were run and analyzed on the BD Accuri cytometer.

2.9. In Vivo T Cell Depletion Studies

CD8 and CD4 cells were depleted by i.p. administration of purified anti-CD8α (100 μg/mouse) and anti-CD4 (100 μg/mouse) depletion antibodies starting a day prior to Neuro2a inoculation and by repeating injections on days 3, 7, and 11 after inoculation. Depletion of CD8+ and CD4+ T-cells were validated using peripheral blood and analyzed by flow cytometry (>95% depletion).

2.10. Sample Sizes and Statistical Analysis

Please refer to the supporting information document for a detailed description of the statistical analyses used in this study.

3. Results

3.1. PBNPs Exhibit pH-Dependent Stability

To determine whether the PBNPs are suitable for i.t. administration of PTT, we measured their properties (Vis-NIR spectra, PTT capabilities, size distributions, and TEM images) at 3 different pHs – mildly acidic (pH 5.5), neutral (pH 7.0), and mildly alkaline (pH 7.4). The mildly acidic pH was selected to mimic conditions typically encountered by i.t. administered nanoparticles, i.e. pH of the tumor interstitium (~5.5).^31,32 Neutral pH was selected to represent the conditions under which the PBNPs are stored (in Milli-Q water), while the mildly alkaline pH was selected to mimic conditions typically encountered by nanoparticles in the blood or lymph (pH ~ 7.4).

The Vis-NIR spectrum of PBNPs at pH 7.0 demonstrated its characteristic absorption band from 650 to 900 nm (λ_max = 705 nm; Figure S1).^23,24,27 We utilized this characteristic absorption band to quantify the degradation of the PBNPs, as degraded PBNPs would be expected to exhibit an attenuated absorption band relative to intact PBNPs. PBNPs incubated at pH 5.5 (pH of the tumor interstitium) exhibited negligible change in their Vis-NIR spectra over seven days (Figure 2A), indicating that the PBNPs were insignificantly degraded at pH 5.5. Similarly, insignificant degradation properties were observed with PBNPs incubated at a neutral pH of 7.0 (Figure 2B). However as the pH of the solution was marginally increased from 7.0 (neutral) to 7.4 (mildly alkaline, mimicking blood and lymph pH), we observed a significant (51%) reduction in their Vis-NIR spectrum peak intensity over the course of seven days (Figure 2C), indicating degradation of the PBNPs. These observations were complemented by photographs of the PBNPs after 7 days (Figure. 2A–C insets), where PBNPs at mildly acidic and neutral pHs exhibited their characteristic blue colour while PBNPs at the mildly alkaline pH exhibited a colorless (“bleached”) appearance.

Fig. 2.

pH-dependent properties of PBNPs. (A–C) Visible-NIR spectra of PBNPs over 7 days at A) pH 5.5, B) pH 7.0, and C) pH 7.4, exhibiting PBNP stability at mildly acidic (5.5) and neutral (7.0) pHs, and decreased PBNP stability at a mildly alkaline pH (7.4). Insets: PBNP photographs on Day 7 at a pH of 5.5, 7.0, and 7.4, respectively. (D–F) Dynamic light scattering analysis of the PBNPs over 7 days (Day 0: blue, Day 7: red) at pH D) 5.5, E) 7.0, and F) 7.4, illustrating detectable PBNP populations at mildly acidic (5.5) and neutral (7.0) pHs over 7 days, and undetectable PBNP populations at a mildly alkaline pH (7.4) on Day 7. (G–I) TEM images of PBNPs on Day 7 at G) pH 5.5, H) pH 7.0, and I) pH 7.4 showing detectable PBNPs at mildly acidic (5.5) and neutral (7.0) pHs, and undetectable PBNPs at a mildly alkaline pH (7.4).

The degradation of the PBNPs at pH 7.4 was likely caused by attack of the characteristic Fe^II-CN-Fe^III bonds of PBNP by the slight excess of hydroxyl ions resulting in the formation of hydroxides and released cyanoferrate ions, as previously observed.^43,44 These findings were corroborated by a study that measured PBNP concentration as a function of pH and time, calculated using optical density measurements at 680 nm, and its measured mass extinction coefficient at this wavelength (Figure S2). Similar to the Vis-NIR observations, the PBNP concentration decreased to ~43% of its starting concentration at pH 7.4 after seven days, indicating degradation of the nanoparticles under these conditions; while remaining essentially unchanged over seven days at pH 5.5 and 7.0. To determine whether the pH-dependent stability of PBNPs had an effect on their function as PTT agents, we measured the PTT capabilities of the PBNPs as a function of concentration (0.01–1 mg/mL) at the two extremes of pH studied: 5.5 and 7.4. We observed that the PBNPs heated to higher temperatures when they were incubated at a pH of 5.5 versus 7.4, and this occurred in a concentration-dependent manner (Figure S3A). This is likely due to the fact that at higher pH, PBNPs exhibit a significant reduction in their PTT capabilities due to their degradation under these conditions, consistent with our earlier findings.

We also conducted a temporal DLS study, which was used to assess nanoparticle size distributions as a function of pH over time. The PBNPs were observed to be stable when incubated at pH 5.5 and 7.0 (constant mean hydrodynamic diameters; Figure 2D and E). In contrast, DLS was unable to detect a nanoparticle population of PBNPs when incubated at pH 7.4 for 7 days (Figure 2F), indicating an attack of the nanoparticles at this pH. Lastly, TEM images of PBNPs at the different pHs confirmed that the PBNPs were intact at pH 5.5 and 7.0 (detectable individual nanoparticles; Figure 2G and H), and degraded at pH 7.4 (no detectable nanoparticles; Figure 2I). Taken together, our findings indicate that we can synthesize PBNPs to exhibit inherent pH-dependent stability, where they are stable under mildly acidic conditions (tumor interstitium pH) and at a neutral pH, and degrade under mildly alkaline conditions (blood and lymph pH), thus rendering them suitable for i.t. administration of PTT for neuroblastoma.

3.2. PTT Reduces Tumor Burden and Increases the Number of Tumor-free Days in a Mouse Model of Neuroblastoma

We then conducted studies to determine the effective i.t. dose of PBNPs to achieve temperatures suitable for thermal ablation of neuroblastoma tumors (Figure S3B), and evaluated whether PBNP-based PTT treatment was efficacious in the neuroblastoma tumor model. For these studies, neuroblastoma tumor-bearing mice were either i.t. injected with 50 μL of 1 mg/mL PBNPs and irradiated with an 808 nm laser (1.875 W/cm² for 10 min) or left untreated (Figure 3; n ≥ 5/group). Mice in the PTT-treated group exhibited tumor eradication immediately after treatment (decreased bioluminescence; Figure 3A) compared with mice in the untreated, control group that exhibited consistent tumor growth (Figure 3B). Aggregate data from multiple tumor growth studies showed that when tumor-bearing mice were treated with PTT, their tumors rapidly shrunk, and they had a mean of 3 tumor-free days before the tumors recurred (Figure 3C). Furthermore, the tumor growth was slower in these mice compared with mice in the untreated, control group, which exhibited rapid tumor growth. Our results indicate that PBNP-based PTT can rapidly reduce tumor burden, increasing the number of tumor-free days while decreasing tumor growth rates in this hard-to-treat neuroblastoma model.

Fig. 3.

Efficacy of PBNP-based PTT in vivo. (A–B) Representative images of a A) PTT-treated, tumor-bearing mouse showing debulking of the tumor mass after PTT and an B) Untreated tumor-bearing mouse showing tumor growth (scale bars = tumor bioluminescence intensity; p/s/cm²/sr). C) Normalized tumor growth curves for untreated (black) and PTT-treated (red) tumor-bearing mice showing slower tumor growth in PTT-treated mice relative to untreated controls (n ≥ 5/group).

3.3. PTT Results in Increased Lymphocyte and T Cell Infiltration to the Tumor Area in Vivo

After determining the ablative properties of PBNP-based PTT, we next we sought to evaluate if it elicited immunostimulatory effects. To this end, we quantified the relative proportions of tumor infiltrating lymphocytes after PTT in vivo. For these studies, neuroblastoma-tumor bearing mice were divided into two groups: PTT-treated and untreated controls (n ≥ 4/group). To measure the tumor expression levels of lymphocytes and specifically T cells after PTT, mice were euthanized 24 and 96 h post-PTT and tumor tissue was isolated and processed for CD45 (lymphocyte) and CD3 (T cell) expression levels. After 24 h, there was no significant difference in lymphocyte and T cell populations in the PTT-treated versus untreated tumors (p > 0.05; Figure S4–B). However, 96 h post-PTT, the tumors in PTT-treated mice exhibited a significant increase in both lymphocyte (p = 0.0294) (average values of CD45+; 9.7% PTT-treated vs. 4.1% untreated; Figure 4A–C) and T cell (p = 0.0424) (average values of CD3+; 6.2% PTT-treated vs. 2.2% untreated; Figure 4D–F) infiltration compared to untreated mice. These results suggest that there is an increased recruitment of T cells to the tumor site four days following. Additionally, PBNPs alone do not increase infiltration relative to untreated controls (Figure S5).

Fig. 4.

Immunostimulatory effects of PBNP-based PTT. Representative scatter plots of CD45+ cells in tumors of: A) Untreated and B) PTT-treated mice. C) %CD45+ cells in the tumors of untreated and PTT-treated mice showing significantly higher percentage of CD45+ cells in tumors of PTT-treated vs. untreated mice (p = 0.0294). Representative scatter plots of CD3+ cells in tumors of: D) Untreated and E) PTT-treated mice. F) %CD3+ cells in the tumors of untreated and PTT-treated mice showing significantly higher percentage of CD3+ cells in tumors of PTT-treated vs. untreated mice (p = 0.0424); (n ≥ 4/group for this study).

We also determined if PTT resulted in a global activation of T cells by evaluating whether splenic T cells in PTT-treated mice exhibited a recall response when co-cultured with tumor cells (Figure S4C). An IFNγ ELISpot assay (Supporting Methods for details) showed no significant differences in recall response between the PTT and control groups, indicating that PTT alone cannot elicit a robust recall immune response important for successful tumor eradication. Taken together, our results suggest that PTT alone can elicit immunostimulatory effects that lead to an increase in the infiltration of lymphocytes and T cells to the tumor area, priming the body for an antitumor immune response. Although these effects are not strong enough to eradicate hard-to-treat cancers such as neuroblastoma (as observed in Figure 3), this strategy can be exploited in combination with an immunotherapy such as anti-CTLA-4 (in photothermal immunotherapy) to treat these tumors.

3.4. Photothermal Immunotherapy Results in Tumor Regression and Long-term Survival of Mice.

To test the efficacy of the photothermal immunotherapy (PBNP-based PTT plus anti-CTLA-4) in treating neuroblastoma, Neuro2a tumor-bearing mice were divided into five groups (Table 1): 1) PTT + anti-CTLA-4-treated, 2) PTT-treated, 3) anti- CTLA-4-treated, 4) PBNP-treated, and 5) Untreated (n ≥ 5/group). A representative temporal image measuring the tumor-specific bioluminescence indicated a gradual decrease in tumor size and the subsequent elimination of the tumor in a mouse treated with PTT + anti-CTLA-4 (Figure 5A). Further, the tumor progression was significantly slower in the PBNP-based PTT + anti-CTLA-4 group when compared with untreated controls (p = 0.0002; Figure 5B). Most importantly, the photothermal immunotherapy resulted in complete tumor regression and long-term survival in 55.5% of the treated mice (at 100 days post-treatment, Figures 5C and S6). The long-term tumor-free survival was significantly higher than that observed (at 100 days post-treatment) for mice treated with anti-CTLA-4 alone (12.5%, p = 0.0293), PTT (0%, p = 0.0305), PBNPs (0%, p = 0.0002), or left untreated (0%, p = 0.0002). Building on these promising results, we next sought to describe the immune effector cells involved in eradicating the tumors following photothermal immunotherapy. We depleted subsets of T cells by systemic administration of antibodies against CD4+ and CD8+ T cells. The effectiveness of the depletion (>95% depletion) was confirmed using flow cytometry before commencing treatment (Figure S7). When mice that were depleted in CD4 or CD8 were treated with our photothermal immunotherapy, they rapidly succumbed to tumor burden, leading to 0% survival at 8 days after treatment (Figure 5D). The tumor sizes (Figure S8) and survival curves (Figure S9) were also analyzed for CD4 and CD8-depleted mice treated with PTT, anti-CTLA-4, or left untreated for comparison, which yielded similar results i.e. no survival >10 days in CD4/CD8-depleted mice for all groups. Our results suggest the potential of photothermal immunotherapy in securing complete tumor remission and long-term survival in a significantly higher proportion of neuroblastoma tumor-bearing mice, and the importance of CD4+ and CD8+ T cells in eliciting these responses.

Table 1.

Groups and treatments used in the study.

Group	Treatment
PTTa+anti-CLTA-4b	PTT on day 0;
	anti-CTLA-4 on days 1, 4, 7
PTTa	PTT on Day 0
Anti-CTLA-4b	Anti-CTLA-4 on days 0, 3, 6
PBNPsc	PBNPs on Day 0
Untreated	No treatment

^aPTT-treated groups receive 50 μL of 1 mg/mL PBNPs i.t., irradiated by an 808 nm laser at 1.875 W/cm² for 10 min.

^bAnti-CTLA-4-treated groups receive 150 μg of anti-CTLA-4 per dose by i.p. injection.

^cPBNP-treated group receive 50 μL of 1 mg/mL PBNPs i.t.

Fig. 5.

Effect of photothermal immunotherapy (PTT + anti-CTLA-4 therapy) on tumor regression and long-term survival in the neuroblastoma mouse model. A) Representative image of a long-term surviving mouse treated with PTT + anti-CTLA-4 showing complete tumor regression (scale bar = tumor bioluminescence intensity; p/s/cm²/sr). B) Normalized tumor growth curves for tumor-bearing mice treated with PTT + anti-CTLA-4 (violet) or left untreated (black). C) Kaplan-Meier survival plots of neuroblastoma mice that were treated with PTT + anti-CTLA-4, PBNPs alone, anti-CLTA-4 alone, PTT alone, or untreated. Mice receiving photothermal immunotherapy showed significantly higher long-term survival (>100 days) compared with mice in the other groups (log-rank test; p < 0.05); (n ≥ 5/group). D) Kaplan-Meier survival plots of neuroblastoma-bearing mice depleted in CD4+ and CD8+ T cells. Depletion of CD4+ and CD8+ cells (n = 5/group) effectively abrogated the therapeutic responses of the photothermal immunotherapy (log-rank test; p < 0.005).

3.5. Long-term Surviving Mice Treated with Photothermal Immunotherapy Exhibit Protection against Tumor Rechallenge

An ideal tumor therapy would be one that not only effectively eradicates tumors but prevents recurrence after successful elimination from the body. Therefore, we next investigated whether photothermal immunotherapy conferred protection in long-term surviving mice that were rechallenged with the original tumor cells (Neuro2a) (Figure 6A–D). Our studies consisted of two groups:1) naïve group: where untreated mice were challenged with 10⁶ Neuro2a cells and 2) rechallenged group: where long-term surviving mice previously treated with photothermal immunotherapy (PTT + anti-CTLA-4) were rechallenged with 10⁶ Neuro2a cells after at least 90 days of tumor-free survival (n ≥ 3/group for this study). Remarkably, all of the photothermal immunotherapy-treated, long-term surviving mice exhibited protection against the tumor rechallenge, and these mice rapidly eliminated the rechallenged tumors (Figures 6A, C and S10), compared with control mice where rapid tumor progression was observed (Figures 6B, D and S10). Further, the rechallenged mice survived for >90 days post tumor rechallenge compared with naïve mice that had to be euthanized due to high tumor burden 12–14 days post-challenge. These data suggest the potential of photothermal immunotherapy in conferring immunity and protection in long-term surviving mice against tumor rechallenge/recurrence.

Fig. 6.

Effect of tumor rechallenge in photothermal immunotherapy-treated, long-term surviving mice. (A and B) Representative images showing protection against tumor rechallenge in A) photothermal immunotherapy-treated mice and B) tumor growth in naïve, untreated mice (scale bars = tumor bioluminescence intensity; p/s/cm²/sr). C) Tumor growth curves after challenge with 10⁶ Neuro2a cells in untreated mice (naïve, black) and long-term surviving photothermal immunotherapy-treated mice (rechallenged, orange) showing protection in the rechallenged group compared to progression in the naïve group. D) Kaplan-Meier survival plots showing significantly higher long-term survival in the rechallenged group compared to naïve mice (log-rank test, p < 0.05); (n ≥ 3/group).

4. Discussion

We have described photothermal immunotherapy, which combines PBNP-based PTT with anti-CTLA-4 checkpoint inhibition (Figure 1) for treating neuroblastoma in an animal model. PBNPs exhibited inherent pH-dependent stability (Figure 2) where they were stable at an acidic pH mimicking conditions observed in the tumor interstitium, and exhibited incipient degradation at higher pHs mimicking the blood/lymph. Harnessing the pH gradient of the tumor interstitium relative to surrounding tissue is an effective strategy to selectively trigger and/or control tumor treatment.^45,46 Our in vitro data demonstrating the pH-dependent stability of PBNPs suggest their potential use in delivering tumor-specific therapies, where the PBNPs remain intact i.t., while rapidly degrading when they enter the bloodstream or lymphatic system. This pH-dependent stability also minimizes potential toxicities associated with the long-term persistence of the nanoparticles in vivo – an important consideration in the field of nanomedicine for clinical translation. These findings complement an earlier study by our group measuring the biodistribution of PBNPs,²⁷ and a previous toxicological evaluation of PBNPs that exhibited no long-term toxicity (at 60 days) after intravenous administration in mice.⁴⁷ Important to note, our studies use over 3 fold-lower PBNPs compared to this earlier report (2.5 vs. 8 mg/kg).⁴⁷

The pH-dependent properties of PBNPs had a significant effect on their PTT capabilities; as their relative temperatures were decreased at blood/lymph pH relative to i.t. pH (Figure S3A). This led us to establish the concentrations of PBNPs i.t. administered to ensure that there were sufficient nanoparticles to effect tumor ablation in our neuroblastoma model (Figure S3B). It is likely that similar optimization studies will need to be carried out should the conditions under which PTT is administered is changed, e.g. superficial versus deeper tumors may require different nanoparticle doses, laser power densities, and/or duration of irradiation. It is important to state here that should the need arise for PBNPs to exhibit longer temporal stability and significantly slower degradation kinetics than that observed using the synthesis scheme described here (especially in applications that require intravenous administration), the PBNPs can be appropriately modified by surface coating with biocompatible polymers such as polyethylene glycol, as previously described.³³

Although growth rates were slowed, PBNP-based PTT by itself was unable to confer long-term survival in the syngeneic Neuro2a model of neuroblastoma (Fig. 3). As described in previous studies in the literature, PTT confers long-term, tumor-free survival in multiple animal models of cancer (e.g. breast cancer)^20,33 premised on the observation that cancer cells are more susceptible to heat than normal tissue because of their elevated metabolic rates.^48,49 However, in the case of this neuroblastoma model, we suspect that PTT did not eliminate all cancer cells because nascent cancer cells likely grew into new tumors, similar to clinical observations of latent cancer populations in neuroblastoma.⁵⁰ We speculate that in tumor models that demonstrated complete remission in these earlier studies,^20,33 any residual tumor cells were likely cleared by a robust immune response generated by PTT in those less aggressive tumors.^3,5

PBNP-based PTT caused an increase in lymphocytic infiltration into the tumor regions (Figure 4). Lymphocytes found in tumors have been shown to be effective at delaying tumor progression, suggesting their potential influence on improved patient prognosis.^51–54 Therefore, an increased population of CD45+ cells (Figure 4A–C) into the residual tumors of PTT-treated mice presents an opportunity to recruit these cells for tumor eradication.⁵⁵ Within this subset of lymphocytes, T cells are also present in increased numbers (CD3+ cells; Figure 4D–F), and their recruitment can generate a T cell-mediated antitumor response.⁵⁶ Additionally, PBNPs alone do not increase infiltration relative to untreated controls, ensuring that the cellular infiltration observed is due to PTT and not simply due to an inflammatory response initiated by the particles alone (Figure S5). However, PTT by itself was not sufficient to elicit a robust tumor recall response (Figure S4C) suggesting the inadequacy of this modality by itself in securing durable tumor responses in treating this mouse model of neuroblastoma. This prompted us to look into complementary immunotherapies that could be combined with PTT to stimulate robust antitumor responses. We therefore selected anti-CTLA-4, a checkpoint inhibitor that has been shown to reverse immunosuppression and elicit robust antitumor responses.^57,58

PTT in combination with anti-CTLA-4 immunotherapy resulted in complete tumor regression and long-term survival in 55.5% of the tumor-bearing mice compared to only 12.5% survival observed in mice treated with anti-CTLA-4 alone and 0% survival observed in all mice treated with PBNPs alone, PTT alone or left untreated (Figure 5). Previous studies using the Neuro2a mouse model have demonstrated higher long-term survival using anti-CTLA-4 alone than that observed in this study (~40–50% vs. 12.5% in our study).^36,59 The difference between these observations can potentially be attributed to the fact that the earlier studies commenced the anti-CTLA-4 immunotherapy when their mice reached tumor sizes of ~1 mm or after a fixed number of days (typically 5–6 days) after tumor inoculation, while we commenced the therapy only after tumors were established and reached a minimum diameter of 5 mm, thus potentially reflecting a significantly higher tumor burden and disease progression in our studies. We attribute the significantly higher long-term survival benefit in the photothermal immunotherapy-treated mice to the priming of an immune response by PTT, which is complemented by a reversal of T cell exhaustion and immunosuppression by anti-CTLA-4. Indeed our studies depleting CD4+ and CD8+ populations confirmed the importance of CD4+ and CD8+ T cells in eliciting improved therapeutic responses (Figures 5D, S8, S9), and are consistent with earlier published studies in this model.^36,59 While these studies represent a good start, a thorough characterization of the underlying immune effects contributed by both PTT and anti-CTLA-4 (innate and adaptive) that drive the antitumor responses of our combination therapy is warranted, and is the focus of ongoing work in our group.

Finally, long-term surviving mice treated with photothermal immunotherapy exhibited protection against tumor rechallenge indicating the development of immunity against these tumors in photothermal immunotherapy-treated mice (Figure 6). Similarly, further studies are necessary to elucidate the underlying immunological mechanisms that elicit these protective responses. To conclude, our study points to the important role that PBNPs may play in the upcoming years in immunoengineering,⁶⁰ where the nanoparticles are used to engineer a suitable immune response to treat advanced cancers such as high-risk neuroblastoma, improving the outlook for these patients.

Abbreviations:

FDA

Food and Drug Administration

PBNP

Prussian blue nanoparticle

PTT

Photothermal therapy

CTLA-4

Cytotoxic T lymphocyte-associated protein 4

Anti-CTLA-4

Anti-cytotoxic T lymphocyte-associated protein 4

PD-1

Programmed cell death protein 1

Anti-PD1

Anti-programmed cell death protein 1

i.t.

Intratumoral/intratumorally

i.p.

Intraperitoneal/intraperitoneally

DMEM

Dulbecco’s Modified Eagle Medium

DLS

Dynamic light scattering

TEM

Transmission electron microscopy

FITC

Fluorescein isothiocyanate

CD45

Cluster of differentiation 45

CD3

Cluster of differentiation 3

MFI

Mean fluorescence intensity

Vis-NIR

Ultraviolet-visible spectroscopy or ultraviolet-visible spectrophotometry

IFNγ

Interferon gamma

ELISpot assay

Enzyme-Linked ImmunoSpot assay

CD8

Cluster of differentiation 8

CD4

Cluster of differentiation 4